DOCUMENT RESUME

ED 364 587 TM 020 814

AUTHOR Maline, Hindi S., Comp.TITLE The National Longitudinal Study of the High School

Class of 1972: Annotated Bibliography of Studies,1980-1992.

INSTITUTION Office of Educational Research and Improvement (ED),Washington, DC. Office of Research.

REPORT NO OR-93-3156PUB DATE Oct 93NOTE 93p.PUB TYPE Reference Materials Bibliographies (131)

EDRS PRICE MF01/PC04 Plus Postage.DESCRIPTORS Academic Aspiration; Annotated Bibliographies; Cohort

Analysis; Databases; Economic Factors; *EducationalAttainment; Education Work Relationship; EthnicGroups; Graduate Surveys; High SchoGls; *High SchoolStudents; Life Events; *Longitudinal Studies;*National Surveys; Postsecondary Education; RacialDifferences; Research Methodology; *Research Reports;Secondary Education; Sex Differences; *VocationalEducation

IDENTIFIERS *National Longitudinal Study High School Class1972

ABSTRACTThe National Longitudinal Study of the High School

Class of 1972 (NLS-72) is the "grandmother" of the longitudinalstudies designed and conducted by the National Center for EducationStatistics, and is probably the richest archive ever assembled on ageneration of Americans. Participants were selected as high schoolseniors in 1972 and were followed through postsecondary transcriptsin 1984. In 1981, Samuel Peng assembled a bibliography of studiesdescribing or using NLS-72 data. This publication extends Peng's workby describing the considerable number of studies published since thefirst bibliography. Works are arranged in the following categories:(1) general descriptions (data sets and populations); (2)methodology; (3) work (employment); (4) secondary education; (5)postsecondary education; (6) educational and occupationalaspirations; (7) educational and occupational attainment; (8) genderand racial-ethnic dil7ferences; (9) vocational technical andvocational education; (10) teachers; (11) marriage, family, and life*.ransitions; (12) economics of education; (13) testing; (14) others;,tnd (15) tabulations. In all, 259 sources are listed and annotatedwith a brief description of the study and its methodology.Publication of this document coincides roughly with the release of anoptical data disk with the NLS-72 data archive, manuals, and samplecomputer programs. (SLD)

***********************************************************************

Reproductions supplied by EDRS are the best that can be madefrom the original document.

********************************************************k**************

SCOPE OF INTEREST NOTICE

The ERIC Facility has assignedthis document for processingto:

In our ludgment. this documentis also of interest to the Clear-mghouses noted to the right.

The Indexing should reflect theirspecial points of view.

NationalLongitudinal Studyof t e High School

Class of 1972Annotated

Bibliography of Studies,1980 1992

U.S. DEPARTMENT OF EDUCATIONOffice gt Educational Research and improvement

EDUCATIONAL RESOURCES INFORMATIONCENTER (ERIC)

rtuS document has been reproduced asreceived from the person or organizationonginating it

C Minor changes have been made to improvereproduction quality

Points of view or opinions stated in this docu-ment do not necessarily represent oftic.aiOEM position or policy

FFICE ofESEARCH

C Dr ;Nu merit ,1

Offim e 1,1 Ed,. tut,,r al, It "rt.! ,

2

BEST COPY AVAILABLE

TheNational

Longitudinal Studyof the High School

Class of 1972Annotated

Bibliography of Studies,1980-1992

Compiled by Mirei S. Ma lineOffice of Research

U.S. Department of Education

,)a

U.S. Department of EducationRichard W. RileySecretary

Office of Educational Research and ImprovementSharon P. RobinsonAssistant Secretary

Office of ResearchJoseph C. ConatyActing Director

October 1993

4

PrefaceThe National Longitudinal Study of the High School Class of 1972 (NLS-72) is the grandmother of thelongitudinal studies designed and conducted by the National Center for Education Statistics (NCES) of theU.S. Department of Education. At this point in time (1993), it is probably the richest archive ever assembledon a single generation of Americans.

Participants in the study were selected when they were seniors in high school in the spring of 1972, and in asupplementary sample drawn in 1973. The records include the "Base Year" survey; follow-up surveys in1973, 1974, 1976, 1979, and 1986; high school records; and postsecondary transcripts (collected in 1984).

Researchers have drawn on this archive since its inception. The history of the Class of '72 from its high schoolyears through its early 30s is widely considered as the baseline against which the progress and achievementsof subsequent cohorts will be measured. The principal comparisons drawn to date have been to the second inthe series of NCES longitudinal studies, the High School and Beyond cohorts of 1980 (seniors and sophomoresin that year). In the future, researchers will also compare these generations to the eighth graders of the NationalEducation Longitudinal Study of 1988.

In 1981, Samuel Peng of NCES assembled an annotated bibliography of studies describing or using NLS-72data (Taylor, Mary Ellen et al. National Longitudinal Study of the High School Class of 1972 Study Revn-tsUpdate: Review and Annotation ED 218 314). The period covered by this document was 1972 through 1981.Our intent in these pages is to update this work by identifying and describing the considerable number ofstudies using the NLS-72 published since Peng' s work.

We do not pretend to have captured everything that has been written about the NLS-72 since 1980. Inparticular, we are probably missing many dissertations that did not turn up on the major radar screens ofpublications and public access documents. Nonetheless, we hope to make life easier for researchers and otherinterested parties in terms of their hunt for relevant literature.

The publication of this document will coincide (roughly) with the release, by NCES, ofa CD-ROM containingthe entire NLS-72 data archive, user's manuals and guides, and sample computer programs to help analystsconfigure the data.

ill5

AcknowledgmentsA nunber of people contributed to the development of this collection. Thomas Calhoun ran the original ERICsearches when he served as a summer intern in the U.S. Department of Education's Office of Research in1991. Cliff Adelman, Aurora D'Amico, and Jeff Owings of the Department's Office of Educational Researchand Improvement provided helpful reviews and/or found additional fugiti-v-e materials. Dan Goldenberg ofthe Department of Education' s Office of Pc licy and Planning and Ron Cervero of the University of Georgiaalso offered useful feedback and reviews. Maggie McNeely provided editorial guidance and layout assistance.John Blake of the Department of Education's Research Library was extremely helpful in locating numerousarticles.

ContentsPreface

Acknowledgements

I. General Description (Data Sets/Populations)

II. Methodology

iii

v

1

51. Data File User's Manuals 52. Data Collection Procedures 73. Analysis Strategies 84. Measurement Property 95. Nonresponse Bias 106. Sample Design and Effects 10

Ill. Work 131. Labor Force Participation 132. Training 133. School-to-Work Transition 14

IV. Secondary Education 171. School/Program Effects 172. Student Characteristics 23

V. Postsecondary Education 251. Access 252. College/Program Effects 273. Financial Aid 324. Patterns of Attendance 355. Persistence 41

VI. AspirationsEducational and Occupational 45

VII. AttainmentEducational and Occupational 49

VIII. Gender and Racial-Ethnic Differences 67

IX. Vocational Technical and Vocational Education 71

X. Teachers 73

XI. Marriage/Family and Life Transitions 77

vii

XII. Economics of Education 81

XIII. Testing 83

XIV. Other 85

XV. Tabulations 87

Author Index 91

8

I. General Description (Data Sets/Populations)1

Burkheimer, Graham J., and Novak, Thomas P.1981. A Capsule Description of Young AdultsSeven and One-Half Years After High School.Washington, DC: U.S. Department of Edu ,ation,National Center for Education Statistics. ED 208794

This study uses NLS-72 data to examine thevocational activities, plans, aspirations, and atti-tudes of NLS participants after leaving high schoolalong with the relationship of these activities to pasteducational experiences and background charac-teristics. Data were drawn from the base yearthrough 1979 follow-up surveys. Postsecondaryeducation, work, marital status, home ownershipand life goals are among the categories explored.

2Eagle, Eva, and others. 1988. A Descriptive Sum-mary of 1972 High School Seniors: Fourteen YearsLater. Washington, DC: U.S. Department of Educa-tion, National Center for Education Statistics. ED299 913

This report provides detailed information onthe activities of 1972 seniors between their highschool graduation and February 1986. Data from theNLS-72 are used to look at educational attainment,work and marriage history, attitudes, and civic par-ticipation. Differences and patterns of change forseveral subgroups (including race-ethnicity, genderand socioeconomic status) are examined.

The first chapter examines the educationalprogression of 1972 seniors through 1986, with anemphasis on degree attainment. The second chapterfocuses on labor force participation between 1979and 1986 and describes patterns )f employment formembers of the 1972 senior cohort. The third chap-ter analyzes change since 1972 in student attitudes,registration and voting, and memberships in volun-tary organizations by demographic characteristicsand educational attainment.

1

Major findings include:

(1) Sixty-six percent of the 1972 high schoolseniors enrolled in some form of postsecondaryeducation by 1986, with most enrolling immediatelyafter high school graduation.

(2) In 1986, 44 percent of 1972 high schoolseniors still expected to continue their education.

(3) Differences in employment patterns ex-isted between men and women but were not stronglyassociated with race-ethnicity.

(4) Women from this cohort were more likelythan men from this cohort to have been parents by1986.

(5) Substantial proportions of the 1972 sen-iors agreed that there are serious problems facingelementary and secowaary schools.

3Fetters, William B., and others. 1984. High SchoolSeniors: A Comparative Study of the Classes of1972 and 1980. Washington, DC: U.S. Departmentof Education, National Cener for Education Statis-tics. ED 248 277

This report uses "Base Year" survey data fromthe NLS-72 and High School and Beyond "BaseYear" cohorts to examine differences in school ex-periences, school performance, student self-percep-tion and values, and plans and aspirations. Thesecond chapter of the report compares the twoclasses with regards to school experiences (e.g.,curricula, courses, extracurricular activities) andstudents' evaluation of their school and the educa-tion they received. Differences in course grades,homework effort, and standardized cognitive testscores are examined in the third chapter. The fourthchapter looks at differences in values, self-concepts,and the extent to which students believe they havecontrol over their own fate. The fifth and final

chapter looks at trends in short- and long-rangeeducational and occupational plans.

Findings show a decline in the percentage ofstudents enrolled in an academic curriculum and asharp increase in the number of seniors who thoughtschools should place more emphasis on basic aca-demic subjects. 1980 seniors rated their schoolshigher than the 1972 seniors with regard to libraryfacilities, teacher interest, and guidance counseling.Declines were seen in both verbal and mathematicsscores as well as in the amount of time spent onhomework. Despite this, there was an increase ingrade point averages by .07 of a letter grade.

4Larsen, Amy S., Watson, Mark E., Duncan, DebraP., Riccobono, John A., and Burkheimer, Graham J.1981. Tabular Summary of the "Fourth Follow-Up"Questionnaire Data. Washington, DC: U.S. Depart-ment of Education, National Center for EducationStatistics.

This report contains tabular summaries of thenumerical responses to the "Fourth Follow-Up"Questionnaire items of 18,630 individuals. Theseitems deal primarily with the survey participants'educational experiences and occupational attain-ments in October 1977, 1978, and 1979, along withtheir continuing or revised plans about education,training, and jobs, as well as measures of theiraspirations, attitudes, and opinions.

The panel of repeated observations will permitthe examination of the relationship between school-ing, work, and other experiences and subsequentcareer choices as well as educational and labor forceparticipation of each of the selected individuals.Such information and the resultant analyses areimportant to those engaged in formulating legisla-tive proposals and educational policy. (From articlesummary)

5 ,--

Lenth, Charles S., and Christal, Melodie E. 1991."National Data Bases and Statistical Resources onHigher Education: An Annotated Guide." New.Directions for Institutionai Research (Charles S.Lenth ed.) c)2: 83-107.

The authors provide a guide to major sourcesof higher education data. More than 50 national,periodic, and publicly available sources, includingthe ns-72, are described.

6McAdams, Kathy, and others. 1981. National Lon-gitudinal Study of the High School Class of 1972.,An Historical Overview and SumrnaQc. Washing-ton, DC: U.S. Department of Education, NationalCenter for Education Statistics. ED 217 051

This report is a brief history of the NLS, alongwith its origins and goals. In addition to the designand content of the sample, the summary also coversthe NLS questionnaire design through the fourthinstrument, the administration of the pretests, "BaseYear" and "Follow-Up" surveys, data preparationand quality, and the use of data by federal agencies,social scientists and others. Materials extracted fromkey NLS documents are provided. Tables illustratestatistical and variables data, and the appendicesinclude complete survey instruments with an NLSbibliography. (Adapted from ERIC abstract)

7Riccobono, John A., ed., and others. 1981. NationalLongitudinal Study of the High School Class of1972: "Fourth Follow-Up" Survey,. Final Methodo-logical Report. Washington, DC: U.S. Departmentof Education, National Center for Education Statis-tics. ED 217 052

The report of the "Fourth Follow-Up" survey(1979) of the NLS-72 is methodological, historicaland descriptive in nature, but contains no analyticalresults. Reference to earlier reports is provided indetermining postsecondary educational and voca-tional activities of young adults in transition years.The six sections present the background and purposeof the NLS, an overview of operations, the develop-ment of the "Fourth Follow-Up" instruments, re-lated materials, and linkages between the fourth andprior "Follow-Up" questionnaire items. Data tablesillustrate the text. (Adapted from ERIC abstract)

8Riccobono, John A., and others. 1980. "The NLSData Base: From "Base Year" 1972 to "Fourth

2 10

Follow-Up" 1980." Paper presented at the AnnualConference of the NCARE, Wake Forest Univer-sity, November 13-14. ED 195 590

A brief description of the survey design of theNLS-72 and a listing of the kinds of research con-ducted using the NLS-72 are included in this report.Topics of research include access to higher educa-tion, persistence and performance in higher educa-tion, financial aid, progress in postsecondaryeducation, vocational education, occupationalchoice, psychological development, secondaryschool effects, consumer and political behavior, mi-nority students, and research methodology.

3

9Taylor, Mary Ellen, and others. 1981. NationalLongitudinal Study of the High School Class of1972 Study Reports Update: Review and Annota-tion. Washington, DC: U.S. Department of Educa-tion, National Center for Education Statistics. ED218 314

This document annotates and catalogues all ofthe NLS-based studies received from data userssince the beginning of the study (1972) through1981, summarizes their major findings and, whereappropriate, makes comparisons among those fmd-ings.

1 1

II. Methodology

1. Data File User's Manuals

10Henderson, Louise, and Levinsohn, Jay. 1981. Na-:tional Longitudinal Study of the High School Classof 1972. School File Documentation. Washington.DC: U.S. Department of Education, National Cen-ter for Education Statistics. ED 313 401 (See alsoED 313 402)

This documentation is intended to assist usersof the school file which was prepared from informa-tion collected for the NLS-72. The school file is acompanion to the NLS-72 student-based data fileand consists of 1,318 records, one from each highschool from which NLS-72 students are sampled. Itcontains information about the environment, staff,procedures, facilities, and curricula of the samplehigh schools. This documentation describes thesample design of the survey and is supplemented byappendices containing detailed information aboutthe variables. Appendix A contains the data collec-tion instruments: the School Questionnaire and theCounselor Questionnaires. Appendix B is a sum-mary listing of the 504 variables in the file. Thelisting provides, for each variable, the variable num-ber and label, the corresponding question label, theposition in the data file, and the corresponding ques-tion label. Appendix C is a listing of the frequenciesof the responses to each question. (ERIC abstract)

11Jones, Calvin, and others. 1986. National Longitu-slinaLS1ady_d_thIligh SchooLCIass of 1972. Post-secondary Education Transcript Study Data FileI.Jser's Manual. Washington, DC: U.S. Departmentof Education, National Center for Education Statis-tics. ED 313 399, ED 274 713

This is the original code book for the Postsec-ondary Education Transcript Sample File (PETS)NLS-72. It describes the machine-readable data filecontaining information on transcripts from NLS-72senior cohort members who reported attending apostsecondary institution after high school. Thesetranscripts were collected in 1984. Information de-

rivable from the file includes: (1) major and minorfields of study; (2) periods of enrollment; (3) coursestaken; (4) credits earned; and (5) grades receivedand credentials earned at 2-year, 4-year and othertypes of postsecondary institutions. Data are organ-ized at the student, transcript, term, and course lev-els. This user' s manual documents the proceduresused to collect this information and provides re-searchers with technical guidance for using the pub-lic release data files. Data editing procedures arediscussed, and the organization and content of thefiles are described. Procedures used to constructsampling weights for use in computing populationestimates are reviewed. Lists of the endorsing insti-tutions, postsecondary school codes in numericalorder, postsecondary school codes in alphabeticalorder, and course subject codes in numerical orderare provided. The data file record layout and rawfrequency distributions are tabulated. (Adaptedfrom ERIC abstract)

12Kolstad, Andrew. 1987. National LongitudinalStudy of the IF r-J3 School Class of 1972 NLS-72

D PI \I II S D. . U I

Washington, DC: U.S. Department of Ecluieation,Center for Education Statistics.

The purpose of this user's manual is to providethe information needed to use the Defense Man-power Data Center (DMDC) military records datafile in corjunction with survey data from the NLS-72. The NLS-72/DMDC military records data filecontains information about the military careers ofabout 1750 men in the NLS-72 sample who enlistedin the military. The information contained in thisdata file includes both measures recorded at the timeof enlistment (such as entry pay grade, ArmedForces Qualifying Test (AFQT) score, demographicmeasures, and educational attainment) and meas-ures that change during the enlistee's military career(such as current pay grade, military occupation, andre-enlistment or separation status). When linked, theresult is a unique analysis file that combines the bestfeatures of both data sources, making possibleanalyses of both pre-service and post-service ca-

5 12

rees of young meninformation that is not avaii-able from military records alone. (From article in-troduction)

13Knepper, Paula. 1987. Addendum to NLS-72 Post-secondary Education Transcript File. Washington,DC: U.S. Department of Education, National Cen-ter for Education Statistics.

This addendum contains information onchanges and corrections to the original NLS-72transcript files. Most problems dealt with creditcounts when BA degrees had been awarded. Prob-lems included duplicate information, incorrecttransfer indicators, incorrect term type, inaccurateinformation on institution type, difficulties convert-ing/standardizing credit unit measures, missingdates, and standardizing grades. Instructions for cer-tain variables, as well as examples of SAS program-ming to assist users in accessing, merging, and usingthe NLS-72 PETS files are also included.

14

Adelman, Clifford. 1992. supplementary Adden-dum: NLS-72 Postsecondary Transcript Files,Washington, DC: U.S. Department of Education,Office of Educational Research and Improvement.

This addendum explains the purpose and re-sults of a second cleaning and revision of the NLS-72 PETS data. The results of these efforts were fourdatasets with the SAS extensions STUDENT3,TRANS4, TERM4, and COURSE4. While not offi-cial NCES datasets, these are recognized as thedefinitive PETS files.

The files include a recoding of the COURSEfile according to a revised taxonomy, corrected in-accuracies in the "major" designated on transcripts,revisions in term dates, adjusted credit equivalen-cies, further elimination of duplicate transcripts, andcorrections of other difficult-to-detect problems inthe original coding (e.g., students with BA degreesawarded by community colleges). New variablesand flags were added to the TRANS file to indicateinstitutional Carnegie type, historically black col-leges, etc.

6

15Riccobono, John, Henderson, Louise B., Burkhe-imer, Graham J., Place, Carol, and Levinsohn, JayR. 1981. National Longitudinal Study: "Base Year"(1972) through "Fourth Follow-Up" (197q) DataFile User's Manual Volumes MI. and III. Washing-ton, DC: U.S. Department of Education, NationalCenter for Education Statistics.

These three volumes are to be used as a guide-book to facilitate use of data files prwared frominformation collected, processed, and systemati-cally stored during the course of the NLS-72. Thevolumes are divided into several sections. The firstprovides users with a general understanding of thefile. Part 2 provides the methodological backgroundof sampling, weighting, and data collection as wellas brief descriptions of the study data collectioninstruments. Part 3 indicates the procedures of filepreparation, including data editing/coding, entry op-erations and development of data quality indices andcomposites. Part 4 overviews the contents and or-ganization of the data file, and describes importantblocks of data. Part 5 provides technical specifica-tions for the data file and defines some potentialprocessing problems. (Adaptcd from Introduction)

16Tourangeau, Roger, and others. 1987. The National

' It I .1 I

1972 (NLS-72) "Fifth Follow-Up" (1986) Data FileUser' s Manual, Washington, DC: U.S. Departmentof Education, National Center for Education Statis-tics. ED 292 890

This manual was produced to familiarize datausers with the procedures followed in data collectionand processing of the "Fifth Follow-Up" survey(1986) of the NLS-72 and to provide documentationfor use of the public release data tapes. The manualrefers to each item of data by its Statistical AnalysisSystem (SAS) variable name as defined in the SASand SPSS-X (Statistical Programs for the SocialSciences) control cards provided in the data file.Appendices include the survey questionnaire, codesused in the data collection, and the data layout.(Adapted from ERIC abstract)

13

17Sebring, Penny, and others. 1987. The NationalLongitudinal Study of the High School Class of1972 (NLS-72). "Fifth Follow-Up" (1986). Teach-ing Supplement. Data File User's Manual, Wash-ington, DC: U.S. Department of Education,National Center for Education Statistics. ED 292891

As part of NLS-72, a Teaching Supplementmail questionnaire was sent to all respondents to the"Fifth Follow-Up" survey who indicated they had ateaching background or that they trained to becometeachers (N=1,517). The instrument focused on thequalifications, experiences and attitudes of currentand former elementary and secondary teachers andqualifications of persons who had completed a de-gree in education or certification but had not taught.This manual familiarizes data users and others withprocedures followed for data collection and process-ing of the supplement, and provides necessary docu-mentation for use of the Teaching Supplement data.The data gathered can be used to investigate issuesrelated to teacher quality and retention such as: (1)ways to attract potential teachers to the profession;(2) how to match teachers' assignments with theiracademic training; (3) the level of teacher satisfac-tion with their working conditions and jobs; and (4)the principal incentives and disincentives for teach-ers remaining in the profession. The appendicesinclude the questionnaire and the data layout. Thelast segment of the manual is the codebook for theTeaching Supplement data. (Adapted from ERICabstract)

See Teachers

2. Data Collection Procedures

18Henderson, Louise B., anA Allen, Danny R. 1981.NLS Data Entry Quality Control: The "Fourth Fel=low-Up" Survey. Washington, DC: U.S. Depart-ment of Education, National Center for EducationStatistics. ED 221 593

The data entry quality control procedures indiscrete data entry tasks of the NLS- 72 "FourthFollow-Up" Survey are examined. Direct data entry

7

terminals were used to key survey questionnaireitem responses, telephone interview corrections, re-spondent background information and supplementalquestionnaire responses into computer disk storage.Data entry error rates were computed on the surveyquestionnaires on the basis of three keyings: selec-tion of a random sample from each batch after theinitial keying of the data, and selected question-naires rekeyed by two additional operators. In theimplementation described, the overall error rate tol-erance established for the NLS survey was not ex-ceeded. The variable error rate over samples andoperators on the selected supplemental question-naires was 0.00040; estimated character error ratewas 0.00023. The telephone interview additions andcorrections, and directory information entry proce-dures are described. (Adapted from ERIC abstract)

19Wisenbaker, Joseph M. 1981. Factors Associatedwith Edit Failure. Washington, DC: U.S. Depart-ment of Education, National Center for EducationStatistics. ED 216 040

The NLS-72 employed a combination of mail,telephone, and personal interview data-gatheringtechniques. The NLS quality control process in-volved editing each returned questionnaire to deter-mine if certain key items were answered and ifresponses were consistent. If the respondent failedto answer these key items properly, the question-naire failed edit and the respondent was contactedby telephone. The purpose of this study was todetermine why so many NLS questionnaires failededit (over 60 percent of the "Third Follow-Up"survey of 1976). The study concluded that: (1)though subgroup differences were found, their mag-nitude is too small to be the basis for concreterecommendations for survey practice; (2) the major-ity of edit failure problems associated with itemizedfinancial questions revolved around the respon-dents' failure to supply answers to each of the re-quested line items; (3) items structured as "check allresponses that apply" are likely to be failed by asubstantial number of respondents; (4) reflexiveitems have a realizable potential for reducing re-spondent burden where repetitive information isrequired; and (5) overall data entry errors were low

14

except for items requiring itemized financial infor-mation. (Adapted from ERIC abstract)

20Wisenbaker, Joseph M., and Kolstad, Andrew J.1981. Factors Related to "Third Follow-Up" SurveyResponses, Washington, DC: U.S. Department ofEducation, National Center for Education Statistics.ED 216 036

The "Third (1976) Follow-Up" survey of theNLS-72 found that there is little difference in groupvariabiThy in survey response when consideringbackground characteristics, while significant differ-ences in response rates are found when consideringthe group patterns of response to prior instruments.Prior response was determined to be the most im-portant predictor of "Third Follow-Up" response.The scope and purpose and descriptive statistics arepresented analyzing response rates, response modesand their quality. The prediction approaches andresults of the variables, a discriminate functionanalysis of mail and personal interview modes, andthe quality of data are discussed, referring to theimportanct of a response payment incentive proce-dure. The study was meant to facilitate future datacollection efforts. Appendices and tables illustratethe analysis. (Adapted from ERIC abstract)

3. Analysis Strategies

21Eckland, Bruce K., Henderson, Louise B.,Tickamyer, Anne R., and Trent, William. 1981.Constructed Education Vaniables Washington, DC:Department of Health, Education, and Welfare, Na-tional Center for Education Statistics.

The purpose of this report is to define fourbasic research variables related to the educationalprogress of the following students: (a) all under-graduate enrollees in academic college programs foreach year between October 1972 and 1976, (b)college graduates in October 1976, (c) postgraduatestudents in October 1976, and (d) earned credithours as of October 1976. The definition of thesevariables appears to be fairly straightforward, butconstruction is complex.

8

Given the considerable effort and costs in-volved in resolving the measurement problems inthese cases, along with the importance of the vari-ables for analytical purposes, the authors stronglyrecommend that they be added to the NLS masterfile either as described in this report or in somemodified form. (Adapted from article abstract)

22Pavel, D. Michael, and Reiser, Mark. 1991. "UsingNational Data Bases to Examine Minority StudentSuccess in Higher Education." New Directions forInstitutional Research (Charles S. Lenth ed.) L9i5-20.

This article discusses using national databasesto examine access and achievement of racial andethnic minority students in higher education. Thearticle specifically looks at longitudinal databasessuch as the NLS-72, High School and Beyond, andNELS:88 and at time-lag cross-sectional surveys ofpostsecondary institutions such as IPEDS andHEGIS.

23Peng, Samuel S., and Jaffe, Jay. 1980. "Third Fol-low-Up" Survi:y Composite Variables. Washington,DC: Department of Health, Education and Welfare,National Center for Education Statistics.

The composition and statistical characteristicsof 14 composite variables from the NLS-72 aredescribed in this report. The composites derivedstem from the seven general areas from the NLS-72:satisfaction with education and training, satisfactionwith work, voluntary participation or "activism,"consumerism, perceived quality of life, political par-ticipation, and factors in choosing graduate school.With few exceptions, the composite variables pro-duced reliability and discriminant validity favorableto their use in social science and education research.(Adapted from article summary)

24Shah, Babubhai V., and others. 1981. "InferencesAbout Regression Models." Washington, DC: U.S.Department of Education, National Center for Edu-cation Statistics. ED 216 044

1 5

Aside from the theoretical issues involving thevalidity of inferences from surveys, the basic prob-lem of producing unbiased estimates of regressionparameters and estimates of the associated standarderrors has been a particularly difficult issue in deal-ing with results from stratified multistage sampledesigns such as the one used in the NLS-72. Thepurpose of this report is to review some appropriateavailable techniques that may be useful in applyingregression models to the NLS-72 data. The firstsection provides a framework for evaluation and anappraisal of some alternate approaches within thisframework. A preferred approach (combining theHorvitz-Thompson estimator and taylorized devia-tion) is compared to the Ordinary Least Squaresapproach through a simulation procedure using ac-tual NLS data. The several results are summarized.Formulae underlying the preferred approach areprovided separately in Appendixes A and B, anddetails of the development and use of a computerprogram to implement the approach are provided inAppendixes C and D. (ERIC abstract)

25Wolf le, Lee M. 1980. Path Analysis with Categori-cal Data. Applications to Ed. ucation, Blacksburg,VA: Virginia Polytechnic Institute and State Uni-versity. ED 189 155

This study uses withdrawal from higher edu-cation to illustrate an educational application of pathanalysis with categorical data. The data are drawnfrom the NLS-72 and the sample includes those whoattended 2-year or 4-year colleges in the fall of 1972.Dropouts are defined as those students who reportedthemselves as not attending school, not graduatingand not reentering school by the time of the 1976"Follow-Up" survey. Results indicate that with-drawal from higher education depends on the addi-tive effects of ability and grades. Race is not a factorin the decision to withdraw.

4. Measurement Property

26Patterson, Barbara J., and Wolfle, Lee M. 1981."Specification Bias in Causal Models with Fallible

9

Indicators." Paper presented at the Annual Meetingof the American Educational Research Association,Los Angeles, CA, April 13-17.

In the bivariate case, measurement error in theindependent variable produces an attenuated esti-mate of the true regression coefficient. In the multi-variate case, the bias which results from specifying,incorrectly, a model with no measurement error willproduce biased estimates which are predictable inneither their direction nor magnitude. Using data oneducational attainment from the NLS-72, this paperexamines the extent of bias inherent in ordinary leastsquares regression estimates when the presence ofmeasui ,ment error is ignored. In some cases, thenature of the bias can be predicted, but the moreusual situation is that measurement error bias isunpredictable. (ERIC abstract)

27Richards, James M. 1983. "Validity of Locus ofControl and Self-Esteem Measures in a NationalLongitudinal Study." Educational and Psychologi-cal Measurement 43(3): 897-905.

The author studied relationships among locusof control and self-esteem relative to selected per-sonal goals (educational aspiration, career goals,marriage and anticipated number of children). Evi-dence regarding the validity of the locus of controland self-esteem measures relative to the goal attain-ment of the above areas was also obtained using the"Base Year" and "1976 Follow-Up" of NLS-72.

Results indicate no meaningful pattern be-tween internal control and positive self-esteem, andgoal attainment. Because locus of control and self-esteem are more strongly related to differential goalattainment, the validity of the NLS-72 measures oflocus of control and self-esteem are questioned.

28Wolfle, Lee M., and Robertshaw, Dianne. 1983."Racial Differences in Measurement Error in Edu-cational Achievement Models " Journal of Educa-tional Measurement 20,(1): 39-49. (See also ED 219429)

Using data from NLS-72, the authors exploremeasurement error differences between black andwhite students. Various statistical tests are used toexamine differences between blacks and whites inmodels of educational attainment and reported pa-rental status.

These tests "indicate that both blacks andwhites relate their reports of parental status to thelatent true score with unit slopes. That is, a one-unitincrease in true status leads, on average, io a one-unitincrease in reported status. However, whites in gen-eral report these measures with less relative enorthan do blacks; as a result, the reliability of measure-ment is greater for whites than for blacks. Finally,these results suggest but do not prove that compari-sons between blacks and whites on models of edu-cational achievement are suspect if they are basedsolely on least-square estimates. Such estimates arebiased by the presence of measurement error; be-cause the extent of measurement error varies by race,interracial comparisons of least-squares structuralparameter estimates may be biased."

29Wolfie, Lee M., and Robertshaw, Dianne. 1981."Latent Causal Structures in Panel Surveys." Paperpresented at the Annual Meeting of the AmericanEducational Research Association, Los Angeles,CA, April 13-17. ED 205 555

Since measurement errors exist in panel sur-veys, LISREL (a procedure for controlling measure-ment error by the analysis of covariance structures)was used in this investigation to determine the sta-bility of the social psychological concept of locus ofcontrol expectancy change with the acquisition ofpostsecondary education. A subset of the NLS-72data for 8,650 white males was used for this study.The measures of the locus of Pontrol, based onRotters' Scale, were used twice, ome in the students'senior year and again 4 years later. The variables ofability and socioeconomic status were also incorpo-rated into the study. Apparently the acquisition offurther education beyond high school does not sub-stantially increase one's internal control expectan-cies. (ERIC abstract)

5. Nonresponse Bias

30Jones, Calvin, and others. 1985 National Longitu-dinal Study of the High Sell& Class of 1972 "Fifth

I I I I 1, I ISO 111 OI

"Third Follow-Up" Survey. Field Test Report.Washington, DC: U.S. Department of Education,National Center for Education Statistics. ED 269465

10

The combined field tests for the "Third Fol-low-Up" survey of High School and Beyond(HS&B) and the "Fifth Follow-Up" survey of theNLS-72 were conducted during the spring f 1985.The overall purposes of the field tests were to evalu-ate questionnaires and procedures to be used for thefull-scale surveys scheduled for 1986. Both theNLS-72 and the HS&B questionnaires were re-viewed for item nonresponse for all items with 4percent or more missing data. Both questionnairesincluded a large number of skip patterns. Empiricalresults concerning response rates to skip patternswere discussed and special problems requiring revi-sion were analyzed. A calendar format was tested tooffer an effective way of capturing periods of inter-rupted schooling and employment, reducing the sizeof the main survey data file without compromisingdata quality. The effectiveness of new and revisedHS&B and NLS-72 items were evaluated by fre-quency distributions and evaluators' observations.The current "Follow-Up" surveys of HS&B andNLS-72 field tests were done by telephone. In addi-tion, a separate questionnaire, the Teaching Supple-ment, was developed for a subset of the NLS-72sample. (ERIC abstract)

6. Sample Design and Effects

31Adelman, Clifford. 1990. "The Data Game: SuccessDepends on Howand About WhomYou Askthe Question." Change (May/June): 44-45.

This article discusses the implications and the"riskiness" of using different types of national

1 7

datasets (cross-sectional versus longitudinal). Theauthor also discusses the advantages of using NLS-72 and the Postsecondary Transcript Sample(PETS), and demonstrates that answers to questionssuch as college completion rates and time-to-degreedepend on who one counts in the universe.

32Benrud, C.H., and Williams, R.L. 1981. Develop-ment of Adjusted School Vhights for the NationalLongitudinal Survey of the High School Class of1972. Washington, DC: U.S. Department ofEducation, National Center for Education Statistics.ED 216 041

The major emphasis of the NLS-72 has beendirected toward individuals in the 1972 graduating

11

class. Consequently, sampling parameters were es-tablished to ensure student representation, and thebulk of data collected were individual-level data.School-level data exist on a separately documenteddatabase, and weights appropriate for analyzingthese data have been computed. The purpose of thisreport is to defme the post-stratification adjustmentprocedure used to develop appropriate school-levelweights, given an initial school selection proceduredesigned to provide student representation. This re-port describes the NLS-72 school sample design,computation of unadjusted student weights (used indetermining weights for individual high school sen-iors), and computation of adjusted weights forschool-level analysis. (ERIC abstract)

1. Labor Force Participation

III. Work

2. Training

3433Black, Matthew. 1980. An Analysis of Youth JobSearch and Geographic Mobility. Princeton, NJ:Mathematica Policy Research.

NLS-72 base year through 1976 follow-upsurveys and the 1970 Census of the Population andHousing 5th Count Summary tape were used toexamine the job search methods of recent highschool graduates, their relocation to a new area, andhow relocation and changes in employment affectsuccessful attainment of career goals. Results of thestudy were intended to help policymakers reduceyouth unemployment and improve the quality ofemployment. The study also examined the role ofthe education system in the school-to-work transi-tion. The study included general background infor-mation on graduates, education policy issues, areview of the literature, determinants and implica-tions of migration and job search activity, theoreticaland probability models of migration and informa-tion on different job search methods.

Findings indicated that recent high schoolgraduates were responsive to labor market signals(such as availability of jobs) and behaved accord-ingly. There appears to be no racial or gender differ-ence in migration behavior. Several factors, such asSES, fewer number of years at high school location,recent moves, marriage (for females only), employ-ment experience and labor force status in the recentpast, and verbal and mathematical aptitude had apositive influence on migration. Migration tended toenhance employment prospects for those who wereinitially unemployed, especially students. For thoselooking for employment, using placement agenciesor union registration was more beneficial than usinginformal search methods such as friends.

See School-to-Work Transition

Alsalam, Nabeel, and Stacey, Nevzer. 1989. Ena=plpy.alraining of Work-Bound Youth: An Histori-cal Review and N43.. Washington, DC: U.S. Department of Labor,Commission on Workforce Quality and Labor Mar-ket Efficiency. ED 317 712

This paper describes work-bound youth bylooking at their history of training, the providers oftraining, and training's effects on salary. A samplefrom NLS-72 was used to look at the education,training and employment experiences of the cohortfrom 1972-1986. ThoFe with no postsecondary edu-cation involvement comprised the work-bound sam-ple.

Some main findings of the study were thatthose who received more formal education (bothdegree and nondegree postsecondary education par-ticipants) also received more training. Those work-bound youth who received more training got highersalaries than their peers with less training. Althoughwomen and men were equally likely to say theyreceived training, women actually spent less time intraining.

Altonji, Joseph G., and Spletzer, James R. 1991."Worker Characteristics, Job Characteristics, andthe Receipt of On-The-Job Training." Industrial andLabor Relations Review 4,5(1): 58-75.

This study uses a subsample of 3,181 respon-dents of the NLS-72 to examine the influence ofworker characteristics and training on employment.The article abstract is as follows:

The authors examined the relationship be-tween the receipt of employer-provided training and

the characteristics of workers and jobs. They foundthat the intensity of training and the duration oftraining were negatively related; the incidence oftraining was slightly higher among women thanamong men, but the amount of training was higheramong men; blacks received somewhat more train-ing than whites; postsecondary education had astrong positive relationship with training; and apti-tude had a positive effect on training. Training doesnot appear to have been affected by high schoolcurriculum or by observable and unobservable fac-tors specific to the individual's high school.

The findings of the study imply "that geo-graphic mobility can be viewed as a producti, eundertaking for promoting upward mobility once ayoung person has successfully entered the laborforce and gained employment."

36Cook, Robert, and others. 1989. Analysis of Appren-

s 1 o sof 1- L. so o: s.Siudy_ofihedlighlehool_Class of 1972, Washing-ton, DC: U.S. Department of Labor, Bureau ofApprenticeship and Training; International Union ofOperating Engineers. ED 306 454

This study examined the effects of on-the-jobor "hands-on" vocational training relative to stand-ard classroom vocational instruction on subsequentemployment, earnings, wages, and job satisfaction.The data were drawn from the NLS-72 base yearthrough fifth follow-up surveys. The study was car-ried out in stages. The first stage used the base yearthrough fourth follow-up surveys and the secondadded the "Fifth Follow-Up" survey.

Findings indicated that apprentices earnedmore than those in the comparison group, both interms of hourly wage and family income. Appren-tices tended to come from families of medium orhigh SES. More apprentices received vocationaleducation, while the comparison group receivedmore general academic instruction. The distributionof grades between the two groups, how ever, wasremarkably similar. Union membership was higherfor apprentices during the period between highschool graduation and 1979. The labor force partici-

pation of apprentices was also higher than the com-parison group.

37Zemsky, Robert, and others. 1982. Training's.

I IIIvided Training and Education: 1969-1981. Task IRepott The Impact of Public Policy on Educationand_Training in the Private Sector. Philadelphia, PA:University of Pennsylvania. ED 265 379

14

This document, comprised of five separate re-ports, presents a set of statistical benchmarks forgauging the growth and development of corporatetraining and education over the last decade. "Com-pany Training, Employment Stability and Employ-ers' Needs and Practices" (Ivar Berg and JaniceShack-Marquez) uses three national surveysthePanel Study of Income Dynamics, the new NationalLongitudinal Survey of Youths, and the NLS-72toexamine company training programs. The first partof this report focuses on the implications that receiv-ing training from the company has on respondents'employment experiences. The second part, whichuses surveys from three Michigan counties, looks atthe relationship lx..ween firms' characteristics andthe human resources needs employers profess tohave.

Findings indicate that, in these national sam-ples, workers do not appear to be in jobs requiringmuch training or have employers who feel theiremployees do not need much additional training.

3. School-to-Work Transition

38Bishop, John, Blakemore, Arthur, and Low, Stuart.1985. High School Graduates in the Labor Market:A Comparison of the Class of 1972 and 1980 Co-lumbus, OH: Ohio State University, National Cen-ter for Research in Vocational Education. ED 268328

This study identified characteristics of highschool experience that were related to initial success

20

or failure in the labor market. The authors hypothe-size that poor school-to-work transition was a resultof poor educational preparation. Students who werenot enrolled full-time in higher education or were notserving in the military are included in the sample.Part-time postsecondary students are also included.Data were taken from the "Base Year" and 1982surveys of the HS&B 1980 cohort and the "BaseYear" and "First Follow-Up" of the NLS-72.

In general, having a job during high school,taking more vocational courses, and having parentswith high income had significant, positive effects onentry wages. High school job (with the exception ofNLS-72 females and HS&B males), higher GPA(for NLS-72 males and HS&B females), greaternumber of semesters of vocational courses(with theexception of NLS-72 males), and higher parentalincome had significant positive impact on annualearnings. Region also appears to have a significantimpact on annual earnings, but a consistent patterndoes not exist. (Article summary)

39Black, Matthew. 1980. An Analysis of Youth JobSearch and Geographic Mobility. Princeton, NJ:Mathematica Policy Research. ED 219 497

For description see entry #33.

See Labor Force Participafion

40Campbell, Paul B., and Basinger, Karen S. 1985.Economic and Noneconomic Effects of AlternativeTransitions Through School to Work. Washington,DC: U.S. Department of Education, Office of Vo-cational and Adult Education. ED 254 638

The study compared the economic and non-economic effects of various combinations of highschool curriculum and postsecondary school-to-work transition patterns. Data for the study wereobtained from the NLS Youth. In some cases, the1979 data from the high school class of 1972 data-base supplemented the NLS Youth data, and for

15

noneconomic outcomes, class of 1972 data wereused exclusively. These data indicated that highschool vocational education is associated with aclear wage advantage for vocational graduates injobs related to their area of training. Postsecondaryeducation also appeared to add to this advantage.Although vocational education brought an increasein labor force participation for white women, nosignificant relationship between vocational educa-tion and employment stability was found for otherdemographic groups. Noneconomic outcomes ofparticipation in a high school vocational educationprogram turned out to be more difficult to assess;however, the noneconomic benefits of participationin postsecondary education were clear. Those whoachieved a postsecondary degree were more likelyto register to vote and to accept as positive thecurrent societal trend in broadening the role ofwoinen in the labor market. The earnings advantageof vocational education was most pronouncedamong white males and did not exist at all for mi-norities of either sex. The policy implications ofthese findings were examined. (ERIC abstract)

See Economics of Education

41Campbell, Paul B., and others. 1984. TransitionPatterns Between Education and Work. Washing-ton, DC: U.S. Department of Education, Office ofVocational and Adult Education. ED 240 272

This study investigated those transitional pat-terns that account for substantial numbers of youngpeople moving from secondary education to em-ployment. Data came from the NLS-72 and theNational Longitudinal Survey of Labor Market Ex-perience, Youth Cohort. Findings indicated thatlarge numbers of students began postsecondary edu-cation but did not earn degrees. Often the interrup-tion was followed by going to work. The availabilityof nearby community colleges led to higher atten-dance. Socioeconomic status, gender, and race wererelated to choice of pathway. Choices were alsoinfluenced by significant others, particularly parentsand close friends. Few students cited high school

2 1

teachers and counselors as important influences. Thechoice of postsecondary education as a pathway wasrelated to the high school experience; the higher thehigh school achievement, the higher the educationallevel the student generally completed. Outcomes otthe nonpostsecondary path were predominantlylower skilled craft and service jobs. For vocational

students, postsecondary work tended to lead to pro-fessional/technical and craft occupations. On-the-job training was the predominant kind ofpostsecondary education. Post-high school training,incomplete pathways, and transitional decisionmak-ing were recommended for policy attention. (ERICabstract)

16

22

IV. Secondary Education

1. School/Program Effects

42Alexander, Karl L., and Pallas, Aaron M. 1983."Private Schools and Public Policy: "New Evidenceon Cognitive Achievement in Public and Privateschools." Sociology of Education 56(October): 170-182.

Recent research by Coleman, Hoffer and Kil-gore (High School Achievement: Public. Catholicand Private Schools Compared. New York: BasicBooks. 1982) on the effectiveness of public andprivate schools may be seriously flawed because ofits neglect of input-level differences in student per-formance in its reliance on cross-sectional testingdata as the criterion measure. Using data from theHigh School and Beyond and the NLS Class of 1972datasets, we examine public-Catholic sector differ-ences within high school tracks for a variety ofcognitive and achievement outcome measures. Evenwithout any controls for sector differences in studentcharacteristics, the public-Cath: differences areall very small. They account for less than 1 percentof the variance in both test scores and in years ofschool completed. When st .xlent selection and back-ground characteristics are controlled, these smalldifferences shrink even further. We thus cannotagree with Coleman, Hoffer, and Kilgore's claimthat Catholic Schools produce better cognitive out-comes than do public schools. This claim is the firstof the "factual premises" that they say would sup-port policies to increase the role of private schoolsin American education. In our view, this premise iswrong, and hence should not be invoked in supportof such policies. (Authors' summary).

43Braddock, Jomills Henry II. 1990. Tracking: Impli-cations for Student Race,Ethniclubgroup,,RaportNo. 1 Baltimore, MD: Center for Research onEffective Schooling for Disadvantaged Students.ED 325 600

This study uses HS&B, NLS-72, the NationalSurvey of Middle School Principals, and the 1986

17

NAEP Young Adult Literacy Survey to depict thestate of curriculum tracking and ability grouping inmiddle and high schools. Effects of tracking onAfrican-Americans, Hispanics, American Indians,Asian-Americans, and white students are included.The NLS-72 is specifically used to examine the stateof tracking as well as the extent of between classgrouping and curriculum tracking.

Findings show tracking to have significantnegative effects for African-Americans, NativeAmericans, and Hispanics. While tracking appearsto positively influence Asian-Americans, the prob-lems of educationally disadvantaged Asian-Ameri-cans may be masked. White students also appear tobenefit from tracking and ability grouping, althoughsignificant numbers of them moved from the aca-demic track to the general and vocational tracksbetween 1972 and 1982. The author feels that coop-erative learning, flexible grouping, within-classgrouping along with cooperative learning and theuse of competency-based curriculum in multigradegroupings are better alternatives to tracking.

44Datcher-Loury, Linda, and Loury, Glenn C. 1983.

Wash-ington, DC: U.S. Department of Education, Na-tional Institute of Education. ED 267 083

I Ij . 1 . VI 1

The authors used a fixed effects model of theimpact of individual high schools on adult earnings.This model improves on past models as it can controlfor selection bias resulting from correlation betweenschool quality and student ability. Data are drawnfrom the NLS-72 base year through 1979 follow-upsurveys and are restricted to a sample of out-of-school white males and females who lived in Chi-cago, Detroit, Los Angeles, New York, Philadelphiaor San Francisco.

Findings reveal large and significant differ-ences in earnings for otherwise identical individualswho attended different high schools. Results alsosuggest that this outcome is not necessarily tilt. "spu-rious result of high ability or wealthier students

2 3

attending the better schools, but instead occurs be-cause of other aspects of the school environment."

See Economics of Education

45Hilton, Thomas L. 1986. Changes in StudentAchievement and Behaviors: A Cross-SectionalAnalysis. 1972-1980. ED 274 670

This study uses data from the NLS-72 andHS&B to examine changes in seniors' achievementbetween 1972 and 1980. The school and studentfactors affecting changes are also examined alongwith policy implications of the results.

Findings show significant changes from 1972to 1980 in student demographics; home and familycharacteristics; student behaviors and attitudes; andtest scores. Some of these findings included anincrease in mean level of parental education, fewerstudy aids in the homes of 1980 seniors, more sen-iors enrolled in general or vocational curriculum in1980, a decline in time spent on homework between1972-1980, and declines on all three achievementtests (math, vocabulary, and reading) and the SAT.The study's fmdings suggest a decreased focus onacademics between 1972 and 1980.

46Lindsay, Paul. 1984. "High School Size, Participa-tion in Activities, and Young Adult Social Partici-pation: Some Enduring Effects of Schooling."Educational Evaluation and Policy Analysis 6(1):73-83.

1979 data from a sample of 8,952 respondentsto the "Fourth Follow-Up" of the NLS-72 were usedto evaluate a model of the effects of school size onparticipation in extracurricular activities and theeffects of this participation on subsequent socialparticipation in young adulthood.

Results support the hypothesis that greater par-ticipation in high school activities was related togreater participation in voluntary social activitieswhen controlling for other influences. Also, studentsfrom smaller schools were more likely to participatein extracurricular activities, this effect being

18

stronger than those of individual background vari-ables, curricular track and class rank. School size,however, did not have a statistically significant di-rect effect on young adult participation.

47Lindsay, Paul. 1982. "The Effect of High SchoolSize on School Participation, Satisfaction, and At-tendance." Educational Evaluation and PolicyAnalysi 4(1): 57-65.

This study looks at the effects of school size onstudent participation, satisfaction and attendance atthe secondary level. The author hypothesizes thatthese factors will be more significant in smallerschools. Four categories of activities are examined:athletics; drama, music and debating; journalism;and student government and political clubs. Studentsatisfaction is examined in terms of satisfaction withthe quality of required courses and with the totalschool experience. Data are drawn from the "BaseYear" survey of the NLS-72 and included partici-pants who completed aptitude tests. Backgroundvariables are controlled for in the analysis.

Overall, the results support the hypothesis thatextracurricular participation is higher at smallerschools. Satisfaction and attendance are also higher.In both cases, differences between school sizes aregreater for women than for men.

48Meyer, Robert, and Wise, David. 1982. "HighSchool Preparation and Early Labor Force Experi-ence." In Tir,Yauth_lahallarktt_Troblom..1ta

Richard B.Freeman and David Wise (eds). Chicago, IL: Na-tional Bureau of Economic Research.

I

This study assesses the impact of academiceducation, vocational training, or work experienceduring high school on labor market outcomes (suchas annual weeks worked and wage rates). Data aredrawn from the NLS-72 "Base Year" through "1976Follow-Up" surveys.

Findings indicate a significant relationship be-tween greater number of hours worked during highschool and number of weeks worked a year after

24

graduation. Those who worked more in high schoolalso received higher hourly wage rates. Job trainingin high school was not significantly related to wagerates or employment after graduation. On-the-jobtraining initially does not affect wage rates butyields significant effects by the "1976 Follow-Up."(Article summary)

49Miller, Jon D. 1985. "Effective Participation: AStandard for Social Science Education." Paperpresented at the Annual Meeting of the Social Sci-ence Education Consortium, Racine, WI, June 5-7.ED 265 083

The study examines the effects of social stud-ies courses on subsequent participation in politicalactivities for blacks and whites. NLS-72 (base yearthrough fourth follow-up surveys) data are used toexplore these effects for those who were not in-volved at all in political activity, those who votedbut were not involved in any other political activityand for those who were involved in one or morepolitical activity other than voting. Level of partici-pation in extracurricular activities, high school pro-gram, and SES are included in the model.

Findings indicate, in general, no direct rela-tionship between social studies courses and partici-pation in political activities. Level of participationin extracurricular activities, enrollment in an aca-demic high school program, and high SES weremore influential. In terms of racial differences, mi-nority students were much more likely either to benot active or to participate in political activities otherthan voting. Nonminorities were more likely to votebut not participate in other activities.

50Navaratnam, K.K., and Oliver, J. Dale. 1990. "AComparison of 1972 and 1980 Secondaly Agricul-tural Education Students." Journal of AgriculturalEducation 31(3): 54-58.

Thi:, study compares senior secondary agricul-tural edu mtion classes of 1972 and 1980 using asample of 373 students from the NLS-72 "BaseYear" survey and 390 students from the High School

and Beyond "Base Year" survey. Three specificissues examined are:

(1) The changes in student background char-acteristics such as race, sex and SES.

(2) The changes in academic ability of stu-dents. (Academic ability is determined by achieve-ment tests administered as part of both NLS andHS&B).

(3) The post high school plans of agriculturaleducation students.

19

Results show that the number of female agri-cultural education students doubled between 1972and 1980. Results also indicate a 21 percent increasein Hispanic/Spanish agricultural education studentsand a 74 percent decrease among the AmericanIndians and Alaska Native group. No changes inSES or academic ability were observed. Between1972 and 1980 there was a 15 percent increase in thenumber of agricultural education students planningto go to a 4-year college, a 17.7 percent increase inthe number of students planning to enter profes-sional careers and an increase in those seniors plan-ning to choose female-dominant jobs. Results of thestudy suggest that the equity movement has elevatedenrollment of female and black students. Areas ofconcern, indicated by the findings, include the de-cline in enrollment of American Indians and AlaskaNatives and the need to attract more females tomale-dominant professions.

51Navaratnam, K.K., and Asche, F. Marion. 1988. "ACross-Sectional Analysis of College PreparatorySeniors of 1972 and 1980." The High School Jour-nal (February/March): 127-134.

The objectives of this study are to analyzechanges in demographic characteristics over time(including sex, race, religion and SES), to estimatethe changes in measured student achievement, andto compare the school experiences, educational andoccupational plans, aspirations, and postschool ac-tivities for both high school seniors of the class of1972 and the seniors of the class of 1980. Theauthors feel that this type of analysis is necessary in

2 5

order to affect policy aimed at improving secondaryeducation.

Using data from both the "Base Year" and"First Follow-Up" of the NLS-72 and High Schooland Beyond senior cohorts, the authors conducted across-sectional analysis comparing college prepara-tory/academic high school students. Findings indi-cate several changes in each variable category overthe study period, although these changes weresmaller than anticipated. Changes include an in-crease in the proportion of females and minorities inthe college preparatory curriculum, modest changesin federal support for special programs for the handi-capped, disadvantaged and limited-English-speak-ing students, and an increase in the number ofcollege preparatory students with high SES. Theauthors suggest that increased participation in extra-curricular activities and out-of-school work accountfor differences in tested academic performance andalso that the rise in student grades may lend supportto the "grade inflation" hypothesis.

52Riordan, Cornelius. 1985. "Public and CatholicSchooling: The Effects of Gender Context Policy."American Journal of Education 518-540.

This study examines the effects of single- andmixed-sex Catholic schools compared to mixed-sexpublic schools. The primary purpose of the compari-son is to assess whether students at single-sexschools outperform those at mixed-sex schools. Thedata are taken from the NLS-72 "Base Year" and"1979 Follow-Up" and include the full number ofrespondents who were white, Catholic or publicschool students enrolled in college, general or busi-ness tracks. The variables used in the comparisonare vocabulary, reading, and math achievementscores and the SAT verbal and math scores.

Findings indicate few ditferences betweenCatholic mixed-sex and public mixed-sex schools.The only steady difference between the two is thatCatholic school males tend to perform better ver-bally than public school students. Results from othercomparisons between mixed-sex schools are incon-sistent, insignificant or favor public schools. Stu-dents in Catholic single-sex schools, however,

20

consistently score higher than public school stu-dents. The author concludes that, because of thesmall differences between the mixed-sex schools,the two types of Catholic schools should be differ-entiated when doing comparisons with publicschools. He also suggests further exploration of whysingle-sex schools may outperform mixed-sexschools.

53Schmidt, William H. 1981. The High School Cur-riculum: It Does Make a Difference. East Lansing,MI: Michigan State University. ED 213 093

The question examined in this paper is whethervariability in the quantity of schooling students re-ceive in different curricular areas is a contributor toobserved differences in achievement, not onlyamong students attending different high schools, butamong students in the same high school. A concep-tual framework enumerates the determinants ofachievement, including school and communitycharacteristics, student background, and quantity ofschooling in the specific curricular areas of mathe-matics, English, foreign language, fine arts, socialstudies and science. The sample used was 9,195 highschool seniors in 725 schools taken from the NLS-72. The results suggest that quantity of schooling hasa positive effect on academic achievement. Themore the achievement is school-related, the larger isthe resulting effect of the quantity of schooling. Thiswas especially true for mathematics. Quantity ofschooling also had positive effects on achievementin science and English; less clear results were foundin the areas of vocabulary and reading comprehen-sion, although quantity of schooling did continue tohave a positive effect on achievement. (ERIC ab-stract)

54Stone, James R. 1984. "Economic Attainment ofSecondary Marketing and Distributive EducationStudents." Paper presented at the Annual Meetingof the American Educational Research Association,New Orleans, LA, April 23-27. ED 242 934

By developing models of economic attainmentusing factors such as job status attainment, unem-ployment, and salary, the study examines the rela-

26

tionship between participation in secondary market-ing and distributive education (MDE), and economicattainment after gaduation. Two samples drawnfrom the NLS-72 data were used for the analysis, ruidconsisted of 1,118 students who took a set amountof courses in MDE and another 3,500 employed inmarketing-related jobs in 1979.

Results showed positive, significant relation-ships between MDE participation and cooperativeeducation, and job status attainment. Other factorsinfluencing job status attainment in marketing weresex, education attainments, GPA and the communitysize where the respondent attended high school.Educational attainment was the strongest predictorof job status attainment and sex was a strong nega-tive predictor. MDE and cooperative education onlyexplain a small portion, however, as 86 percent ofthe variance is explained by factors not included inthe model.

55Trent, William. 1981. Contrasts. Trends and Impli-cation, of Student Course Enrollments and Extra-curricular Membership in Desegregated HighSchools. Baltimore, MD: Johns Hopkins UniversityCenter for the Social Organization of Schools. ED229 491

This paper examines academic program en-rollments, course enrollments, and extracurricularmemberships in racially and ethnically mixed highschools in order to determine the extent of participa-tion by black and white students. The major issue iswhether, in desegregated schools, black and whitestudents are resegregated through these mecha-nisms.

The study is based on data collected from 1,318schools throughout the nation for the NLS-72. Thestudy finds that white students in desegregatedschools participate less in extracurricular activitiesand clubs than do white students in segregatedschools, and that black male students in desegre-gated schools participate more. Both black male andblack female students in desegregated schools par-ticipate more in athletics, drama, and music activi-

ties, but white students participate more in honoraryclubs.

For academic program and course enrollments,the study finds that black students are less likely thanwhites to be enrolled in academic or college prepara-tory programs in segregated and desegregatedschools. In desegregated schools, however, blacksand whites are about equally likely to be enrolled inmost courses, but the use of tracking or ability group-ing procedures significantly reduces the opportuni-ties for cross-racial contact in those courses. (Articleabstract)

56Trent, William, and McPartland, James M. 1981."Race Comparisons of Student Course Enrollmentsand Extracurricular Memberships in Segregated andDesegregated High Schools." Baltimore, MD:Johns Hopkins University Center for the Social Or-ganization of Schools. ED 210 406

21

The purpose of the study was to examine theparticipation of black and white students in aca-demic programs, courses, and extracurricular activi-ties in racially and ethnically mixed high schools.The authors also looked at whether in-school segre-gation occurs within desegregated schools on thebasis of those factors. Data are taken from NLS-72.Participation rates are examined by race and resi-dence (north, south) for both segregated and deseg-regated schools. A comparison ot participation ofblack Fad white students from the same school isalso done to examine where opportunities for cross-racial contact exist as well as where patterns of overor underrepresentation occur.

Findings are similar to Trent (1981) (see entry#55). White students enrolled in desegregatedschools tend to participate less in extracurricularsthan their peers in segregated schools. Both blackmales and females participate more in desegregatedschools, especially in athletics, music, and drama.Blacks and whites are equally likely to be enrolledin most courses in segregated schools. The existenceof tracking, however, sign' icantly hinders the op-portunity for cross-racial contact.

2"

57Wiley, David E., and Harnischfeger, Annegret.1980. Curricular Resource Allocation and CourseExposure in Secondary Schools. Chicago, 1L: M-LGroup for Policy Studies in Education. ED 218 301

Data from the NLS-72 "Base Year" survey,and from a school record form filled out by schoolofficials describing course-taking information forthe last 3 years of high school by curricular area foreach student' s academic record, are used to presentthe degree of exposure students received in variouscurricular areas.

Findings indicate large and varied differencesin academic instruction across groups (ethnicity andgender). Also, pupils enrolled in nonvocational pro-grams actually received more vocational instructionand resources. Females received more vocationalinstruction than males and black students receivedmore vocational resources in academic programsthan whites, but less in vocational or general curric-ula. A major factor influencing academic prepara-tion for further education and employment was highschool course track membership. "This study con-cluded profound track-induced barriers to postsec-ondary access are built up due to the strong variationin course exposure over tracks. This occurred re-gardless of the ultimate aspirations of the students."(Adapted from ERIC abstract)

See Vocational Technical and Vocational Educa-tional

58Wiley, David, E., and Harnischfeger, Annegret.1980. Highaehool Learning, Vocational Tracking.and What Then? Chicago, IL: M-L Group for Pol-icy Studies in Education. ED 212 847

After analyzing those data from the base yearthrough third follow-up surveys of the NLS-72 per-taining to vocational education, researchers arrivedat a number of conclusions. Included among thesewere the following: (1) race, sex, and academicperformance are key factors for selection of voca-tional education programs, with minority students,

22

poor achievers, and females being the most frequentenrollees in vocational education; (2) because voca-tional education funds are allocated by local enroll-ment and program costs rather than by programenrollment, vocational education resources andservices are diffused far beyond those enrolled invocational education; (3) those in vocational pro-grams work 40 percent more than do those in non-vocational programs; (4) far fewer vocational-trackstudents receive postsecondary schooling than doacademic-track students; and (5) high schoolcoursework and work hours related strongly to post-secondary work hours and wages. (ERIC abstract)

59Wolfle, Lee M. 1987. "Enduring Cognitive Effectsof Public and Private Schools." Educational_jk::searcher16: 5-11.

In response to several articles assessing theeffectiveness of public and private schools, theauthor extends the debate about effects, specificallyof single-sexed and mixed-sexed Catholic schoolsversus mixed-sex public schools. Using the NLS-72"Base Year" and "Fourth Follow-Up" surveys, thelong-term impact of Catholic schooling on math andvocabulary achievement are examined.

"Base Year" test battery, as well as the numberof semesters of high school and college math, wereused to assess math and vozabulary achievement.Results of the study indicate that short-term effectsof Catholic schooling do not endure as no significantlong-term changes in vocabulary achievement areseen. Also, compared with public school students,students attending Catholic school experience a de-cline in mathematics achievement over time.

The author suggests that limitations of thestudy include the lack of sufficient measures ofdeveloped ability prior to entering high school andthe effects of placement of test scores in the model.If test scores, (used in this model as control vari-ables) were placed prior to entry into either Catholicor public schools, the total impact of Catholicschooling on math would be negative and therewould be no difference in vocabulary achievementbetween Catholic and public schools.

28

2. Student Characteristics

60Hansen, David 0., and Ross, Peggy J. 1980. "Edu-cation, Significant Others and Farm-Reared Adoles-cents." Paper presented at the Annual Meeting ofthe North Central Sociological Society, Akron, OH,May 1-3. ED 211 292

This study examined the role of significantothers in the educational and occupational decisionsof farm-reared youth. Specific questions addressedinclude:

(1) Compared to others, how important areeducators in influencing the choice of high schoolcurriculum?

(2) Compared to others, how important areeducators in influencing post-high school plans?

(3) To what extent do educators influence thechoice of post-high school career activities?

(4) What factors predict variations in the influ-ence of educators and other significant others oncurriculum choice and post-high school plans?

23

Results show that family and friends exertedmore influence than teachers and guidance counsel-ors over career choice. Educators, however, focusedon encouraging educational plans and discouragedmilitary service and employment. These results sug-gest that teachers and counselors do not 'cool-out'farm youths. Level of encouragement was related toSES, ability and gender, with those of high SES andability more encouraged to attend college and girlsbelieving that significant others had more influenceover their choice of curriculum.

29

V. Postsecondary Education

1. Access

61Alexander, Karl L., Pallas, Aaron M., and Holupka,Scott. 1987. "Consistency and Change in Educa-tional Stratification: Recent Trends Regarding So-cial Background and College Access." SocialSitatificatiLaathiliability L: 161-185.

After a 30- or so-year period of rapid growth,college enrollments and high school-to-college con-tinuation rates have leveled off. The present chapterexamines current conditions of postsecondary ac-cess in light of these developments. Enrollmentpatterns from the NLS-72 (base year through firstfollow-up surveys) and the HS&B 1980 senior co-horts (base year through first follow-up surveys) arecompared to evaluate changes in racial-ethnic, so-cioeconomic, and gender constcaints upon collegeattendance. Academic resource measures (i.e., testperformance, high school grades, and high schoolcurriculum) are also examined in order to assess thepatterning and magnitude of such disparities whenacademic qualifications are equivalent.

Attendance among members of the more re-cent cohort is about 5 percent less than for the earliergroup, but this overall average decline obscurespatterns of improved and diminished opportunities.Attendance among women actually has increasedsome, with the largest gains being at the lowersocioeconomic levels. In contrast, attendanceamong men has fallen off sharply, especially amonglow resources and low SES whites and blacks. As aresult of these divergent trends, women now attendcollege at higher rates than men, which reverses alongstanding pattern of male advantage. Minorityyouth (i.e., blacks and Hispanics) also attend athigher rates than whites at all but the highest SESstrata, and when academic resource differences arecontrolled, the minority advantage appears at allpoints of comparison. Socioeconomic differentials,however, are persistent, and reflect the charac-teristic disadvanrage associated with low status ori-gins. Finally, attendance has not declined among

youth with the highest academic qualifications, noris there much evidence of ascriptive inequalityamong them. The implications of these results forour understanding of current conditions of educa-tional stratification are discussed in the concludingsection. (Article abstract)

62Alexander, Karl L., and Holupka, Scott. 1987. "So-cial Background and Academic Determinants of2-Year versus 4-Year College Attendance: Evi-dence from Two Cohorts a Decade Apart." Ameri-can Journal of Education 9: 56-80.

This paper examines determinants of access to2-year versus 4-year colleges among college-goingyouth. Using the NLS-72 and High School andBeyond "Base Year" and "First Follow-Ups," theauthors evaluate changes in racial-ethnic, SES andgender constraints on enrollment patterns and howaccess to different college types is affected by stu-dent background characteristics and academic re-source measures (test performance, high schoolgrades and high school curriculum). Issues regard-ing the relationship between type of college at-tended and economic/social outcomes are alsoaddressed.

The three main findings are:

(1) The effects of student background on ac-cess to 4-year schools are small. The SES differ-ences found in the initial ol3prvation diminishgreatly once correlated differences in academic cre-dentials/qualifications are added to the analysis.Blacks had somewhat greater probability of attend-ing 4-year colleges than whites, and Hispanics areoverrepresented in 2-year schools.

(2) Comparisons between the 1972 cohort andthe 1980 cohort indicate little falloff in opportunitiesto attend 4-year colleges.

(3) Evidence of pre-college preparation is agood predictor of access to 4-year colleges.

2530

63Gardner, John A. 1987. Transition from HighSchool to Postsecondary Education: Analyticalstudies. National Center for Research in VocationalEducation. Columbus, OH: Ohio State University.(See also ED 280 370)

This study focuses on factors affecting accessto postsecondary education. The major emphasis ison aspects of personal background and institutionalpolicy that influence educational aspirations andenrollment in postsecondary education. The influ-ences of family income and availability of financialaid on decisions to attend college, as well as methodsused to finance postsecondary education are alsoexamined. Data from the NLS-72 and High Schooland Beyond are used in the analysis.

Results provide information on educationalexpectations, rates of enrollment, institutional char-acteristics, programs selected, and methods of stu-dent financing. Information on policy implicationsand areas of further study is also included. Majorfindings include:

(1) Little change was evidenced between 1972and 1980 in the overall level of education that highschool seniors expected to attain.

(2) Higher academic achievers, students fromhigh SES families, and students from families withhigher incomes are more likely to aspire to at leastthe bachelor's degree, and less likely to expect onlyhigh school graduation.

(3) Those seniors who expected at least abachelor's degree preferred to attend 4-year col-leges rather than junior colleges. Blacks are muchless likely than Hispanics and slightly less likelythan whites of the same gender to expect only highschool graduation.

(4) Rates of attendance or enrollment immedi-ately following high school graduation have fallenover the last decade.

(5) Socioeconomic status has a stronger im-pact on attendance for whites or Hispanics of eithergender than for blacks.

26

(6) Hispanics are more likely than blacks orwhites to attend 2-year institutions.

(7) Scholarships, grants and loans are usedmore often by students of higher ability than by thoseof low ability.

(8) As expected, Pell grants, the most fre-quently used source of aid, are used more often byblacks, students from low- income families, and lowacademic aptitude students.

64Levin, Bernard H., and Clowe, Darrel A. 1987."Competition Between Community Colleges andPostsecondary Proprietary Schools: Reality orMyth?" Journal of Studies in Technical Career2(4): 317-324.

T in study uses data from the NLS-72, a sampleof 487 students attending public postsecondary in-stitutions and a sample of 145 proprietary studentsfrom the overall sample of students stating they werein a vocational curriculum, to test the hypothesis thatcommunity colleges compete with postsecondaryproprietary schools. The authors compare studentson the basis of race, sex, socioeconomic status,parental influence and peer influence.

Findings show that proprietary school studentswere more likely than community college studentsto be white, female and middle class. Parental andpeer influence appear to be the primary reason stu-dents attended proprietary schools. Data show, sta-tistically and conceptually, significant differencesbetween the two groups and therefore do not supportthe hypothesis that competition exists between thetwo types of institutions.

65Manski, Charles F., and others. 1983. CollegeChoictiaAmericn. Cambridge, MA: Harvard Uni-versity Press.

The processes of choosing and being acceptedby a college are analyzed based on data from theNLS-72. Econometric modeling and descriptive sta-tistics are provided on: student behavior in selectinga college, choice of school/nonschool alternatives,

31

decisions by colleges concerning grant aid alloca-tions, the enrollment effects of the Basic Educa-tional Opportunity Grant Program, graduates'postsecondary school and early work experiences,and individual and family attributes that affect ap-plication and acceptance. The predictive validity oftest scores and high school performance is alsoexamined. Among the findings are: most highschool graduates would be admitted to some 4-yearcolleges of average quality were they to apply; ap-plicants do not necessarily prefer the highest qualityschool; high school class rank and Scholastic Apti-tude Test (SAT) scores are equally important incollege admissions; federal scholarship aid has hadonly a small effect on enrollments at 4-year collegesbut a much stronger effect on attendance at 2-yearcolleges; and the attention paid to SAT scores inadmissions is commensurate with the power ofscores in predicting persistence to attain a collegedegree. (Adapted from ERIC abstract)

66St. John, Edward P., and Noe 11, Jay. 1989. "TheEffects of Student Financial Aid on Access toHigher Education: An Analysis of Progress withSpecial Consideration of Minority Enrollment."Research in Higher Education 3.(1 (6): 563-581.

Using NLS-72 "Base Year" and "First Follow-Up" and High School and Beyond 1980 and 1982cohorts, the report analyzes the effects of both aidoffers and type of aid on enrollment decisions bothby college applicants in general and minority stu-dents. The authors' model compares the influenceof the type of aid package on enrollment decisionsfor different students.

Major findings include:

(1) All types of aid packages had a positiveimpact on enrollment decisions by college appli-cants in all three high school classes.

(2) All types of aid had positive influence onenrollment by minority students.

(3) Background effects, such as mother' s edu-cation, academic track in high school, high gradesand high postsecondary aspiration had a strong in-

27

fluence on college enrollment in all three classes.Findings also indicate gains in access to higher edu-cation between the 1970s arid 1980s.

(4) Student aid appears to have a strongerimpact on access for minority students than forwhites, although all types of aid had a positiveimpact on the enrollment decisions by all threegroups. The size of the affect of loans is less thanthat of grants/scholarships in 1980 and 1982.

qee Financial Aid

67Thomas, Gail E., and Braddock, Jomills Henry, II.1981. Determining the College Destination of BlackStadonts. Atlanta, GA: Southern Education Foun-dation. ED 215 031

Drawing from the NLS-72 base year throughfourth follow-up surveys, the authors examined fac-tors determining the college destination of blackstudents. Additional information on college charac-teristics was taken from the Office for Civil Rights1972 Survey data. Effects of several variables wereexamined in terms of enrollment at (1) 2-year versus4-year colleges, (2) public versus private colleges,(3) predominantly black versus predominantly whitecolleges, and (4) highly selective versus less selec-tive colleges.

Findings show that standardized test perform-ance and high educational expectations were majorpredictors of attending a private college for all sexand regional groups. High school curriculum was animportant predictor of enrollment for males andfemales from the South. The location of the collegesstudents attended was also related to the demo-graphic characteristics of their high schools.

2. College/Program Effects

68Anderson, Kristine L. 1986. "College Contexts, Stu-dent Integration, and Educational Attainment." Pa-per presented at the Annual Meeting of theAssociation for the Study of Higher Education, SanAntonio, TX, February 20-23. ED 268 874

32

This study focuses on structural and organiza-tional features of institutions in relation to studentpersistence. The analysis incorporates ideas fromTinto' s theory (1975) and status attainment models,and looks at persistence in terms of the influencescolleges have on social and academic integration,academic performance, goal commitment and edu-cational attainment.

The sample of 2,700 males and 2,700 femalesconsists of all NLS-72 respondents who enteredcollege directly after high school and were in aca-demic programs. Persistence was looked at in termsof academic/graduate institutions versus vocationalsub-baccalaureate institutions, size and complexity,and gender.

Background variables all exhibited significanteffects on such variables as quality of college at-tended, institutional size, and degree of integration.Academic skills and achievement goals had strongereffects than SES. They also had significant effectson experiences and achievement. Quality of thecollege chosen appears to increase goal commit-ment and attainment while SES and integrationseem to lead to lower goal commitment.

Regarding college characteristics, the authorhighlights the importance of considering more thana single college dimension "since no one factor fitobserved patterns in the data" or effected all studentsin the same way.

See Persistence

69Anderson, Kristine L. 1984. Race Differences in theEffects of College Characteristics on EducationalAttainment. Washington, DC: U.S. Department ofEducation, National Institute of Education. ED 256249

The study examines patterns of attainment forblack and white students attending a variety of insti-tutions in the early 1970s. The sample is composedof 795 black and 6,581 nonblack respondents to theNLS-72. The author specifically looks at the follow-ing questions:

(1) Are there institutional characteristics thatare particularly important in helping black studentsperform adequately, persist, and graduate from col-lege?

(2) Are the performance, persistence, andgraduation of black students particularly dependenton institutional characteristics?

(3) Are integration into college life, facultysupport, and avoidance of conflicting roles moreimportant for black students?

Findings show that high school, race, and aca-demic preparation affect black students' college se-lection while SES and goals affect the decisions ofwhite students. Black students who attended pre-dcminalitly black institutions received higher gradesand were more likely to persist to a second year andobtain a bachelor' s degree when compared to blackswho attended predominately white schools. Find-ings also show that black students have higher attain-ment at cohesive, small, private, less vocationalcolleges with low SES composition.

28

See Gender and Racial-Ethnic Differences

70Anderson, Kristine L. 1984. The Effect of CollegeType and Characteristics on Educational Attain-nient. Washington, DC: U.S. Department of Educa-tion, National Institute of Education. ED 256 206

The study examines the impact of institutionaltypes and of their structural and compositional char-acteristics on a sample of students who entered col-lege in the early 1970s. A sample of 7,376 entrantsto 2- or 4-year colleges, enrolled in academic curric-ula, by fall of 1975, from the NLS-72 "Base Year"and "First Follow-Up" are used in the analysis. Thestudy specifically focuses on educational attain-ment. College and university variables consist ofprivate versus public control, level (university, 2-year or 4-year college) and various other specificcharacteristics such as ethnicity/gender composi-tion, percent of minority and graduate students, sizeof enrollment, and tuition.

33

Findings regarding college and university typeshow that private college and university entrantsreceive high grades and have high levels of facultycontact and satisfaction. Those attending public in-stitutions have a small advantage in overall attain-ment but tend to have lower grades and satisfactionlevels and the least amount of faculty contact. 2-yearcollege students tend to receive higher grades buthave a lower chance of graduating. Colleges with adiversity of programs and majors appear to havehigher :ates of degree completion and overall attain-ment, although students have somewhat lowergrades and lower levels of satisfaction. Collegeswith high ability student populations and low SEScomposition have positive effects on degree com-pletion, achievement and attainment goals. SEScomposition, however, appears to be more influen-tial than ability composition. Minority compositionhas positive effects on grades but not on degreecompletion or attainment. It also has positive affectson persistence early on in the college career.

See AttainmentEducational and Occupational

71Anderson, Kristine L. 1984. College Effects on theEducational Attainment of Males and Females.Washington, DC: U.S. Department of Education,National Institute of Education. ED 256 205

This study examines college experiences andeducational attainment of students by merging dataon postsecondary institutions and NLS-72 data. Ex-periences and attainment are examined in terms ofinstitution type and college characteristics (such asaverage SAT score, percentage of incoming stu-dents with family income below $6,000, proportionof major areas in vocational fields, institution size,highest degree granted, proportion of part-time stu-dents, educational and general expenditures per stu-dent and average tuition and fees). Respondents toNLS-72 who entered academic programs in a 2-yearor 4-year college directly after high school gradu-ation, and who completed the base year through1979 follow-up surveys are included in the analysis.Separate analyses are performed for males and fe-males.

Findings indicate that attending schools with ahigher proportion of vocational programs, a higherproportion of part time students, higher costs andhigh SES composition is most detrimental to fe-males' persistence and graduation rates. For men,larger size, higher proportions of part-time students,and high per student expenditures with low tuitioncosts were most detrimental.

29

The author suggests that more research be doneon the measurement and analysis of college charac-teristics other than those most closely tied to "selec-tivity" or "quality." There is also a need foradditional information on the role of different stand-ards for academic performance at different kinds ofinstitutions. The author proposes further exploringdifferences in involvement of students with the stu-dent role rather than with roles as spouse, parent orworker and examining the college outcomes forvarious types of students, such as nontraditionalstudents.

See AttainmentEducational and Occupational

72Clowe, Darrel A., and Levin, Bernard H. 1980."How Do 2-Year Colleges Serve Recent HighSchool Graduates?" Community College Review2(3): 24-35.

This study examines patterns of postsecondaryschooling for a sample of NLS-72 respondents.Patterns are looked at in terms of students' race,SES, aptitude, and also by type of college first at-tended. The most significant finding of the study isthat blacks and whites have the same educationalaspirations.

73Hilton, Thomas L., and Schrader, William B. 1985.Siudcnts_pf_thcilumanititsiniht19.10-9.Anda9.8.4a.Washington, DC: National Endowment for the Hu-manities. ED 270 017

This report is a longitudinal study comparingthe educational paths of 1972 and 1980 senior hu-manities majors. Data from the NLS-72 for 1972seniors who received a bachelor' s degree in thehumanities by October 1979 and for a comparison

34

group of 1980 seniors from High School and Be-yond are used in the analyses. Specific questions forthe 1972 group included: (1) By what pathways didthey progress to their degree?; (2) How did thehumanities students differ from students who hadearned a bachelor' s degree in other fields by Octo-ber, 1979?; (3) How did the humanities studentsdiffer in 1972 from students who majored in otherfields?: and (4) How did the humanities studentsdiffer in 1972 from humanities students in 1980?

Findings indicate that humanities majors hadhigher test scores and grades than students majoringin other areas. In addition, the net number of studentsmajoring in the humanities increased from the fresh-man to senior year for the 1972 cohort. No evidenceexisted indicating limits in postbaccalaureate educa-tional and career opportunities for humanities ma-jors. Finally, results -how that humanities serve theeducational needs of a wide range of undergradu-ates.

74Conaty, Joseph C., Alsalam, Nabeel, James, Estelle,and To, Duc-Le. 1989. "College Quality and FutureEarnings: Where Should You Send Your Sons andDaughters to College?" Paper prepared for the 1989Annual Meeting of the American Sociological As-sociation.

The authors use data from the "Fifth Follow-Up" of the NLS-72 to examine the impact of collegecharacteristics (as an indicator of college quality) onfuture earnings of college graduates. The influenceof variables such as student expenditures, studentbody composition, college type as well as studentbehavior while at college (academic achievement,curriculum, use of time and effort) and gender dif-ferences on income are explored. A sample of 1,320men and 1,009 women with college degrees, whosegraduate institution is identified, who took at least60 credit hours at that school, and who worked foran employer at least 20 hours a week in 1985 is usedin the analysis. Data on institutional characteristicsare taken from HEGIS and data on college experi-ence are taken from PETS.

General findings indicate that the impact ofcollege on future earnings is small, especially for

men. What students do while they are at collegematters more than which college they attend.

75Hunter, Vincent Curtis. 1986 Academic Achieve-ment and Self Attitudes: A Longitudinal Study withBackgound Controlled. ED 303 526

30

Data from NLS-72 are used to determine therelationship between student achievement and self-attitudes (such as self-concept and locus of control).A sample of 4,628 participants in the study (baseyear through third follow-up surveys) is included inthe analyses. The specific question addressed waswhether self-referent attitudes have lasting effectsover time on educational achievement, independentof the effects of background and attitude.

Findings show achievements and attitudeshave a small, significant relationship after back-ground influences have been controlled. This rela-tionship generally remains stable over time.

76James, Estelle, Alsalam, Nabeel, Conaty, Joseph C.,and To, Duc-Le. 1989. "College Quality and FutureEarnings: Where Should You Send Your Child toCollege?" Economics of the Education Industry22(2): 247-252.

This study examines the influence of collegequality (here defined as future earnings) on malecollege graduates. A sample of 1,241 males who hadreceived a college degree, whose graduating institu-tion was identified, who took at least 60 -.redit hoursin the identified school, and who worked for anemployer in 1985 is used in the analysis. Institutionaldata are drawn from HEGIS and college experiencedata are taken from PETS.

Findings indicate that characteristics such asselectivity and private East coast location have posi-tive impacts on earnings. Expenditures per studenthad insignificant effects. College experience vari-ables, such as GPA and major, explain more of thedifference in earnings than do measured familybackground, ability, and college characteristicscombined.

35

77Kanouse, David E., and others. 1980. Effects ofPostsecondary Experiences on Aspirations.Attitudes. and Self-Perceptions. Washington, DC:Department of Health Education and Welfare. ED214 430

The effects of postsecondary experiences onthe development of young people's aspirations, at-titudes, and self-conceptions during their first 4years after high school were studied. Attention wasdirected to self-esteem; orientations toward work,family, and community; sex-role attitudes; educa-tional expectations; career aspirations; satisfactionwith career progress; and locus of control.

To determine the effects of career and educa-tional choices on the outcomes of education, com-parisons were made among groups that followeddifferent tracks after high school graduation, includ-ing postsecondary education, entry into the civilianlabor force, military service, full-time homemak-ing, and unemployment. An individual effectsmodel was used to adjust observed outcomes over a4-year period for initial differences in ability, highschool achievement, family background, and otherfactors. In addition, changes in outcome measuresover time were analyzed.

Theoretical perspectives concerning self-con-cept and its relationship to life choices made inyoung adulthood are briefly considered, and theNLS-72 database and study methodology, includingthe development of a "Career Aspiration Index," areexamined. The main finding is that the initial differ-ences between young people who enter various post-secondary tracks are generally much greater thanrelative changes that emerge thereafter. Postsecon-dary experience generally had less effect on atti-tudes and aspirations than did factors that led to theirinitial choices. (Adapted from ERIC abstract)

78Robertshaw, Dianne and Wolf le, Lee M. 1981. "TheCognitive Value of 2-Year Colleges for Whites andBlacks." Integrated Ethic Ad.= 12(3-6): 68-71. Seealso paper presented at the Annual Meeting of theAmerican Educational Research Association, NewYork, NY, March 19-23.

Data drawn from the "Fourth Follow-Up" ofthe NLS-72 are used to evaluate verbal and quanti-tative skills of students. Achievement test perform-ance for 1,273 whites and 327 blacks who (1)received certificates or licenses for completion ofnoncollege-related vocational programs; (2) whograduated from 2-year colleges in a vocational edu-cation program; (3) who graduated from a 2-yearcollege general education program; or (4) whograduated from 4-year college programs, is com-pared to those who did not continue on to postsec-ondary education.

Results for white students indicate that themean level of both verbal and quantitative achieve-ment of those who received postsecondary educa-tion was significantly different from theachievement of those who did not, even when con-trolling for SES and 1972 test scores. Verbal testperformance rose between 1972 and 1979 whilemath scores experienced a general decline. For blackstudents, the same pattern was found, although theirmean scores were considerably lower than the scoresof their white peers with similar educational back-ground. Also, when comparing scores of those withsome postsecondary education to the scores of thosewith none, the only significant difference is foundfor black, 4-year college graduates.

31

These findings suggest that math competencemay decline or remain the same over time, regard-less of the type of education obtained, and that2-year colleges contribute to the cognitive develop-ment of their students, although the benefits may begreater for white students. However, limitationssuch as lack of a more comprehensive measuringinstrument and lack of larger samples of communitycollege graduates (especially blacks) need to betaken into account.

79Sharp, Laure M., and Weidman, John C. 1989."Early Careers of Undergraduate Humanities Ma-jors." Journal of Higher Education fla (5): 544-64.

The study assesses the extent to which collegemajor plays a decisive role in the early career pat-terns and subsequent occupational attainments ofthose entering the workforce with a bachelor's de-

3 E;

gree in the humanities. In addition, the study alsoexplores the effects of family background charac-teristics, self-concept, personal goals, aspirations,academic performance in college and quality of theinstitutions from which the graduate receives thedegree, on the process of early career attainment forundergraduate humanities majors.

The authors used the NLS-72 base yearthrough the 1979 follow-up surveys to describe theemployment status of and jobs held by collegegraduates who had been in the labor market for, atmost, 3 years since receiving their undergraduatedegrees. Findings indicate little difference betweenhumanities graduates and those in other fields interms of prestige, pay, and extent to which their jobsinvolved working with ideas during the early stagesof their careers. Davi, then, do not appear to supportthe prevailing notion that humanities majors, com-pared to thog majoring in other areas, are at adisadvantage in the labor market, although humani-ties majors were somewhat less likely to work full-time and more likely to work part-time. Also, usingDuncan' s socioeconomic index of occupationalstatus (SEI), the authors found that men and womenwho majored in humanities and liberal arts were ina: lower prestige cluster (with women doing some-what better than men). Findings also show consis-tent differences in earnings between men andwomen, with men having a steady earnings advan-tage in all fields (although the gap is narrowest ineducation).

See AttainmentEducational and Occupational

80Sharp, Laure M., and Weidman, John C. 1986. EarlyCareer Patterns of Undergraduate Majors in theHumanities. Research Report Number 1. Pittsburgh,PA: School of Education University of Pittsburgh.ED 274 258

A forerunner to #79, this exploration also con-sidered first jobs and career attainments of collegegraduates with baccalaureate degrees in the humani-ties soon after receipt of their degrees. Outcomes forhumanities majors were also compared to outcomesfor social science, business, and education majors,

as well as graduates who majored in interdiscipli-nary studies in the liberal arts and sciences. Theauthors concluded that while a sizable proportion ofhumanities graduates have not been launched onprofessional careers during the early postcollegeyears, their work experiences have not been discour-aging and have differed little from the work experi-ence of their contemporaries who majored in otherfields. Detailed findings are presented, along withappendices that explain study variables and provideinformation on the number of bachelor's and mas-ter' s degrees conferred, by major, for U.S. collegessince academic year 1968-69. (Adapted from ERICabstract)

32

See AttainmentEducational and Occupational

3. Financial Aid

81Brown, Kenneth G. 1980. "Dropout Rates: A Lon-gitudinal Analysis of Student Loan Users ComparedWith Users of Other Forms of Financial Assistance.AIR Forum 1980 Paper." Paper presented at theAnnual Forum of the Association for InstitutionalResearch, Atlanta, GA, April 27-May 1. ED 189 956

This study focuses on differences in collegedropout rates between loan users and users of otherforms of aid. The units of analysis are: loans only,loans in combination with other aid, other aid only,and aid from families only. Data were drawn fromthe 1ILS-72 base year though 1974 surveys. Find-ings show dropout rates to be higher in the first yearof college than in the following 2 years. Studentsrelying on loak. s only or family aid only had higherdropout rates than students using loans in combina-tion with other types of aid. Low SES students hadhigher dropout rates when relying exclusively onloans or on family aid than did high SES students.(Adapted from ERIC abstract)

82Hansen, W. Lee. 1982. Economic Growth and EqualOpportunity: Conflicting or Complementary Goalsin Higher Education. Washington, DC: NationalInstitute of Education. ED 227 761

37

This study examines the effects of studentfinancial aid on changing the composition of thepopulation attending or planning to attend college.A discussion of policies from the 1960s to the early1980s and their effects is included. Data from theNLS-72 as well as the High School and Beyond andthe Current Population Survey are used to determinethe effects of aid on enrollment rates.

Results of the study failed to show that theincrease in available aid from 1972 to 1980 changedthe plans of college-going high school seniors.While increases in enrollment rates occurred forblacks and women, it is difficult to attribute thesechanges to increases in aid. The author feels that itis hard to explain why enrollment and plans to enrolldid not change as expected considering an increasein student financial aid. He suggests that aid simplyacted as a means to transfer funds, therefore reduc-ing the burden for parents and students and at thesame time failing to increase or change the mix ofcurrent enrollment.

83Leslie, Larry. 1984. "Changing Patterns in StudentFinancing of Higher Education." Journal of HigherEducatioua(3): 313-346.

This study examines trends in student financ-ing of higher education. Specifically, the authorlooks at parent contribution, tuition price changes,students' ability to finance their education, changesin debt reliance, the role of grants and scholarshipsin student financing, and the relative benefits tovarious groups of students.

The author uses data from the NLS-72 baseyear through 1976 surveys, and the CooperativeInstitutional Research Program (CIRP) surveys toinvestigate any changes occurring over time inamount financed by students enrolled in higher edu-cation and in sources of student funds. Differencesamong various student groups (e.g., racial-ethnicgroups, part-time and full-time, male and female) inrelationship to type of institution attended are alsoexamined.

Major findings from the NLS-72 data include:support from families decreases after students'

freshman year; part-time students contribute a largerportion of their personal finances to cover educa-tional expenses and receive little aid; including fam-ily support, over 80 percent of student financing isfrom private sources; overall student access has beenenhanced by lower net prices; students who enroll invocational-technical programs in high school tend topay a higher proportion of their net college costscompared to those who were in a general academictrack; Whites and Asian-Americans self-finance thegreatest amount while other minorities receive morestudent aid; student aid tends to favor lower abilitystudents; and equity improvements have been pri-marily between low-income and middle-incomefamilies in regard to net price.

33

84St. John, Edward P. 1989. "The Influence of StudentAid on Persistence." Journal of Student FinancialAid19.(3): 52-68.

This study looks at the influence of student aid,specifically of the different types of aid packages, onyear-to-year persistence. The model used looks atpersistence in terms of social background, academicability and achievement, high school and collegeexperience, and financial aid types. The NLS-72seniors (n=4,485; base year through 1979 follow-upsurveys) and the HS&B 1980 (n=4,178; base yearthrough third follow-up surveys) and 1982(n=3,965; base year through third follow-up sur-veys) cohorts who attended college are included inthe analysis.

Findings indicate that the impact of loans onpersistence has changed over time. For example,loans had negative effects in the 1970s for the first-to-second year transition and positive effects in the1980s. In most models, attending college full-time,grades received in college, and beginning at a 4-yearschool as a freshman were positive and significantpredictors of persistence. Postsecondary plans werealso significant in all models studied. College expe-riences had an influence as well although there weresome differences across years and cohorts.

85St. John, Edward P., and Masten, Charles L. 1990."Return on the Federal Investment in Student Finan-

38

cial Aid: An Assessment for the High School Classof 1972." Journal of Student Financial Aid 24(3):4-23.

This study develops a model for estimating thereturns from gains in educational attainment that areattributable to student financial aid. The authorsview financial aid as an investment in human capitaland the returns examined here are expressed ascash-flow ratios (the amount of money spent by thefederal government on a certain cohort divided bythe tax revenue on that investment for the samegroup). Data used in the study were taken from a1983 Census report on lifetime earnings and fromthe NLS-72 base year through 1986 follow-up sur-veys.

Findings indicate that financial aid was a prof-itable investment during the 1970s. The net presentvalue of each federal aid dollar in the 1970s is $4.30.Under every set of economic assumption3 (worstcase/optimistic), investment in aid has a substantialrate of return.

86St. John, Edward P., and Noe 11, Jay. 1989. "TheEffects of Student Financial Aid on Access toHigher Education: An Analysis of Progress withSpecial Consideration of Minority Enrollment."Research in Higher Education1Q(6): 563-581.

For description see entry #66.

See Access

87Sanford, Timothy R. 1980. "The Effects of StudentAid on Recent College Graduates." Research inHigher Edu mtion .12(3): 227-24Also see Sanford,Timothy R. 1980. "The Impact of Student Loans onGraduate Education." Paper presented at the AnnualForum of the Association for Institutional Research,Atlanta, GA, April 27-May 1. ED 189 936

The study examines the impact of undergradu-ate financial aid on subsequent life choices for asample of 3,136 NLS-72 respondents who receiveda BA degree by October 1976. Effects of the typeand amount of loan received on attending graduate

or professional school, choosing a job, forming afamily and developing values were explored.

34

Results show that graduates with loans wereslightly more likely to go to graduate or professionalschool regardless of loan amount. Contrary to expec-tations that loans would have negative impacts onsubsequ,:nt life choices, students who received loanswere similar to other graduates. Overall, aid in gen-eral did not have negative effects but the authorcautions that the sample examined in the study is ofrecent college graduates and perhaps the long-termimpact of aid should be examined in the future.

88Terkla, Dawn Geronimo. 1984. "Does Financial AidEnhance Undergraduate Persistence?" Paper pre-sented at a joint meeting of the American Educa-tional Research Association Division J and theAssociation for the Study of Higher Education, SanFrancisco, CA, October 28-30. ED 253 127

The purpose of this study is to explore theinfluence of financial assistance on persistence inhightr education. Financial assistance is defmed aswork study, grants or loans. Dropout from highereducation is defined as any student who entered aprogram at a 2- or 4-year school by October 1972,who had not obtained a BA or AA degree by May1979 and who was no longer enrolled by May 1979.The data from this study are drawn from a sample of4,838 respondents to the base year through fourthfollow-up surveys of the NLS-72.

Findings show that SE3, race, and sex had nodirect influence on persistence. Only high schoolGPA and educational aspirations had aronger directeffects than did aid. Financial aid had the thirdstrongest direct effect and the fifth strongest totaleffect on persistence. When controlling for othervariables, slightly more than half the students receiv-ing aid were likely to complete their degree.

See Persistence

89Wagenaar, Theodore C. 1987. Changes in

*I I . I : I

ED 284 481

3

Using data from NLS-72 and High School andBeyond 1980 cohort, this study examines changesin postsecondary education choices, in college ap-plication rates, in financial aid application andaward rates, as well as changes in college attendancerates between 1972 and 1980. Comparisons aremade by institution type, first, second, and thirdchoice institution as well as by various backgroundvariables (including sex, region, and race).

Findings regarding changes in applicationrates, choice of institution, institutional charac-teristics, factors related to college choice, accep-tance rates, financial aid, percentage of those whoreceived grant, loans, or work study, and changes inattendance are included in the report.

Findings related to application rates indicattan increase in the percentage of seniors applying topostsecondary institutions, in the number of womenapplying, and a significant rise in the applicationrates among African-Americans. Changes in type ofinstitution selected as a first choice school as wellas in the quality of institutions selected were negli-gible between the 2 years. Financial aid, however,increased in importance as a factor in selecting acollege. The majority of students in both years at-tended their first choice institution.

Regarding financial aid, more students appliedfor aid in 1980 than in 1972. Students were offeredmultiple sources of aid, particularly in the form ofgrant and loan combinations with grants and loansaccounting for over two-fifths of total aid offered inboth years at first choice institutions.

See Patterns of Attendance

4. Patterns of Attendance

90Adelman, Clifford. 1988. "Transfer Rates and theGoing Mythologies." Chan= (anuaty/February):38-41.

This article examines attendance patterns andcredentialing at 2-year colleges and the role of time

35

in the educational experiences of students enrolledat 2-year colleges. Student background and changesin students' aspirations are also examined. Datafrom the NLS-72 and Postsecondary EducationTranscript Study were used for the analysis.

Eight major fmdings from the article include:

(1) One out of five individuals who attend2-year colleges eventually receives an Associate'sDegree from a 2-year college.

(2) One out of three individuals who attead2-year vocational or technical schools eventuallyreceives either an Associate's Degree or certifi-cate/liceLse. (The credentialing rate from this typeof instivition is higher than for all 2-year institutions,but the type and mix of credentials are different.)

(3) One out of five individuals who attended2-year colleges eventually attends a 4-year college,irrespective of whether a degree is earned at eithertype of institution.

(4) The vast majority of those who attend 2-year colleges are "lockstep" students (i.e., enteringwithin a year of high school graduation).

(5) Slightly over half of those who earn Asso-ciate's Degrees do so "on time" (within 29 monthsof entering an Associate's Degree program).

(6) Four-year colleges award nearly one out ofevery five Associate's Degrees, yet their "de factotransfer rate" (from the lower to the upper division)is no better than that of public 2-year and privatejunior colleges.

(7) One-fourth of all students who attend 2-year institutions earn less than one-semester's worthof credits.

(8) Less than 2 percent of 2-year college atten-dces enrolled after they received the Bachelor'sDegree.

See AttainmentEducational and Occupational

91Burkheimer, Graham J., and Jaffe, Jay. 1981."Highly Able Students Who Did Not Go ToCollege." Washington, DC: U.S. Department ofEducation, National Center for Education Statistics.ED 212 682

This paper uses NLS-72 data to explore thefollowing three questions: (1) What are the reasonsthat highly able students do not attend college?, (2)Are factors related to college attendance among ablestudents the same as those for students of otherability '.evels?, and (3) What happens to highly ablestudents who do not go to college in terms of earlycareer attainment and family life? The NLS-72sample consists of respondents who completed thebase year through third follow-up surveys and in-cludes 4,052 students of high ability, 7,000 of aver-age ability and 4,788 of low ability.

Results indicate similarities among college-going students of all ability levels. They are gener-ally characterized as not married by the secondOctober after graduating from high school, havinghigher average SES levels, higher high school classranks, more math and science courses, and highereducational expectations. Both groups, college andnoncollege going, worked the same number of hoursand earned roughly the same salary. Few tembersof the noncollege-going group felt they hadachieved their career goals and less of them hadentered professional ortechnical careers. Bothgroups, however, reported only mild satisfactionwith their jobs.

See AttainmentEducational and Occupational

92Cervero, Ronald M., and Kirkpatrick, Thomas E.1990. "The Enduring Effects of Family Role andSchooling on Participation in Adult Education."American Journal of Education 22(1): 77-93.

The study explores reasons why adults partici-pate in adult education, specifically examining thosepre-adult factors identified in the theoretical litera-ture as well as effects of prior educational attain-ment. The authors also look for any influences thesefactors have on participation in credit and noncredit

forms of adult education. Data are drawn from thebase year through 1986 follow-up sti...v.;ys of theNLS-72. Pre-adult factors included father's educa-tional attainment, amount of high school and/or col-lege, type of high school program, class rank in highschool, and educational aspirations and attainment.

36

Findings strongly support the "proposition thatparticipation in adult education is the outcome of aset of processes occurring throughout the life span."The authors suggest that future research should fo-cus on identifying other factors, such as barriers toparticipation, as well as on the pre-adult factorsincluded in this study. They also suggest that futureresearch should "proceed on the assumption thatdifferent types of adult education may have differentantecedents."

See Marriage/Fami:y

93Eagle, Eva, and Carroll, C. Dennis. 1988.Postsecondary Enrollment. Persistence, and Attain-ment for 192212E0. _ans11932±liglISchaskaatilk.ates. High School and Beyond _NationalLongitudinal Study. Washington, DC: U.S.Department of Education, National Center for Edu-cation Statistics. ED 304 036

The report t. ,es data from the NLS-72 and the1980 and 1982 high school seniors of the HighSchool and Beyond Study as well as the Postsecon-dary Transcript Study (1984), to compare patternsof enrollment, persistence, and attainment amongthese cohorts for several years following highschool. Patterns were analyzed in terms of sex, raceand socioeconomic status.

Findings on enrollment indicate that:

(1) Rates of enrollment in postsecondary edu-cation for high school graduates increased betweenthe 1972 cohort and the 1980 cohort, but dt.clinedfor the 1982 cohort.

(2) For all three cohorts studied, enrollmentwas higher for whites than for blacks or Hispanics,and there was a positive relationship between enroll-ment and socioeconomic status.

41

(3) The proportion of students who first en-tered postsecondary education during the secondyear after high school graduation was higher for the1980 and 1982 cohorts than for the 1972 cohort.There were new enrollees each year after graduationin each cohort with a higher enrollment for the 1980cohort, regardless of sex, race-ethnicity or socioeco-nomic status.

(4) Rates of persistence during the first year ofschool did not vary among cohorts for students whoenrolled immediately in postsecondary education.

(5) The rate of persistence through 4 years ofpostsecondary education was lower in each succes-sive cohort for blacks. Among whites and Hispanics,fewer 1982 or 1980 graduates persisted through 4full years than 1972 graduates, although there wasno significant decline between the last two cohorts.

(6) Rates of attainment for the 1982 and 1972cohorts were about the same, except that the 1982cohorts had higher attainment rates among blacks,Hispanics, and low SES students.

See AttainmentEducational and Occupational

94Eagle, Eva, and Schmitt, Carl. 1990. Patterns andTrends of Stopping Out from Postsecondary Educa-tion: 1972. 1980 and 1982 High School Graduates,Ell_TABI_NatioaaLLangitadinaL.Stud,y_1222.,HighSghs2a.and_augad. Washington, DC: U.S.Department of Education, National Center for Edu-cation Statistics. ED 314 509

Data from all NLS-72 surveys and the Postsec-ondary Education Transcript file and High Schooland Beyond 1980 senior and sophomore cohortsbase year through third follow-up surveys are usedto examine patterns of stopping out and dropping outfrom postsecondary education. The analysis focuseson time of entry, educational outcomes, and selectedstudent characteristics.

Major findings include:

(1) Students who delayed entry into Nstsecon-dary education were more likely to drop out thanthose who entered immediately after high schoolgraduation.

37

(2) The drop out rate was higher at 2-yearinstitutions than at either 4-year, or less-than-2-yearinstitutions.

(3) Stopping out was more common at 2-yearand 4-year institutions than at less-than-2-year insti-tutions.

(4) Males and females generally stopped out ofpostsecondary education at the same rate.

95Eagle, Eva, and Schmitt, Carl. 1990. Patterns and

-IS b.. -I s essa 1

tion: 1972. 180. and 1982 High School Graduates,I I t I

HighSchooLiand_Beyond. Washington, DC: U.S.Department of Education, National Center for Edu-cation Statistics. ED 314 508

Data from the NLS-72 base year through fifthfollow-up surveys and Postsecondary EducationTranscript File, and High School and Beyond 1980senior and sophomore cohorts base year throughthird follow-up surveys, are used to describe patternsand trends of delayed entry into postsecondary edu-cation. Differences in program type, type of institu-tion, and time of entry into postsecondary educationare examined. Information on enrollment and meanlength of delay is included in several tables.

Major findings include:

(1) Students entering 2-year or 4-year institu-tions delayed entry less than those who enteredinstitutions offering less than 2-year programs.

(2) A higher proportion of students who en-tered 4-year institutions entered in the fall term

4 2

immediately after high school graduation, as com-pared to students in other types of institutions.

(3) Compared to the 1972 cohort, a higherproportion of the postsecondary students from the1980 and 1982 graduating classes enrolled in insti-tutions offering programs of less than 2 years.

96Eagle, Eva, and Schmitt, Carl. 1990. Patterns andConsequences of Delay in PostsecondaryEducation: Degree Attainment for 1972. 1980. and1982 High School Graduates, Survey Report. Na-tional Longitudinal Study 1972. High School andBeyond. Washington, DC: U.S. Department ofEducation, National Center for Education Statistics.ED 314 507

This report, using data from the NLS-72 baseyear through fifth follow-up surveys and Postsecon-dary Education Transcript File, and High Schooland Beyond 1980 senior and sophomore cohortsbase year through third follow-up surveys, describespatterns of delay in entering 4-year, 2-year, andless-than 2-year postsecondary institutions for 1972,1980 and 1982 high school graduates. The reportdoes not compare cohorts (findings have not beenstandardized to account for differences in the oppor-tunity to enter postsecondary education) but com-pares students within each cohort by institution type,SES and gender.

The report examines -two typcs of enrollmentpatterns: delayed entry (defined as enrollment laterthan October of the high school graduation year),and stopping out (defined as the practice of leavingschool for more than 2 months or a summer periodand returning to school after that break). The reportexamines delay from an institutional perspective,which classifies students' enrollment with respect tothe institution type as well as their time of entry orthe continuity of their enrollment.

Findings reveal diverse enrollment patterns ofstudents from the three cohort groups.

(1) Over 40 percent of high school graduatesfrom these cohorts enrolled in 4-year institutions by

38

1986, 30 percent enrolled in 2-year institutions and10 percent in less-than- 2-year institutions.

(2) A significant number of students enrolledin postsecondary education later than the fall of theyear they graduated from high school and the ratesof delayed entry were related to institution type.Students seemed more willing to delay beginningshorter postsecondary programs than longer pro-grams.

(3) Fewer students delayed by stopping outthan by entering late. The stopout rate was alsorelated to institution type with the lowest stopout ratefound at less-than-2-year institutions. Stopout ratesfor the 1980 and 1982 cohorts were higher for stu-dents at 4-year than at 2-year institutions, while thestopout rate for the 1972 cohort was higher at 2-yearthan at 4-year institutions.

(4) Students who delayed entering or whostopped out from each type of postsecondary insti-tution were significantly less likely to earn a degreefrom such an institution or, in the case of less-than-4-year institutions, to transfer to 4-year institutions.

(5) Attainment rates were lower for stopoutsthan for delayed entrants.

(6) Students from high SES backgrounds wereless likely to delay entry into 4-year institutions thanwere other students. A higher proportion of low SESstudents in 4-year institutions delayed entry than dideither medium or high SES students.

(7) There is no relationship between genderand the likelihood of delay among students at anypostsecondary institution.

97Eagle, Eva, and Schmitt, Carl. 1990. Consequencesof Delay in Postsecondary Education: Degree At-.tainmextia.1.222-1212._and_121i2aligh_SsimolGraduates . E.D . TABS . National Longitudinal

. Washington,DC: U.S. Department of Education, National Cen-ter for Education Statistics. ED 314 506

leo .54 : es'

4 3

Data from the NLS-72 base year through fifthfollow-up surveys and Postsecondary EducationTranscript File, and High School and Beyond 1980senior and sophomore cohorts base year throughthird follow-up surveys, are used in this E.D. Tabu-lation to present data on the effects of delayed entryinto postsecondary education on degree attainment.Patterns of delay are compared for the three cohortgroups by pattern of enrollment, type of institution,and selected student characteristics.

Findings show:

(1) Delays in beginning postsecondary educa-tion lower the rate of degree attainment.

(2) Students who begin postsecondary educa-tion immediately after graduating from high schoolare more likely to complete a postsecondary degreethan those who delay entry.

(3) Students at 4-year institutions who delayedinitial enrollment were more likely to attain theirdegree than those who delayed enrollment at othertypes of institutions.

98Grubb, W. Norton. 1991. "The Decline of Commu-nity College Transfer Rates. Evidence from Na-tional Longitudinal Surveys." Journal of HigherEducation f2(2): 194-216.

This study uses the NLS-72 and HS&B sur-veys and PETS files to examine transfer rates ofcommunity college students. The first section looksat overall trends in transfer rates. Section II exam-ines rates in terms of gender, race-ethnicity, familybackground, high school performance and postsec-ondary education aspirations. The third section in-cludes information on transfer rates of BAcompleters. The final section presents implicationsof the results for programs seeking to enhance trans-fer. Possible causes of changes in transfer rates arealso included.

General findings show a decline in transferrates between 1972 and 1980. The author suggeststhat the rise of community college vocational pro-

39

grams may be the reason behind some of the decline.Changes in transfer rates did not appear to be relatedto changes in the composition of students in termsof their racial-ethnic background, SES, ability, aspi-rations, and high school track. Findings also indicatea decline in the rates of BA completion within 4years after high school between 1972 and 1980. Thelikelihood of completing the BA, however, increasesover time. The author concludes that declining trans-fer rates may be due to many small factors includingchanging demographics, decline in achievementduring high school, poor career counseling in highschool, the "apparent weakening of academic Asso-ciate degree programs as routes to transfer, an in-crease in 'milling around' in all of postsecondaryeducation, and declining federal aid." He also sug-gests that recent innovations, such as articulationagreements, may enhance transfer rates.

See Persistence

99Hilton, Thomas L., and Schrader, William B. 1987.Pathways to Graduate School: An Empinical Study

Princeton, NJ.GRE Board Research Report No. 82-21R/ETS Re-search Report 87-41.

-I SI \ SI .5" I I . . .

This study explores the pathways students takefrom high school to graduate and professionalschool. Differences in characteristics between thosewho went on to graduate school and those who wentto professional school are compared. All NLS-72respondents who earned a bachelor's degree by fall1979 and were enrolled as full-time graduates for atleast one semester between fall of 1975 and 1979 areincluded in the sample. Respondents are examinedat five key points in the educational process: as highschool seniors, as new college students, 2 years aftercollege entrance, as bachelor' s degree recipients,and as enrollees in graduate or professional school.The authors examined the impact of such factors ashigh school curriculum, type of postsecondary insti-tution attended, being "on track" toward a degree 2years after high school, the field in which the BAdegree was earned, and attendance in graduateschool. Gender and ethnic differences were alsoexamined. An analysis of prospective graduate stu-

4 4

dents in the 1980s compares HS&B 1980 seniorswith 1972 seniors in regard to numbers of graduatestudentr, ability, and educational aspirations.

Overall findings indicate slight differences be-tween patterns and choices of students enteringgraduate school and those enrolling in professionalschool. Both groups took a more academic curricu-lum in high school. Women and men followed simi-lar paths to graduate school although fewer womenenrolled in professional school. The authors foundsignificant relationships between SES and educa-tional attainment, as well as bctween type of collegefirst entered and both earn a BA and enroll in gradu-ate or professional schools. They believe it is possi-ble to predict graduate or professional enrollment asearly as high school using their model althoughmeasures taken after high school graduation aremore highly related to enrollment in graduate orprofessional schools.

See AttainmentEducational and Occupational

100Jackson, Gregory A. 1986. "MISAA (the MiddleIncome Student Assistance Act), The Fall of Saigon,and College Choice, 1972 to 1980." ASHE 1986Annual Meeting Paper. Washington, DC: U.S. De-partment of Education, National Institute of Educa-tion. ED 268 867 (See also ED 275 224)

This report compares changes in high schoolgraduates' college choices between 1972 and 1980.Specific questions examined are (1) How did theimportance of different influences on college choicechange over time?; and (2) How did these changesinteract to affect college entry rates? Data focus oncollege and university enrollment 15 months follow-ing high school graduation (fall 1973 for the NLS-72seniors and 1981 for the HS&B 1980 seniors).

Overall, the rate of college going was stablebetween 1972 and 1980. College choice processesalso remained steady over time. When majorchanges in participation did occur, and only forsubsets of the overall student population, theytended to result from policy or social changes, suchas Pell Grants, BEOGs, and federal programs aimed

40

at increasing college participation among underrep-resented groups.

101Jackson, Gregory A. 1986. Workable, Comprehen-sive Model of College Choice. FinaLand Technical&put Cambridge, MA: Harvard University. ED275 224 (See also ED 268 867)

This study, similar to entry #100, analyzedNLS-72 and High School and Beyond to assesschanges in college choice patterns over time. Theauthor specifically looked at relationships betweenstudent attributes and college choice.

Main findings show a change in SES variables(an increase in parent education and income),changes in academic variables (including a decreasein the number of students in academic programs) andan increase in the number of students receiving aid.Even with those change patterns, college choiceremained stable between 1972 and 1980. Collegeparticipation rates also remained essentially un-changed during this time.

102Robertshaw, Dianne, and Wolfle Lee M. 1980. "Dis-continuities in Schooling and Educational Attain-ment." Paper presented at the Annual Meeting ofthe American Educational Research AssociationBoston, MA, April 7-11. ED 186 488

Based upon prior research on status attain-ment, the authors examine the differences betweenthose who delayed and those who interrupted post-secondary education 4 1/2 years after high school. Asample of 3,300 white male respondents of the NLS-72 is used in the investigation.

Findings show that students who were 20 yearsold or older when they graduated from high schoolin 1972 were less likely to go or to postsecondaryeducation than those who graduated "on time." Re-sults for those who actually attended postsecondaryinstitutions show that levels of ability and gradesaffect both entrance into postsecondary education

4 5

and amount of education. Also, higher educationaland occupational aspirations have a positive effecton education. Findings also show differences ineducational attainment based on type of institutionattended. In general, those who attain more educa-tion are those who enter college without delay andremain there through graduation.

See AttainmentEducational and Occupational

103Robertshaw, Dianne, and Wolf le Lee M. 1983. "Dis-continuities in Schooling and Educational Attain-ment." Higher Education .12: 1-18.

This study expands the authors' previous re-search (see entry #102) and includes black males,and white and black females in the sample studied.Results for the additional groups were similar tothose for white males. Factors causing educationaldiscontinuities were the same for white males andfemales, although the effects of these factors oneducational attainment were twice as great forwomen. Social background and father's occupationaffected educational progress for black males but didnot affect white males. The effect of delaying thestart of postsecondary education had a greater influ-ence on blacks than on whites while interruptingeducation influenced whites more.

See AttainmentEducational and Occupational

104Thomas, Gail E. 1981. "Choosing a College Majorin the Hard Technical Sciences and the Professions:A Causal Explanation." Washington, DC: U.S. De-partment of Education, National Institute of Educa-tion. ED 206 829

A causal framework was used to explore fac-tors that affect the enrollment of 4-year collegestudents in the hard technical sciences (math, sci-ence, engineering) and the professions (pre-law,medicine, dentistry). The factors studied were fam-ily status, standardized test performance, highschool rank, educational expectations, high schoolmath and science preparation, college charac-

teristics, and college grades. The NLS-72 was thedatabase used in a path analytic approach to evaluatedeterminants of college major field choice. Findingsshowed that sex and high school major field inten-tions were the two most important determinants.Being female exerted a large direct negative effecton majoring in the hard and technical sciences andthe professions. The intention to major in these fieldsat the high school level was positively related toactually pursuing a college major in these fields.Findings from past studies regarding the importanceof standardized test performance and high schoolmath and science preparation were not strongly sup-ported by present data. (ERIC abstract)

41

105Wagenaar, Theodore C. 1987. Changes in Postsec-ondaiyEducational Choices: 1972 to 1980. Wash-ington, DC: U.S. Department of Education, NationalCenter for Education Statistics. ED 284 481

For description see entry #89.

See Financial Aid

5. Persistence

106Anderson, Kristine L. 1986. "College Contexts, Stu-dent Integration, and Educational Attainment.ASHE 1986 Annual Meeting Paper." Paper pre-sented at the Annual Meeting of the Association forthe Study of Higher Education, San Antonio, TX,February 20-23. ED 268 874

For description see entry #68.

See College/Program Effects

107Grubb, W. Norton. 1991. "The Decline of Commu-nity College Transfer Rates. Evidence from NationalLongitudinal Surveys." Jounnal of Highet Educa-1101.62(2): 194-216.

For description see entry #98.

4.

See Patterns of' Attendance

108Hilton, Thomas L., and Lee, Valerie E. 1988. "Stu-dent Interest and Persistence in Science: Changes inthe Educational Pipeline in the Last Decade." Jour-nal of Higher Feiucatioa 52(5): 510-26.

The authors make predictions about "who willdo science" at the end of this decade by incorporatingboth cross-sectional and longitudinal data to outlinepatterns of persistence in science and math education(MSE).

Cross-sectionally comparing two longitudinalstudiesNLS-72 and High School and Beyond(starting with data from 1982 HS&B cohort so thetwo groups would be at comparable educationallevels)the authors assessed the proportion of stu-dents expressing an interest in majoring in MSE atsix transitional points along the educational pipeline,the net loss at each of these points (movement intoand out of the pipeline), and any differences ininterest in science between the two cohort groups.

Results indicate a small net loss of MSE majorsfor the total population and all subgroups, except forblacks from 1980 to 1982. The greatest loss occurredduring the transition from high school to college,"reflecting both loss to science and loss to highereducation." Losses occurred less frequently oncestudents enrolled in college as a result of high per-sistence in MSE, not as a result of additional num-bers of students in MSE.

For the 1982 cohort, the proportion of the totalsample moving into MSE between the sophomoreand senior years was virtually identical to the pro-portion moving out of MSE. The greatest loss for the1972 cohort occurred in the transition from highschool to college (data for sophomore to senior yearis not available in NLS-72). Data from the 72 cohortsample was used to look at "leaks" occurring at thepostsecondary level (postsecondary data were not

42

yet available for the 82 cohort). Loss from first yearto junior year and from third year to graduationoccurred due primarily to students dropping out ofcollege rather than to switching out of an MSEmajor. The second greatest loss occurred betweenundergraduate school and full-time graduate school.The authors conclude that the future MSE popula-tion will continue to be primarily white and male,but with a substantial growth in the number of mi -

norities and women.

109Hilton, Thomas L. 1982. Persistence in HigherEdu-cation. College Board Report No. 82-5. New York,NY: College Entrance Examination Board; Prince-ton, NJ: Educational Testing Service. ED 227 737

The author uses NLS-72 "Base Year," "First"and "Second Follow-Up" surveys to examine atten-dance patterns of persisters and dropouts from col-lege and vocational-technical schools as well as totest a model of persistence. The basic hypothesis ofthe study was that persistence (defmed here as therelationship of educational starus and educationalaspirations over time) in higher education resultsfrom the impact of a "chain of career decisions bythe individual student."

Findings show that 26 percent of high schoolseniors continued their education in either 2- or4-year colleges without interruption from 1972-1974. A slightly higher percentage of white studentsthan blacks continued after high school although,among the group that continued, the drop out rate didnot differ significantly between black and whitestudents or male and female students.

Findings of the causal model indicate that re-tention in higher education is influenced primarilyby prior educational status, which was influenced, inturn, by high school curriculum. Persistence wasalso influenced by academic performance. Unex-pectedly, variables such as race, sex, SES and num-ber of hours worked did not affect persistence whencontrolling for other causal variables.

4 -7

110Munro, Barbara Hazard. 1981. "Dropouts fromHigher Education: Path Analysis of a National Sam-ple." American Education Research Journal .1.a(2):133-141.

Using a sample of 6,018 NLS-72 respondentswho entered 4-year colleges as full-time students in1972, the author tests Tinto' s model of student per-sistence. The model looks at the impact of back-ground characteristics and the levels of social andacademic integration (extent of student involvementin the social and academic communities at the insti-tution) on student persistence.

Results indicate that the effects of ethnicity andSES were indirect. Overall, goal commitment had

the greatest effect on student persistence. Both stu-dent and parent educational aspirations had strongereffects on goal commitment than academic integra-tion. Academic integration had stronger effects oninstitutional commitment than did social integration.

111Terkla, Dawn Geronimo. 1984. "Does Financial AidEnhance Undergraduate Persistence?" Paper pre-sented at a joint meeting of the American Educa-tional Research Association Division J and theAssociation for the Study of Higher Education, SanFrancisco CA, October 28-30. ED 253 127

For description see entry #88.

43

See Financial Aid

4 b

VI. AspirationsEducational and Occupational

112Brown, Kenneth G. 1982. "Postsecondary Plans ofHigh-School Seniors in 1972 and 1980. Implica-tions for Student Quality. AIR Forum." Paperpresented at the Annual Forum of the Associationfor Institutional Research, Denver, CO, May 16-19.ED 220 060

This study compares the postsecondary educa-tional aspirations of black and white high schoolseniors in 1972 and 1980 by using NLS-72 and HighSchool and Beyond data. The study looked specifi-cally at plans to attend vocational school, vocationalprograms at a community college, academic pro-grams at community colleges and 4-year colleges oruniversities. These plans were examined in terms ofaptitude levels (lower, middle, and upper).

Findings show more students at both high andlow aptitude levels planning to go to 4-year schoolsin 1980 than in 1972, while a smaller percentageplanned to attend vocational schools. Black studentsshowed an increase in their plans to attend commu-nity colleges.

Findings regarding aptitude show an increasein the percentage of blacks in the top and middleaptitude levels and a decrease in the lowest level.The author warns, however, that factors such asnonresponse rates, and declines in ACT and SATscores, make it difficult to reach definitive conclu-sions on aptitude levels based on the above-men-tioned result alone.

113Falkowski, Carolyn K., and Falk, William W. 1983."Homemaking as an Occupational Plan: Evidencefrom a National Longitudinal Study." Journal ofVocational Behavior 22: 227-242.

This study compares occupational choices,background, and social-psychological variables ofthose who expect to become homemakers and ofthose who do not. In general, the authors expect thatthose who choose homemaking have fathers withlower occupational status, nonemployed mothers,

rural residence, are in the nonacademic track, andhave lower scholastic performance. The sample con-sists of 463 black and 4,227 white women whocompleted both the "Base Year" and "First Follow-Up" of the NLS-72.

Results show a higher number of white womenexpect to become homemakers. Differences be-tween blacks and whites were also found in locus ofcontrol (blacks perceived more external locus ofcontrol), social psychological variables (blackwomen had higher self-esteem) and life importancecategories (large differences on money and workitems). Race aside, the authors' expectations werecorrect in that women from rural areas whose fathershad lower occupational status, whose mothers neverworked outside the home, and who were enrolled innonacademic tracks with low scholastic perform-ance were more likely to become homemakers.

114Holsinger, Donald B., and Chapman, David W.1984. "Students' Occupational Aspirations andChoice of College Type " College Student Journal111: 87-93.

Changes in career aspirations during collegefor men, women, Mexican-American and Anglostudents and the differences in the changes for thoseat 2-year and 4-year colleges were examined usingthe base year and first through third follow-up sur-veys of the NLS-72. A general decline in occupa-tional aspirations during college occurs although thedecline was greatest for Mexican-American womenand Anglo men. Students attending 2-year collegesexperienced a much greater decline in occupationalaspirations than did 4-year college students. Find-ings raise implications for advising and counselingat high schools and at the 2-year college level asinformation needs appear to differ by student back-ground and across institutional type.

115Lindsay, Paul, and Knox, William E. 1984. "Conti-nuity and Change in Work Values Among Young

45 4J

Adults: A Longitudinal Study." American Journalof Sociology 22(4): 918-931.

The authors develop a model to examine therelationships among intrinsic and extrinsic workvalues, education and work. They expand previouswork on this topic by incorporating four additionalfactors: (1) how already existing values influencechoice, selection, and participation in education; (2)whether educational experiences have socializingeffects on values; (3) how education may influenceoccupational placement; (4) how social status attrib-utes may be related to values and work. The stabilityof values over time is also assessed. The studysample includes 9,208 respondents to the NLS-72who reported a work history during the 12 monthsprior to the 1979 survey.

Findings indicate the complexity of the occu-pational selec tion process. Educational attainmentand prior existing values exhibit influence on futurejob characteristics while SES is a strong predictor ofeducational attainment. Findings also lend somesupport for the idea that values are influenced by thesocializing effects of work experience (occupa-tional-socialization hypothesis). Evidence stronglysupporting the value-stability hypothesis (work val-ues remain stable over time), was also found. Genderwas found to be a major influence as it related toinitial work values, educational attainment and moststrongly to income.

116Ohlendorf, George W., and Rafferty Donna M.1982. The Educational and Occupational Aspira-tions of Louisiana Rural High School Students: AComparative Study. Baton Rouge, LA: LouisianaState University, Agricultural Experiment Station.ED 218 020

The study uses samples of students from theSouthern Youth Study (n=453) and the NLS-72(n=4,738) to gain greater understanding of the edu-cational and occupational aspirations of rural stu-dents in Louisiana. The cohorts studied allow forcomparisons on regional and national levels.

Results show a decrease in educational aspira-tions of Louisiana students between 1968 and 1972.Educational aspirations of the 1972 cohort were alsolower than those of both the southern region and theU.S. The authors believe that higher occupationalaspirations of the rural students indicate that theymay have been related to employment opportunitiesin Louisiana at the time. Differences in aspirationover time may also be attributed to the social activ-ism of the 1960s and 1970s when students oftenpreferred alternative goals.

46

117Smith, Michael Francis. 1982. A Comparison ofPostsecondary Educational Plans of Black andWhite High School Students in 1972 and 1980.Doctoral Dissertation, University of Arizona, Publi-cation No.: AAC8217473

The educational plans of black high schoolseniors over the past decade were examined andcompared to the plans of white students by analyzingselected base year data from the NLS-72 and HS &B .

Expectations to participate in postsecondaryeducation were essentially the same for all black andwhite seniors between 1972 and 1980. Amongwhites, womens' expectations exceeded mens' in1980. When levels of educational expectation wereexamined, significant differtaces were apparent forboth races. Large declines occurred in the 4-yearcollege category though they were partially offset byincreases in the graduate school category. Smallincreases also occurred among both blacks andwhites in the community college category. No cleartrends were seen in the vocational school category.Black expectations were greater than those of whitesin the combined upper two categories with the moststriking difference seen among low-aptitude andlow-SES students.

Immediate plans for 2- or 4-year college atten-dance changed significantly for both blacks andwhites with more students indicating 4-year prefer-ence in 1980. Overall, more blacks than whitesplanned 4-year attendance in both survey years, withthe largest differences between races found on thelow- and middle-aptitude levels.

Preferences for planned field of study alsochanged for both blacks and whites between surveyyears. The largest increase for both races occurredin the business category, with the next largest expan-sion appearing in engineering. The largest decreasesoccurred in education and the social sciences, withthe downturn being greater among blacks thanwhites. Black and white seniors in the middle, andupper-aptitude levels were more likely to select thebiological and physical sciences, engineering, andsocial sciences while low-aptitude seniors weremore likely to choose vocational studies, health-re-lated studies, and education. (Dissertation abstract)

118Smith, Wayne E. 1982. "Factors related to the Per-formance of 2-Year College Transfer Students. AIRForum 1982 Paper." Paper presented at the AnnualForum of the Association for Institutional Research,Denver, CO, May 16-19. ED 220 052

In light of the predicted decline in the college-going population, this study examines and comparespostsecondary education plans of 1972 and 1980black and white high school seniors. Plans are ex-amined in terms of school categories (vocationalschools, community college vocational and aca-demic programs and 4-year college or university)and aptitude levels (lower quartile, middle two quar-tiles, and upper quartile). The three samples used inthe study are (1) all students who answered base yearsurveys of the NLS-72 and HS&B students whoresponded to postsecondary planning questions, (2)a subset of all black students included in set 1, and(3) a subset of all white students included in set 1.

47

Major findings include a larger percentage ofmore able students planning to attend 4-year institu-tions, a smaller percentage of the overall populationplanning to attend vocational schools, an increase inthe percentage of African-American students in thetop quartile of aptitude and a decrease in percentageof those in the lowest quartile. The author doescaution that the latter results may be due to differ-ences in nonresponse rates for NLS and HS&B andmay be better explained by evidence from SAT andACT tests.

119Yang, Shu-O W. 1981. "Rural Youths' Decision toAttend College: Aspirations and Realizations." Pa-per presented at the Annual Meeting of the RuralSociological Society. Guelph, Ontario, Canada, Au-gust 19-23. ED 207 765

The influence of peers, teachers, counselors,and parents on college attendance decisions of ruralyouth is analyzed using the NLS-72 "Base Year" and"First Follow-Up." Other variables, such as highschool grades, family SES, number of siblings, andstudents' beliefs regarding their ability to handlecollege level study are also included in the analysis.

Findings show that high school grades, andboth parents' and peers' influence are significantlyrelated to college aspirations and actual college at-tendance. Mothers' expectations had the strongesteffect. Self-confidence in completing college studydid not contribute as a factor in students' aspirationsalthough it did have a positive impact on actualcollege attendance.

5

VII. AttaivmentEducational and Occupational120Adelman, Clifford. 1991. Women at Thirtysometh-ing: Paradoxes of Attainment. Washington, DC:U.S. Department of Education, Office of Educa-tional Research and Improvement.

This study describes the educational careersand labor market experience of women in the classof 1972 through the time they were 32 years old("Fifth Follow-Up"), and sets its analysis in thecontext of national economic development and thesocial utility of equity.

Women' s academic performance in highschool was far stronger than that of men, yet, at thesame time, both their educational aspirations andplans were lower than those of men. Nonetheless,they continued their education after high school atthe same rate as men, were rewarded more withscholarships for postsecondary education, and com-pleted college degrees faster than did men. Using thecleaned PETS data, the study reveals more subtlepatterns of differences in curricular paths than pre-viously assumed and shows that women's gradepoint averages in college were higher than men's nomatter what field they studied. As a result of theirundergraduate achievements, the educational aspi-rations of women changed considerably, with dra-matic increases aspiring to graduate degrees. Fromage 18 to 32, the women of the class of 1972 devel-oped more positive attitudes toward education thandid men, and came to believe that they truly bene-fited from schooling.

These benefits, however, did not hold up in thelabor market, where evidence of women's superioreducational performance and commitment was dis-counted. Between ages 25 and 32, for example, amuch higher percentage of women than men expe-rienced genuine unemployment, no matter what de-gree they earned. In only 7 of 33 major occupationsdid women achieve pay equity with men by age 32.All these comparisons are between men and womenwho did not have children by age 32. Despite the

discouraging pattern of earnings differentials, ahigher percentage of women than men found theireducation relevant to their work, and, among bache-lor's degree holders, more women than men came towork "a great deal" with ideas, the engine of aninformation economy. Women also took a morepositive attitude than did men toward working con-ditions, relationships on the job, and new skills de-velopment. They were, in short, more enthusiasticand potentially productive workplace participants atthe same time they were under-rewarded. The studyconcludes that both women' s knowledge and theirwillingness to share that knowledge in the workplaceis critical to the nation's future and should be re-warded so that all may benefit.

A lengthy list of references and 23 tables ofdata compiled from the surveys and PETS files areincluded at the end of this report. A second editionof this monograph, including minor revisions in twotables, was published in July 1992.

See Gender and Racial-Ethnic Differences

121Adelman, Clifford. 1990. Light and Shadows onCollege Athletes. Washington, DC: U.S. Depart-ment of Education, Office of Educational Researchand Improvement.

Using the NLS-72 base year through 1986follow-up surveys and the NLS/PETS data, theauthor examines the long-term educational and labormarket careers of college varsity athletes. The var-sity group, made up of football and basketball only(n=134), was compared to 5 other groups: varsityathletes in sports other than football and basketball(n=92), performing arts students (n=310), intramu-ral sports participants (n=629), nonathletes(n=2,400), and everybody else (n=4,535). Allgroups were drawn from the PETS sample, and allgroups attended college in patterns dominated by4-year schools.

49 5.2

Major fmdings include:

(1) Compared with other groups, varsity foot-ball and basketball players enter college with rela-tively poor high school records and test scores andthe highest ratio of vocational to academic subjectsin high school, yet graduate from college at only aslightly lower rate.

(2) The principal reasons for the outcome men-tioned above are that varsity football and basketballplayers (a) are more likely than students in othergroups to receive scholarships and (b) are morelikely to enter college directly from high school.They also appear to have access to a "safety net" thatis not available to other students.

(3) Those varsity football and basketball play-ers who graduate take longer to do so than nonath-letes, earn lower grades, and pursue less demandingcurricula.

(4) Varsity football and basketball playersearned one out of every seven credits in a de factosports curriculum. Compared with the other groups,these students were more likely to have taken reme-dial and introductory courses, personal service anddevelopment courses, and vocational courses intrades and office support.

(5) By age 32, ex-varsity football and basket-ball players had the highest rate of home ownershipand the lowest rate of unemployment of all groups,along with earnings 10 percent above the mean forall former 4-year college students.

(6) Former varsity football and basketballplayers were the least likely of any group to claimthat their higher education was relevant to theirwork, were the least likely to work with ideas "agreat deal," and were the most likely to have lowerstatus occupations at age 32 than they had aspired toat age 19.

50

122Adelman, Clifford. 1988. "Transfer Rates and theGoing Mythologies." Change (January/February):38-41.

For description see entry #90.

See Patttrns of Attendance

123Alexander, Karl L., Riordan, Cornelius, Fennessey,James, and Pallas, Aaron M. 1982. "Social Back-ground, Academic Resources, and College Gradu-ation: Recent Evidence from the NationalLongitudinal Survey." Amarican_kilmaLQfIsiaca:don 911: 315-333.

This study examines the effects of social back-ground factors (race, gender, and SES) and aca-demic resources (standardized test scores, classrank, and curriculum) at the time of secondaryschool graduation on rates of completion of bacca-laureate degrees. Data were drawn from the baseyear through the 1979 follow-up surveys of theNLS-72 (n=1,916 whites and 1,204 blacks).

Findings indicate that "opportunities for com-pleting college, given that one has already completedhigh school, are influenced primarily by individualacademic merit and that social background handi-caps are less than might have been anticipated."Results show virtually no differences in rates ofdegree completion between males and females,slight differences by race (blacks, generally, havinghigher rates of degree completion) and notable dif-ferences by SES. Academic characteristics, how-ever, appear to be the strongest predictors of degreecompletion.

124Anderson, Kristine L. 1988. "The Impact of Col-leges and the Involvement of Male and FemaleStudents." Sociology of Education fa (July): 160-177.

53

This study focuses on the structural and organ-izational determinants of educational attainment andreviews research and theoretical models of determi-nants of educational careers of college entrants, in-cluding social and academic backgrounds, collegecharacteristics, and students' roles and experiencesafter they enter college. Using the "Base Year" and"First" and "Fourth Follow-Up" surveys of the NLS-72, the author draws upon Tinto's (1975) statusattainment models and integrates the determinantsinto a causal model in order to examine their com-bined influences on men and women.

Results indicate the continuing effects of SES,aptitude, high school preparation, goal commitment,student role involvement, gender differences, andachievement in college. High school grades are astrong predictor of college grades and the degree ofstudent role involvement. Findings also illustrate theimportance of institutional cohesiveness, the nega-tive effect of institutional integration on women'scollege grades and the strong support for the impor-tance of students' involvement in the student role.For both men and women, goal commitment is amore important determinant of attainment than insti-tutional integration or commitment.

See Gender and Racial-Ethnic Differences

125Anderson, Kristine L. 1984. The Effect of CollegeType and Characteristics on Educational Attain-ment. Washington, DC: U.S. Department of Educa-tion, National Institute of Education. ED 256 206

For description see entry #70.

See College/Program Effect

126Anderson, Kristine L. 1984. Institutional Differ-ences in College Effects. Final Report,. Washington,DC: U.S. Department of Education, National Insti-tute of Education. ED 256 204

Participants in the NLS-72 who entered anacademic program by fall 1974 are used in analysesto determine the impacts of type of postsecondaryinstitution attended on educational attainment,

achievement goals, and occupational attainment netof relevant individual predictors (such as socialstatus, academic preparation and early goal levels).Individual survey data on the higher education insti-tutions attended by participants were merged withthe NLS data. Institutional data include control (pub-lic versus private), level (university, 4- or 2-yearschool), gender and ethnic composition, percentageof students at the graduate level, percentage of vo-cational major areas, percentage of campus residentsand part-time students, enrollment size, expendi-tures per student, and tuition and fees.

51

Findings show a wide range of educationaloutcomes by institution type, although this may beattributed to differences in the students attendingthose institutions. In general, however, more aca-demically selective, smaller, less vocationally ori-ented colleges with higher levels of structuralintegration exhibit positive effects on student attain-ment. Two-year institutions had the most negativeeffects on educational attainment.

127Anderson, Kristine L. 1984. College Effects on thQEthuatimaL_Attainmcnt_DfisialaL_amL_Etmalca.Washington, DC: U.S. Department of Education,National Institute of Education. ED 256 205

For description see entry #71.

See College/Program Effects

128Burkheimer, Graham J., and Jaffe, Jay. . 1981."Highly Able Students Who Did Not Go To Col-lege." Washington, DC: U.S. Department ofEducation, National Center for Education Statistics.ED 212 682

For description see entry #91.

See Patterns of Attendance

129Dawkins, Marvin P. 1986. "A Longitudinal Analysisof Variations in Mobility Goals Across Race-SexGroups." Western Journal of Black Studies "QM:34-43.

5 zi

The study examines race and sex differencesin the stability of mobility goals over a 7-year periodfollowing high school graduation and also assessesthe impact of race, sex, social class and academicability on 1979 mobility goals. The data are drawnfrom the NLS-72 "Base Year" survey and "FourthFollow-Up," and include a sample of 3,333 whitesand 3,119 blacks. Mobility goals are defined byvariables "reflecting progress toward occupationalattainment" (such as satisfaction with progress to-ward achieving expectations). Academic ability isbased on results of the academic aptitude test score;administered to respondents in the "Base Year" sur-vey.

Findings indicate consistent sex differences inoccupational expectations. Males more frequentlynamed professional, crafts, technical and manage-rial/administrator categories while females indi-cated expectancies for professional, clerical,homemaker and school teacher roles. Regardingrace, blacks had lower levels of educational attain-ment 7 years out of high school and generally weremore dissatisfied than whites with their progresstoward attainment of occupational goals. With theaddition of social class background and academicability in the analysis of race and sex effects onmobility goals, findings point to the relatively strongimpact of race both independently and in interactionwith other factors. Sex does not have a significantimpact on indicators of mobility but does interactwith race to affect satisfaction with career progress.These results provide further evidence showing thatrace and sex remain important factors in the devel-opment of mobility goals and in the process of careersocialization.

See Gender and Racial-Ethnic Differences

130Daymont, Thomas N., and Andrisani, Paul J. 1984."Job Preferences, College Major and the GenderGap in Earnings." Journal of Human Resources12(3): 408-427.

This report analyzes gender differences inearnings among recent college graduates. Theauthors hypothesize that differences in preferencesor tastes, and in preparation for various types of work

52

measured prior to entering the labor market, explainmuch of the earnings gap between recent male andfemale college graduates.

Data from the NLS-72 base year through 1979follow-up surveys are used to distinguish betweenthe effects of differences before and outside the labormarket and differences within the labor market. Re-sults indicate that preferences in occupational rolesand in choice of college major are different for menand women. These two differences account for ap-proximately 1/3 to 2/3 of the gender gap in hourlyearnings 3 years after college graduation. Althoughsimilar studies interpret the unexplained portion ofthe gender gap in earnings as due to labor marketdiscrimination, the authors suggest that the resultsmay be over or underestimations due to selectivity,"the chilling effect of past discrimination," compa-rable worth-type issues, and other omitted variables.

See Gender and Racial-Ethnic ijifferences

131Dunkelberger, John E. 1984. S,114,jaefiningandAchieving Life Goals: A Process of Human Re-source Development. Termination of Report of Co-up.erativzikgionaLEIDie&i. ED 242 450

Project S-114 (1977-1983) consisted of threeobjectives which dealt with rural Southern youth andwhich focused on educational and occupational as-pirations and attainment results for males and fe-males, blacks and whites. Each objective involvedusing a distinct dataset. In 1979, Objective 1 wascompletedthe fourth wave of a longitudinal panelstudy begun in 1966 was carried out by interviewingalmost 1,000 young adults (then about 30 years old)raised in rural areas of six Southern states. Data wereanalyzed in areas such as education, occupation,residence and family. Objective 2 was modified andultimately used in the National Longitudinal Studydata for a wide variety of special analyses. Objective3 involved the analysis of a survey which was mailedto almost 3,300 agriculture students.

Objective 2, which used NLS-72, specificallyexamined several factors related to occupational at-tainment, such as patterns of both educational and

occupational goals, and the movement of womeninto traditionally male occupations. (Adapted fromERIC abstract)

132Eagle, Eva and Carroll, C. Dennis. 1988. Postsecon-dary Enrollment. Persistence. and Attainment for1972 1980. and 1982 High School Graduates. HighSchool and Beyond National Longitudinal Study.Washington, DC: U.S. Department of Education,National Center for Education Statistics. ED 304036

For description see entry #93.

See Patterns of Attendance

133Eck land, Bruce K., and Henderson, Louise B. 1981."College Attainment Four Years After HighSchool." Washington, DC: U.S. Department ofEducation, National Center for Education Statistics.ED 215 643

This report focuses on those NLS-72 respon-dents who went to college, dropped out, returned,and graduated on schedule. Special attention is givento both students entering 2- and 4-year colleges andpart-time students. The report also looks at the influ-ence of SES, gender, and race-ethnicity on the col-lege attainment process. Differences in outcome bytype of student, SES, race, sex, and ability wereexamined for each point in the college attainmentprocess. Data were taken from the base year and thefirst three follow-up surveys of NLS-72 (1972-76).

Findings indicate that students attending 2-year schools are less likely to graduate with a BA onschedule. Also, the difference in graduation ratesbetween 2- and 4-year schools was nearly the samefor each SES and ethnic group. SES had a significantimpact at almost every stage of the education cycle,even when ability was controlled for. SES appearsto influence full- or part-time status, persistence, andreturn to college.

Effects of race on the college attainment proc-ess were mixed. Some findings related to race in-clude: more blacks than whites delayed entering

college, blacks were equally likely as whites toreturn to college if they dropped out, and morewhites than blacks attended 2-year schools. A verysignificant finding regarding race was the reversalof black disadvantage when ability was controlledfor.

134Epps, Edgar G., and Jackson, Kenneth W. 1985.Educational and Occupational Aspirations and EarlyAttainment of Black Males and Females. Atlanta,GA: Southern Education Foundation. ED 263 818

53

Using a modified version of the Wisconsinmodel of status attainment, the authors try to ascer-tain specific mechanisms which generate status (de-fined here as occupational and educationalattainment) among African-Americans. Data fromthe NLS-72 "Base Year" and "1979 Follow-Up,"and High School and Beyond 1980 "Base Year"survey and "First Follow-Up" are used in the analy-sis. The Wisconsin model stresses social-psycho-logical influences. The authors, however, feel thatmodels such as this are ineffective for explaining theattainment of black students because of inadequatemeasures of the school dimension.

Findings suggest that, for black students, aspi-rations may be a more significant factor for educa-tional attainment than measured achievement. Theyalso show the strong influence of school factors onstatus attainment processes for both black males andfemales. While findings provide a general pattern ofeducational attainment, the authors believe that amodel combining background, social-psychologi-cal, and school factors was insufficient for under-standing occupational attainment for black students.The authors suggest the possibility that factors ,,,uchas discrimination and structure of opportunity maybe missing from the model.

See Gender and Racial-Ethnic Differences

135Fun, Jane D. 1980. "The R elationship of HighSchool Reading and Mathematics Achievement andEmployment Experiences: Implications forMinimal Competency Testing." Paper presented atthe Annual Meeting of the National Council on

. ,5

Measurement in Education, Boston, MA, April 8-10. ED 187 737

This study uses the NLS-72 ("Second Follow-Up" survey) to examine data regarding the relation-ship between reading and math achievement, andemployee experience variables (work experiences,occupational category, job satisfaction, income, andsteadiness of employment). The sample consists ofrespondents who were of average and below averageachievement.

Findings indicate that there is little relationshipbetween high school reading and math achievementlevels and employee experience variables. Theauthor suggests that further research is necessary inthis area to ensure that appropriate, minimal compe-tency objectives are set to prevent labeling people asincompetent "on the basis of the examination oftasks which bear little relationship to life experi-ences."

136Gerstein, Marti, Lichtman, Marilyn, and Barokas,Judy U. 1988. "Occupational Plans of AdolescentWomen Compared to Men: A Cross-Sectional Ex-amination." The Career Development Quarterly 31(March): 223-230.

The central purpose of the study is to examineshifts in the occupational plans of adolescent womencompared with those of men, using data from theNLS-72 and High School and Beyond. The trend ofentering into high-status and exiting low-status ma-jor occupational categories is examined, and entryinto "male-dominant" and exit from "female-domi-nant" occupations is compared.

Results indicate a slight shift in occupationalplans for women, although many still planned toenter traditionally female fields. Data suggest that,in general, many occupational categories will re-main dominated by a single sex. The authors includeimplications and suggestions for individuals provid-ing career counseling for students, such as offeringcurrent educational/occupational information andspecific labor market data, and facilitating studentgoal achievement through psychological encourage-ment.

, See Gender and Racial-Ethnic Differences

54

137Gordon, Howard R.D. 1985. "Analysis of the Post-secondary Educational Attainment of AgriculturalEducation Graduates of the High School Class of1972." hurnaLDLIQ.catiQuaLlslucatismiltsearch1Q(2): 11-18.

This study examines the effects of participa-tion in agricultural education in high school on sub-sequent postsecondary educational attainment.Specifically, the author seeks to determine the de-gree of association between the number of semestersin agricultural education taken at the secondary leveland the number taken at the postsecondary level.Also, possible differences in postsecondary attain-ment among agriculture graduates (four or moresemesters completed), semi-agriculture graduates(one to three semesters completed), and nonagricul-ture education graduates (zero semesters completed)is examined.

Four main findings are:

(1) Agricultural education graduates partici-pated less frequently in postsecondary programsthan either semi-agricultural or nonagricultural edu-cation graduates.

(2) When the differences in postsecondary par-ticipation for each group are subjected to multipleregression analysis, using selected demographic andpersonal data as predictor ..ariables, the differencesbased on participation in agricultural education allbut disappear.

(3) When evaluating program results in termsof subsequent achievement of graduates, effects ofbackground variables must be taken into account.

(4) More emphasis should be placed on coun-seling students to consider continuing their educa-tion beyond high school and on providingagricultural education students with information re-garding postsecondary opportunities in their areas ofstudy.

5

138Gordon, Howard R.D., and Camp, William G. 1984."Analysis of the Occupational Attainment of Agri-cultural Education Graduates of the High SchoolClass 1972." Paper presented at the National Agri-cultural Education Research Meetings, New Or-leans, LA, December. ED 259 109

The purposes of this study are (1) to determinethe differences in the job status attainment amongagricultural education (four or more semesters com-pleted), semi-agricultural education (one to threesemesters completed), and nonagricultural educa-tion high school students who were employt. inagricultural occupations for each year from 1972 to1979, taking varying backgrounds into account; and(2) to ascertain differences, still controlling for back-ground, in job status attainment for agricultural edu-cation , semi-agricultural education andnonagricultural education high school students. Asample of 6,585 respondents from the NLS-72 areused in the analysis. The authors use the Duncanscale score, an indirectly derived continuous vari-able, as a measure of socioeconomic attainment.

Initial analyses reveal that participation in ag-ricultural education has negative effects on the statusof graduates for 3 of the 7 reported years. Whenbackground variables are controlled for, however,results change dramatically. The authors concludedthat "agricultural education program evaluations aretoo complex to be addressed adequately by suchsimplistic analyses as the straightforward examina-tion of graduate incomes or placement records."They also conclude that participation in agriculturaleducation training in high school had relatively neu-tral effects on the future job status attainment ofgraduates in either agricultural or nonagriculturalcareers.

139Grubb, W. Norton. 1989. The Causes and Conse-quences of Enrollments in Higher Education:Evidence From the National Longitudinal Study ofthe Class of 1972. Final Report. Berkeley, CA: In-stitute for the Study of Family, Work, and Commu-nity. ED 318 371

55

The purpose of this study is to examine theprobability of enrolling in postsecondary education,the likelihood of completing various types of creden-tials, and the impact of those credentials (includingcourse credits earned by those who do not completecredentials) on wage rates, earnings and on othereducational and noneducational adult outcomes.Data were drawn from the NLS-72 base year throughfifth (1986) follow-up surveys and the Postsecon-dary Education Transcript file.

The first chapter provides an overview of bothenrollments in and completion of postsecondaryeducation by institution type and for different groupsof students. Results are based on both the NLS-72data and comparable data from High School andBeyond. The second chapter expands upon enroll-ment patterns described in chapter one and presentsa multivariate analysis of the causes of enrollment.The third chapter presents results from the analysesof wage rates, earnings and other adult outcomes.The effects of field of study on outcomes and thereceipt of credentials from different kinds of post-secondary institutions is also examined. The finalchapter provides an explanation of the results.

Findings indicate that several forms of postsec-ondary education fail to increase either wage ratesor earnings. Results also show significant differ-ences between BA degrees and all other forms ofpostsecondary attainment in terms of benefits re-ceived (e.g., economic returns). Other findings showan increase in the tendency to drop out of postsecon-dary education for groups frequently viewed as atrisk of failure in the labor market (minorities, lowSES, and those with poor academic records in highschool) and an increase in the rate at which adultsare enrolling in higher education (although difficultyexists in detecting either economic or noneconomicbenefits).

140Grubb, W. Norton. 1989. Access, Achievement.Completion, and "Milling Around" in Postsecon-dary Vocational Education. Washington, DC: U.S.Department of Education, National Assessment ofVocational Education. ED 315 548

The author looks at student behavior at thepostsecondary level in terms of degree completion,dropout and "milling around" (frequently switchingmajor area of study). The research is in response toincreased access to higher education of minoritystudents and to implications for completion and oc-cupational attainment. Postsecondary vocationalstudents attending community colleges, technicalinstitutes and private vocational schools are high-lighted. Data are taken from the NLS-72 and theHS&B 1980 cohort. Data are used to examine initialenrollments in higher education; fields of study inpostsecondary education; progress through institu-tions of higher education and the tendency to dropout; completion of credentials; the special problemsof minority students; the nature of transfers amonginstitutions; course enrollments and credits earnedby noncompleters; late entrants into higher educa-tion, and employment during postsecondary educa-tion. Data are also used to compare postsecondaryvocational institutions and vocational students inpostsecondary education.

Findings indicate an expansion of postsecon-dary vocational education, especially in communitycolleges, but a substantial number of students fail tocomplete a program. The author cautions that inter-pretation of these results is difficult because of sev-eral questions concerning the motivation behindstudent behavior: (1) How many students are reallydropouts? (2) How many are completers in terms oftheir own goals? and (3) How many are experiment-ers?

See Vocational Technical and Vocational Educa-tional

141Hill, Susan, and Owings, Maria. 1986. CompletionTime for Bachelor' s Degrees. Washington, DC:U.S. Department of Education, Center for EducationStatistics. ED 276 376

This report compares bachelor degree recipi-ents who began college by the fall of 1972 andearned degrees within 4 years (traditional pattern),with recipients who either (1) took longer than 4years to graduate, (2) began college later than the fall

56

after high school graduation, or (3) both. The influ-ence of aptitude, test composite score, high schoolcurriculum (academic, general or vocational), SES,and region of high school (Northeast, North Central,South, or West) were examined for a sample ofNLS-72 students whc received bachelor's degreesby December 1984 (n.4,440). The data came fromthe first public release (uncleaned) PETS files.

Women were more likely to follow the tradi-tional pattern of degree attainment. Those whoscored in the highest quartile of the NLS aptitudetest, who took a college preparatory curriculum inhigh school, who came from the Northeast, and whowere of high SES also tended to follow the tradi-tional pattern. In addition, students majoring in so-cial sciences, science, and mathematics generallyfollowed the traditional pattern.

Of those who received their degree later than4 years after high school graduation, 70 percentstarted by the fall after high school but took longerthan 4 years to finish. Delayed entrants (those wait-ing until 3 or 4 years later) were equally likely asthose entering immediately after high school to earntheir degree within 4 academic years.

142Hilton, Thomas L., and Schrader, William B. 1987.Pathways to Graduate School: An Empirical StudyRased on National Longitudinal Data. Princeton, NJ:GRE Board Research Report No. 82-21R ETS/Re-search Report 87-41.

For description see entry #99.

See Patterns of Attendance

1.43

Horn, Laura, and Carroll, C. Dennis. 1989 Enroll-ment. Completion. Attrition. and VocationalCourse-Taking Patterns in PostsecondaryEducation: A Comparison of 1972 and 1980 High

I S S j . .

Schoolfind Beyond. National Longitudinal Study of1972. Washington, DC: U.S. Department of Educa-tion, National Center for Education Statistics. ED309 820

This report compares the postsecondary edu-cational experiences of students who entered a 2-year institution the summer or fall after high schoolgraduation. The Postsecondary Educational Tran-script file of the NLS-72 and High School and Be-yond 1980 seniors are used to compare students onthe basis of enrollment, completion (defined in rela-tion to degree attainment), dropout rates and voca-tional course-taking patterns by gender, SES,racial-ethnic groups, and parental education.

Major findings include:

(1) Rates of enrollment in 2-year institutionsimmediately after high school graduation werehigher in 1980 than 1972. This was true regardlessof sex, race-ethnicity, or parental education. Enroll-ment rates were also higher for students of low andmiddle socioeconomic status.

(2) Hispanic students enrolled in 2-year insti-tutions at higher rates than black students in 1972and at higher rates than either bla,:k students or whitestudents in 1980. Black students enrolled at lowerrates than white students in 1972 but had approxi-mately the same rates as white students in 1980.

(3) Males, white students, students whose par-ent(s) had a bachelor's degree or higher, and highSES students of the 1980 cohort had lower rates ofdegree completion than in the 1972 cohort. Con-versely, Hispanic students from the 1980 cohort hadhigher completion rates than those in the 1972 co-hort.

(4) White students had higher completion ratesthan either black students or Hispanic students in the1972 cohort. In the 1980 cohort, white and Hispanicstudents had approximately equal completion ratesand both were greater than black student completionrates.

(5) Dropout rates were higher in the 1980cohort than in the 1972 cohort regardless of sex,socioeconomic status, or parental education. OnlyHispanic students had no decline in dropout ratesbetween the two cohorts.

57

(6) Approximately equal proport;ons, regard-less of sex, race-ethnicity, parental education orSES, of students entered vocational programs in1972 and 1980.

(7) Overall, no decline in degree completionfor vocational students between the 1972 and 1980cohorts existed, regardless of sex, race-ethnicity,parental education or SES.

See Vocational Technical and Vocational Educa-tion

144Knepper, Paula R. 1990. Trends in PostsecondaryCredit Production. 1972 and 1980 High School.Graduates. Washington, DC: U.S. Department ofEducation, National Center for Education Statistics.

This study examines differences between thehigh school graduating classes of 1972 and 1980 interms of their postsecondary progress and comple-tion. Data are drawn from the NLS-72 and HS&B1980 senior cohort. The "Base Year" survey andPETS are used for each cohort. Because of timeconstraints with the 1980 PETS data, only the first 41/2 years of the 1972 PETS data are included.

Some of the main findings include:

(1) While almost a third (31 percent) of the1972 cohort who entered postsecondary educationright out of high school completed a bachelor' sdegree within the 4 1/2-year period, less than aquarter (22 percent) of the 1980 cohort did so in the

same length of time.

(2) Overall, baccalaureate degree completionin 4 1/2 years for the 1980 cohort dropped about 10percentage points below that of the 1972 cohort. This

same 10 percentage point drop was evident for allgroups, regardless of race, sex, or socioeconomicstatus (SES). By sex, men dropped from 30 to 21percent, and women dropped from 33 to 22 percent.By race, whites dropped from 33 to 24 percent,blacks dropped from 22 to 11 percent, and otherminorities dropped from 20 to 12 percent. By SES,

6

those in the lowest group dropped from 21 percentto 11 percent, those in the middle dropped from 27to 19 percent, and those in the highest group droppedfrom 40 to 30 percent.

(3) Almost 10 percent more persons in the1980 cohort than in the 1972 cohort took longer than9 months to complete the freshman year (67 percent,up from 58 percent).

(4) For those who finished the BA, 11 percentmore in the 1980 cohort than in the 1972 cohortfinished within 45 months (84 percent, up from 73percent).

145

Knepper, Paula R. 1989. Student Progress in Col-lege: NLS-72 Postsecondary Education TranscriptStudy. 1984. Washington, DC: U.S. Department ofEducation, National Center for Education Statistics.ED 309 710

This study examines the average amount oftime taken to complete each level of postsecondaryeducation, including the BA, the number of studentswho take longer than the expected time to advance,and goal attainment. These issues are examined interms of sex, race, and SES and their relationship topersistence and progress. Data were taken from theNLS-72 and PETS for students who had at least onetranscript.

Some main findings include:

(1) Over 40 percent of the NLS-72 seniorsearned a BA degree within 12 years after highsc hool.

(2) In general, more than half of all BA com-pleters took longer than the expected 45 months tofinish.

(3) Men were more likely to take longer at eachlevel than women.

(4) Type of postsecondary institution attendedwas related to completion time. For example, stu-dents who attended 4- year private colleges had a 33

percent greater likelihood of completing the BA ontime compared to students who entered similar pub-lic colleges.

(5) Transfer between schools usually resultedin longer completion time.

146

Lichtman, Marilyn, and Rothschild, Susan J.S. 1982."Career and Education Patterns of Rural Women."Paper prepared for presentation at the Annual Meet-ing of the American Educational Research Associa-tion, New York, NY, March 19-23. ED 214 725

This study uses the base year through 1979follow-up surveys of the NLS-72 to examine theeffects of geographic location (rural, in this case) onwomens' status attainment. This study compares theeducational and occupational expectations and at-tainments of those who went to high school in ruralareas and remained following high school with theexpectations and attainments of those who wereeducated in rural areas but moved to other types ofcommunities following high school.

Findings show strong evidence that rural resi-dence after high school graduation is a significantelement in the educational and occupational patternsof women. Women residing in rural areas have lowereducational aspirations and are less likely to workoutside the home after high school. Actual attain-ment, however, is low for both groups.

147Lyson, Thomas A. 1986. "Migration Selectivity andEarly Adult Attainments." Rural Sociology 51(3):328-342.

This study examines two interrelated issues:(1) the relationship between background charac-t!ristics and personal attributes, and the decision tomigrate to or from a rural area; and (2) how thedecision to move from, into or remain in a rural arearelates to early adult social and economic attain-ments. Six populations (rural nonmigrants, rural-to-rural migrants, rural-to-urban migrants, urbannonmigrants, urban-to-urban migrants, and urban-to-rural migrants) were drawn from the NLS-72"Base Year" and subsequent follow-ups. Discrimi-

58 6'4

nant analysis measured the extent of differencesamong the six groups-.

Findings related to the first question show that,for men, the biggest distinction in background char-acteristics among the six groups was farming. Plansto farm played an important part in the decision tomigrate. For women, a farm background was impor-tant in characterizing rural-reared and urban-rearedpersons, although a larger proportion of women withfarm backgrounds migrated to urban areas. Findingsrelated to the second question indicate significantlydifferent patterns of early adult social and economicattainment. These findings show that "a rural orurban context can be seen as a set of parametersthrough which and by which career plans are shaped,molded, and eventually crystallized.., researchers,therefore, must account for the effects of variationsin situational circumstances on the hypotheses, mod-els, and theories being tested."

148McClelland, Katherine. 1990. "Cumulative Disad-vantage Among the Highly Ambitious." Sociologyof Education 63: 102-121.

This study examines the early educational at-tainments and changes in career expectations of asub-sample of respondents to the NLS-72. The baseyear and first through fourth follow-up surveys areused in analysis. Using Bourdieu and colleagues'theories of "cumulative disadvantage," the authorshows, after controlling for differences in measuredacademic ability, that men and those from upper-white-collar homes are more likely to survive the7-year period following high school on track and toachieve their initial goals. Results also indicate that,for women and those from nonprivileged back-grounds, marriage is an especially strong transmitterof cumulative disadvantage while attendance at anelite college or university is beneficial to all.

149Mertens, Donna M., and Gardner, John A. 1982.Educational and Employment Experiences of theYounger Adult Worker. Washington, DC: U.S. De-partment of Education, Office of Vocational andAdult Education. ED 215 210

59

Three large-scale surveys were used to exam-ine the effects of vocational education on educa-tional and occupational attainment: ihe YoungerAdult Worker (YAW), the 1966-78 National Longi-tudinal Surveys of Labor Market Experiences forYoung Men and Young Women and the NLS-72.

The study's results strongly indicated sexstereotyping in both enrollment in vocational pro-grams and subsequent employment. The findingsalso provide strong evidence that participating inmarketing education programs in high school re-sulted in higher earnings for men. Participation inbusiness, health, home economics or agriculturalprograms were associated with lower earnings.Women who took trade or business curriculum hadhigher earnings. Postsecondary vocational educa-tion is associated with higher earnings for both menand women.

In general, completing a vocational curriculumin high school was negatively associated with com-pleting traditional postsecondary education. How-ever, completing a vocational curriculum waspositively related to completion of nontraditionalpostsecondary education such as apprenticeshiptraining or employer-sponsored training.

See Vocational Technical and Vocational Educa-tion

150Powell, Brian, and Steelman, La la Carr. 1990. "Be-yond Sibship Size: Sibling Density, Sex Composi-tion, and Educational Outcomes." SaCiaL_EQE.C.ta

62(1): 181-206.

This paper looks at the effects of sibling den-sity (number of siblings within specific time inter-vals) and sex composition of the sibship (number ofbrothers and sisters) on academic performance. Dataare from the "Base Year" and "First Follow-Up" ofHS&B and the "Base Year" survey of the NLS-72.

Several findingspositive associationsamong family income, parental educational level,race and educational performancecoincide withthe results from previous studies of siblings. The

6 (4, 1

authors found high sibling density associated withlower test performance, with math scores beingmore affected than verbal scores. Sex compositioneffects indicate that having sisters is less damagingto classroom performance than having brothers is,although sex composition of the sibship has no sig-nificant effect on standardized test scores.

See Marriage/Family

151

Robertshaw, Dianne, and Wolfle, Lee M. 1980."Discontinuities in Schooling and Educational At-tainment." Paper presented at the Annual Meetingof the American Educational Research Association,Boston, MA, April 7-11. ED 186 488

For description see entry #102.

See Patterns of Attendance

152Robertshaw, Dianne, and Wolf le, Lee M. 1983."Discontinuities in Schooling and Educational At-tainment." Higher Education 12: 1-18.

For description see entry #103.

See Patterns of Attendance

153Rothschild, Susan J.S., and Lichtman, Marilyn.1980. Factors Related to White. Black. and HispanicWomen's Mathematics Attainments: A DescriptiveStudy. ED 195 570

The study examines the relationship betweenfamily background, high school experience, self-concept, and significant others' perceived influenceon college plans, expectations (educational and pro-fessional), expected college major, college mathe-matics experience, sex-role orientation, familystatus, and the study of mathematics for black, whiteand Hispanic women. Data are drawn from the baseyear through 1976 follow-up surveys of the NLS-72with samples sizes as follows: n=6,384 whitewomen, n=1,193 black women and n=384 Hispanicwomen.

Findings show educational expectations andnumber of semesters of high school math to beinfluential for all three groups examined. Black andwhite women showed similar correlation patterns inthat neither background characteristics, marriage,self-concept, sex-role orientation nor significantothers' perceptions on college plans exhibited strongeffects on the study of math. The children/familystatus variable had a significant, negative relation-ship to the study of math for Hispanic women, whileeducational expectations, number of semesters inmath and high school program were highly relatedto the study of math. The self-concept measures andsex-role orientation showed a slight relationship tothe study of math. Hispanic women were also theonly group for which background variables, specifi-cally fathers' education, showed any effects on thestudy of math.

60

See Gender and Racial-Ethnic Differences

154Seyfarth, John T., and Ranson, James T. 1989. "Oc-cupational Stayers and Strayers: Career and SalaryExperiences of Current and Former Teachers." Pa-per presented at the Annual Meeting of the AmericanEducational Research Association, San Francisco,CA, March 27-31. ED 314 817

This study utilizes data from the "Fifth Follow-Up" of the NLS-72 to better understand career deci-sions of teachers. The following four questions areexamined:

(1) In what occupations were respondents tothe 1986 survey found?

(2) How many persons from the cohort whobecame elementary or secondary teachers later tookjobs in other fields, and what occupations did theychoose?

(3) How many persons who initially enteredother occupations subsequently took teaching jobsin elementary and secondary schools, and what oc-cupations did they leave in order to teach?

(4) How did the salary experiences of personswho remained in elementary and secondary teaching

63

compare with the expeiiences of those who nevertaught, those who left teaching for other occupa-tions, and those who entered teaching from otheroccupations?

Findings show that 60 percent of the sample ofthose who ever taught were teaching in 1986. Thosewho left teaching for other jobs tended to receivehigher salaries. Those who persisted in their firstoccupational choice tended to have higher salariesthan job changers. Those who left teaching tendedto choose sales, clerical or managerial jobs. Three-fourths of those switching to teaching came from theprofessional/technical, managerial/administrative,sales, or clerical fields.

See Teachers

155Sharp, Laure M., and Weidman, John C. 1989."Early Careers of Undergraduate Humanities Ma-jors." Journal of Higher Education 6.Q(5): 544-564.

For description see entry #79.

See College/Program Effects

156Sharp, Laure M., and Weidman, John C. 1986. EarlyCareer Patterns of Undergraduate Majors in the Hu-manities. Research Report No. 1. Pittsburgh, PA:School of Education, University of Pittsburgh. ED274 258

For description see entry #80.

See College/Program Effects

157Smart, John C., and others. 1985. "Self-Concept andCareer Orientation Developmental Patterns Duringthe College Years." Paper presented at the AnnualMeeting of the American Educational ResearchAssociation, Chicago, IL, March 31-April 4. ED 261620

The purposes of the study are (1) to determinethe degree to which changes in adolescents' self-

concept and career attitudes are related to their levelof involvement in postsecondary education, and (2)to describe the extent to which these relationshipsare consistent for males and females. The sampleconsists of 10,759 NLS-72 respondents with com-plete data on sex, 1972 family SES, postsecondaryeducational attainment level, self-concept, and ca-reer orientation. A comparison group of respondentswith no exposure to postsecondary education is alsoincluded in the analyses.

"General self-image maturation pattern for allrespondents indicated a consistent increase over the7 years following high school. Those respondentswith higher levels of involvement in postsecondaryeducation had a more positive self-image at eachof...5 time intervals. However, differences betweenthe groups was greater in 1979 than initially in 1972,suggesting that exposure to the collegiate experiencetended to reinforce and increase the more positiveself-image held initially by college- bound youth.Exposure to postsecondary education experiencesappears to have a strong relationship with the devel-opment of a more positive self-image, a modestrelationship with enhancing self-confidence, but af-fects men and women differently in terms of theirrespective career orientation development. Forwomen, a strong, positive relationship between levelof postsecondary education attainment and careerorientation scores exists, while for men the two areessentially unrelated."

158Smart, John C., and Ethington, Corinna A. 1985."Early Career Outcomes of Baccalaureate Recipi-ents: A Study of Native 4-Year and Transfer 2-YearCollege Students." Paper presented at the AnnualMeeting of the Association for the Study o' HigherEducation, Chicago, IL, March 15-17. ED 259 630

This paper examines conceivable differencesin the job stability, status, and satisfaction of three6roups of BA recipients with varying levels of atten-dance at 2-year colleges. The sample consists of1,609 NLS-72 respondents who were BA recipientsin 1976 with complete data on academic ability,family SES, intended occupational status, number ofyears employed between 1976 and 1987, gender,

61 6 4

number of years enrolled in a 2-year college, 1972-76 job stability, and job status.

Results "clearly suggest that potential short-term detrimental effects of 2-year college atten-dance, if they exist, are not immutable and thatstudents are not forever disadvantaged by their in-itial decision to begin undergraduate schooling in2-year colleges in terms of subsequent career behav-iors."

159

Stone, James R. 1985. "Measuring Marketing Edu-cation' s Impact on the Acquisition of EducationalCredentials: A Creative Analysis." Journal of Vo-cational Education Research12(4): 33-48.

This study investigates the net influence ofseveral variables (background, vocational educationparticipation, and school characteristics) on post-high school attainment of marketing education stu-dents. The author's model suggests that educationalattainment for marketing education students is posi-tively influenced by mother's education, race, gen-der, aptitude, participation in cooperative education,and educational aspirations. Data are taken from the"Fourth Follow-Up" of the NLS-72.

Results suggest that high ability nonwhitesmay be overrepresented in marketing education, thatmales tend to gain more education after high schoolthan females, and that student participation in coop-erative education has a significant negative impacton post-high school educational attainment. Partici-pation in marketing education, however, has no in-fluence on post-high school educational attainment.

As a result of these findings, the author recom-mends that research examine the impact of partici-pation in vocational education programs onpost-high school outcomes and include measureswhich are both program specific and treated as con-tinuous variables. The author also suggests that thosevocational education students who participatethrough a cooperative methods class (cooperativeeducation) be encouraged to continue their post-highschool activities.

62

160Velez, William, and Javalgi, Rajshekhar G. 1987."Two-Year to Four-Year College: The Likelihoodof Transfer." American Journal of Education 26:81-94.

This paper looks at the effects of several pre-dictor variables on transfers from 2-year to 4-yearcolleges. Variables include those related to aca-demic processes (GPA and high school track), psy-chosocial processes (influence of peers and friends),background characteristics (SES, ethnicity, relig-ion), and institutional integration (participation inwork-study, living on campus, fraternity or sororitymembership).The sample includes those NLS-72respondents enrolled in 2-year colleges in academicprograms (approximately 1,407 students).

Results show that males have higher transferrates. Other factors positively effecting transfer in-clude having high SES, being Jewish, being in acollege preparatory track in high school, being influ-enced by parents, and having high occupational andeducational goals and high levels of institutionalintegration.

161

Wagenaar, Theodore C. 1984. Occupational Aspira-tions and Intended Field of Study in College. Ox-ford, OH: Miami University. ED 262 695

This study examines the selection and attain-ment of an occupation using data from the NLS-72and High School and Beyond. Specific researchissues include examining (1) changes which haveoccurred between 1972 and 1980 in seniors' occu-pational aspirations and intended field of study; (2)the success of seniors in attaining their occupationaland field of study plans; and (3) factors related to theselection of an occupation and a field of study forthe 1980 cohort.

General findings include:

(1) Decreased interest in teaching and an in-creased interest in managerial/proprietor occupa-tions between 1972 and 1980. These were the mostsubstantial changes during this time period.

6 5

(2) Increased interest in computer/engineer-ing/architecture fields and decreased interest in edu-cational fields were the most substantial changes infield of study choice between 1972 and 1980.

(3) Sex and educational aspirations were themost important factors in selecting an occupation.

(4) Sex, aptitude, educational aspirations, sci-ence coursework, and work values were the mostimportant factors in selecting a field of study.

(5) Those selecting teaching, and educationfields were less likely to be in the academic curricu-lum in 1980 than in 1972.

Policy implications for educators and federalofficials are also included in this report.

162Waite, Linda J., and others. 1986. The Effects ofParenthood on the Career Orientation and Job Char-acteristics of Young Adults. Bethesda, MD: Na-tional Institutes of Health, National Institute of ChildHealth and Human Development. ED 294 044

This study looks at several hypothese:, aboutthe relationship between becoming a parent, and jobcharacteristics and career expectations of men andwomen. Specifically, the study asks:

(1) How do those who become parents differfrom those who delay or forgo children before thefirst birth, in terms of expectations and achieve-ments?

(2) Does parenthood change the work-relatedexpectations and behaviors of young adult men andwomen?

Data for the study are taken from the NLS-72.Included are respondents for whom marital and pa-rental status were known in 1976 and 1979. Theuniverse included approximately 2,807 women and2,477 men who were ever-married nonparents in1979, and 3,350 women and 2,190 men who wereever-married parents by that year.

Results point to the existence of differences(even before the first birth) between men and womenwho became parents and those men and women whodelayed having children. Those planning to be moth-ers, in general, are less career oriented, with greaternumbers expecting to be homemakers. Prior to be-coming pregnant, however, their occupational andsocial expectations are not low. After pregnancy,employment decreases dramatically for women andmany indicate homemaking as their primary activ-ity. The differences between soon-to-be fathers andother men is small in terms of work commitment,current occupational status, career expectations andhourly wages. In general, results show "a slight,early advantage for fathers in occupational status,followed by a reversal of this advantage during latepregnancy and the first 2 years of the child's life."

63

See Marriage/Family

163West, Jerry. 1989. The Postsecondary VocationalEducation of 1980 High School Seniors. The 2-YearAssociatt DfArtspegree. Survey Report. Washing-ton, DC: U.S. Department of Education, NationalCenter for Education Statistics. ED 309 821

This report examines factors which differenti-ate between students who complete vocational As-sociate of Arts degrees and those who completeacademic Associate of Arts degrees. Data from thePostsecondary Education Transcript files of highschool seniors included in the High School andBeyond Study of the class of 1980 and the NLS-72were used to study the characteristics of studentswho received an associate degree or certificate in avocational education program within 4 years of highschool graduation.

Main findings include:

(1) Only 12 percent of the HS&B graduateswho attended any postsecondary institution com-pleted a vocational degree or certificate program.

(2) Among the 1980 high school seniors whocompleted a postsecondary program within 4 yearsof high school graduation, vocational credentials

were earned at a rate four times that of academicdegrees.

(3) High School and Beyond students com-pleted vocational degrees at a higher rate than NLSstudents, but completed academic degrees at a lowerrate.

(4) Most associate vocational degrees werecompleted at public 2-year colleges.

(5) The most popular field of study for voca-tional program graduates was business, followed bytechnical and engineering studies and health forassociate degree recipients, and trade, industry, andhealth for certificate completers.

(6) Forty percent of all credits earned by voca-tional degree completers were in academic subjects.

(7) Students with certificates earned slightlymore than half as many credits as students complet-ing vocational degree programs.

(8) The postsecondary credit patterns for thosewho completed vocational certificate programs andfor those who completed academic degree programsdid not change much between the early-to-middle1970s and the early-to-middle 1980s.

See Vocational Technical and Vocational Educa-tion

164Wolf le, Lee M. 1985. "Postsecondary EducationalAttainment Among Whites and Blacks." AratricanEducational Research Journal 22(4): 501-525.

This study examines the existence of similari-ties and/or differences in the process of educationalattainment for blacks and whites. The author firstdetermined whether educational attainment wassimilar for blacks and whites and then examined the

64

extent and sources of educational inequality and howthose inequalities were "expressed through interven-ing variables in the acquisition of postsecondaryeducation attainment."

Data for the study are drawn from the base yearthrough fourth follow-up of the NLS-72. The sampleconsists of 6,825 whites and 433 blacks who havecomplete information on the 13 variables includedin the analysis. Major findings of the study include:

(1) The processes of educational attainment isthe same for blacks and whites.

(2) The process of educational attainment de-pends modestly on social background.

(3) Ability is the most important determinantof placement in an academic program.

(4) To the same extent, curriculum differentia-tion in high school has a major impact on subsequenteducational attainment for both blacks and whites.

See Gender and Racial-Ethnic Differences

165Wolfle, Lee M., and Lichtman, Marilyn. 1981.Measurement Error Models of Educational Attain-mein_far_Whitts_iath_Mfalf.Blacksburg, VA: Virginia Polytechnic Institute andState University. ED 207 905

This study begins with the premise that differ-ences in educational attainment between whites andMexican-Americans may be the result of either so-cial or educational attainment processes. Data aredrawn from the NLS-72 for a sample of white (av-erage n=11,743) and Mexican-American (averagen=493) respondents. The model of educational at-tainment used in the analysis is based on the idea thatattainment is related to father's occupation andstatus, mother's education, number of siblings, sex,ability, academic preparation and college plans.

6 "

Results indicate that more similarities than dif-ferences exist between the groups and that the dif-ferences are relatively unimportant.

166Adelman, Clifford. 1991. Women at Thirtysometh-ing: Paradoxes of Attainment. Washington, DC:

/U.S. Department of Education, Office of Educa-tional Research and Improvement.

For description see entry #120.

65

See AttainmentEducational and Occupational

VIII. Gender and Racial-Ethnic Differences

167Anderson, Kristine L. 1988. "The Impact of Col-leges and the Involvement of 1' Tale and FemaleStudents." Sachkgysi_EgInsangn a (July): 160-177.

For description see entry #124.

See AttainmentEducational and Occupational

168Anderson, Kristine L. 1984. Race Differences in theEffects of College Characteristics on EducationalAttainment. Washington, DC: U.S. Department ofEducation, National Institute of Education. ED 256249

For description see entry #69.

See College/Program Effects

169Behuniak, Peter Jr., and Gable, Robert K. 1980. MeInfluence of Sex and College Major on Students'Attituues About Future Employment and Satisfac-ticamith_folleg4. Washington, DC: National Sci-ence Foundation. ED 191 417

A sample of 5,624 NLS-72 respondents whowere undergraduate seniors in the areas of business,education, humanities, fine arts, physical and bio-logical sciences, math, and the social sciences wasused to examine sex-related differences in the rela-tionship between student attitudes and choice ofundergraduate major. The authors looked specifi-cally at differences across majors in attitudes aboutfuture work and satisfaction with academic pro-gram, and the nature of the relationship betweenstudents' attitudes and their satisfaction with col-lege.

Findings indicate that attitudes about futureemployment vary across majors and differ for menand women. Satisfaction with college appears to berelated to attitudes about future work with gender asa moderating factor.

170Dawkins, Marvin P. 1986. "A Longitudinal Analy-sis of Variations in Mobility Goals Across Race-SexGroups." Western Journal of Black Studies 1(1(1):34-43.

For description see entry #129.

67

See AttainmentEducational and Occupational

171Daymont, Thomas N., and Andrisani, Paul J. 1984."Job Preferences, College Major and the GenderGap in Earnings." Journal of Human Resources12(3): 408-427.

For description see entry #130.

See AttainmentEducational and Occupational

172Epps, Edgar G., and Jackson, Kenneth W. 1985.Educational and Occupational Aspirations andEarly Attainment of Black Males and Females,Atlanta, GA: Southern Education Foundation. ED263 818

For description see entry #134.

See AttainmentEducational and Occupational

173Gerstein, Marti, Lichtman, Marilyn, and Barokas,Judy U. 1988. "Occupational Plans of AdolescentWomen Compared to Men: A Cross-Sectional Ex-

6 9

amination." Career Development Quarterly 3_6(March): 223-230.

For description see entry #136.

See AttainmmtEducational and Occupational

174Gruber, James E. 1980. "Sources of SatisfactionAmong Students in Postsecondary Education."American Journal of Education asT. 320-330.

Using a sample of 3,700 black and white malerespondents to the NLS-72 who are either voca-tional/technical, junior college, or university stu-dents in October 1973, the author examines theeffects of several variables on educational satisfac-tion. The author looks specifically at how thesefactors relate to educational satisfaction.

Findings indicate that self-esteem has a verystrong influence for blacks, who had a greater self-esteem than whites. The type of school had no effecton satisfaction for white students and a small nega-tive effect on satisfaction for black students. Overall,the best predictor of satisfaction for both white andblack students was academic achievement.

175Lyson, Thomas A. 1984. "Sex Differences in theChoice of a Male or Female Career Line. An Analy-sis of Background Characteristics and Work Val-ues." Work and Occupations: An InternationalSociological Journal I1(2): 131-146.

This study examines a set of NLS-72 (baseyear through 1976 follow-up) respondents who se-lected either a predominantly male or a predomi-nantly female major. The article summary is asfollows:

Using data from the NLS-72, this article iden-tifies similarities and differences in social back-ground characteristics, academic ability, and workvalues among men and women in sex-atypical ma-

jors and sex-atypical curricula. The research wasdesigned to examine two implicit assumptions in theliterature about the choice of a sex-atypical major:(1) women in traditionally male areas manifest char-acteristics more closely associated with men in thesesame areas than with women in traditionally femaleareas; and (2) men in traditionally female curricu-lums are more like women in these areas than theyare like men in traditionally male areas. Resultsshow that men and women in sex-atypical majors aremore like their sex peers than like their curriculummates. The implications of these findings for main-taining gender inequalities in the labor market arediscussed.

68

176Rothschild, Susan J.S., and Lichtman, Marilyn.1980. FactWomen' s Mathematics Attainments- A DescriptiveStudy. ED 195 570

I. -I I U/I :

For description see entry #153.

1 411 i.t

See AttainmentEducational and Occupational

177Thomas, Gail E., and Gordon, Samuel A. 1983.

. 1 S S 1 11 1

Black. White. and Hispanic Students, Report No.344. Baltimore, MD: Johns Hopkins University,Center for Social Organization of Schools. ED 235733

This paper examines the effects of college typeand major on subsequent educational and occupa-tional attainments of a sample of students who at-tended postsecondary educational institutions in1972-73. The sample consisted of white, black, andHispanic participants in the base year through fourthfollow-up surveys of the NLS-72 who attended post-secondary institutions, including 2-year/vocationalschools and 4-year schools, in 1972-73. Separateanalyses are done by race, gender, and institutiontype for educational attainment and by race andgender for occupational attainment.

7 0

Findings of a preliminary analysis showed col-lege type and major as having significant, positiveeffects on educational attainment while college typehad a small but significant effect on occupationalattainment. College major, grades, educational ex-pectations, and high school grades all had significanteffects on educational attainment for 4-year collegestudents. College major did not have significanteffects on educational attainment for any race or sexgroups at 2-year colleges. Neither college type normajor had significant effects for ethnic groups ormales. Educational attainment appeared to be themost important determinant of occupational attain-ment.

178Wilder, Gita Z., and Powell, Kristin. 1989. SexDifferences in Test Performance. A Survey of theLiterature. College Entrance Examination BoardReport No. 98-3. ED 308 231

This report examines current data on sex dif-ferences in test performance and several of the hy-potheses and evidence concerning possible causesof the differences. Topics covered include: examin-ing differential performance of men and women onvarious tests; exploring various correlates andcauses of the reported differences (such as biologi-cal, social, psychological, demographic, educationalfactors and test characteristics); and identifying ar-eas for continuing and future research. Researchcovered in the report addresses undergraduate ad-missions tests, validity studies, national studies, ver-bal ability tests, quantitative ability tests, andvoluntary testing programs. Specific testing pro-grams discussed include NAEP, NLS-72, HighSchool and Beyond and the SAT.

Data from both graduate and undergraduateadmissions tests support the hypothesis that men arebetter at quantitative tasks, however, data does notsupport the hypothesis that women hold a consistentadvantage over men on verbal tasks. Overall, find-ings indicate that differences between men andwomen are gradually diminishing. The topic, how-

ever, continues to be one of concern due to both thesocial consequences of differences in performanceand the effects of test performance on motivation,attitudes and behaviors.

See Testing

179Wolf le, Lee M. 1985. "Postsecondary EducationalAttainment Among Whites and Blacks." AmerionEducational Research Journal 22(4): 501-525.

For description see entry #164.

See AttainmentEducational and Occupational

180Wolf le, Lee M. 1983. "Effects of Higher Educationon Ability for Blacks and Whites." Paper presentedat the Annual Meeting of the American EducationalResearch Association, Montreal, Canada, April 11-15. ED 230 148

This study examines the impact of college onknowledge of vocabulary and math as measured byscores from the "Fourth Follow-Up" of the NLS-72.Data are drawn from a sample of 1,738 whites and451 blacks.

Findings indicate that the impact of college onboth math and vocabulary is similar for whites andblacks and that postsecondary education, in general,increases performance in both vocabulary andmathematics. The best predictors of test scores werethe scores of the same tests from high school, al-though correlations show that all predictor variablesare positively related to vocabulary and math skills.

181Yang, Shu-O W. 1981. "Rural-Urban Differences inSex Role Orientations." Paper presented at the An-nual Meeting of the Rural Sociological Society.Guelph, Ontario, Canada, August 19-23. ED 207764

69 71

Data from NLS-72 are used to test two hy-potheses: (1) rural young adults have more tradi-tional sex-role orientations than do urban youngadults; and (2) the rural-urban differentials in sex-role expectations will not be supported after control-ling for major social characteristics which haveeffects on sex-role attitudes. The "Base Year" sur-

70

vey provides background information, and otherdata is taken from the "Third Follow-Up."

Data support the first hypothesis, but not thesecond, although the difference between urban andrural youth in sex-role orientations is relativelyweak.

72

IX. Vocational Technical and Vocational Educational

182Conroy, William G., Jr. 1980. "The EconomicEffects of Vocational Education." VocEd 55(7):39-41.

The author discusses the economic benefits ofvocational education. Value is examined in terms ofeconomic and social benefits. A section of the articlelooks at what various studies and databases, includ-ing the NLS-72, appear to say about the value ofvocational education in terms of earnings differen-tials for students of different ability and by highschool curriculum.

See Economics of Education

183Grubb, W. Norton. 1989. Access.,_Adnel.Qment,Completion. and "Milling Around" in Postsecon-dary Vocational Education. Washington, DC: U.S.Department of Education, National Assessment ofVocational Education. ED 315 548

For description see entry #140.

See AttainmentEducational and Occupational

184Gustman, Alan L., and Steinmeier, Thomas L. 1982."The Relation Between Vocational Training in HighSchool and Economic Outcomes." Industrial andLabor Relations Review 16(1): 73-87.

The two primary purposes of this study are (1)to determine whether enrollment alone in a voca-tional program had an impact on one of the labormarket outputs; and (2) to determine whether theactual number of semester hours of vocationalcoursework taken had an impact on one of the labormarket outputs (mean weekly earnings 3-4 yearsafter high school, mean weeks employed, hourlywages). Data were drawn from the NLS YouthSurvey (n=167 participants with 12 years of schooland are 21 years old) and the NLS-72, "1976 Fol-low-Up" survey (n=2,405 participants). Respon-

dents enrolled in higher education or the militarywere excluded from the overall sample.

Findings show no significant differences be-tween black or white male graduates of vocationalprograms and those in general programs in terms ofoutcome measures. NLS Youth data show, for whitewomen, a significant, positive effect of business-commercial programs on weekly and annual earn-ings when compared to those earnings in generalprograms.

185Horn, Laura, and Carroll, C. Dennis. 1989. Enroll-ment. Completion. Attrition. and VocationalCourse-Taking Patterns in Postsecondary Educa-tion: A Comparison of 1972 and 1980 High SchoolGraduates Entering 2-Year Institutions. HighSchool and Beyond. National Longitudinal Study of1972. Washington, DC: U.S. Department of Edu-cation, National Center for Education Statistics. ED309 820

For description see entry #143.

See AttainmentEducational and Occupational

186Mertens, Donna M., and Gardner, John A. 1982.Educational and Employment Experiences of theYounger Adult Worker. Washington, DC: U.S. De-partment of Education, Office of Vocational andAdult Education. ED 215 210

For description see entry #149.

See AttainmentEducational and Occupational

187West, Jerry. 1989. The Postsecondary VocationalEducation of 1980 High School Seniors. The 2-Year Associate of Arts Degree. Survey Report.Washington, DC: U.S. Department of Education,National Center for Education Statistics. ED 309821

For description see entry #163.

See AttainmentEducafional and Occupational

188Wiley, David, E., and Harnischfeger, Annegret.1980. Curricular Resource Allocation and CourseExposure in Secondary Schools. Chicago, M-LGroup for Policy Studies in Education. ED 218 301

For description see entry #57.

See School/Program Effects

72

189Woods, Elinor M. 1980. "National LongitudinalStudies and Data Sets." VocEd 51(7): 35-38.

This article reviews the effectiveness of sev-eral nationally representative longitudinal studiesand datasets in examining the effects of vocationaleducation. The five datasets are Project TALENT,Youth in Transition, the National Longitudinal Sur-veys of Labor Market Experience (NLS Youth), theNLS-72, and the National Longitudinal Survey ofYoung Americans (NLS New Youth Cohort). Theauthor' s two main conclusions were that no uniquetype of vocational education program exists and thatvocational education students are much differentthan those who enroll in other programs (in terms ofSES, ability level, and education after high school).

7 4

X. Teachers

190Benton, Cynthia J. 1985. "Predicting OccupationalPersistence: A Comparison of Teachers and FiveOther Occupational Groups." Paper presented atthe Annual Meeting of the American EducationalResearch Association, Chicago, IL, March 31-April 4. ED 255 522

This study examines the effects of severalvariables in predicting occupational turnover in the3-year period following postsecondaty graduation.Teaching, nursing, social work, accounting, engi-neering and computer science are the six fieldsexamined. Data are taken from the NLS-72 baseyear through fourth (1979) follow-up surveys.

Results show unequal distribution by genderwith more women in teaching, nursing, and socialwork. Occupational turnover is lowest for nursing(15 percent) and highest for social service (74 per-cent). Turnover for the technical fields is between39 percent and 54 percent; turnover for teaching is35 percent. The author believes these figures reflectnational norms of high job turnover early on incareer development. In terms of persisters within afield, quality of life, needs, commitment, and expec-tancy are important predictors of persistence forteachers, who seem to be motivated by intrinsicrewards. Extrinsic factors, such as size of work-place, income, and job satisfaction are more impor-tant predictors of persistence in the other five fields.

191Bethune, Stuart B. 1983. "Recruitment to TeachingDuring College: Results from the NLS." Paperpresented at the Annual Meeting of the AmericanEducational Research Association Montreal, Can-ada, April. ED 228 234

This study examines the effects of backgroundcharacteristics and social-contextual influences onstudents' decisions to enter or leave undergraduateprograms in teacher education. A comparison ofstudent behavior in both similar (traditionally fe-male) and dissimilar (traditionally male) fields isincluded. Specifically, patterns of movement be-

73

tween the freshman and senior year are observed.The sample includes 848 white, female respondentsto the NLS-72, who began 2- or 4-year colleges inthe fall following high school (1972) and remainedcontinually enrolled through October 1976.

Results indicate that freshmen women choos-ing to major in education tended to be of lowerability, lower level of high school performance,lower SES; to have lower self-concept; to attend lessselective colleges; and to come from rural areas.Ability, class background, and selectivity of collegeattended had the strongest, direct, negative effectson choosing education. Education majors were notmuch different from students in similar fields (homeeconomics, health, and public services) in terms ofability, background, residence, self-concept, highschool performance and selectivity of college at-tended. Those students leaving education most oftenchose fields in the social sciences. Those enteringeducation after the freshman year came primarilyfrom the humanities and fine arts fields.

192Hafner, Anne, and Owings, Jeffrey. 1991. Careersin Teaching: Following Members of the HighSchool Class of 1972 In and Out of Teaching. Wash-ington, DC: U.S. Department of Education, Na-tional Center for Education Statistics. ED 336 386

This study looks at career patterns of teachersor potential teachers in terms of their backgrounds,education, teaching qualifications and experiences,attitudes regarding teaching and job satisfaction.Variables which may predict retention in the fieldare also identified. A sample of 1,000 respondentsto the "Fifth (1986) Follow-Up" of the NLS-72 whohave ever taught elementary or secondary school orwho were trained but never taught were used in theanalysis.

Findings indicate that 55 percent of the sampletaught for 10 consecutive years; 21 percent of thegroup never entered teaching while 37 percent leftteaching within the first 5 years. A significant num-ber of individuals moved in and out of teaching with

an average annual attrition rate of 7 percent. Thereis no evidence, however, of an increase in the num-ber of teachers leaving the field. Variables whichappear to be good indicators of retention are thosemeasuring human capital. Number of years teach-ing, job satisfaction, teaching in a public school,number of continuing education activities and num-ber of education credits also predict retention in thefield.

193Heyns, Barbara. 1988. "Educational Defectors: AFirst Look at Teacher Attrition in the NLS-72."Educational Researcher (April): 24-32.

This paper presents preliminary analysis, spe-cifically focusing on issues of teacher attrition andretention, of the special supplement included in the"Fifth (1986) Follow-Up" of the NLS-72 for presentor former teachers or for those who were trained tobe teachers. Findings indicate that retention amongteachers in the last decade has increased and thatability and qualifications of teachers have not de-clined over time. Based on the results of the study,the author makes three "fairly controversial claims":

(1) Teacher turnover is higher than levels ofattrition suggest.

(2) Former teachers are quite similar to currentteachers, except that they have typically been em-ployed in school-work settings that may be aboveaverage. Also, patterns of entry and re-entry into theteaching profession may indicate erratic careers forthe "best" teachers (as judged by their academicqualifications).

(3) Results of the study have implications forthe direction of educational policies designed toenhance teacher professionalism or to link schoolreforms to the retention of teachers.

194Lyson, Thomas A., and Falk, William W. 1984."Recruitment to School Teaching: The Relation-

74

ship Between High School Plans and Early AdultAttainments." American Educational Researchkurnal a( 1 ): 181-193.

This study examines the nature of the relation-ship between teaching plans and actually becominga teacher. Data from the NLS-72 base year throughfourth follow-up surveys are divided into six ana-lytic subgroups: (1) achieversthose planning toteach and who were teaching 7 years later; (2) de-fectorsthose who reported teaching plans but heldprofessional/managerial positions 7 years later; (3)dropoutsthose expecting to become teachers butheld nonprofessional/managerial jobs; (4) con-vertsthose expecting to have professional mana-gerial positions but ended up as teachers 7 yearslater; (5) climbersthose not reporting professionalmanagerial occupational expectations but heldteaching positions in 1979; and (6) professionalsthose who did not hold teaching plans in 1972 butheld nonteaching professional/ managerial positionsin 1979. Within these groups, the authors examinefactors involved in plans to become a teacher, suchas race, sex, residence, place, academic ability andsocial class origin.

Results show that almost half the teachers ihthe 1979 NLS sample had not reported teachingplans in 1972. Also, over three-fourths of thoseplannipg to enter teaching in 1972 were not teaching7 years later, suggesting that occupational plans atage 17 or 18 may not be good predictors of earlyadult occupational attainment and also may reflectmarketplace demands for teachers between 1972and 1979. Because more than half of the respondentsplanned to be in a profession lower in prestige thanteaching, the authors suggest, especially for women,nonwhites and those from rural areas, that teachingmay be used as an "upward boundary of adolescentachievement."

195Nelson, Howard F. 1985. "New Perspectives on theTeacher Quality Debate: Empirical Evidence fromthe NLS." Journal of Education Research 28(3):133-140.

7

The study seeks to evaluate the academic abil-ity of teachers due to concerns about the quality ofteacher education and the decline in teacher quality.A total of eight aptitude measures were used toevaluate the academic ability of those who taughtelementary and secondary schools between fall of1976 and fall of 1979 (including SAT, H.S. rankpercentile and the NLS-72 ability tests).

Principal findings indicate that (1) teachersemployed in nonpublic schools have higher verbaltest scores, with the loss of the most able teachersoccurring primarily in the nonpublic sector; (2)while less than 3 1/2 percent of college graduatesmajoring in education come from the bottom quarterof their high school class, they are just as likely tofind jobs as teachers and to get paid as much as otherteachers; and (3) less than 25 percent of those whobecome teachers planned to major in education ashigh .chool seniors. "The latter finding, in particularsuggests some public policy may be fallaciouslyguided by widely cited research based only on thestudy of prospective education majors among col-lege-bound seniors."

These results have several implicatic ns regard-ing policies for improving teacher education andtraining. "First, reliance on the SAT for measuringability, failure to weight survey data, and focusingon the test scores of college-bound high school sen-iors cause overestimation of the severity of theteacher quality problem. Second, this study providesevidence that the problem of retaining the best teach-ers may be overstated. Third, policies directed atimproving the academic ability of new teacherswould be more effectively focused on recruiting able

college students to teaching rather than on collegeentrance."

The author includes several limitations of thisstudy such as the poor job market for teachers at thistime, so the NLS-72 "data cannot effectively be usedto determine whether or not the dim job prospectskept more of the bright students out of the professionand left only the most committed and able studentsto seek teaching careers, or caused education facul-ties to le(mer admissions standards."

196Sebring, Penny, and others. 1987. Tht_NatismalLongitudinal Study of the High School Class of 1972(NLS-72). "Fifth Follow-Up" £1986), Teaching

. U.S. Depart-ment of Education, National Center for EducationStatistics. ED 292 891

.-I 11 1 I. 14

For description see entry #17.

See Data File User's Manual

197Seyfarth, John T., and Ranson, James T. 1989. "Oc-cupational Stayers and Strayers: Career and SalaryExperiences of Current and Former Teachers." Pa-per presented at the Annual Meeting of the AmericanEducational Research Association, San Francisco,CA, March 27-31. ED 314 817

For description see entry #154.

See AttainmentEducational and Occupational

75 7 7

XL Marriage/Family and Life Transitions

198Cervero, Ronald M., and Kirkpatrick, Thomas E.1990. "The Enduring Effects of Family Role andSchooling on Participation in Adult Education."American Journal of Education 92(1): 77-93.

For description see entry #92.

See Patterns of Attendance

199Goldscheider, Frances Kobrin. 1985. MeasuringLife-Cycle Transition in Young Adulthood; ACross-Validation of Data from Three ProspectiveSurveys. Santa Monica, CA: Rand Corporation.ED 266 384

This document examines three important na-tional surveys that provide periodic longitudinaldata on the transition to adulthood during the late1960s and 1970s. The surveys are the NationalLongitudinal Study of Labor Market 17.xperience(NLS Youth), initiated in 1966 and 1968; the PanelStudy of Income Dynamics, initiated in 1968; andthe NLS-72. Section I of thi paper provides anintroduction to the three datasets and describes twoprojects in the Rand Corporation Labor and Popula-tion Program which build on the complementaritiesof these datasets. Section II outlines the specialcharacteristics of each survey that affect the con-struction of appropriate cohorts for comparison andcompares them on such background characteristicsas age at survey, race and parental education. Sec-tion III presents the constructions of the four meas-ures of life-cycle progress: first marriage, departurefrom the parental home, school enrollment, andmilitary service. Section IV shows how the fourmeasures compare acrosc surveys. Section V dis-cusses the results and tneir implications for thecomplementary use of these datasets and for theprospective collection of information on life-cycletransitions. Background tables are appended andreferences are given. (Mapted from ERIC abstract)

200Haggstrom, Gus W., and others. 1984. atangeainthe Lifestyles of New Parents. Bethesda, MD: Na-tional Institutes of Health, National Institute ofChild Health and Human Development. ED 260 813

This study examines how becoming a parentaffects the lives of young men and women duringthe 7-year period following their graduation fromhigh school. The study is based on data from theNLS-72. The research focuses on two subpopula-tions of the NLS: the over 8,000 participants whobecame parents by late 1979 and the approximately5,000 who married but were childless at that time.Primary interests of the investigation were the edu-cation and employment activities of the youngadults. The study also explored the affects of parent-hood on participation in political affairs and com-munity groups, marital stability, financialwell-being, and several attitudinal measures, includ-ing educational and career expectations. The princi-pal fit 'ing was that young couples who becameparents in the 1970s, like those in previous decades,modified their lifestyles in accordance with tradi-tional family rolesthe women serving as home-makers and mothers and the men serving asbreadwinners. On the whole, these young adultshave met the demands and responsibilities of par-enthood without changing their views of who theyare or what they wish to accomplish. (ERIC ab-

stract)

201Haggstrom, Gus W., and others. 1981. TeenageParents: Their Ambitions and Attainments. Be-thesda, MD: National Institutes of Health, NationalInstitute of Child Health and Human Development.ED 216 273

This report, using data from the base yearthrough 1976 follow-up surveys of the NLS-72,presents the results of a study of the impact of earlyfamily formation on young people's lives. Follow-ing a discussion of methodology and data, the report

77 7S

examines the effects of early family formation onpostsecondary education, career development,women' s orientation toward homemaking, depend-ency on parents and society, economic charac-teristics, fertility and attitudes. By distinguishingbetween early parenthood and early marriage with-out children, this report reveals that many of theeffects that might be attributed to early parenthoodalso show up as effects among nonparents whomarried at approximately the same time. Althoughthe materials stress that the data apply to parenthoodin the late, rather than early, teens, the report chal-lenges the notion that early parenthood, by itself, hasa dest uctive impact on the ambitions and attain-ments of teenagers. The report concludes that teen-age parenthood should be regarded as only oneaspect of a broader process of early family forma-tion. The six appendices provide definitions of vari-ables, a model for estimating effects, backgroundfactors affecting early parenthood, a regression in-dex, and sex-role attitude measures. A bibliographyis also included. (Adapted from ERIC abstract)

202Kobrin, Frances. 1981. High School Seniors andHigh School Dropouts: An Evaluation of Life CycleBias in the National Longitudinal Study of the HighSchool Class 1972. Santa Monica, CA: Rand Cor-poration. ED 219 394

This study uses the March 1972 Current Popu-lation Survey (CPS) and the NLS-72 data to notedifferences in several life-cycle dimensions, includ-ing marital, parental, and employment status. Theauthor chose to use CPS because the NLS data aresomewhat biased (only high school graduates areincluded in the dataset).

Results show small differences between highschool seniors and dropouts in terms of employmentpatterns. Findings also indicate that more dropoutswere married and had children by the spring of theirsenior year and that more of them no longer livedwith their parents.

203Owings, Jeffrey, and others. 1980. "The Impact ofMotherhood on the Career Pattern of WorkingWomen: Differences in the Life Histories and

Career Status of Employed Mothers and WorkingWomen Without Children." Paper presented at theAnnual Meeting of the American EducationalResearch Association, Boston, MA, April 1. ED 201830

78

This study compares employed mothers withworking women who do not have children alongdimensions reflecting their life histories and labormarket experiences. It uses data from the NationalLongitudinal Survey of Youth and the NLS-72 baseyear through 1976 follow-up surveys. The analysisfocused on differences in socioeconomic back-ground, educational attainment and goals, occupa-tional status and aspirations, and personal attitudesof employed mothers and employed women withoutchildren. A regression analysis was undertaken todetermine which variables would be predictors ofthe dichotomous variable indicating whether awomen is a mother, when their overlapping effectsare controlled. Statistical analyses did not oontradictthe suggestions of theorists in vocational develop-ment that employed mothers have many obstacles tosurmount as their careers develop. They were morelikely to come from lower socioeconomic levels, tohave completed fewer years of education, and to beavailable to work only part-time. (Adapted fromERIC abstract)

204Powell, Brian, and Steelman, Lala Carr. 1990. "Be-yond Sii,ship Size: Sibling Density, Sex Composi-tion, and Educational Outcomes." Social Forces62(1): 181-206.

For description see entry #150.

See AttainmentEducational and Occupational

205Rindfuss, Ronald R., and Stephen, ElizabethHervey. 1990. "Marital Noncohabitation: Separa-tion Does Not Make the Heart Grow Fonder " Jour-nal of Marriage and the Family 52 (February):259-270.

This study uses both the 1976 and 1979 follow-

up surveys of the NLS-72 and Census data to exam-ine factors that may explain marital noncohabitation

7

and its possible consequences. The article abstract isas follows:

A variety of situations may arise in whichhusbands and wives live in separate households,temporarily or permanently, for reasons other thanmarital discord. Such living arrangements may havefar-reaching implications, and perhaps the most im-portant question is whether living apart from one' sspouse leads to divorce. The percentage of currentlymarried persons living apart in the United States ishighest for ages 18-24 and for blacks. The two mostcommon identifiable reasons for husbands andwives not living together are military service andincarceration. We found that those living apart fromtheir spouses in 1976 were nearly twice as likely toexperience a marital dissolution within 3 years, com-pared with persons cohabiting with their spouses.

206Rindfuss, Ronald R. 1991. "The Young Adult Years:Diversity, Structural Change, and Fertility." De-mography 2$(4): 493-512.

This study looks at the young adult years of asample of NLS-72 participants in terms of familyactivities, school/work activities, the impact of onesphere of activities on the other and the extent thatactivities from each overlap. Effects of the order ofthese activities are also investigated. Fertility ratesfor the sample are also examined.

Findings indicate that the young adult years are"demographically dense" (high incidence of demo-graphic action such as marriage, moving, leavingschool, childbearing) and diverse in terms of thenumerous roles young adults experience. Resultsalso show that more young adults are "disorderly"in the school/work sphere in that they did not followexpected routes of school then work or continuouswork. Disorderly sequences in the school/worksphere are more common for women than men.Trends in marital dissolution and out-of-wedlockbirths indicate an increase in disorderly sequences inthe family sphere. Examining fertility notes an in-crease for all age groups between 1974 and 1987.Also, the fertility rates for unmarried women in-creased faster than the fertility rates for marriedwomen.

207Rindfuss, Ronald R., Swicegood, C. Gray, andRosenfeld, Rachel A. 1987. "Disorder in the LifeCourse: How Common and Does it Matter?"American Sociological Review (December) 52:785-801.

79

This study examines the transition from earlyadulthood to parenthood by assessing the impact ofthe ordering of family and nonfamily activities onparenthood. Data are taken from the NLS-72 baseyear through 1979 follow-up surveys. The articleabstract is as follows:

Many researchers conceptualize the majortransitions in the life course as occurring in an or-derly progression. Using data from the NLS-72 andits follow-ups, we fmd considerable disorder in non-family events. By the time they had been out of highschool 8 years, over half of the men and women inthe 1972 class had sequences of states that deviatedfrom the "normal" (e.g., they returned to school ormoved to a category not usually included in life-course research, such as employment). To what ex-tent does this disorder matter? We answer thisquestion with respect to the transition to parenthood.A simple measure of disorder did not uniformlyaffect this transition. Rather, disorder has heteroge-neous effects. Particular sequences of activities,some orderly, some not, affect when people firstbecome parents. For example, delaying the effect ofeducation is less powerful if that education has beer,interrupted by work or some other activity.

208Teachman, Jay D. 1990. "Socioeconomic Resourcesof Parents and Award of Child Support in the UnitedStates: Some Exploratory Models " Journal of Mar-riage. and the Family (August): 689-699.

Data from the "Fifth Follow-Up" of the NLS-72 are used to examine factors affecting the deter-minants of child support awards and amounts. Thearticle summary is as follows:

The American system of family and kinship isundergoing considerable change. Decreasing stabil-ity in the nuclear family has called into question therole of parents in providing for children across sepa-

rate households. A significant proportion of absentparents do not make economic contributions to thewell-being of their children. As a consequence, chil-dren living in single-parent homes run a greater riskof living in poverty than children in two-parenthomes. This study examines the determinants ofchild support awards and amounts for a sample ofever-divorced women from the NLS-72. The argu-ment is divorce influences the amount and award ofchild support. The results indicate that socioeco-nomic resources are significant predictors of childsupport outcomes.

209Teachman, Jay D., and Polonko, Karen. 1990. "Ne-gotiating Divorce Outcomes: Can We Identify Pat-terns in Divorce Settlements?" Journal of Marriageand the Family 52: 129-139.

This study uses data from a sample of 570white and 94 African-American respondents fromthe "Fifth Follow-Up" of the NLS-72 to describeobserved patterns in divorce outcomes. The articleabstract is as follows:

As nuclear families have become increasinglyless stable over the past quarter of a century, marriedcouples have been faced with the difficult task ofnegotiating the terms by which marriage is ended.For parents, divorce involves questions revolvingaround the care and sustenance of children. In thisstudy, we investigate whether observed divorce set-tlements suggest that parents make trade-offs be-tween several dimensions related to their children,including custody, visitation, child support, and themarital property. Using data from the NLS-72, wefind evidence consistent with the notion that parents

80

make trade-offs. We also suggest a frameworkwithin which these results can be interpreted.

210Waite, Linda J., and others. 1986. The Effects ofParenthood on the Career Orientation and Job Char-acteristics of Young Adults. Bethesda, MD: Na-tional Institutes of Health, National Institute of ChildHealth and Human Development. ED 294 044

For description see entry #162.

See AttainmentEducational and Occupational

211Waite, Linda J., Haggstrom, Gus W., and Kanouse,David E. 1985. "The Consequences of Parenthoodfor the Marital Stability of Young Adults." Ameri-can Sociological Review al (December): 850-857.

The study examines the effects of a first birthon short-run stability of marriage compared to sta-bility with estimates of what would have happenedwithout the occurrence of a birth. Data are takenfrom NLS-72 for 2,901 women and 1,979 men whobecame parents between October 1972 and October1979 and whose months of marriage and entry intoparenthood were known. A comparison group con-sisted of 2,807 women and 2,474 men who marriedbut did not have children through October 1979 (orOctober 1976 if no response to the "1979 Follow-Up") and who were either divorced or separated.

Results show that only between 5 percent and8 percent of parents actually divorced or separatedcompared to 20 percent who would have divorcedor separated in the absence of parenthood, indicatingthe influence of children on marital stability.

XII. Economics of Education

212Altonji, Joseph G. 1990. Accounting for UncertainEducational Outcomes in Estimating the Return toEducation. New York, NY: National Center onEducation and Employment; Santa Monica, CA:Rand Corporation. ED 333 820

A prototype model was developed to estimatethe impact of several variables on the internal rateof return to starting college. The model arose fromobservations of a nonlinear relationship betweeneducation and earnings, that many students whobegan college never finished, that large differencesin earnings exist between college and high schoolgraduates across fields of specialization, and thatmany high school and family background charac-teristics appear to affect earnings. The model wasthen applied to data from the NLS-72. Results foundthat black males had a lower rate of return on 1 yearof college than white males (5.4 percent versus 5.9percent), which appears, however, to be due not torace but to other characteristics. Females also had alower rate of return than men. It was noted that noneof these differences are explained by dissimilaritiesin the observed characteristics of males and females.Other research using this model has shown a higherreturn on college attendance for women over men.It is suggested that further research and refinementof the model is necessary before strong conclusionscan be developed. (Adapted from ERIC abstract)

213Benson, Charles S., Ibser, Maya H., and Klein,Steven G. 1991. Ecnnomic Returns to VocationalEducation and Other Types_of_Qcopational Train-ing. Berkeley, CA: University of California,Berkeley.

This study examined the economic returns tovocational education and other types of occupa-tional training in terms of labor market outputs(wage and salary earnings and human capital in-creases) and other outputs such as social returns(reduction of taxpayer transfers to unemployed, im-

proved living conditions, reduced crime and home-lessness) for high school graduates who did not goon to college and for students who attended college.Data are drawn from the NILS-72, the National Lon-gitudinal Survey of Labor Market Experience (NLSYouth), and High School and Beyond.

Findings indicate that vocational education ex-hibits its greatest outcomes if and when it is followedby employment in a related field. Care must be takento examine the effects of vocational education bytype of program, race, and gender because of differ-ences in outcomes. (Article summary)

214Campbell, Paul B., and Basinger, Karen S. 1985.Economic and Noneconomic Effects of AlternativeTransitionsThroughSchoolio_Work. Columbus,OH: Ohio State University, National Center forResearch in Vocational Education. ED 254 638

81

For description see entry #40.

See School-to-Work Transition

215Conroy, William G., Jr. 1980. "The Economic Ef-fects of Vocational Education." yacad 5_5(7): 39-41.

For description see entry #182.

See Vocational Technical and Vocational Educa-tion

216Datcher-Loury, Linda, and Loury, Glenn C. 1983The Effects of School Quality on Earnings. Wash-ington, DC: U.S. Department of Education, Na-tional Institute of Education. ED 267 083

For description see entry #44.

See School/Program Effects

217Inger, Morton, and Larsen, Eric. 1990. "The Deci-sion to Start College: Factoring In the Uncertainty."L IS .11 Hell 11-$

Brief No. 11, November 1990. ISSN 0889 8049

Using information from two papers by JosephAltonji ("The Demand for and Return to EducationWhen Education Outcomes are Uncertain," Tech-nical Paper No. 21. New York: National Center onEducation and Employment, 1991, and "Accountingfor Uncertain Educational Outcomes in Estimatingthe Return to Education," Technical Paper No. 15.New York: National Center on Education and Em-ployment, 1990.), the authors discuss the importanceof considering the effects of uncertainty of educa-tional outcomes on the decision to pursue postsec-ondai: education. Included in the discussion areAltonji ':. calculation of rates of return that highschool gra -luates might expect if they choose topursue higher education and differences in returnbetween several classificatiuns of high school stu-dents such as men and women or students of differ-ent ability levels.

Altonji' s findings demonstrate greater eco-nomic benefits for women who start college com-pared with men who start, but show greater expectedbenefits from college completion for men. In terms'of family background and coursework, having afavorable family background or being in an aca-demic high school curriculum affects economic re-turns of men. For women, simply attending collegeaffects economic returns. Altonji also found signifi-cant differences in returns between high and low-

ability men, with low-ability men receiving feweconomic benefits from starting college comparedto those who entered the labor market directly afterhigh school. The authors conclude that Altonji'smodel can be useful in examining the effects ofseveral variables on the probability of realizing cer-tain educational outcomes.

82

218Kolstad, Andrew. 1982. Does College Pay? WageRates Before and After Leaving School. Washing-ton, DC: U.S. Department of Education, NationalCenter for Education Statistics. ED 224 438

This study looks at the relationship betweeneducational choice and future salary levels usingdata from the NLS-72. The sample consisted of17,366 respondents to the base year through 1979follow-up surveys.

Findings reveal a strong relationship betweeneducational attainment and salary, as higher startingsalary was associated with greater amounts ofschooling. In general, men who did not attend post-secondary alucation have a greater advantage incatching up to men who attended college thanwomen who did not attend college have in receivingsalaries equivalent to women college graduates. Forall levels of education, however, mens' wages werefound to be higher than womens'. Findings alsoindicate crossover points, or points where wages arerelatively equal for those attending postsecondaryeducation and those who did not. The age of cross-over was 25 for men and 22 for women.

XIII. Testing

219Bekhuis, Tanja CH.M. 1988. "The Estimation ofTrue Scores for Tests Not Taken: A SimulationStudy." Paper presented at the Annual Meeting ofthe Eastern Educational Research Association, Mi-ami Beach, FL, February 24-27. ED 304 477

The author tests an Educational Testing Serv-ice (ETS) procedure, which is based on item re-sponse theory and estimated true scores on tests nottaken. The reading, vocabulary, and mathematicstests of high school seniors from the NLS-72 and theHigh School and Beyond 1980 and 1982 seniorcohorts are used in the analysis. The author con-cludes that the procedure should remain experimen-tal. "In particular, the method should not be used forassessing cohort differences until the likely direc-tion and magnitude of bias, given the constraints ofany specific study, are well understood."

220Crouse, James. 1986. "Should a Million and a HalfStudents Be Required to Take the SAT Next Year?"Phi Delta Kappart LI (5): 345-352. See also Crouse,James. 1985. "Does the SAT Help Colleges MakeBetter Decisions?" Harvard Educational Review55(2): 195-219.

This article, adapted from a research article(Crouse, James. 1985. "Does the SAT Help Col-leges Make Better Decisions?" Harvard Educa-tional Review 55.(2): 195-219.), argues that the useof SAT scores does not help colleges make moreaccurate admissions decisions in terms of academicperformance and graduation rates of students theyadmit. Using a sample of 2,470 students from NLS-72 who attended college, the author conducted apolicy comparison (admissions policies using highschool rank versus high school rank plus SATscores) and examined the rate of accuracy in admis-sion decisions, the effect of each policy on thenumber of false positive and false negatives andoverall excellence of the college (e.g., freshmangrades and college completion rates). Results showhigh correlations between policies using high schoolrank alone and those using both high school rank and

SAT scores for admissions in predicting academicperformance and college completion.

83

221Hilton, Thomas L., and others. 1984. "Score ChangePartitioning Analysis of the NLS and HS&B TestData." Paper presented at the Annual Meeting ofthe Evaluation Network/Evaluation Research Soci-ety, San Francisco, CA, October 10-13. ED 271 508

Since the mean score for a sample composedof several subgroups can be viewed as the sum ofthe mean of each subgroup weighted by the propor-tional size of the subgroup, then the mean change ina time periodin this case, from 1972 to 1980isthe sum of the differences between the means foreach subgroup, with each mean weighted by itsproportional size. Thus, the mean change for thetotal sample is influenced both by mean change inthe score for each groupin this case, a test scoreand by change in the proportional size of thegroup. A decline or gain in mean test scores has botha subgroup mean component and a population shiftcomponent. In addition, how much a particular sub-group may contribute to the total change can beshown to depend on what might be called itsweighted distance from the original total mean. Howthese simple concepts can be used to decomposescore gains or declines is demonstrated by means ofdata from the seniors in the NLS-72 and HighSchool and Beyond. The presentation should be ofboth methodological and substantive interest.(ERIC abstract)

222Rock, Donald A., and others. 1985. PsychometricAulysis of the NLS and the High School andae,-yond Test Batteries. A Study of Excellence in HighSchool Education: Educational Policies. SchoolQuality. and Student Outcomes. Washington, DC:U.S. Department of Education, National Center forEducation Statistics. ED 314 452

The authors examined changes in testedachievement using cross-sectiol.al comparisons be-tween the NLS-72 seniors and the 1980 High School

8 4

and Beyond senior cohort. They also examined theextent and correlates of cognitive growth betweenthe sophomore and senior year in high school.

The overall analysis was designed to:

(1) Estimate possible effects on performanceof differing administrative procedures from one co-hort to another and to determine whether the testswere measuring the same things across cohorts oronly for different administrations within the samecohort.

(2) Estimate item statistics, reliabilities, andindices of precision of measurement for the separatetests and to determine how these measures varyacross sex and ethnic groups.

Item analysis statistics, computed for all testsin all administrations, were used to investigate theappropriateness of the tests in terms of difficulty,expected gains in score, and reliability.

Findings offer little empirical support for theidea that the tests measure different abilities fordifferent subgroups or genders. Findings also showlarge gains in vocabulary and reading over the 2-year schooling period. White students showedgreater gains than blacks or Hispanics, and Hispanicwomen showed greater gains than Hispanic men.

223Rock, Donald A., and others. 1985. Factors Associ-ated with Decline of Test Scores of High SchoolSeniors 1972 to 1980. A Study of Excellence_inHigh School Education. Educational Policies.School Quality. and Student Outcomes. Washing-ton, DC: U.S. Department of Education, NationalCenter for Education Statistics. ED 271 517

Data from the senior cohort of the NLS-72 andthe 1980 senior cohort of High School and Beyondare used to examine changes in academic achieve-ment and to identify school as well as student factors

84

related to these changes. The study also seeks tofocus on factors that can be changed via educationalpolicy.

Findings include declines for all three achieve-ment tests (vocabulary, reading, and mathematics)between 1972 and 1980 with the largest declinesoccurring in vocabulary and reading. Four school-level changes appear to have played a major role inthis decline. They are: (1) a greater likelihood ofbeing in a general or vocational curriculum ratherthan the academic curriculum; (2) a drop in thefrequency with which students report taking "tradi-tional" college preparation courses such as foreignlanguages, science and/or courses requiring labora-tory work; (3) a decrease in the amount of home-work done; and (4) an increasing dissatisfactionamong the students with the lack of emphasis onacademics in the schools. Significant changes werealso seen in the characteristics of high school sen-iors, their homes and families, their schools, andtheir attitudes and behaviors.

Policy implications, based on the results of thestudy, include: (1) increasing required courses in the"New Basics" for nonacademic students; (2) up-grading and reviewing course content and instruc-tional methods to ensure a more rigorous content;(3) assigning more appropriate amounts of home-work in all courses; and (4) continuing to providespecial prog,.ams for educationally and economi-cally disadvantaged students, the handicapped andlanguage minority students.

224Wilder, Oita Z., and Powell, Kristin. 1989. SeDifferences in Test Performance: A Survey of theLiterature. College Entrance Examination BoardReport No. 98-3. ED 308 231

For description see entry #171.

See Gender and Racial-Ethnic Differences

85

XIV. Other

225Adelman, Clifford. 1990. A College Course Map:Taxonomy and Transcript Data. Based on the Post-secondary Records 1972-1984 of the High SchoolClass of 1972, Washington, DC: U.S. Departmentof Education, Office of Educational Research andImprovement. ED 326 153

This document is the result of a 2-year project(1988-90) to revise the taxonomy of the 1985 editionof the Classification of Instructional Programs(CIP) to describe courses. Supported, in part, by theNational Science Foundation and the U.S. Depart-ment of Labor, 1,037 course categories were empiri-cally derived from the 485,000 course titles in theNLS-PETS and were reviewed by teams of collegefaculty. These categories were simultaneously usedto recode the NLS-72 COURSE file. A CollegeCourse Map describes this process ai:4 its decisionrules, and shows what course categorie were de-leted, modified and added to the 1985 CIP. Follow-ing the taxonomy, the volume includes tablesindicating the percentage of students completingundergraduate courses in 1,037 course categories;percentage of students completing undergraduatecourses in 103 aggregate course categories; and thedistribution of all completed courses by type ofinstitution. Additional data covers basic demogra-phy and high school backgrounds of students in theNLS-PETS; highest degree earned; college scholar-ship support; basic demography of 2- and 4-year andtransfer attendance patterns; date of bachelor's de-grees and associates degrees; course of study anddegrees earned beyond the bachelor's; and collegeexperiences and attainments.

226Adelman, Clifford. 1992. Tourists in Our QwqLand: Cultural Literacies and the College Curricu-lum Washington, DC: U.S. Department of Educa-tion, Office of Educational Research andImprovement.

This study uses the base year through fifthfollow-up surveys of the NLS-72 and the PETS files

85

to assess the extent of college students' exposure todifferent cultures, societies and intellectual tradi-tions. The author found three basic categories ofcultural information: Western culture and society,non-Western culture and society, and minority andwomen's studies. By examining course-taking pat-terns, the study concluded that:

(1) The sheer amount of time this generationspent studying accounting, physical education,nursing, and electrical circuits, for example, dwarfsthe amount of time it spent in all the streams ofcultural information put together.

(2) This generation's exposure to cultural lit-eracies other than that of Western societies wasextremely limited, and the bulk of its exposure toWestern traditions was confined to introductory-level courses.

(3) Doctoral degree-granting institutions werethe principal providers of information on non-West-ern culture and society, while comprehensive col-leges were the principal providers of information ondomestic minority cultures.

(4) The curriculum of students at elite colleges(3 percent of all bachelor's degrees in the NLS-72)was so different from that followed by the othercolleges (97 percent) that it was irrelevant to discus-sions of the diffusion of cultural information.

(5) Demography was curricular destiny bothin expected and unexpected ways: as an example ofthe expected, women comprised 80 percent of en-rollments in Women's Studies courses; as an exam-ple of the unexpected, white students were lesslikely than minority students to study history of anykind.

The conceptual framev, ork for the study drawson E.D. Hirsch, Jr.' s, notions of "cultural literacy,"the "schemata" and "scaffolding" theories of read-ing, and distinctions between the s,-holarly canonand the pedagogical canon. The study also examines

6

(and criticizes) other sources of information on whatcollege students study, including catalog analyses,deans surveys, enrollment surveys, course sched-ules, and syllabi.

227Adelman, Clifford. 1989. Cultural Literacies in theCollege Curriculum: The Record of a Generation.Washington, DC: U.S. Department of Education,Office of Educational Research and Improvement.ED 318 377

This study, a precursor to entry #226, usesNLS-PETS data to explore the role of higher educa-tion in the diffusion of cultural information. Fourstreams of information, drawn from the NLS-72archive, are identified as primary cultural languageand secondary stores of language corresponding todemographic, cultural interest and specialist sub-groups. The four stores are represented by coursecultures in (1) minority and women's studies; (2)popular culture/media; (3) non-Western culture andsociety; and (4) Western culture and society. Thepaper also examines the character of institutionsproviding cultural literacy to the general college-go-ing population as well as the demography of enroll-ments in the four clusters.

Findings related to demographics showed thatonly one in five BA degree recipients had no post-secondary exposure to Western cultural informa-tion. Students majoring in applied sciences andengineering showed the lowest rates of participa-

86

tion. Differences in course-taking patterns by gen-der were slight. In general, the author concludedthat, as a result of this generation's limited exposureto secondary cultural literacies, "it certainly seemsunsuited to participate in the diversity of worldculture and economy."

228Adelman, Clifford. 1987. "Basic Statistical Portraitof American Higher Education." Prepared for theSecond Anglo-American Dialogue on Higher Edu-cation. ED 290 410

Previously published data, as well as the Post-secondary Transcript Sample (PETS) of the NLS-72and the "Fifth Follow-Up" Survey (1986) of theNLS cohort, are used for this statistical study ofhigher education in America. The data focus on theformal higher education sector in America (under-graduate education in particular). The topics cov-ered include: institutions (classifications andenrollment); state frameworks (enrollments, capac -ity, and support); students (aspirations and access);faculty; governance (state, institution, faculty); andassessment. Major themes from these data include adecline in social sciences and rise of applied fields,a rise in comprehensive colleges, a decline in edu-cational aspirations of high school graduates in theUnited States, and a reappearance of student dis-tance from learning. Numerous tables and graphsillustrate the major noints. The NLS-72 data playsecondary roles.

8"

XV. Tabulations

The following is a list of tabular summaries usingNLS-72 data.

229Carroll, C. Dennis. 1984. High School and BeyondTabulation: Time on Homework for High SchoolSeniors in 1972. Seniors in 1980. Sophomores in1980. and Seniors in 1982. Washington, DC: U.S.Department of Education, National Center for Edu-cation Statistics.

230Eagle, Eva. 1988. Educational Experience of the1972 Senior Class. Berkeley, CA: MPR Associates.

231Eagle, Eva. 1988. Comparison of Data Sets: Post-secondary Enrollment. Persistence. and Attainmentfor 1972. 1980. and 1982 High School Graduates.Berkeley, CA: MPR Associates.

232Fitzgerald, Robert. 1988. Civic Participation andAttitudes of the 1972 Senior Class. Berkeley, CA:MPR Associates.

233Geist, Mary B. 1987. NLS-72 "Fifth Follow-Up"Classification Variable Set. Washington, DC: U.S.Department of Education, Center for Statistics.

234Geist, Mary B. 1987. NLS-72 Child Care Alterna-fives for Parents Who Work: NLS 1986 "FifthFollow-Up" Washington, DC: U.S. Department ofEducation, Center for Statistics.

235Gifford, Antoinette. 1988. Family Formation Pat-terns of the 1972 Senior Class. Berkeley, CA: MPRAssociates.

87

236Knepper, Paula. 1988. Profile of 1972 High SchoolGraduates Who Were Vocational Students at thePostsecondary Level. Washington, DC: U.S. De-partment of Education, Center for Statistics.

237Knepper, Paula. 1988. Entry and Persistence ofPostsecondary Vocational Students Who WereHigh School Graduates h. 1972. Washington, DC:U.S. Department of Education, Center for Statistics.

238Knepper, Paula. 1988. Field of Entry. Timing. andCompletion for Postsecondary Vocational StudentsWho Were High School Graduates. Washington,DC: U.S. Department of Education, Center for Sta-tistics.

239Knepper, Paula. 1988. CrtditalarnffilyItar_andLevel of Attainment for Postsecondary VocationalStudents Who Were High School Graduates in1972. Washington, DC: U.S. Department of Edu-cation, Center for Statistics.

240Knepper, Paula. 1988. Credits Earned by Field ofStudy and Level of Attainment for PostsecondaryVocational Students Who Were High SchoolGraduates in 1972. Washington, DC: U.S. Depart-ment of Education, Center for Statistics.

241Knepper, Paula. 1988. Enrollment and Aid Status ofPostsecondary Vocational Students Who WereHigh School Graduates in 1922. Washington, DC:U.S. Department of Education, Center for Statistics.

242Knepper, Paula. 1988. Progress and Persistence inPostsecondary Education for 1972 High SchoolSeniors. Washington, DC: U.S. Department of Edu-cation, Center for Statistics.

243Knepper, Paula. 1988. Progress_and Persistence inPostsecondary Education for 1972 High School

Washington,DC: U.S. Department of Education, Center for Sta-tistics.

-I eke 1111- -a . : D

244Knepper, Paula. 1988. Progress_and Persistence inPostsecondary Education for 1972 High SchoolSeniors Who Did Not Complete a BA Degree.Washington, DC: U.S. Departnit,nt of Education,Center for Statistics.

245Kolstad, Andrew J. 1984. NLS-72 Tabulation.,WageRates and Earnings After High School orCollege by Gender, Schooling. and Starting Pay.Washington, DC: National Center for EducationStatistics.

246MPR Associates. 1988. The Postsecondary Educa-tion of 1972 Seniors Completing Associate of ArtsDegrees and Certificates Berkeley, CA: Author.

247MPR Associates. 1988 The Postsecondary Educa-tion of 1972 Seniors Completing Academic Associ-ate of Arts Degrees. Berkeley, CA: Author.

248Owings, Jeffrey A. 1986. Pasisecondary Credits

111 11 1 11 1 1 1 1

Who Received Bachelors' Degrees. Washington,DC: U.S. Department of Education, Center for Sta-tistics.

249Owings, Jeffrey A. 1986. Postsecondary Credits

81.9 1 Id 1 10 1 1 II ISeniors Who Received Bachelors' Degrees. Wash-ington, DC: U.S. Department of Education, Centerfor Statistics.

88

250Owings, Jeffrey A. 1986. Postsecondary CreditsEarned by 1972 High School Seniors Who ReceivedBachelors' Degrees. Washington, DC: U.S. De-partment of Education, Center for Statistics.

251Salganik, Laura. 1987. Patterns of Student andCourse Credit Transfers Among Postsecondary Stu-dents in the High School Class of 1972. Washington,DC: Pelavin Associates.

252Salganik, Laura. 1988. Course Enrollment Patternsat 4-Year Colleges. Washington, DC: Pelavin Asso-ciates.

253Salganik, Laura, and Tal, Joseph. 1988. ProgressionPatterns for BA Recipients. Washington, DC:Pelavin Associates.

254Schmitt, Carl. 1988. Highest Educational DegreeAttained by 1972 High School Seniors by Sex. Race,

e- e ,i gee s' 1111 11 1 0

Spring 1976. Washington, DC: U.S. Department ofEducation, Center for Statistics.

255Schmitt, Carl. 1988. Highest Educational DegreeAttained by 1972 High School Seniors by Sex. Race,Type of High School and Type of Community as ofSpring 1979. Washington, DC: U.S. Department ofEducation, Center for Statistics.

256amitt, Carl. 1988. Highest Educational Degree

Attained by 1972 High School Seniors by Sex Race,Type of High School and Type of Community as ofSpring 1986. Washington, DC: U.S. Department ofEducation, Center for Statistics.

c.)

257Stowe, Peter. 1985. Application Rates of 1972 and1980 High School Senioi s to Postsecondary Institu-tions Washington, DC: U.S. Department of Educa-tion, Center for Statistics.

258Stowe, Peter. 1984. NLS22 aabulaticur_CgaduataStudent Borrowing. Washington, DC: U.S. Depart-

89

ment of Education, National Center for Educa-tion Statistics.

259Tuma, John. 1988. Labor Force Experiences ofthe 1972 _Senior Class. Berkeley, CA: MPRAssociates.Numbers following names refer to item numbers.

Author Index

Adelman, Clifford 14, 31, 90, 120-122, 166, 225-228Alexander, Karl L. 42, 61, 62, 123Allen, Danny R. 18Alsalam, Nabeel 34, 74, 76Altonji, Joseph G. 35, 212Anderson, Kristine L. 68-71, 106, 124-127, 167, 168Andrisani, Paul J. 130, 171Asche. F. Marion 51Barokas, Judy U. 136, 173Basinger, Karen S. 40, 214Behuniak, Peter Jr. 169Bekhuis, Tanja CH.M. 219Benrud, C.H. 32Benson, Charles S. 213Benton, Cynthia J. 190Bethune, Stuart B. 191Bishop, John 38Black, Matthew 33, 39Blakemore, Arthur 38Braddock, Jomills Henry 43, 67Brown, Kenneth G. 81, 112Burkheimer, Graham J. 1, 4, 15, 91, 128Campbell, Paul B. 40, 41, 214Carroll, C. Dennis 93, 132, 143, 185, 229Cervero, Ronald M. 92, 198Chapman, David W. 114Christal, Melodic E. 5Clowe, Darrel A. 64, 72Conaty. Joseph C. 74, 76Conroy. William G. Jr. 182, 215Cook, RLbert 36Crouse, James 220Datcher-Loury, Linda 44, 216Dawkins, Marvin P. 129, 170Daymont, Thomas N. 130, 171Duncan, Debra P. 4Dunkelberger, John E. 131Eagle, Eva 2, 93-97, 132, 230, 231Eckland, Bruce K. 21, 133Epps, Edgar G. 134, 172Ethington, Corinna A. 158Falk, William W. 113, 194Falkowski, Carolyn K. 113Fennessey, James 123Fetters, William B. 3Fitzgerald, Robert 232Furr, Jane D. 135Gable, Robert K. 169Gardner, John A. 63, 149, 186Geist, Mary B. 233, 234

Gerstein, Marti 136, 173Gifford, Antionette 235Goldscheider, France Kobrin 199Gordon, Howard, R.D. 137, 138Gordon, Samuel A. 177Grubb, W. Norton 98, 107, 139, 140, 183Gruber, James E. 174Gustman, Alan L. 184Hafner, Anne 192Haggstrom, Gus W. 200, 201, 211Hansen, David 0. 60Hansen, W. Lee 82Harnischfeger, Annegret 57, 58, 188Henderson, Louise 10, 15, 18, 21, 133Heyns, Barbara 193Hill, Susan 141Hilton, Thomas L. 45, 73, 99, 108, 109, 142, 221Holsinger, Donald B. 114Holupka, Scott 61, 62Horn, Laura 143, 185Hunter, Vincent Curtis 75Ibser, Maya H. 213Inger, Morton 217Jackson, Gregory A. 100, 101Jackson, Kenneth 134, 172Jaffe, Jay 23, 91, 128James, Estelle 74, 76Javalgi, Rajshekhar G. 160Jones, Calvin I I, 30Kanouse, David E. 77, 211Kirkpatrick, Thomas E. 92, 198Klein, Stephen G. 213Knepper, Paula 13, 144, 145, 236-244Knox, William E. 115Kobrin, Frances 202Kolstad, Andrew J. 12, 20, 218 245Larsen, Amy 4Larsen, Eric 217Lee, Valerie E. 108Lenth, Charles S. 5Leslie, Larry 83Levin, Bernard H. 64. 72Levinsohn, Jay 10, 15Lichtman, Marilyn 136, 146, 153, 165, 173, 176Lindsay, Paul 46, 47, 115Loury, Glenn C. 44, 216Low, Stuart 38Lyson, Thomas A. 147, 175, 194Manski. ..rles F. 65Masten, Charles L. 85McAdams, Kathy 6

McClelland, Katherine 148McPartland, James M. 56Mertens, Donna M. 149, 186Meyer, Robert 48Miller, Jon D. 49MPR Associates 246-247Munro, Barbara Hazard 110Navaratnam, K.K. 50, 51Nelson, Howard F. 195Noe 11, Jay 66, 186Novak, Thomas P. 1Ohlendorf, George W. 116Oliver, J. Dale 50Owings, Jeffrey 192, 203, 248-250Owings, Maria 141Pallas, Aaron M. 42, 61, 123Patterson, Barbara J. 26Pavel, D. Michael 22Peng, Samuel S. 23Polonko, Karen 209Powell, Brian 150, 204Powell, Kristin 178, 224Rafferty Donna M. !16Ranson, James T. 154, 197Reiser, Mark 22Riccobono, John A. 4, 7, 8, 15Richards, James M. 27Rindfuss, Ronald R. 205-207Riordan, Cornelius 52, 123Robertshaw, Dianne 28, 29, 78, 102, 103, 151, 152Rock, Donald A. 222, 223Rosenfeld, Rachel A. 207Ross, Peggy J. 60Rothschild, Susan J.S. 146, 153, 176Salganik, Laura 251-253Sanford, Timothy R. 87Schmidt, William H. 53Schmitt, Carl 94-97, 254-256Schrader, William B. 73, 99, 142Sebring, Penny 17. 196

92

Seyfarth, John T. 154, 197Shah, Babubhai V. 24Sharp, Laure M. 79, 80, 155, 156Smart, John C. 157, 158Smith, Michael Francis 117Smith, Wayne E. 118Spletzer, James R. 35St. John, Edward P. 66, 84-86Stacey, Nevzer 34Steelman, Lala Carr 150, 204Steinmeier, Thomas L. 184Stephen, Elizabeth Hervey 205Stone, James R. 54, 159Stowe, Peter 257-258Swicegood, C. Gray 207Taylor, Mary Ellen 9Teachman, Jay D. 208-209Terkla, Dawn Geronimo 88, 111Thomas, Gail E. 67, 104, 177Tickamyer, Anne R. 21To, Duc-Le 74, 76Tourangeau, Roger 16Trent, William 21, 55, 56Tuma, John 259Velez, William 160Wagenaar, Theodore C. 89, 105, 161Watson, Mark E. 4Waite, Linda J. 162, 210, 211Weidman, John C. 79, 80, 155, 156West, Jerry 163, 187Wilder, Gita Z. 178, 224Wiley, David E. 57, 58, 188Williams, R.L. 32Wise, David 48Wisenbaker, Joseph M. 19, 20Wolfle, Lee M. 25, 26, 28, 29, 59, 78, 102, 103, 151,

152, 164, 165, 179, 180Woods, Elinor M. 189Yang, Shu-O W. 119, 181Zemsky, Robert 37

9 2t, 1 IN NFI4NMEI.1 I RINTIlo 01,11,L 114 1146

United StatesDepartment of Education

Washington, DC 20208-5573

Official BusinessPenalty for Private Use, $300

BEST COPY AVAILABLE

OR 93-3156

93

Postage and Fees PaidU.S. Department of Education

Permit No. G-17

Fourth Class Book Rate