Elmore, Richard; Even, Johanna; Kenyon, Susan - US ...

DOCUMENT RESUME

ED 428 463 EA 029 730

AUTHOR Abelmann, Charles; Elmore, Richard; Even, Johanna; Kenyon,Susan; Marshall, Joanne

TITLE When Accountability Knocks, Will Anyone Answer?INSTITUTION Consortium for Policy Research in Education, Philadelphia,

PASPONS AGENCY National Inst. on Educational Governance, Finance,

Policymaking, and Management (ED/OERI), Washington, DC.REPORT NO CPRE-RR-42PUB DATE 1999-00-00NOTE 59p.

CONTRACT R308A60003AVAILABLE FROM CPRE Publications, University of Pennsylvania, Graduate

School of Education, 3440 Market Street, Suite 560,Philadelphia, PA 19104-3325 ($10).

PUB TYPE Reports - Research (143)EDRS PRICE MF01/PC03 Plus Postage.DESCRIPTORS *Accountability; Administrator Attitudes; *Educational

Environment; Elementary Secondary Education; *InstitutionalAutonomy; Organizational Climate; *Policy Formation; SchoolCulture; *Teacher Attitudes

ABSTRACTThis report examines the problem of accountability from the

perspective of schools, rather than from the perspective of external policiesintended to influence schools. The objective was to learn about how people inschools actually think about accountability in their daily work. Data werecollected through case studies in 20 schools, and the sample wasintentionally constructed to maximize variability in conceptions ofaccountability. Results indicate that all the schools had distinctivesolutions to the problem of to "whom" they were accountable. The lack of aguiding external-accountability environment explained a lack of uniformity inperceptions and, in most cases, solutions to the question of accountabilitywere tacit, unarticulated, informal, and arose more from individual beliefsand values of teachers and administrators than from external sources. Thedefault solution to accountability was characterized by individual teacherresponsibility, where personal discretion overruled organizationalexpectations. Other schools exhibited discernible effects of collectiveexpectations. A third group, exhibiting collective expectations, gelled intohighly interactive, relatively coherent, informal and formal systems by whichteachers and administrators held each other accountable for their actions.The study confirms the view that schools develop internal normativestructures that are relatively immune to external influences. (RJM)

********************************************************************************* Reproductions supplied by EDRS are the best that can be made *

* from the original document. *

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

CPRE Research Report

When Accountability Knocks,Will Anyone Answer?

111

Charles AbelmannRichard Elmore

with

Johanna EvenSusan Kenyon

Joanne Marshall

BEST COPY AVAILABLE

I I

U.S. DEPARTMENT OF EDUCATIONOf ice of Educational Research end Improvement

E CATIONAL RESOURCES INFORMATIONCENTER (ERIC)

This document has been reproduced asreceived from the person or organizationoriginating it.

O Minor changes have been made toimprove reproduction quality.

Points of view or opinions stated in thisdocument do not necessarily representofficial OERI position or policy.

CONSORTIUM FOR POLICY RESEARCH IN EDUCATION

The Consortium for Policy Research in Education (CPRE) unites five of the nation's top researchinstitutions in an exciting venture to improve student learning through research on education reform,policy and finance. The members of CPRE are the University of Pennsylvania, Harvard University,Stanford University, the University of Michigan, and the University of Wisconsin-Madison.

CPRE conducts research on issues such as:

education reformstudent and teacher standardsstate and local policymakingeducation governanceschool financeteacher compensationstudent incentives

To learn more about CPRE, please call (215) 573-0700, and then press 0 for assistance.

In addition to conducting research as described above, CPRE publishes reports and briefson a variety ofeducation issues. The Consortium also sponsors regional policy workshops for state and localpolicymakers.

CPRE's website can be found at: http://www.upenn.edu/gse/cpre/

CPRE Research Report Series

Research Reports are issued by CPRE to facilitate the exchange of ideas among policyrnakers andresearchers who share an interest in education policy. The views expressed in the reports are those ofindividual authors, and are not necessarily shared by the U.S. Department of Education, CPRE, or itsinstitutional partners. This publication was funded by the National Institute on Educational Governance,Finance, Policymaking and Management, Office of Educational Research and Improvement, U.S. Departmentof Education (grant number OERI-R308A60003).

Copies of this report are available for $10.00 each, prepaid. Prices include book-rate postage and handling.Make checks payable to Trustees of the University of Pennsylvania. Sorry, we cannot accept returns, creditcard orders, or purchase orders. Sales tax is not applicable. To obtain copies, write to:

CPRE PublicationsUniversity of PennsylvaniaGraduate School of Education3440 Market Street, Suite 560Philadelphia, PA 19104-3325

Quantity discounts are available. For more information, please call (215) 573-0700.

When Accountability Knocks,Will Anyone Answer?

Charles Abe !mannRichard Elmore

with

Johanna EvenSusan Kenyon

Joanne Marshall

CPRE Research Report SeriesRR-42

Consortium for Policy Research in EducationUniversity of Pennsylvania

Graduate School of Education

© Copyright 1999 by the Consortium for Policy Research in Education

When Accountability Knocks, Will Anyone Answer? Abelmann and Elmore

Contents

List of Figures

Biographies vii

Acknowledgments ix

The Problem: External Accountability and School Variability1

A Working Theory of School-Site Accountability: Responsibility, Expectations andAccountability 2

Responsibility 3Expectations 4Accountability 4Accountability for What, to Whom and How? 7

Atomized Accountability: Individual Responsibility Dominates 8Phoenix Charter School 8

To Whom Are You Accountable? The Teacher-Student Relationship at Phoenix 8Teacher Responsibility for What? 10

Student Learning 11Order in the Hallways and Classrooms 11Students Well-Being 12

Other Schools 13Gateway Elementary School 13Stevens Middle School 15Hutchinson High School 15

Summary 16

The Emergence of Collective Accountability: Expectations Influence Responsibility . . . . 17St. Aloysius Elementary School 17

Expectations Shape Teachers' Work 19Teachers' Sense of Responsibility Reflects Expectations 22

Other Schools 23North Beach High School 23Tatuna Point Elementary School 24

Summary 26

CPRE Research Report Series, RR-42

5


Internal Accountability: The Alignment of Responsibility, Expectations, andAccountability 26

Turtle Haven Pilot School 27External Context 28Expectations 29Expectations Influence Responsibility 31Consequences: What Happens When You Don't Meet Expectations 32Internal Accountability 33

Other Schools 33Saint B's Elementary School 33Pine Creek Elementary School 35

Summary 37

Conclusion 38

References 45

Appendix 1: Interview Protocols 47

End Notes 51



List of Figures

Figure 1: Interactions and Alignment 5

Figure 2: Phoenix Charter School 8

Figure 3: St. Aloysius Elementary School 17

Figure 4: Turtle Haven Pilot School 27

CPRE Research Report Series, RR-42 v


Biographies

Charles Abe lmann is currently an education specialist with Eastern and Southern Africancountries at the World Bank. His research interests include the role of state policy in improvingeducation, the role of education in economic development, and the current desegregation effortsin public higher education systems.

Richard F. Elmore is professor of education and chairman of the Department of Administration,Planning, and Social Policy at the Graduate School of Education, Harvard University. He is alsoa Co-Director of the Consortium for Policy Research in Education. His research focuses onstate-local relations in education policy, school organization, and educational choice. He teachesregularly in programs for public sector executives and holds several government advisorypositions. He holds degrees in political science from Whitman College and the ClaremontGraduate School, and a doctorate in educational policy from the Graduate School of Education atHarvard University. He has recently completed a book on the relationship between schoolrestructuring and classroom practice.

Johanna Even, a doctoral candidate at the Harvard Graduate School ofEducation, isconcentrating in Administration, Planning, and Social Policy. She has worked with RichardElmore and Charles Abelmann on a qualitative research study of teachers and accountability. Forher dissertation she will explore issues of accountability in bilingual educationfocusing on therole of the principal.

Susan Kenyon is Project Coordinator and a member of the research staff on the accountabilitystudy currently underway at CPRE-Harvard University. She holds a Masters in City and RegionalPlanning from Cornell University, where she focused on education and community development.Susan is currently an Ed.M candidate, with a concentration in Risk and Prevention, at theHarvard Graduate School of Education.

Joanne M. Marshall is a doctoral candidate at the Harvard University Graduate School ofEducation, where she has worked with Richard Elmore and Charles Ablemann in researchingteacher accountability. Her dissertation focuses upon the experiences of conservative Christianstudents who attend public schools.

CPRE Research Report Series, RR-42 vii

8


Acknowledgments

The authors would like to acknowledge the work ofMary Metz. The structure of this report,using a series of lead cases with supporting cases around each theme, is informed by andmodeled after Metz's chapter in The Contexts of Teaching in Secondary Schools, edited by M.McLaughlin, J. Talbert, and N. Bascia. The authors would also like to express their gratitude toFred Newmann for his thoughtful comments on this report.

This research reported in this report was supported by a grant (No. OERI-R308A60003) to CPREfrom the National Institute on Educational Governance, Finance, Policymaking and Management,Office of Educational Research and Improvement, U.S. Department of Education. The viewsexpressed in this report are those of its authors and are not necessarily shared by the U.S.Department of Education, CPRE, or its institutional members.

CPRE Research Report Series, RR-42 ix

When Accountability Knocks, Will Anyone Answer?Abelmann and Elmore

The Problem: ExternalAccountability and SchoolVariability

pressure for increased school accountabil-ity is a distinctive hallmark of the presentperiod of educational reform. Account-ability, as presently defined in state and

local educational policy, includes four majorideas: the school is the basic unit for the deliveryof education and hence the primary place whereteachers and administrators are held to account;schools are primarily accountable for studentperformance, generally defined as measuredachievement on tests in basic academic subjects;school-site student performance is evaluatedagainst externally-set standards that define accept-able levels of student achievement as mandated bystates or localities; and evaluation of schoolperformance is typically accompanied by a systemof rewards, penalties, and intervention strategiestargeted at rewarding successful schools andremediating or closing low-performing schools(Ladd, 1996).'

These accountability policies are typically directedtoward individual schools or teachers, and in-creasingly, students, as in Texas, New York,Virginia, and Florida where exit exams or profi-ciency requirements are central to educationalreform policies. Coupled with these new account-ability systems, states and localities often arepursuing policies such as charter schools andchoice programs that move schools outside theexisting bureaucratic structure and are intended tosharpen the focus on academic quality and studentperformance. Growing political and fiscal pres-sure on schools lies behind this conception ofaccountability. The political pressure stems fromthe increasing visibility of school performance asa policy issue at the state and local levels and theincreasing capacity of states and localities tomeasure and monitor student achievement. Thefiscal pressure derives from heightened awareness

about educational expenditures as a component ofstate and local budgets. Further, the results of theThird International Mathematics and ScienceStudy (TIMSS) and the National Assessment ofEducational Progress have fueled public concernover what American students are taught and know,in comparison with students from other countries.2Taken together, these pressures have createdstrong incentives for elected state legislators andlocal school board members as well as localadministrators to take a continuing interest inschool performance.

Nested within these developing external account-ability systems are real schools: schools that havetheir own distinctive organizational characteristicsand problems; schools that have unique studentpopulations; schools situated in diverse andparticular communities; and schools with theirown institutional histories. The reality of particu-lar schools belies the pressure for uniformitybehind the emerging external accountabilitysystems. External accountability systems assumea world in which all schools are held to the sameexpectations for student performance. The worldthat school administrators and teachers see,however, is bounded by their particular settings,by their own conceptions of who they are, whothey serve, what they expect of students, and whatthey think of as good teaching and learning.

The long-term fate of educational reform, as it ispresently conceived, lies largely in this tensionbetween the uniform requirements ofexternalaccountability systems and the particularities ofreal schools. The new educational accountabilitysystems will succeed or fail to the degree that theyare designed with knowledge of how schools varyin their own conceptions of accountability.

Part of what we hoped to learn in this study, thefirst phase of a five-year research project, was thelanguage of accountability as it is used andoperationalized in schools. Therefore, we havechosen not to adopt the more precise definitions


10

When Accountability Knocks, Will Anyone Answer? Abe !mann and Elmore

of accountability present in the literature onschool reform, but to leave the definitions as openas possible.

This study is focused primarily on schools andhow they construct their own conceptions ofaccountability. We chose this focus for conceptualand practical reasons. First, we are interested inunderstanding how teachers, administrators,students, and parents think about and behave towardaccountability issues in schools, apart from how theyrespond to new external accountability systems.Schools function, in part, as accountability systems intheir own right, and these systems are worth under-standing in and of themselves. Second, we areinterested in learning, from the variations we observeamong schools, about the range of responses thatschools of various types formulate to the problem ofaccountability. To the degree that schools vary intheir responses to the accountability problem, welearn something about how conceptions of account-ability are formed and how they change in the dailylife of schools. Third, we are ultimately interested injoining our research on school-level accountabilitywith research on external accountability systems tounderstand the sources of school-site variation inresponse to state and local accountability structures.

A Working Theory of School-SiteAccountability: Responsibility,Expectations and Accountability

Our research on school-site accountability wasexploratory and formative in nature.' Our objectivewas to learn as much as we could about how peoplein schools actually think about accountability in theirdaily work. To do this, we conducted case studies ina diverse sample of 20 schools, roughly halflocated ina major metropolitan area on the east coast of theUnited States and roughly halflocated in anothermetropolitan area on the west coast. The schoolsample was intentionally constructed to maximize thelikelihood that schools would vary in their conceptions

of accountability. For example, we chose publiccomprehensive elementary and secondary schools,Catholic parochial elementary and secondary schools,independent private schools, charter schools, andpublic schools operating under special administrativearrangements. We also chose schools on the basis ofvariations in communitiesschools serving predomi-nantely affluent or poor communities, as well as urbanand suburban locations. And we chose schools onthe basis of their size and the diversity of their studentpopulation. It is important to note that none of theseschools were located in a strong external accountabil-ity environment. In this exploratory study, we lookedat schools in states and districts where strong ac-countability was just coming on line.

We spent the equivalent of two weeks in each school.Two researchers at each site observed classes,conducted focus groups with parents and students,and interviewed teachers and faculty. The interviewprotocol we used in conversations with teachers andadministration (included in Appendix 1) was basedupon a working theory described below. The proto-col includes direct, indirect, and hybrid questions,labeled according to how explicitly the accountabilityissue is addressed. In general, researchers reliedupon the indirect and hybrid questions, and found thatresponses to these questions flowed more freely thanwith the direct line of questions. Interviews withteachers were preceded by observation of a math orEnglish lesson. This common point of departureprovided the basis upon which to ask teachers "towhom, for what, and how" are you accountable inyour daily teaching practice.

To structure our field research in these case studyschools, we developed a relatively simple workingtheory which we have continued to elaborate overthe course of our research.4 It continues to be aworking theory in the sense that we will rework itas we understand more about how schools grapplewith accountability.

The working theory begins from a set of four keypremises. The first premise is that schools actu-

CPRE Research Report Series, RR-42 2

11


ally have conceptions of accountability embeddedin the patterns of their day-to-day operations,whether they acknowledge these patterns explic-itly or not.' In order for schools to function, in otherwords, they have to establish channels, both formaland informal, through which individuals and the schoolas a whole may provide an account of behavior.How, for what, and to whom this account is givenmay vary from school to school. The second premiseis that these school-site conceptions ofaccountabilityare organic; they are built out of the raw material ofhuman interactions around the work of teaching andlearning and running an organization. Though it maynot be explicitly articulated, we assume that basicnotions of what it means to be a schoolassumptionsabout how schools, in the most general sense, oper-atesare influential upon teachers', administrators',parents' and students' conceptions of accountabilityin their particular context.' Schein (1992) describesthis group culture as "a pattern of basic, sharedassumptions that the group learned as it [solves] itsproblems of external adaptation and internal integra-tion . . . " A school's conception of accountability,then, can be revealed in the way teachers, administra-tors, students, and parents talk about fundamentalissues of schooling. The third premise is that partici-pants in schools are active agents in the creation ofthe conceptions ofaccountability under which theyoperate, and they can be active agents in changingthese conceptions. Whether consciously aware of itor not, teachers, administrators, students, and parentsact out their conceptions of accountability in theirdaily work; these conceptions, while relatively stable,can be changed, either in response to external pres-sure or out of intentional action at the school level. Afourth premise is that formal, external accountabilitysystems are only one among many that influence aschool's internal conception of accountability.Schools form their conceptions ofaccountability froma variety of sources, including individual teachers' andadministrators' beliefs about teaching and learning,their shared conceptions of who their students are, theroutines they develop for getting their work done, andexternal expectations from parents, communities andthe administrative agencies under which they work.

Our working theory posits a set of relationshipsamong three factors: individual conceptions ofresponsibility; shared expectations among schoolparticipants and stakeholders; and internal andexternal accountability mechanisms. An indi-vidual school's conception ofaccountability, inour view, grows from the relationship amongthese three factors (Wagner, 1989).

Responsibility

Individuals who are parties to schoolingteach-ers, administrators, students, and parentshavetheir own personal values that define their respon-sibilities toward others. Teachers, for example,may have strong views about their personalresponsibility for student learning, or the degree towhich students and their families share thisresponsibility. Administrators may feel personallyresponsible for influencing teachers' instructionalpractice in particular ways, or they may locateresponsibility for instructional practice primarilywith teachers. The distinguishing characteristic ofresponsibility, in other words, is that it is personaland individual in nature and it stems from thevalues and beliefs of individuals. Individualconceptions of responsibility may come from anumber of sourcesfrom the life experience andmoral background of the individuals, from theireducation and training, from their beliefs aboutthe social determinants of student learning, andfrom their interaction with others. From theperspective of our working theory, we do notassume that individuals' conceptions of responsi-bility come mainly from their work environmentor from formal accountability systems. Instead,subscribing to Lortie's (1975) assertion thatteaching occurs primarily in isolation, we assume thatorganizational and external influences may play a partin teachers' perceptions of their role, but that indi-vidual values are certainly influential.

For example, individual English teachersmay havestrong beliefs about what constitutes a good essay,


;. 2

When Accountability Knocks, Will Anyone Answer? Abe lmann and Elmore

what constitutes a good book for students to read,what might be an acceptable number of books for astudent to read in a year, or what might be an accept-able amount of homework to assign in a given week.They may also have strong beliefs about the capaci-ties of their students to learn certain things. Further,teachers may include among their responsibilities,students' emotional and physical-well-being, and insome cases individuals may even perceive this re-sponsibility as taking priority over curriculum require-ments. Beliefs may be shared among English teachers,or they might lie in the domain of individual teacherdiscretion and vary widely among English teachers.To the degree that beliefs lie in the domain of indi-vidual discretion, and relate to one's individual beliefsabout his or her own behavior, we call them responsi-bility.

Expectations

Expectations, by contrast, are collective in natureand they characterize the shared norms and valuesof school participants developed to get the workof the school done. They are formed out ofrelationships among individuals, and they operatein often powerful ways to shape individuals'behavior and values. For example, first gradeteachers may have shared conceptions in a givenschool about how fluently first graders should bereading by the end of the school year. Or, theymight have expectations of how much noise istolerable from their colleagues' adjoining class-rooms, or of what constitutes good student deco-rum in the hallways. Parents may expect teachersto treat their children in certain ways in the class-room or to prepare their children for certain post-school futures. And, teachers may have expectationsregarding the amount of time parents should spendsupervising homework. Teachers and administratorstogether may form certain expectations about whatacademic work students from "their" community arecapable of doing; these expectations may or may notbe shared by students and their families.

The distinctive feature of expectations is that they arecollective in natureshared among individualsalthough not necessarily with complete consensusamong all the individuals in a given school. Further,expectations are beliefs about others 'behavior,though individuals may include themselves within thecollective for whom they hold these expectations.Certain expectations might be widely shared amongall partiesteachers, administrators, students, andparents or expectations might vary among groups orfactions within a school. Differentgroups ofteachers,for example, might have different expectations of whatconstitutes adequate student performance or decorumin the classroom. Teachers might have one set ofexpectations for students, and parents might haveanother. So the fact that expectations are shareddoesn't necessarily mean that they reflect a consensusamong all parties in a given school.

Accountability

Accountability mechanisms are, literally, thevariety of formal and informal ways by whichpeople in schools give an account of their actionsto someone in a position of formal authority,inside or outside the school. Some accountabilitymechanisms are internal to schools. Principals,for example, may require teachers to providecopies of their lessons, to write a daily scheduleon the blackboard in their rooms, or to be avail-able for supervisory duty in hallways, play-grounds, or lunchrooms. Some accountabilitymechanisms are external to schools. Schooldistricts may administer periodic student assessments,for example, and use the resulting data to influencewhat teachers teach. Accountability mechanisms,whether internal or external, take a wide variety offorms. They might be explicitlyforma/ in character, aswhen written in a school handbook or district or statepolicy. They might also be relatively informal, aswhen a principal communicates to teachers that theyshould keep the noise level down in their classrooms,then engages in explicit monitoring ofclassrooms.Likewise, accountability mechanisms vary consider-



ably in their consequences for success or failure. Theconsequences might be communicated, with relativelylow stakes, such as a principal's approval or disap-proval communicated to a teacher for something thathappens in that teacher's classroom. Or accountabil-ity mechanisms might carry relatively high stakes, aswhen a principal publicly praises or criticizes ateacher for disciplinary practices or when a districtpublishes in the local newspaper student academicperformance data by schools.

In the context of our working theory, accountabil-ity carries a specific meaning. When we askedpeople in schools about accountability, we wereinterested in: accountability for what; how they wererequired to give an account of their actions; and theconsequences or stakes for failing to do so.

In our working theory, responsibility, expecta-tions, and accountability operate in a close mutualand reciprocal relationship with each other, andthis relationship takes a variety offorms in differentschools. This relationship is captured in Figure 1.Individual conceptions ofresponsibility may influencecollective expectations, or alternatively, collectiveexpectations may influence individual conceptions ofresponsibility. Similarly, individual conceptions of

Figure 1. Interactions and Alignment

Responsibility

Accountability

ResponsibilityExpectations,

AlignmentInternal

of

Expectations

Accountability

responsibility or collective expectations may influenceformal or informal accountability systems, or viceversa. In Figure 1, we mean to convey that a givenschool's response to the problem of accountability isthe result of how it resolves the tensions, inconsisten-cies, and complementarities between individuals'personal values, their shared expectations, and themechanisms by which they account for what theydo.

Implicit in the model presented in Figure 1 is thenormative view that schools are likely to havemore powerful internal accountability systemsformal or informalif the values and normsembodied in these systems are aligned withindividual conceptions of responsibility andcollective expectations in the school. That is,internal accountability systems are likely to bepowerful in their influence over individual actionsto the degree that they are consistent with thevalues represented in individual responsibility andcollective expectations.

Alignment can be produced in a variety ofwaysfor example, by deliberately choosing people whoshare a common set of values to participate in theschool or by deliberately using the structures andprocesses of the organization to socialize peopleto a set of common views. To the degree thatindividual responsibility, expectations, and inter-nal accountability systems are not aligned, one canexpect various degrees of incoherence amongindividual beliefs and collective norms, andrelatively weak internal accountability systems.

We have said nothing yet in this analysis aboutwhat individuals or schools consider themselvesto be responsible or accountable for. To say thatthere is a high degree of alignment betweenresponsibility, expectations, and accountability isto say nothing specific about the purposes for whichthe school is aligned. Schools could, for example,have a high degree of alignment about values thatstress student academic performance, or they could


145


have alignment about values that stress order anddiscipline in the classroom and hallways, but little orno agreement on academic goals. Alignment, then,refers to the consistency and strength of agreementinside the school, not the subject of that agreement.

Also implicit in the Figure 1 model is a normativeview about the relationship between externalaccountability systems and the internal life ofschools. If the power of internal accountabilitysystems is a function of the alignment of responsi-bility, expectations, and internal accountabilitymechanisms, then the power of external account-ability systems is a function of the alignmentbetween the norms and values represented in thesesystems and the internal accountability mecha-nisms of schools. The effect of external account-ability systems is mediated by internal account-ability mechanisms. Schools might, for example,have a high degree of internal alignment aroundvalues and expectations that are quite inconsistentwith the requirements of local or state account-ability systems. Or alternatively, schools may notbe aligned around individual responsibility,collective expectations, and internal accountabil-ity, and, therefore, respond incoherently to rela-tively clear guidance from local or state account-ability systems. In other words, how a schoolresponds to external accountability systems islargely determined, not by the details of thoseexternal systems, but by the degree of alignmentbetween the schools' internal accountabilitymechanisms and the requirements of the externalaccountability system.

For example, a school might have relatively weakcommon expectations for teachers and studentsand relatively weak internal accountability struc-tures. In such a school, teachers' conceptions oftheir work would be largely driven by their indi-vidual sense of responsibility. As represented inFigure 2 on page 8, the responsibility area woulddominate our Venn diagram, and there would bevery little overlap between the circles. Similarly,internal accountability measures, if they exist at

all, would have relatively little influence. Thisschool would be atomized, that is, fragmented intoindividual or very small units. Teachers wouldform their expectations for students and theirideas about what and how to teach, largely out oftheir individual conceptions of responsibility.This school's response to any external account-ability system, we predict, would reflect its inter-nal incoherence. The requirements of the externalsystem would be translated into idiosyncraticvalues and practices by individual teachers.

Another type of school might have relativelystrong common expectations about certain sharednorms, and these expectations might be alignedclosely with teachers' conceptions of personalresponsibility. The graphic representation of thisscenario is seen in Figure 3 on page 17. Expectationsdominate the diagram, but to a lesser degree thandoes responsibility in Figure 2, and witha morebalanced relationship between the responsibility andexpectations areas. A school might arrive at this stateby recruiting teachers who already share a commonview of teaching and learning and by creating internalstructures and processes through which teachersshare their personal beliefs and develop commonexpectations of each other. These shared expecta-tions might be similarly extended to parents andstudents, by recruitment of like-minded clients or byactive socialization. In some instances, coherencemight be achieved by a community activelyimposing its values on the school, through sus-tained parent involvement or political influence inthe recruitment of teachers and administrators.Such a school might have either weak or stronginternal formal accountability mechanisms. Theschool might simply operate on a daily basis, andteachers might define their work, based on sharedexpectations that are aligned with their sense ofpersonal responsibility, with relatively few ex-plicit rules or procedures designed to hold indi-viduals accountable for their work. Or a schoolmight extend its agreement at the level of respon-sibility and expectations into a relatively explicitinternal accountability system of rules and proce-

CPRE Research Report Series, RR-42 15 6


dures that provide a basis for teachers and stu-dents to account for their actions. This scenario isrepresented by Figure 4, on page 27, where thestrength of alignment between the three areas, andparticularly between responsibility and expecta-tions, functions as an informal accountabilitysystem.

This type of relatively cohesive school is charac-terized by a high degree of alignment betweenindividual responsibility and collective expecta-tions, and can possibly be complemented by arelatively explicit internal accountability system.Such a school might, in our working theory,respond to an external accountability system in anumber of ways, including: accepting and inter-nalizing it; rejecting it and developing defensesagainst it; or incorporating just those elements ofthe system that the school or individuals deemrelevant. Response to the introduction of anexternal accountability system would, we assume,depend upon the degree of alignment between thepurposes of the external accountability system andthe internal norms of a school.

Accountability for What, to Whom andHow?

A final part of our working theory addresses theissue of the purposes behind accountability sys-tems. Most formal external accountability sys-tems are predicated on the assumption thatschools should be held accountable mainly forstudent academic performance. Viewed from theschool level, however, the picture is far more com-plex. We addressed the issue ofpurpose by posing,in each of our schools, the question: For what areyou accountable, to whom, and how? Schools arecharacterized, not surprisingly, by a wide variety ofanswers to thefor what question, and the variousanswers they give to thefor what question, notsurprisingly, have very different implications for howthey answer the to whom and how questions.

In some schools, for example, teachers have explicittheories about the relationship between the character-istics of the children and the communities they serveand for what they as teachers are personally respon-sible or for what they are collectively accountable.Some teachers, for example, believe that their re-sponsibility and their formal accountability is, andshould be, heavily shaped by the socio-economicbackground ofthe children they teach. Children livingin poverty, they argue, require social supports in theclassroom and in the school that children not living inpoverty do not require. When asked for what theyare formally accountable, these teachers were apt torephrase the question using the language ofresponsi-bility in place of accountability. Some teachersanswered thefor what question by stating their beliefthat they are responsible for providing a safe, nurtur-ing environment for children. Other teachers believedthat the socio-economic backgroundof the childrenthey teach should have less importance in determiningfor what they are responsible or accountable. Theyanswered thefor what question by stating their beliefthat they are responsible for students' academicperformance or their future success in school. Theseanswers to thefor what question have very differentimplications for how teachers answer the to whomand how questions. Teachers who see themselves asprimarily responsible for providing a nurturing envi-ronment, for example, are more likely to say they areaccountable to the children and their families.

In the following sections, we have organizedobservations from our first year of exploratoryfieldwork into three categories, based upon the-schools' responses to the problem ofaccountabil-ity. We reiterate that our findings are limited by thefact that the schools in the exploratory study were notlocated in strong external accountability environments.Within each section, we feature a lead case, followedby several supporting cases that represent variationson the theme of the lead case. The categories are byno means exhaustive of the characteristics we ob-served in these schools, norare the distinctionsbetween categories quite so pronounced as they may


167


appear in this format. No school is an absolute caseofjust one theme, but some are more typical thanothers in reflecting that theme.'

Atomized Accountability:Individual ResponsibilityDominates

For many teachers, the idea of accountability haslittle or no tangible reality in their daily work.They operate essentially as solo practitioners inisolated classrooms, relatively detached from theinfluence of outside forces. In this section, weexamine four cases where this daily reality ofisolation dominates conceptions of accountability.The lead case, Phoenix Charter School, which webelieve typifies the theme of this section, is arelatively new inner city elementary school,operating under a charter from the state. Phoenixstudents are disproportionately poor and minority.Phoenix may be unique because of its charterstatus and its corporate sponsorship, but in manyways it is similar to the other schools in our studythat serve urban populations. Figure 2 indicates,by the relative size and independence of indi-vidual responsibility, that there is little internalalignment in this category of schools, and indi-vidual discretion is the primary mode of account-ability. Gateway, a small urban Catholic K-8school situated across the street from a housingproject, serves, like Phoenix, a heavily minorityand disadvantaged student population. Stevens isa large urban middle school, with a reputation forbeing relatively successful with its working classand disadvantaged student population. Hutch-inson isa large nineteenth-century public high school with aonce-proud academic legacy and a recent history ofstudent behavior problems. In all of these schools,accountability begins and usually endsat the levelofindividuals, particularly individual teachers.

Phoenix Charter School

Figure 2.

Phoenix Charter School

InclividualResp

Co

onsibility

Oorllfxpectationsective

entability

ToWhom AreYou Accountable? TheTeacher-Student Relationship atPhoenix

Proponents of charter schools often claim thatthey are "the most accountable" kinds of schools.Schools that apply for and are granted chartersmust persuade their governing agency (the city orstate) that they are able to teach children; in returnfor which they receive funding and autonomyfrom many state and local regulations. Most charterschools are reviewed annually through site visits andreports, and those who fail to meet the terms of theircharter are subject to its revocation. In this sense,some say they exemplify a relatively clear and explicitkind of external accountabilityone focused onparent choice and state oversight.

We might therefore expect that teachers at Phoe-nix would be concerned with this formal account-ability; that they would be concerned with makingsure their students' test scores increase, or withteaching the curriculum mandated by their spon-soring Corporation, or with their ongoing evalua-tion by their administrators. But, in the case ofPhoenix, they are not. Teachers barely mentionedthese kinds of formal accountability mechanisms in thecourse of our interviews. Instead, they spoke with


17


passion and enthusiasm about being most "account-able" to the one group that has no formal power inschools at alltheir students. In speaking about theirrelationships with students, they tended to use theterm "accountability" to refer to what we have calledpersonal responsibility. For Phoenix teachers,"accountability" is largely defined in terms of theirindividual responsibility toward students, rather thanany formal or informal set of rules or procedures bywhich they account for what they do.

For example, one teacher, when asked to whom heis accountable, responded: "Kids. Twenty-eightkids . . . that's why I'm here. That's why we're allhere. So they can get educated. Get them readyfor what they can expect when they get older."Such a response defies the traditional notion ofaccountability as a reciprocal relationship withconsequences, because although teachers claimaccountability to their students, students in thisand other schools have very little, if any, formalinfluence. The teacher-student relationship isinherently one where teachers are given authorityover their students, a position made fast by theschools' in loco parentis function. In return forthe authority granted to teachers, teachers acceptresponsibility for their students. While studentsmay certainly complain about their teacher, or actin a way that makes her job easy or difficult, theydo not themselves exercise any authority over theteacher or hold her accountable in any meaningfulsense of that term, or claim responsibility for theteacher' s actions. This most essential of schoolrelationships is thus one-sided: teachers acceptpersonal responsibility for the students entrustedto them, but that responsibility is unreciprocated.

Students are minors, so society grants authorityand responsibility to their parents or guardians.One would therefore expect that parents wouldrepresent their children in the teacher-studentrelationship: teachers might not be accountable totheir students, but they could be accountable to theirstudents' parents. Some Phoenix teachers, when

asked, did mention students' parents as the people towhom they were accountable, but didnot feel thatparents were accountable to them in return. Whilethere was a core of very involvedparents, mostPhoenix teachers expressed frustration with the levelof parent involvement. One teacher cited her non-attended parent conferences: "Last year I called them. I would come in on a Sunday and nobody wouldshow up. So that was kind of . . . sad. I just ex-pected parents to care a little bit about their child'seducation. It's not like I ask them for too mucheither." Other teachers who said they were account-able to parents said they assumed that parental non-involvement implied satisfaction. One teachersum-marized his relationship with parents:

The fact that I don't really hear from themprobably indicates that they're satisfied withwhat is happening. I make [laughs] thatassumption . . . I don't have the time to callparents when kids are acting up and sus-pended five, days in a row from class. And Iwish that they would call me. I mean, it's theirkid; they should be calling me and letting meknow if they want to know about their kid.They need to call me. So I wish they wouldcall more. But any time I do call them, they'reusually pretty supportive.

A Phoenix administrator pointed out that she haslearned over the years that public schools cannothold parents accountable for anything. There arealways going to be parents who will beuninvolved, she said. This lack of parental in-volvement again presents teachers with a one-sided, unreciprocated relationship, which keeps themfrom being fully accountable to parents and fromaccepting parental accountability in return: themajority of parents simply aren't there. Becauseteachers are alone in the teacher-student-parentrelationship, they assume more responsibility in it andfor it.


is9


Given the primacy of the teacher-student relationshipand its one-sidedness, it is not surprising that whenwe asked Phoenix teachers, "To whom are youaccountable?" many answered, "Myself." When weasked one teacher "who checks" to see if she is doingher job, she said, "No one, but I know. I don't wantthem [her students] to be lost." This attitude extendsnot only to teaching, but also to its auxiliary functions,such as record keeping, about which one teachersaid: "No one has ever checked it, and, to tell youthe truth, I'm not very organized about it. I mainly ammost accountable to myself. If I am not doing what Iam supposed to be doing, then I am failing, and Ihave a problem with that." Other teachers spokeof self-checks such as their ability to "sleep atnight," or to "look in the mirror." Again, whatthese teachers were calling "accountability" iswhat we have defined as their own responsibility.

Based upon the organizational structure ofschools, administrators are the obvious people towhom teachers are accountable. Administratorshire, evaluate, and fire teachers; in return adminis-trators are expected to provide supplies, curricula,and support. However, administrators are ex-cluded from the basic teacher-student relationshipupon which the work of schools is founded. Perhapsthis outsider status explains why many teachersmentioned being accountable to their administratorsonly occasionally, and then only after mentioningstudents, parents, or themselves. This ambivalence isreflected even in attitudes toward evaluation, whichone would expect to be the consequence that givesaccountability its bite. One teacher said of his admin-istrators' evaluation: "I don't see them [administra-tors] coming up and saying, 'This is what you need towork on.' I feel I'm doing a good job. The honesttruth is that I really don't care [laughs] if they approveor not. I feel what I'm doing is correct."

Other teachers told us of quietly disregardingadministrative mandates such as curriculum. Forexample, although the Corporation curriculumdoes not include spelling, one group of fourregular classroom teachers and a special education

teacher told us that they teach spelling anyway,using materials they purchased themselves. Oneteacher said:

I'm traditional . . . They need to learn how tospell. So I start off with things around theroom, and science words, and now I've boughta spelling book . . . boring. But they needthat. They need that background They . . .

need a stronger way to decipher words.

Another teacher agreed: "Spelling isn't a curricu-lum in our school. I believe children need tomemorize ten spelling words a week. So that'ssomething that I do extra, and I give them packetsto do it." These teachers also said that there is"no time" to teach spelling, so they send it homewith their students even though, one reported, theyare "not supposed to."

Because teachers have little interaction withadministrators other than the evaluations that mostof them disregard, and because they do have a lotof interaction with their students, it is logical that theiraccountability to administrators is weaker than is theirsense of responsibility to their students. As notedearlier, when there is weak internal accountability andweak expectations of teachers, teachers' sense ofresponsibility rules.

Teacher Responsibility for What?

When we asked Phoenix teachers for what theyare accountable, their replies fell into three maincategories: students' learning, order, and students'well-being. In answering questions about theiraccountability, however, teachers frequently referredto their own sense of responsibility for learning, order,and well-being. The extent to which all teachersdescribed feeling responsible for these areas placesthem in both the individual responsibility and collectiveexpectations categories as defined in our study.



Student Learning

Student learning is the most obvious function ofschools, and the school people mentioned it as amatter of course. Teachers spoke sincerely andeloquently about their work with students as thisPhoenix teacher did:

Mostly I feel like I'm accountable to mystudents; I'm here to teach them, to make surethat they're learning what they're supposed tolearn, and to present it to them in the best waythat it 's going to get to them.

Another teacher said:

I'm supposed to teach them, and I plan toteach them. Like now I'm doing report cards,and when I write something down, I expect tobe able to stand behind it and say, 'I did mybest to teach this child' and 'I did my bestassessing this child' through tests or observa-tion or cooperative learning and everythinglike that.

And another teacher said:

My job is to teach the curriculum, to suit allthe children in my classroom, regardless oftheir learning ability. So that 's my responsi-bility. I need to find a way to teach everybodyso that they . . . [are] basically on grade level.

These three teachers use the language of responsi-bility. Their comments were very I-centered: "I feel"and "I'm here" and "I plan." There is an impliedassumption that they are "supposed" to do this, thatthe "job" requires it, and that someone might askthem to stand behind their assessment, but when weasked teachers where they got their ideas of what itmeans to be a teacher, they spoke not of their admin-istrators, or their teacher education programs, or theircolleagues, but oftheir own families, their ownteachers, and their core values. One teacher said:

I grew up in a family that was veryI startedworking when I was 14. They believed inwork, they believed . . . every summer, everyholiday, you went to work with the rest of thefamily. You did your part . . . I've always beenraised, myself to . . . you do the best at whatyou do, or don't do it. Find something else todo . . . So as far as teaching, this is huge. I'mteaching 28 children . . . You think of it kind oflike a privilege. My God, I've got these littleminds, little brains, and I can fill them with allthis good stuff and hope that they take some-thing with them to the next grade.

Teachers' language revealed their intense desire tofind "the best" way to teach "all the children."These teachers' responses were common amongPhoenix teachers; so common that they confirmedthe observation of another teacher that Phoenixhas its own culture, one of: "The kids are going toprogress. And you're going to make sure thathappens."

Order in the Hallways and Classrooms

For student learning and progress to take place,everyone at Phoenix believes that order is abso-lutely necessary, and this is the second area forwhich teachers claimed responsibility. Theirresults are immediately noticeable: Phoenix isbright and clean and free of graffiti. Students sit acertain way on rugs (cross-legged), line up aparticular way when leaving (each student stand-ing in a square floor tile), and stand a certain way inthe halls (arms behind backs). The school has acitizenship program of rewards and punishmentswhich formalizes the emphasis on student behav-ior, and most teachers have their own point sys-tems as well. When asked what to look for in aprospective job candidate, most teachers immedi-ately mentioned the candidate's ability to disci-pline. One teacher elaborated:


20


I would say it has to start with discipline. Iwould either just tell [prospective candidates]point blank what we do here about disciplineand what our expectations are as far as dresspolicy; and no talking in the halls; and whensomeone's talking, pencils are in the pencilholders; and people are not leaving their seatswithout permission; and using the bathroomall as a class at the same time. I would eitherjust tell them straight up that's what we do, orI would ask them first what their approach toschooling is. And if they start talking abouttheory and great curriculum and stuff likethat, and do not talk about the nuts and boltsof what you have to deal with during theschool day.. . . then I would probably thinktwice . . . I would say . . . "This is how we doit. And ifyou don't like it, then, ifyou're more

. . touchy feely, and let the kids have this sayand that say, then this isn't really going to bethe place for you . . . Eventually you can getthe kids to where they can do that. But ini-tially it's got to be discipline, discipline,discipline.

At times, it appears order becomes an end initself, rather than a means toward the end oflearning. For example, we observed a lesson atPhoenix where directions were given after eachmath problem, "Chalk down! Chalk up!" and foreach step in clean up, "Collect paper towels. Putyour slate in the middle, on top of the box. Bringme the box . . . Table 2, go wash your hands." Atthe end of class, the line for lunch had to beredonelights off, students sent back to theirseats, free time at the end of the day taken away,and the line re-formed. Each of these directionstakes class timein giving them and in followingthem. One could argue that time spent giving andfollowing such directions saves time later asstudents learn procedures and can move from taskto task efficiently. But the lesson we observedwas mid-year, and the directions did not seem

necessary to that lesson. There seems to be noreason a lunch line would have to be re-formedexcept to maintain order as an end in itself

The expectation that teachers will maintain order isone ofthe few expectations with which staff associateand anticipate profession consequences. They citedunsuccessful teachers who were not asked to returnbecause their classrooms were "crazy" or administra-tors had to intervene frequently. By implication,teachers know they are doing a good job if theirclassrooms are quiet and administrators do notintervene. We asked one of the Phoenix administra-tors about this emphasis on order, and she offeredtwo explanations: first, such order is necessary inorder to maintain safety within the building, especiallyin case of fire; second, such order teacheschildrenhow to behave in society, which is necessary for themto be successful.

Students' Well-Being

Students' well-being is a broad concept thatencompasses and depends on academic learningand discipline. That is, student well-being isnecessary in order for students to learn, and theirlearning will improve their well-being. The urbanteachers in our study shared what can be calledalmost a sense of mission to improve their stu-dents' livesa mission that crosses teachers'race, gender, and class lines.

Some of the Phoenix teachers worried about theirstudents' survival. One teacher said of his stu-dents: "I . . . hope for them to live to see theiradulthood . . . by not making a bad decision thatwill cost them their life. That's what I fear themost because they are inner city kids." We heardthis teacher and others in his cluster repeatedlytalk to and about their students in terms of making"good decisions," a phrase usually referring tostudent behavior. One teacher called this kind ofawareness "preparing students for life." We askedhim how he would "teach life" in the lesson weobserved and he said:



[Today] I didn't go off on a tangent, saying,"If you don't follow directions, then you'll get

firedfrom a job," or whatever. Today it didn'treally come up. But if there was a conflictthere in the class today, then I might havegone off on, "Well, you handle this situationout on the street, then you might end up deador you're going to be locked up. lfyou handlethat situation that way on the job, with a co-worker or your boss, you' might get firedfromyour job. Or ifyou're at school, you might getkicked out of school. If you're not turningyour work in on time, you're going to getfailed; you're going to get F's in college, andyou're going to get kicked out of college."That sort of thing.

Other teachers talked of teaching children "differ-ent values from home," such as not resolving aconflict through hitting. Some teachers talked ofbeing role models (especially teachers of color),or of meeting students' emotional needs. At timesthese responsibilities lead to ambiguity about theteacher's role. One teacher spoke of being bothan emotional support and a disciplinarian. Whenasked what her students expect of her, she said:

Too much, actually. They think I'm theirfriend. They think I'm their mother. [Sheimitates:] "Ms. Dawson, can you unbutton my. . . ", "Ms. Dawson, can you do that?" Andthey would feel sick until I would say, "It'sokay, sweetie," and give them a hug, and thenthey're fine for the rest of the day. But then itgets in the way of discipline, because when Isay, "Okay, now study for your science test,"they're around me, they're giving me a mas-sage . . . and then I'll go crazy and then I'llstart yelling and [she demonstrates :] "Get inline. I'm not your friend. You need to get inline right now." [Imitates student.] "Geesh! Iwas just doing this!" And then, I have a kid,Nikia . . . who writes me letters . . . I shouldlove her more, and why do I love this otherkid? . . . And I say, "Look, Nikia, I'm your

teacher, not your friend." [Imitates Nikia.]"You can be my friend andmy teacher at thesame time." And I said, "No, I can't." . . Ihave two kids who lost their mothers . . . andthey both . . . desperately need a femalesomebody.

Another teacher, wishing for more school socialworkers, talked about the tension between teach-ing students reading and acknowledging theirdifficult home lives:

Schools are becoming more than a place tolearn . . . Some of these kids come here at sixin the morning, early morning, and they arehere until seven-thirty when theirparents pickthem up and . . . the only time they are goingto get counseling or anything is [in] schooland I think the role of school needs to belooked at and how it should be changed. Idefinitely think more counselors. Half thesekids have a parent in jail or a sibling. Theycome from neighborhoods where they can't gooutside and then they expect them to readthese silly books? . . . And I do think it isimportant for us, as teachers, not to excusetheir behavior from where they came from, butto understand it . . . Why is it important toread this book? What do we want out of it?And to really focus on: What we have gainedby doing this? This is really hard for a kid tosee.

These teachers claimed responsibility for modify-ing their teaching practice in keeping with theirstudents' needs, whether those needs are aca-demic, social, or psychological.

Other Schools

Gateway Elementary School

One might expect teachers at Gateway, an inner-cityCatholic school, to talk about accountability in


2 213


religious terms: of being accountable ultimately to Godor, on a more earthly plane, to the archdiocese. Butthey do not. Neither is religion an emphasis forlearning: the principal reported that teaching studentsreligion is not as important as "educating them so thatthey'll be able to better themselves in life . . . andsharing values with them." Perhaps this lack ofemphasis on religion is due to the fact that, althoughthe new principal and one teacher are nuns and mostof the staff is Catholic, most of the studentsare notCatholic.

In this school where teachers were concerned withtheir students' very survivalfor example, "I praythat they can make it through the summer withoutgetting killed"the teachers seemed to feel soresponsible for their students that they were defensiveabout anything that referred to their students' pooracademic performance, be it low test scores or lettergrades. On report card day, several teachers toldtheir students, "Not everyone is an A student," and"Being average is OK as long as [you] are trying."While these remarks were probably reassuring tostudents, they did not reflect the priority on learningthat the principal desired.

At Gateway, when asked about responsibility,every teacher spoke about caring for the children.One teacher said: "I think most of us are here forthe welfare of kids." She continued to talk abouttheir welfare as follows:

We are aware of the fact they're here foreducation. On the other hand, many kids arecoming from homes with alcoholism and thelast thing they care about is an adjective. Soi f I get hysterical about an adjective, I'mreally doing them harm. So their welfarecomes first before their educational process,whatever. I think we want them to be happy,believe that in an atmosphere of happiness,friendliness, making friends, safety here, thatthere 's not going to be violence in the schoolyard, that there's not going to be drugs in thebuilding, that they are safe and that they know

that there are people here who really careabout them, because I wouldsay that we do.We really do. And then secondly we want themto succeed in high school and in college.

In the past, reported the principal, Gatewaystressed the importance of safety and supportrather than teaching and learning. This tensionbetween support and learning was revealed in astory she related about sharing Gateway's low testscores with her staff, and what she perceived to betheir response: "Ho-hum . . . Well, it's an inner-city child who has no family, no motivation, isconstantly underfed, tends to sleep in the class-room, and it's very difficult to reach them." Notonly were the teachers more concerned with theirstudents' affective needs, but they also did notbelieve that the tests were a worthy measure oftheir students' learning. One teacher commentedabout the tests: "We don't do anything with them[the tests]. They do not relate to a lot of whatthese youngsters know. And they [the students]are not readers, so it is very difficult, I think, tohave them do as well as they should."

The principal's concern about academic learningwas supported by the archdiocese, which providesa formal curriculum and teaching guidelines. Butthe teachers were unevenly concerned with whatthe guidelines were or what they were to cover bythe end of the year.

Gateway is a school with little formal accountabil-ity to anyone outside the classroom, very lowteacher expectations for academic potential, andvery real teacher concerns about students' sur-vival. Teachers seemed to define their roles as thatof parent instead of teacher, responsible forstudents' well-being and accountable primarily tothemselves, with no theory about how to combineattention to affective needs with academic learning.


2314


Stevens Middle School

Stevens is another school with little cohesion amongthe staff. When asked about accountability, oneteacher responded, "[It's] individual all the way."Another teacher, when asked who is accountable tohim, said, "I think the students are accountable to me,but who cares? Really, who cares? Except me."Here, too, there was a historical lack of formalaccountability. The school has scored well on tests inthe past, but the tests have variable impact on class-rooms. Instead, teachers reported autonomy overtheir practice and content. Like other teachers in ourstudy, they said they were accountable to themselves,to their students, or both. There was an emergingsense of formal accountability to the principal, be-cause the district's new educational reform planinfluences his retention, but this formal accountabilitywas based on the staff s "trust and loyalty for the pasttwenty years."

Administrators and teachers at Stevens agreed thattheir responsibilities reach beyond the schoolroomdoor, but there was little commonality in how theyfelt this responsibility should be met. One staffmember said:

Middle school is a special kind of place . . . werealized that we have to service the wholechild because some of the parents can't,they're not able to. We just can't focus on theintellect here, and that 's just part of the wholemiddle school emotional development . . .

that's a big part of middle school education. . . just helping them through these years.

Other teachers referred to preparing studentsacademically for high school, teaching organiza-tional skills, and helping students to enjoy learn-ing. When asked what influences what sheteaches, one teacher commented:

What I want my students to have as back-ground. Their futures, I think, [are] whatinfluence what I teach. I want them to havewhat they need to succeed beyond me and if itmeans doing a lot of rote kinds of things sothat in the future when they need to use thatkind of information for whatever comes next,they have it.

She said that she is preparing students "for thekind of education that [she] expects them to get inhigh school, based on [her] own experiences at[one of the city's exam schools], which was a veryacademically oriented program."

If accountability exists at Stevens, it is based on"a set of tacit assumptions that teachers knowwhat to do, that the principal knows what they aredoing, and that he knows they are doing a good job."Again, when teachers are isolated and there areneither clear expectations nor accountability withconsequences, teachers' responsibility rules.

Hutchinson High School

Hutchinson, a large urban comprehensive highschool, is similar to Phoenix in that order prevails, butthe emphasis at Hutchinson emerged from a recenthistory of disorder that escalated to fatalities. Theteachers expected the administration "to regaincontrol of the hallways, corridors and classroomsfrom the ruffians who ran wild about the building" andin return the principal made it clear that teachers wereexpected to take equal responsibility for "establishinga safe and orderly school environment."

Teachers at Hutchinson spoke explicitly about theimportance of order and civility. Their commentsabout accountability and teaching, however,reflected the same ambiguity and isolation asthose of teachers in similar schools. There was aformal teacher handbook, but few teachers oradministrators referred to it. The teachersdownplayed their accountability to the administrators.


2 4


One teacher said: "Nobody is going to checkto seewhat I am doing, but the headmaster will check to seeif I have adequate control, whatever, more manage-ment things and techniques and that type of thing."Another teacher reported "very little collegiality,"while still another said that "We impose our ownstandards." When asked where the standards comefrom, she said, "They're within us."

In addition to their responsibility to discipline,teachers described responsibility to look after theirstudents in a shepherding manner. Nearly allHutchinson teachers spoke of their responsibilityfor students' welfare, defined as, "staying out oftrouble, staying healthy, and doing what's neededto graduate and either get a job or gain admissionto college." One teacher said that his job was tomotivate his students and show them that he caredabout them.

Teachers' perceptions of their students' back-grounds are extremely important, and, whencombined with lack of accountability and lack ofcollegiality, shift the focus to the teacher's per-sonal responsibility. Teachers claimed responsi-bility for the welfare of their students both outsideand inside the classroom, and spoke of being moti-vated by their own experiences of good and badteachers. As at the Phoenix Charter School, whencurricular standards interfered with the teacher'ssense of what was right for her students, the teacherasserted her own opinion of the students' academicneeds. One teacher spoke of a new standard: "Thereis no way in hell I will teach Algebra 2 to kids who donot understand general math! . . . All students canlearn, I agree with that, but I don't think they cancome from middle school and be thrown into asituation where here we're setting themup for failure.Start with the first grade, keep them with us, andmaybe they'll succeed."

Summary

The schools described in this section share a commonsolution to the problem of accountability. They

delegate to the individual teacher most decisionsabout to whom the school is accountable, for what,and how. Accountability in these schools boils downto individual teachers' sense of responsibility. All ofthe schools had some pro forma internal accountabil-ity systems, albeit weak ones, such as teacher hand-books or prescribed curricula. All of the schoolsexisted within some kind ofextemal accountabilitystructurecharter laws, archdiocese curriculumframeworks, or local curriculum standards. But theseaccountability structures exercised no effectiveinfluence over individual teachers' sense of to whomand for what they were accountable.

Teachers in these schools tended to define theirsense of accountability entirely in terms of theirown sense of personal responsibility to what theyperceived as students' needs, both affective andacademic. Their responses are as notable for whatthey do not mention as for what they do. They didnot mention formal accountability systems, whicharewhat interests most school reformers. Instead theytalked about responsibilityto their students and tothemselves. They did not mention how they wereheld accountablebecause informal or formalsystems of accountability, even where they existed,had no reality in their daily lives. Regardless ofrecentchanges in state and local accountability systems,regardless of teacher evaluations, regardless of parentinvolvement, even regardless of the charter schoollaw, which is supposed to increase accountability,these teachers were still largely left to decide, basedon their own values, what and how to teach.

The beliefs of teachers that exercised the greatestinfluence on their sense of responsibility were thoserelated to the social backgrounds of their students.Teachers in these schools, in effect, decided on theirown what kind of education was appropriate forstudents from backgrounds they regarded as disad-vantaged. They spoke of "these children," with clearopinions as to what was required for children fromdisadvantaged backgrounds. The teachers stressedorder in the classroom and their own conceptions ofstudents' well-being, at the expense of academic


25 16

When Accountability Knocks, Will Anyone Answer? Abe lmann and Elmore

performance. These views were unchallenged, eitherby their colleagues' expectations or by externalaccountability systems, because these influences wereweak relative to the teachers' personal values.

The Emergence of CollectiveAccountability: ExpectationsInfluence Responsibility

In the previous section, we focused on schoolswhere individuals' conceptions of responsibilitydominated collective conceptions of accountabil-ity. Our sample of schools also included schoolswhere teachers' work was heavily influenced bythe expectations of other teachers, administratorsor community members. Strong expectations caninfluence and shape what a teacher, administrator,parent or student feels responsible for in his or herwork.

In this section, we highlight three schools charac-terized by strong mutual expectations. The leadcase in this section, St. Aloysius ElementarySchool, is one of the very few schools in oursample that was focused primarily on teaching andlearning. The graphic representation of thisschool in Figure 3 highlights the prominence ofcollective expectations and also reflects therelationship between responsibility and expecta-tions, due in large part to the principal's practiceof hiring candidates whose teaching philosophymatched her own. St. Aloysius Elementary is asmall Catholic school with a growing enrollmentthat is located in an affluent section of a universitycity. The focus on instruction is not the result ofany external formal directive or accountabilitysystem, but rather the combination of a strongschool leader and high expectations for students.Of particular note in this case is the way teachersproject and interpret parent expectations. Theassumption at St. Aloysius Elementary is that allparents have the same high expectations as thoseexpressed by the vocal parents of high socio-economic status.

The second case we present, North Beach High,also has a strong leader but the focus is on attain-ment, assuring that all students graduate. NorthBeach High is located in a blue-collar suburb of amajor city, the demographics of which haverecently begun to change from predominately Irishand Italian to a substantial Asian population.About 15-30 percent of the student population atNorth Beach is Asian, mostly Chinese. The thirdcase, Tatuna Point Elementary, is a K-6 schoollocated in the heart of an affluent suburb that isalmost exclusively white and Asian. The caseillustrates how powerful a force parents can be insetting high expectations for teachers and provid-ing the support that goes with those expectations.The parental presence at Tatuna Point Elementaryovershadows to a large extent the labor of teachersand administrators.

As we analyze the schools in this section, weshow how expectations can shape teachers' work. Insome cases the principal plays a central role, while inothers the community is ofmore importance.

St. Aloysius Elementary School

Figure 3.

St. Aloysius Elementary

IndividualResponsibility

CollectiveExpectations

Accountability


26

1 7

When Accountability Knocks, Will Anyone Answer? Abe [mann and Elmore

St. Aloysius Elementary serves a racially and ethni-cally diverse population of students in kindergartenthrough the eighth grade. The school has experienceda recent influx of Korean students, maintains a steadypopulation of Haitian students, and in any given yearserves several transitory European and South Ameri-can students whose parents come to the region tostudy. About 50 percent of the students are white, 40percent African-American and Haitian, six percentAsian, and four percent Latino/a. Less than onepercent of students are eligible for Title I services asdetermined by family income. Seventy-five percent ofthe students are Catholic. St. Aloysius Elementary'steaching staff, consisting of one teacher per grade, isentirely white. By the principal's account, she hasstudents whose parents work several blue collar jobs,and others whose parents are high-status profession-als and "university parents."

St. Aloysius Elementary is Mrs. Sharp's firstprincipalship. Since she assumed the principalshipfive years ago, the school has undergone analmost complete turnover in teaching staff. Onlythree of the current 17 staff members at St.Aloysius Elementary were there when Mrs. Sharparrived. She attributes this turnover primarily tonatural attrition through retirement, maternityleave and continuing education, but acknowledgedthat in some cases teachers chose to leave, havingidentified themselves as misfits with the school or withher approach to education. She said that at the endof her first year, "By virtue of things I said, peoplecame to understand what I valued. And so, whensome of those people left at the end of that year, Iwas able to hire people." She added that, "each timethat's happened, I've hired someone whose sense ofeducation and philosophy is very much in keepingwith my own." While the Catholic Schools Officepublishes a list of teachers available for hire, Mrs.Sharp was wary of relying upon that list because sheknew of at least one person on the list who she saidwould be "very inappropriate" in a setting withchildren. She said that she is fortunate that she isgiven a great deal of latitude by the church pastor to

hire teachers of her choice, although he has official,final authority.

All but one of the current teachers is Catholic, andthe majority are young professionals with fewerthan ten years' experience. Those teachers withmore experience reported a good working rela-tionship with their younger colleagues, saying thatwhere there might have been tension, instead therewas give-and-take with mutual learning. Becausemost teachers were hired within the past fiveyears, several in the same year, salaries andseniority are relatively uniform across the staff.Some teachers noted that they have an unusuallycollegial staff and attribute this relationshippartially to the fact that so many of them came toSt. Aloysius Elementary at the same time andlearned the ropes together. The teachers earnapproximately $10,000-$14,000 less than entry-level public school teachers in the area and manywork second jobs in the evening and on week-ends. Mrs. Sharp said she tries to be sensitive totheir work schedules when organizing meetings orschool events. She has also authorized teachers totutor students privately after school, which shesaid "provides them with the additional incomethey need." She added that, "It's also enabled usto reach out to the segment of the community thathad never been a part of our school beforethe ESLchildren."

The school' s immediate surroundings includeupper-end real estate, a few shops and restaurants,and within walking distance, a large university. Morethan half of St. Aloysius Elementary's students live inthe same zip code as the school, while the rest live inmostly suburban towns, some up to 45 minutes away.There was no playground in the small school yard atthe time of our visit, but a fund drive is underway topurchase equipment. The building is well maintained,with prominent displays of student work, mostlycompositions and test papers.



Not all students who apply to St. Aloysius Elementaryare admitted. If it appears that a child's needs cannotbe met at St. Aloysius Elementary, Mrs. Sharprecommends another school within the archdiocesethat she believes can better suit those needs, be theybehavioral, academic or other. When pressed on thispoint, Mrs. Sharp described an informal understand-ing between schools, and in the archdiocese, thatdifferent schools have different missions and "forgood reason." She described schools where theacademic performance was not equal to that of St.Aloysius Elementary, but explained that those schoolsprovide a real service to the inner city and immigrantcommunities which she perceives to have differentneeds.

Non-English speakers are fully integrated intoclassrooms at St. Aloysius Elementary. Theteachers have developed informal committees togather ESL materials and to share ESL teachingtechniques. Teachers report that their classesbenefit from exposure to other cultures andlanguages, and that they are able, though withsome difficulty, to devote the necessary attentionto both their native English and non-Englishspeaking pupils.

The St. Aloysius Elementary student population isracially and ethnically diverse, but it is less di-verse in terms of socio-economic background.Fewer than one percent of students are eligible forTitle I services as determined by family income.Not all children come from wealthy families, butfew come from abject poverty, so they are lesslikely than students in other schools describedhere to suffer the range of social, physical, andemotional risks associated with living in poverty.Unlike other schools in our sample, teachers at St.Aloysius Elementary made no reference to stu-dents' home lives or living environments as anobstacle to teaching or student learning. This islargely due to the students' predominantely middle toupper-class status, but it is also reflective ofa norm atSt. Aloysius Elementary that values individual respon-sibility for teaching and learning, and a perceived

intolerance for scapegoating ofany kind, even whenpresented with legitimate challenges to learning.

Tuition at St. Aloysius Elementary was $2,150 in1996-97, approximately twice that which is chargedby some inner-city Catholic schools in thearea.Some financial assistance is available in the secondyear ofattendance for families meeting the school'sneed criteria, but the admissions process is con-ducted without knowledge of families' economiccircumstances. When asked to describe how shebelieved St. Aloysius Elementary is perceived in thecommunity, Mrs. Sharp said she and thepastor boththink the school is viewed as "an inexpensive privateschool." She believed that many view the school asan alternative to the prestigious and expensive secularprivate schools in the area, and that those schools areSt. Aloysius Elementary's competition.

Expectations Shape Teachers' Work

Mrs. Sharp has had relatively free reign from thechurch pastor to exercise discretion in hiring staffand managing the school budget. Mrs. Sharp wasvery happy with her current teaching staff anddescribed them with terms such as "profesSional,""skilled" and "collegial." When asked what shelooks for in new hires, she said that they shouldbelieve "all children can learn, [and be someone] wholooks for the ways in which they learn and will have amultiplicity of activities .. . a person of good will andvalues. "

She attributed much of the coherence within theschool to having hired people whose philosophiesof teaching matched her own, but was quick to saythat she was not a directive principal, and that thestaff has developed into a cohesive group largelyon its own. As an example, Mrs. Sharp said thatthe teachers requested that one of the four facultymeetings per month be devoted entirely to colle-gial discussion related to curriculum and peda-gogy. She also noted that last year, the staff agreed toseek additional accreditation, beyond that awarded


2 8


by the archdiocese, because they knew they could getit and because the coherence and academic quality itrequired "would appeal to people looking for a goodschool for their children."

Without exception, teachers described an atmo-sphere of high expectations at St. Aloysius El-ementary. Some stressed a high priority on"reaching every child" and "making sure no one isleft behind," while others referred to a serious andsupportive environment where everyone is ex-pected to put forth excellent work. Teachers diddescribe a range of abilities within their class-rooms, and the particular challenge of teachingESL students, but none referred to this as anobstacle to teaching. Rather, they described aschool culture where teachers are expected toimprovise and to "reach everyone."

This belief in high expectations for all childrenapplies to both academic and social learning at St.Aloysius Elementary. Academically, children areexpected to achieve at the highest level possiblefor them. Teachers said, with varying degrees ofcertainty, that they believed every student canlearn the skills taught at their grade level, and inmany cases students exceed those expectations.When asked if all of her students could learn theskills expected of their grade level, one veteranteacher responded: "If I see they're havingtrouble, I'll tutor. They're given the time for help.We just stay with it until they know it." Shecontinued, "I never worry [about them going tothe next grade] because they always know whatthey're doing." These comments reflect theteacher's philosophy, and her expectation thatstudents will "stay with it" too. Students praisedthis particular teacher for her willingness to giveextra help, for her unbending belief in them, andfor her equally unbending expectation that theywill learn and retain what she teaches them.

Teachers recognized that students have varyingability levels, and they described the challengesthey face in teaching ESL students, and that these

students face in learning. The teachers maintained"high expectations" for the ESL students byinsisting on the highest degree of effort while, insome cases, adjusting performance expectations.For example, the fifth through eighth gradeteachers developed an ESL program that definedwhat teachers expect of their non-English speak-ing students. One teacher told us: "We expectthem tO increase their English understanding andcomprehension of English. We expect them tomaintain math skills and improve. We listed a setof criteria that we're going to expect from ESLstudents."

Student report cards in the upper grades at St.Aloysius Elementary have a column for perfor-mance and another column for effort. There is noeffort column in the lower grades, but teacherswrite comments which include a description ofstudent effort, behavior and progress. Only oneteacher described occasionally inflating letter (ornumber) grades based upon student effort orextenuating circumstances that might be particu-larly challenging for an individual child.

When asked how they were able to maintain highexpectations for all children, despite the range ofstudent abilities and preparation, several teacherssaid that they did not expect identical work fromevery child, but performance grades accuratelyreflected the range in student products, and effortgrades (or comments) focused on the expectationthat every student do his or her absolute best.Although not every child can produce exemplarywork, those children putting forth their best effortscan be rewarded with an "A for effort." Teachersindicated that students were not graded in com-parison with one another, or on a curve, but oftenon the basis of rubrics. One example of a rubricwas a scoring sheet the teachers developed for theScience Fair. Students first received an informa-tion sheet explaining what was expected of them.Several weeks later they received a sheet statingthat their topic would be due on a certain date, andtheir outline due on another date. Finally, they


29


received the actual scoring sheet that listed all thecriteria upon which they would be graded.

Most faculty members assumed students wouldfinish high school, some of them graduating fromprestigious high schools, and that the "vast major-ity" would go to college.

When asked what parents expected of them andthe school, teachers responses seemed to placeequal emphasis on instilling Christian values andon challenging the students. One teacher, compar-ing St. Aloysius Elementary to another schoolwhere she taught, commented:

[The previous school] was much more work-ing class, very few of the parents had gone tocollege and education was not number one oneverybody's list of priorities . . . Whereas here,I think people really respectyour pushingtheir kid to do their best and I like that . . . Thestudents are much more motivated [here]. Theparents are much more supportive . . . and thestudents, quite honestly.. . . my students seemto be smarter and more interested in doingwell and living up to the expectations that Ihave set for them, they have set for them-selves, that their parents have set for them as

Another teacher described the school as being inthe business of "educating the whole child," andsaid that parents expected that teachers would bethere before and after school "modeling thatphilosophy" for the students.

Several parents verified this assessment ofparentexpectations. One parent said that she worried, atfirst, that the school might be too much pressurefor her son, but that she's discovered he thrives inthe "challenging environment." Other parents ofyounger children expressed a desire to get theirchildren an "early start on their education," and inan environment that is disciplined and orderly.

When asked what parents expected, Mrs. Sharpremarked:

I think all the parents are setting high expecta-tions even though some of them may not beable to articulate them very well. I feel verystrongly that they all want high academicexpectations for their children. They all wanttheir children to be good human beings . . .

There is a group that will voice that morestrongly than many others will and so yes, wedo respond, we hear them and we considerhow we're going to respond to them. For theparents who may be less able to articulateexpectations or maybe less aware of thequality of education that is being provided totheir children, I think they recognize there'ssomething special. I try to deal with theparents on an individual basis, as opposed toa movement . . . We have our parking lotbrigade here, we have a few parents . . . I'vegot a few teachers here that are very adept atdiffusing that. The very unity ofour philoso-phy helps that.

The parents we met in a focus group described avariety of expectations for the school. Some ofthe parents' expectations focused specifically onteachers, but overall, they seemed not to differen-tiate individual people's roles. One father said:

I guess I certainly expect the school to edu-cate. You know, academic education is cer-tainly what they start out with, and the schoolseems to do that well. And the only way thatcan happen is if the environment in the classesallows the children to do that and I think thatcomes from the expectations that the teachershave of the children and I think that is adifference that seems to . . . that is a differencebetween this school and some of the publicschools that I've heard kind of anecdotal talkof In some of the public schools, some of the


3 021


kids seem sort of lost somehow. And thatdoesn't seem to happen so much here at [St.Aloysius Elementary].

Parents also described wanting frequent commu-nication from teachers and the school, and want-ing the school to be responsive to their children'sindividual needs. They remarked that teachers atSt. Aloysius Elementary do not teach the samelessons repeatedly, but seem to vary their teach-ing. Parents noted that the teachers continue theirown professional development, something theparents valued and expected, and believed todistinguish St. Aloysius Elementary from publicschools. Teachers perceived that parents havehigh expectations of them and of the school.

Speaking to the question of expectations ofteachers, and the school as a whole, a parent whowas highly involved in the school commented onthe effects of social-class, and proximity to theuniversity:

The higher the level the parent educationally,the higher level the child will reach . . . theparents are going on to post-graduate workand then they're expecting at least theirchildren will get to that level and I do think itlifts the place . I think it is a very goodinfluence on the school that that's there. It'slike strings from above pulling you up.

This comment speaks both to the way parents atSt. Aloysius Elementary were perceived by staff,and to the expectations those parents communi-cated in various was to the staff.

Teachers' Sense of ResponsibilityReflects Expectations

The staff at the school responded to what wasexpected of them by the principal, parents andtheir colleagues. Those collective expectationsaffected teachers' personal sense of responsibility.

Their sense ofprofessional responsibility was largelyinformed by the schools' collective interpretation of itscommunity' s needs and expectations.

Without exception, St. Aloysius Elementary'steachers expressed feeling responsible for thelearning of every individual child, and for main-taining high expectations for all children. Every-one described feeling responsible for their stu-dents' social development. In some cases this wascharacterized as religious teaching and in othersas training in good manners and behavior.

When asked for what she felt responsible, oneteacher responded definitively:

Well first, academics. To make sure that thechild is learning what they should be learning.That they are on level. If they are above level,that they're being challenged. If they're belowlevel, that they're receiving the extra help theyneed. As far as, like, socially, teaching themthe right and wrong . . . Even though noteveryone is Christian, you're teaching themabout God and loving each other and workingtogether. So, I want to develop them . . . Myresponsibility is to develop their mind, aca-demically, develop their soul or spirit, mor-ally, and [help them] to be able to survive inlife.

A first-year teacher's comments about her per-sonal sense of responsibility cover most of thepoints raised by her colleagues:

I feel that I am responsible for teaching themthe tools so that they are accountable, so thatthey are responsible for themselves, theirwork I think I'm responsible for setting ahigh level of expectations so that they knowthat's what they're expected to meet, so that Idon't accept mediocre work I definitely feelresponsible for that . . . for setting a tone inhere that's serious yet light-hearted enoughthat they feel comfortable enough to interact


3 122


with me and they're not scared of the teacher.I feel responsible for sending home informa-tion so their parents are up to date as towhat 's going on exactly in this class every day.I feel responsible for them going home everyday.. . . and knowing what they've learnedsothat they can't say, "nothing" when theirparents ask because they know [my] ears burneven i f I'm millions of miles away.

Several teachers described doing whatever wasnecessary to help a child learn to his or her fullpotential"to do the best that he or she can."Some teachers implied that they "just get toknow" students' abilities and work habits, but noone offered an explicit explanation of how theygauge a student's potential, or what is his or her"best."

Those teachers who said they felt responsible for"individualized learning" and "educating thewhole child" explained that this means makingoneself available to students for extra help beforeand after school. Others emphasized teachingchildren with varying learning styles differently.When asked to expand upon this idea, one teachermade a clear distinction between students whosegrades suffer because they do not do their work,and those who "don't get it" and do poorly ontests:

. . . f you're choosing not to turn in yourhomework, not to do your assignments, I thinkthe responsibility falls on you [the student]. Iftest grades are the big problem, I think some-times I'd look to me. Why isn't, if half theclass is not understanding what 's on the test,that's my fault I think And then I'd look athow I was teaching or what I was neglectingto teach or what was the method I was usingthat wasn't reaching half the class.

We asked teachers what responsibility they felt tocompensate for problems students might experienceout of school, at home or in their communities. The

teachers indicated that this was not a big issue at St.Aloysius Elementary, but described sensitivity to suchproblems as part of their jobs. Examples offeredusually related to marital problems between parents,or parents whose work schedules prevented themfrom being as involved with their children as otherparents. They implied a willingness to addresschildren's social, emotional or physical well-being aspart oftheir "whole child" orientation, but they clearlydid not view themselves as solely or primarily respon-sible for these non-academic areas.

Other Schools

North Beach High School

North Beach High School has just over 1,200students and is located in what one teacher calledan upwardly mobile blue-collar community. Themajority of the student population is of Irish orItalian descent. About 25 to 30 percent of thestudents are Asian, mostly Chinese, who haveproficient language skills; high school studentswho are not proficient in English attend the otherhigh school in town. The school operates as onelarge family. As one respondent said, "It's like abig family, you know, and I think a lot of thingsget done sometimes based on relationships as op-posed to a structure." In interviews, two teachersidentified the principal and assistant principal as the"mother and father" of the school. Many of the 175adults who work in the school are graduates orparents of graduatesboth the principal, assistantprincipal, and three deans graduated from the school.North Beach High School teachers and administia-tors grew up together, attended the same schools andchurches, and shared cultural traditions. Membershipin this close-knit community includes a sense ofobligation to take care of all of its members. In thecase of schooling, this means that children will gradu-ate from high school.

North Beach High School is committed to seeing thatall students graduate and, essentially, does not let anystudent drop out. The assistant principal explained,


3223


"Basically, we hang onto kids forever . . . we workwith the one to two percent who drop out, we workwith them forever and we try and try. . . . becausewe're going to pay now or later." The principaladded, "I'd like to have a dollar for every kid whodropped out of school and who came back andearned it." North Beach High School reports a dropout rate of 1.3 percent. The assistant principalwasconcerned because it had "creeped up" from 1.1 to1.3 percent. She said that she would "even violatethe attendance policy on the side of kids" to keep thethem in school. The student advisor system is tied intothe homeroom structure and assures that one adultconsistently touches base with every child each day ofhis or her high school years. Students have the samehomeroom teacher for four. years. The homeroomteacher goes to graduation and gives their homeroomstudents their diplomas.

The school administration is more laissez faire oninstructional issues. The principal had confidencein the subject knowledge of his teachers andexpected them "to perform their best, realizingthey're all differentdifferent personalities,different vocabularies sometimes .

,1

Tatuna Point Elementary School

Tatuna Point is an elementary school, located in arelatively affluent suburb, where academic expec-tations are high and student achievement is takenseriously. Second graders talk about going to IvyLeague colleges and teachers know that is whatmany parents expect. As one lower grade teacherexplained, " . . . standards are very high academi-cally; it's expected by the parents, the community,and the staff, and we work hard to meet thoseexpectations." Parents are involved in almostevery aspect of the school. Parents are vocalabout expressing demands, and they are active inproviding support for what they demand. Parentsknow how to articulate their demands loudly and areready to take the steps necessary to achieve themwhether by voicing their discontent to teachers,

administrators or district officials, or by organizing andparticipating in formal institutions that control andregulate school activities. As one parent said:

There's a lot ofparent involvement . . . Parentsare willing to be very vocal and express theirconcern, or shall we say "whine." Most of theparents, because they're highly educated, puta real premium on education, and thereforeexpect a lot from the schools . . . expect highperformance from the schools, and thereforemake a lot of demands on the staff on theprincipal, on the curriculum. But on the otherhand, the majority of those parents say, "Iwant this, but what can I do to help you getit?" . . . My take is that they're willing to backup their demands, ifyou will, with support,either financial or hands on.

The parents do in fact back up their demands.According to the principal, last year the PTAcounted over 8,000 hours of volunteer time, notincluding those who forgot to sign in. On anygiven day, parents can be found assisting theschool secretary, working in classrooms, shelvingbooks, staffing the computer lab, or helping coacha sport or other activity. The PTA donates$50,000 yearly in capital goods suchas playgroundand computer equipment. In 1991, a group ofparents established the Tatuna Point ElementaryEducational Foundation, a non-profit foundationdedicated to raising funds to provide additionalresources for the improvement of the quality ofeducation. Each family at the school is asked tomake a cash contribution of $350. Approximately 40percent of school families give to the Foundation.The funds have supported additional teachers,consultants, and classroom aides. Students at theschool also raise money through an annual walk-a-thon sponsored by the Foundation. Parents are alsovery active on the school council. The principalexplained, "The Foundation buys people, the PTAbuys stuff. . . and the school council keeps it allcoordinated and going together." This parental effortovershadows the work of the teachers. One parent


3 324


noted that as a newcomer to the school, she heardmuch about the parentsthe PTA, the Foundation,and the site councilbut very little about teachers.

Despite the school's academic program and relativelyhigh performance levels, parents at Tatuna PointElementary questioned whether their children weresufficiently challenged academically. Parents werefrustrated by what they claimed was a lack of indi-vidualized attention. As one parent explained, " .

the teacher [ought to] get to know each child andknow how to deal with each child separately insteadof expecting all thirty-two to do the same thing."Most parents were confident in what they believedwas best for their child's education and would exerttheir influence in the classroom, in the principal'soffice or at the district level to ensure that their childbenefitted from a high-quality education tailored to thechild's particular needs. The principal explained,"They know how to use the system. They know howto access the system . . . If they didn't like what I did,they know who to go to. They know who is myboss. They're not shy about calling the superinten-dent if they have a problem."

The Tatuna Point Elementary Education Founda-tion has been instrumental in getting classroomaides for the lower grades. The aides are closelymonitored by the Foundation's board of gover-nors. The Foundation, the school site council, andsome independent parents conducted an evaluation ofresource teachers and classroom aides to assesswhether students received more individualized atten-tion as a result of the increased support. Teacherswere requested to keep a log of aides' time every dayfor one week and classrooms were formally observedby a team of parents. The evaluation report stated,"In the primary grades, 77 percent of aide time wasspent working with small groups or individual studentsand 23 percent was spent in clerical duties. In theupper grades, 38 percent of aide time was spent inworking with small groups or individual students and62 percent was spent in clerical duties." Parentsfound the upper-grade condition unacceptable andrequested that the principal discuss the issue with the

upper-grade teachers to assure that classroom aideswere used more appropriately. One teacher ex-plained if they pay for the aides, they can dictate howwe use the aides. The same teacher summarized theparent sentiment by saying, "We're not going to fundit unless you're doing it our way."

Parents also disagreed with the local district'sopposition to tracking students according toability. There are gifted and talented classes butthey are limited to the top two percent of children.Parents were frustrated by the limited availablespace and felt that their children remained in asystem designed for "less able" students. Parentsat Tatuna Point often bypassed the school leader-ship to go directly to the district to assure theirchildren's placement in the accelerated program.

Although teachers appreciated and welcomed theparticipation and support of parents in classes, andvalued the additional resources they brought to theclassroom, they also resented when parents madedeterminations about how teachers should do theirjobs. The parent activism was certainly felt by theteachers and the principal. Teachers knew that ifthey were not doing what parents expected, theywould hear about it, either verbally or in writing,by way of the principal or even the superinten-dent. One upper-grade teacher commented, "Ithink my greatest pressure comes from the parentsand from what they're asking of me or what theirexpectations are of me." The same teacher said, "Iam happy when I have no parent letter in mybox."

Teachers at Tatuna Point worked long hours andthe work was certainly influenced by the demandsof parents, whether expressed directly or throughthe principal. In spite of outside pressures fromparents, district requirements or administrativemandates, the teachers still asserted that theiractions were first and foremost driven by whatthey viewed as the children's best interests. Theteachers did appear to do what was expected ofthem, but the principal argued that they did so


3 4


from self-motivation and personal dedication. As theprincipal explained, "The accountability structure hereis very often self-imposed by the teachers."

Summary

The schools in this section demonstrate a differentsolution to the problem of accountability. Thefirst group of schools essentially turned all ac-countability problems into matters of individualteacher responsibility. These schools have alldeveloped, in somewhat different ways, a rela-tively powerful culture of expectations that shapesindividuals' views around a common purpose.These schools operate without highly visibleinternal accountability structures, but they accom-plish many of the same purposes through expecta-tions. At St. Aloysius Elementary, Mrs. Sharp hasbeen highly influential in constructing a commu-nity of teachers, students, and parents focused onacademic learning, largely through the influenceof strong expectations. At North Beach High School,the culture of common expectations comes from boththe school leadership's ethic of a"family" environ-ment, and from the cohesive culture of the localcommunity which is transmitted to the school throughthe staff, who are natives ofthe community them-selves, and graduates of the school. Expectations atNorth Beach are focused mainly on attainmentgetting students to stay in school and to graduaterather than on academic learning. The expectations atTatuna Point Elementary seem to originiate largelyfrom aggressive and demanding middle-class parentswho pressure teachers and administrators, the lattersee themselves as somewhat beleaguered but heavilyinfluenced by these parental expectations. Theexpectations of parents in Tatuna Point Elementaryare beginning to translate into an incipient accountabil-ity structure evolving in the Foundation's stronginfluence over the expenditure of its funds.

In all three cases, collective expectations exercisea heavy influence on teachers' individual concep-tions of their responsibilities. Mrs. Sharp deliber-

ately selects teachers who share her views that allstudents can learn; the teachers at St. AloysiusElementary share a view that de-emphasizes familybackground as a determining factor in student learn-ing, instead emphasizing student and teacher effort.At North Beach High School, collective norms aboutthe custodial role of schools and the importance ofattainment heavily influence the way teachers think oftheir work with students. And at Tatuna PointElementary, teachers internalize the norms of competi-tive academic achievement or risk the disapproval ofparents.

Internal Accountability:TheAlignment of Responsibility,Expectations, and Accountability

In the schools we have examined so far, individualconceptions of responsibility and collectiveexpectations tend to guide the actions and motiva-tions of teachers. These factors appear to operatein a way that is incidental to any formal arrangementsor consequences that are visible within the school.Teachers may feel responsible for maintaining order inclassrooms and hallways, and this sense of responsi-bility may be translated into shared expectations, forexample, but in many schools there are no visiblearrangements for enforcing this obligation and little inthe way of direct consequences for failing to meetthose expectations.

In this section we discuss three schools in which astrong internal accountability system has emerged,which appears to influence the actions of mem-bers of the school community. These threeschools operate within quite different externalaccountability policy structures, yet within eachschool, teachers (and parents in the case of one of theschools) are held accountable for meeting a set ofshared expectations. Regardless of the differences inthe external policy structures, in these schools ac-countability is a strong internal operating principle.



We consider how expectationscan shape an internalaccountability system within these three schools.While not a formal external policy mechanism, theinternal accountability system appears to stronglyinfluence teacher behavior, and corresponds closelywith teachers' understanding of their personal respon-sibility. Unlike these discussed previously, the schoolsdiscussed in this section are characterized by visibleaccountability structures withconsequences for failureto meet set expectations.

These schools illustrate the idea outlined in ourworking theory that internal accountability sys-tems are likely to influence individual actions ifthey are closely aligned with individual responsi-bility and collective expectations. These threeschools vary in the content of their shared expec-tations, but they are similar in that the alignmentof personal responsibility with shared expecta-tions, combined with some consequences, has ledto an internal accountability system that actuallyaffects actions and behavior.

The internal accountability systems in these threeschools appear to have the greatest impact onbehavior, but they still operate within an externalpolicy structure. The degree to which the externalpolicy structure affects behavior appears to berelated to the degree of alignment between theexternal policy and the internal accountability system.If there is a conflict, the internal system appears tohave a greater influence on behavior.

The lead case is Turtle Haven, an urban pilot school,locally-chartered elementary school, serving a highproportion of minority and disadvantaged stu-dents. Turtle Haven demonstrates the emergenceof internal accountability in an environment ofshared expectations for high-quality academicwork, as represented by the high degree of align-ment in Figure 4.

Saint B's is an urban Catholic elementary school,serving a predominantely working-class studentpopulation. St. B's demonstrates an internal account-

ability system extended to include parents. PineCreek, an urban elementary school serving predomi-nantely poor white students, demonstrates the align-ment ofresponsibility, expectations, and internalaccountability around relatively low expectations forstudent academic work.

Turtle Haven Pilot School

Rgure 4.

Turtle Haven Pilot School

IndividualResponsibilit

,44Info

Accountabili

CollectiveExpectations

Turtle Haven Pilot School is a small elementaryschool located in a low-income community in alarge eastern city. As a pilot school, or a locally-chartered school, Turtle Haven is administrativelyaffiliated with an urban school system ofover60,000 students, but exempted from many of thedistrict regulations. The building itself is not owned bythe school districtthe school is housed in an unusedwing of a parochial school. The building is old and indisrepair, but the display of student work in thehallways and classrooms livens up the atmosphere.

Although located in an old building, this is abrand-new school, only in its second year of opera-tion. The city's pilot schools were foundedas part ofan agreement with the local teachers' union in aneffort to provide models of excellence which wouldspread to all of the city schools. Pilot schools areexpected to be "models of innovation," and theiradvocates believe they will lead to improved student

CPRE Research Report Series, RR-42 36BEST COPY AvAILABLE

27


performance in other city schools. Because pilotschools are expected to provide models for other cityschools, the models are supposed to be replicablesystem-wide, although the city has no formal plans forreplication.

The Turtle Haven Pilot School was originallypromoted for its "technology-based curriculum,active parent participation, and individualizedinstruction." The school opened with grades Kl(areadiness class for four-year olds), K2 (the secondkindergarten year), first grade and second grade,then during its second year expanded to includethird grade. The school also operates an ex-tended-day program, providing after-schoolservices for children until 5:30 in the evening.

There are currently 200 children enrolled, 57percent of whom are eligible for free or reducedmeals. Turtle Haven participates in the city 'schoice system, and as a result, students are bussedfrom many different parts of the city. About 65percent of the students are African-American, 25percent Anglo, nine percent Latino/a, and onepercent Asian and Native American.

The classrooms at Turtle Haven have many similari-ties. Desks are arranged in clusters of fouror five.Each room has a meeting area with a big blue gymmat on which children sit. In each classroom, theteacher posts a morning messagea handwrittengreeting outlining the day's activities. Children appearto be actively engaged in work, frequently working ondifferent projects at the same time. Most classroomsare not quiet, but the noise appears to be the produc-tive sound of children working.

As a pilot school, Turtle Haven is exempt fromunion regulations governing the hiring of staff.Hiring is conducted directly at the school site, andas a result, the staff characteristics are somewhatdifferent from those of a typical city school.Many of the teachers are youngin their mid-twentieswith little prior teaching experience. Theteaching staff is diverse, with six African-American

teachers, six Anglo teachers, two Latino teachers,and one Asian teacher.

External Context

Turtle Haven has a unique external accountabilitycontext. Having been granted a pilot status, theschool is evaluated every three years, and intheory can be closed if it has not met the goals ofthe pilot school initiative. It is not yet clear howsuch evaluations will work. Staff at the schoolexpressed confidence, however, that whateverevaluation mechanism is used they will certainlymeet and exceed the district' s expectations. Thethreat of losing pilot school status did not have astrong influence on the teachers' understanding ofaccountability. In this way, Turtle Haven is similar toPhoenix Charter School in that the accountabilityarrangement under which it operates is not a heavyinfluence upon conceptions of accountability, insofaras it offers greater freedomfrom accountability.

Turtle Haven also operates within the context ofdistrict-wide mechanisms designed to hold all cityschools accountable. Since hiring a new superinten-dent, the district has embarked on an ambitioussystemic reform initiativeseveral components ofwhich relate directly to accountability. For example,this reform includes the implementation of"CitywideLearning Standards" which outline specific expecta-tions for each grade level. Children must demonstrateperformance in relation to the standards by creatingthe "products" designated for each grade level.These products are performance-based projectsdirectly related to the objectives of the learningstandards. The district also administers the Stanford9 achievement test each year. This test was selectedbecause it was most closely aligned to the learningstandards.

Although the district has developed standards anddesignated products, it is still not clear to teachershow these products will be used to judge a school'ssuccess. Most of the teachers at Turtle Haven were

CPRE Research Report Series, RR-42 3 7 28


working to help all students meet these standards, butthey were unclear whether or not the district wouldcollect the products, and if so, how the district wouldevaluate the products. The district has not providedschools with a detailed rubric by which to evaluate theproducts, so it is difficult for teachers to determine"how good is good enough?" when evaluating studentproducts. This confusion was not unique to TurtleHaven, but common among the public schools inthis district included in our sample. Turtle Havendiffered from other public schools, however, in itsefforts to incorporate the standards and products,vague as they were, into the school' s self-gener-ated academic program.

Nor was it clear to teachers how standardized testingwould be used to hold schools accountable. Severalteachers did mention that they felt accountable forstudent performance on these tests, but others saidthey felt more accountable to parents than to thedistrict for a student's test performance. One teacherexplained, "Formal assessments, like the Stanford 9,parents want their children to excel on those types oftests. We don't teach to the test, but we have to beaccountable for how the children do on the test."

The district has begun to design mechanisms forholding schools accountable, but these mecha-nisms are not yet visible at the school site level.There are very few rewards and sanctions thatrecognized a school's performance. Therefore, wecharacterize the external accountability context asrelatively weak. Within this weak external ac-countability structure, however, Turtle Haven hasdeveloped its own internal accountability systemwith a strong set of expectations closely alignedwith personal responsibility.

Expectations

They (expectations) taken together create sortof a school culture . . . that's very defining anddistinctive and says, this is what we 're about.(Turtle Haven teacher)

Turtle Haven's principal, Maly Carter, has made avery deliberate effort to make her expectations clearto teachers. It is no coincidence that classrooms lookvery similar. At the beginning of the year, she distrib-uted a list of "components or things" that should befound in every Turtle Haven classroom. This listincluded physical objects, such as a morning messageboard and student work posted on the walls, andactivities such as choice time and morning meeting. Inthis document, she outlined her expectations aboutparent involvement. Each teacher is expected to holdfour family events during the year, and 100 percentattendance is expected. Teachers must also holdparent conferences and communicate regularly withparents by letter and phone.

Interviews with Turtle Haven teachers indicatedthat most of the staff have internalized theseexpectations and make every effort to live up tothem. The responses of most teachers regardingthe principal' s expectations closely matched whatthe principal told us are her expectations. Severalteachers listed expectations that repeated the writtenguidelines Ms. Carter had presented at the beginningof the year. Among the expectations they believedMs. Carter had, teachers mentioned choice time, useof the Responsive Classroom model, morning meet-ing, and planning successful family events.

One teacher, after listing many ofthe expectationsoutlined in Ms. Carter's written guidelines, explainedthe importance of these clear expectations. Shereferred to the "components or things" as:

Structures that support expectations we 'yeagreed we would have. I feel that 's important.It's one thing to say, theoretically, it's your jobto involve families. It's another thing to say,here 's a way we expect you are going to do it,which is to have a certain number of class-room events. That makes it concrete andmakes it so it's clear.


3 8


The specific expectations noted by the teachersappear to be directly connected to Ms. Carter'sbroad and general expectations of teachers and thelearning experiences they provide for their stu-dents. On this subject, Ms. Carter commented:

One thing is around curriculum developmentand being able to construct a learning envi-ronment that engages all children, that allowschildren to grow and develop and learn inways that . . . in creative ways, in thoughtfulways, in ways where the teacher can take on alot of different roles. Not just standing andgiving information, but being a facilitator andthis sort of person that guides their under-standing

In addition to this expectation of academic stu-dent-centered instruction, Ms. Carter also men-tioned the importance ofa social curriculum andcommunity-building. All teachers were expectedto use the Responsive Classroom model, and wereprovided with the necessary training.

Ms. Carter sees community-building a's her mostimportant responsibility as a principal. Sheexplained,

What's most important is instructional prac-tice and curriculum, but before that is commu-nity, is building community. A very significantlearning community. But then everything elsefalls under that . . . I see my work as a princi-pal as being more of a community activist, andmore of teaching children what is possible inhealthy communities than what is alreadyexisting. Giving them another model, anothercontext in which to see the world, and whichto see themselves.

The teachers' understanding of these generalphilosophical expectations appeared to be closelyaligned with what Ms. Carter told us. For ex-ample, when asked what Ms. Carter expects of her,one teacher responded, "You are expected to think

about how what goes on in your room emanates fromkids' interests." Other teachers responded thatteachers were expected to "teach in a meaningfulway, and to have high expectations of kids and toteach social behavior too." One teacher said Ms.Carter thought that, "the social curriculum is on equalfooting with the academic curriculum." Severalteachers referred to the expectation thatthey use theResponsive Classroom model. One teacher ex-plained the model's emphasis: "Teaching kids how tolisten to each other, how to ask questions, how to bekind, how to be helpful, how to be responsible, andthat's a huge part of the day."

Teachers at Turtle Haven work in an environmentof clearly articulated administrative expectations.The expectations are communicated throughwriting, through informal conversations, andduring staff meetings and professional develop-ment activities. Our observations in the schooland interviews with teachers revealed that mostteachers made every effort to align their teachingpractice to these expectations.

Ms. Carter has clearly articulated her expectationsfor pedagogical practice and technique. She alsohas expectations related to the expectations theteachers have of their students. Although theschool works with a large percentage of poor andminority students, a population for which otherschools in our sample set relatively low standards,the teachers at Turtle Haven believe that allchildren are capable of learning at high levels. Teach-ers in other schools we studied explained low studentperformance by arguing that "these kids have so manyneeds, it is more difficult to teach them." We did notencounter this argument at Turtle Haven.

Ms. Carter explained her expectations for studentperformance:

I have higher expectations (than the City) . . .

When I first came to [City] and I lookedatthe standards the teachers were using, I wasappalled. I mean I was really . . . This was


3 9


before the new standards came out. So I tookthe standards from [Neighboring City] whichI thought were more comparable to what mythinking around what kids should be able todo in first grade and used those. Yes, I thinkmy expectations for learning are higher ingeneral for students not just in first grade, butin general. I don't look at a child and assesswhat they're capable of I feel like so much ofthat goes on in [City].

Teachers at Turtle Haven appeared to mirror Ms.Carter in her philosophy about expectations ofstudents. All teachers strongly resisted the beliefthat children from disadvantaged families andcommunities were less likely to succeed. At ameeting with the entire school staff and theresearchers working on this project, the teacherswere asked the source of their expectations forchildren. One teacher referred directly to Ms.Carter's conviction that all children are capable ofgreat things, and mentioned that this belief leadsteachers to challenge all students. At this samemeeting, two Turtle Haven teachers spokestrongly against the tendency in inner-city schoolsof characterizing families and communities as thesource of the problem. One teacher called thischaracterization "vindictive and victimizing" andenabling educators to make excuses. Anotherteacher said, "Inner-city kids do not need to besaved, they need to be treated fairly and withrespect."

Expectations Influence Responsibility

In this context of strong expectations, how do expec-tations influence or work in conjunction (or conflict)with individual teachers' sense of personal responsi-bility? Unlike schools discussed earlier in this report,Turtle Haven appears to have a close alignmentbetween personal responsibility and the principal'sexpectations. In some cases, it seems Ms. Carter'sexpectations shaped an individual's understanding ofresponsibility, while in other cases, it appears that Ms.

Carter has hired teachers who already possess asense ofpersonal responsibility that is closely alignedwith her expectations.

For example, many teachers mentioned feelingpersonally responsible for their students' socialdevelopment. One third grade teacher, when askedfor what she felt responsible, referred to many"study skill" related areashelping children todevelop the ability to work independently, and todevelop their confidence and listening skills. She alsomentioned a responsibility for teaching students torespect each other, cooperate and respect adults.Although she mentioned that these were school-wideexpectations, she also felt very responsible for thempersonally.

In this example, there was a clear alignment betweenthe teacher's sense of personal responsibility andtheexpectations outlined by Ms. Carter. Another teacherdiscussed the importance of such an alignment: "Notthat we're trying to prove something to make Marypleased, but that we really believe and we buy intothe practice that all children can and will learn if youset the stage, and you set high expectations. And ifthese are your ideals, not rhetorical ideals, that youbelieve in, you have a place here."

This alignment within the Turtle Haven school wasalso apparent in the case of the one teacher who sawa conflict between her sense of personal responsibilityand Ms. Carter's expectations. This particular teachersaw her philosophy of teaching as more traditionalthan that of Ms. Carter and the other teachers in theschool. She believed that:

Before they're allowed to go, they should begiven the basic steps . to have self -controlto be able to know that, . . . 7 have somethingto do. I'm going to take my time to read thedirections. Do I understand?' Like i f I set upcenters and let them move through centers,they have to know that at the end ofa certainamount of time . . . these things have to beturned in. And my class just isn't there yet to


4 031


just go at centers. And I'm viewed as beingvery traditional because of that. In terms ofbehavior, I really feel that there should becontrol in the classroom. I can't teach ifeveryone is talking at one time.

Consequences: What Happens WhenYou Don't Meet Expectations

For the teacher quoted above, there was a conflictbetween her personal sense of responsibility and theexpectations set by Ms. Carter. For alignmentbetween expectations and personal responsibility tofunction as an internal accountability system, theremust be consequences if the alignment does not existor if an individual fails to meet the expectations. Forthis particular teacher, this lack of alignment led her toleave the school. This decision appeared to havebeen made jointly by the teacher and Ms. Carter.Ms. Carter explained the decision, "That was ameeting of the minds. And it's a good leaving. It's agood leaving because it's not a good match."

The teacher explained:

I'm told that I'm too traditional, that I expectthe kids to sit too much. That they're used tobeing out of their seats and that they shouldhave more choices, more choice time. So, thatwas the decision we came to since we don'thave the same philosophy. That we would justlet it go.

To understand how consequences transform expecta-tions into an informal accountability system, we askedseveral teachers the following question: "Whathappens at Turtle Haven if teachers do not meet Ms.Carter's expectations or conform to the schoolculture created by those expectations?" Mostteachers believed that a person who did not meetexpectations would first receive a great deal ofsupport from the principal and other colleagues.Teachers seemed reluctant to say if such a teacherwould eventually lose his or her job. Most agreed,

however, that a teacher who did not meet the expec-tations would not be happy at the school and wouldeventually "try to weed themselves out" of the school.As one teacher explained,

The administrator would initiate a lot ofsupport kinds of structures to try to help thatperson meet expectations, and that 's some-thing that would go on for a long time. Even-tually, if things were not able to come togetherand there was a sense in the community that acertain number of children were not able toget the kind of education that we say we'recommitted to providing, then I think at acertain point the issue would be, we have tothink about whether somebody belongs here ornot.

Ms. Carter was less reluctant to describe theprocess by which she addressed the issue ofteachers not meeting her expectations. We askedseveral questions about what would happen if ateacher did not meet the expectation regardingparental attendance at the four family events.

Ms. Carter: I would say ifyou've tried every-thing, if you've truly tried and you put allthose things in place. You gave them enoughtime, you did the calls and all that kind ofstuff you tried different times of the day.There's a lot that you would have to be doing.Then I would be very concerned about theteacher. I would be very concerned. Andthat's happening as we speak

Interviewer: Are they going to be back nextyear?

Ms. Carter: Right now, i f I had to decidetoday . . . No. And that's what the teacher'sbeen told. No.



Ms. Carter explained that deciding to let a teacher gois one of the most important decisions she makes asan administrator:

I don't feel there are a whole lot of decisionsthat I make alone in this school, But my job isto identi.ni the best educators for children.And to hold every teacher accountable forhigh-quality work for kids and families. Andit can be done. I don't accept a whole lot ofexcuses.

Internal Accountability

At Turtle Haven, the principal has established astrong set of expectations that guide teacherbehavior. These expectations are both groundedin an educational philosophy and relate to specificpedagogical practice. In some instances theseexpectations shaped personal responsibility inteachers, while in other cases it seems that indi-viduals were hired because their sense of personalresponsibility matched Ms. Carter's expectations.The autonomy in hiring at the pilot school hasbeen crucial in building this alignment betweenexpectations, personal responsibility and internalaccountability, although we did not observe thisalignment uniformly across the charter schools in ourstudy.

In the few instances where there was a conflictbetween personal responsibility and administra-tive expectations, or where a teacher was ineffec-tive in meeting expectations (even though shemay have felt responsible for similar goals), thefunctioning of the internal accountability systemwas clear. There were consequences for teacherswho failed to meet the expectations that had beenestablished. Consequences connect expectationsand personal responsibility to shape an internalaccountability system at the Turtle Haven PilotSchool.

Other Schools

Saint B's Elementary School

Internal accountability does not necessarily affectteacher behavior alone. At Saint B's CatholicSchool, we saw how strong expectations, establishedby administrators and teachers, aligned with personalresponsibility on the part of parents, created aninternal accountability system designed to holdparents accountable for involvement in their children'seducation.

St. B's Catholic School serves 600 students inkindergarten through grade eight. The school islocated in M-town, an incorporated city affiliatedwith one of the largest metropolitan areas in thecountry. Residents of M-town are mostly lower-middle class. Admission to St. B's is competitive.Before being admitted to kindergarten, students musttake an admission test that evaluates their fine motorskills and knowledge of letters, colors, shapes andnumbers.

The school is operated by the St. B Catholic Parish,and the parish pastor, Father L., is ultimately respon-sible for the school. In practice, however, it is theschool's principal, Sister A. who makes most of thedecisions for the school. Like Turtle Haven, theschool operates within a relatively weak externalaccountability system. The school is a member of thearchdiocese of the metropolitan wea, but archdioceseofficials emphasized that their influence over St. B'swas only advisory, and that the archdiocese officialshad no direct authority over the school. Despite theadvisory nature of this relationship, however, both theschool principal and many parents asserted that thearchdiocese in fact does have control over the school.

The most direct way in which the archdiocese influ-ences St. B's is the "scope and sequence" curriculum.Teachers are expected to cover this curriculum in theirclassrooms through the course of the year. Thearchdiocese also receives financial reports from the


42


schools and publishes guidelines on a variety of issuesincluding safety, governance, and curriculum. Despitethese guidelines, archdiocesan officials insisted that St.B Catholic School may not be ordered to followthese guidelines, although it is expected to do so, andusually does.

Within this external context, St. B Catholic School hasdeveloped mechanisms for ensuring that externalguidelines and internal expectations are met. Forteachers, these expectations are generated primarilyby the principal, Sister A. All of the teachers inter-viewed for this case study described their perfor-mance in terms of Sister A's expectations. She visitseach classroom weekly to inspect lesson plans and toreview the students' agenda books (pamphletsdistributed to all students which include school rulesand other relevant information and have weeklycalendars with space for students to note their assign-ments). In addition to these weekly walk-throughs,Sister A also evaluates all teachers annually using astandardized format including formal observation anda written report.

These internal mechanisms at St. B's exist to holdteachers accountable; more striking are the struc-tures designed to hold parents accountable. AtTurtle Haven the internal accountability systemworked to hold teachers accountable for parentalinvolvement, but at St. B's, parents are helddirectly accountable for their involvement inschool activities. As a competitive private school,St. B has the leverage to create and enforce rulesgoverning parental involvement in the schoolcommunity.

At the beginning of each school year, St. B Catho-lic School holds a mandatory parent meetingduring which school rules and policies are re-viewed. Failure to attend this meeting results in a$25 fine charged to the student's tuition bill.Parents are also expected to sign an annual con-tract indicating that they agree to follow theschool rules. These policies include a dress code and

other rules that parents agree to enforce, and aparental agreement to supervise homework and signtheir child's agenda book. Each familymust alsoagree to contribute 25 hours of volunteer work.Parents can, however, choose not to volunteer,instead making an additional fmancial or in-kindcontribution to the school.

This year St. B instituted a new policy designed toincrease communication between parents andteachers. All parents of children in the firstthrough third grades, without exception, arerequired to pick up their children in the classroomat the end of the school day. Parents are fined $1for every minute they are late.

It appeared that these expectations of parentalinvolvement were closely aligned with the par-ents' understanding of their own responsibilities.For example, one parent mentioned that parentinvolvement in the school was one of the aspectsshe valued most about St. B. She explained:"There's more accountability here. You have toparticipate and what I like about this participationis that you get to know the other parents.Whereas in public schools it's more of a drop-off,baby-sit kind of deal." In reference to the 25hours of mandatory volunteer work, she ex-plained, "I like that because you're not takingthings for granted . . . Whichever way you canhelp you can put in those hours. You knowthey're there. That is your obligation." At leastfor this parent, there appeared to be an alignmentbetween her own sense of responsibility and theexpectations of the school.

At St. B, the edge that turns expectations into aninternal accountability system is the fines imposedupon parents when they fail to meet expectations.The accountability system works most stronglyhere for parents, not teachers. Perhaps that isbecause the competition for admission to theschool is greater than the competition for jobs atthe school. Admission to St. B is extemely competi-



tive, but the school administiation has had difficultyrecruiting qualified teachers.

Pine Creek Elementary School

Expectations that lead to an internal accountabilitysystem do not necessarily originate from adminis-trators. At Pine Creek Elementary School, astrong set of expectations surround the subject ofthe students' capabilities and needs. These expec-tations function within the context of external policy,shaping personal responsibility and creatingan internalaccountability system.

Pine Creek Elementary School serves over 400students from early childhood through grade five.Pine Creek is located in the city of Flagston, asuburb of a mid-sized eastern city. The school islocated across the street from a large housingdevelopment where most of the students live.More than 75 percent of the students qualify forfree or reduced price lunch.

In 1994 Pine Creek Elementary School wasdesignated a Blue Ribbon School of Excellence bythe United States Department of Education. Theschool was selected because it was one of fourschools in the state with the largest relative gain infourth grade reading scores. Not only did PineCreek demonstrate significant growth, but theschool scored significantly higher than otherschools in its comparison school band.

Staff development at Pine Creek is designed tohold teachers accountable for meeting self-se-lected goals. Teachers meet in grade-level teamsevery Tuesday afternoon and work toward goals theyhave set as a team. At the end of the year, each teammust present its progress to the principal who in turnreports to the main office. One teacher said shebelieved this system of staff development served tohold them accountable, "I think it's accountability. It'sholding us as teachers accountable to the goals thatwe've set, and certainly often times it's harder to

show what you've done if it's not something concrete,if there is not something written to hand to some-body."

Staff development team meetings are also used todiscuss strategies for improving performance onstandardized tests. Pine Creek has been designatedas a Title I schoolwide project. The principalasserted that in order to maintain the additionallevel of funding associated with this designation,they must show 5 percent annual growth on theStanford Achievement test.

Within the context of these expectations forstudent performance defmed by the Title I law exists aset of collective expectations that relate to students'abilities. Unlike Turtle Haven, Pine Creek teachersdefine their work and expectations for students inrelation to their understanding of the children'sbackground. As one teacher explained, "Unfortu-nately, I've heard too many say around here that whatwe do here gets undone when they go home. Thebaggage that our kids carry with them when theycome to school, sometimes it's unbelievable that theyfunction as well as they do. So, a lot of them have ahard time, because there's nothing to look forward toafter school gets out." Another teacher describedthestudents' parents as follows: "They are one-parentfamilies; they are all on welfare, they don't have anymoney for breakfast; they don't have any money forsub[sidized] lunch, and there is not very goodparenting."

These perceptions of the students, parents, andcommunity are often closely linked to teachers'academic expectations of students. Teachersexplained that because of their backgrounds,students were not able to achieve as much aca-demically as other students in Flagston. Forexample, one teacher explained that she wasparticularly proud of the progress the school hasmade in test scores because of the home environ-ment of most students.


4 4


How the teachers perceived their students and theirbackgrounds appeared to be the most influentialfactor in shaping teacher practice and attitudes. Forexample, the current principal has encouraged teach-ers to establish learning centers within classrooms.One teacher expressed frustration with this encour-agement: "We found out that a lot of these types ofchildren .. . we've learned this through workshopsand we've been told . .. that this type of child that wehave here needs more structure . . structure everyinch of the way.. .. you know, less confusion and lotsof structure that they thrive on."

Personal responsibility at Pine Creek is shaped bythese expectations of students. Teachers per-ceived the children as extremely needy, bothsocially and academically. As a result, manyteachers expressed a responsibility almost like therole of foster parent rather than the role of teacher.As one teacher said: "Kids today are coming inwith a lot more issues, they are not being dealtwith at home, and you need to deal with them.You wear many hats I think. That's your jobtodo it, and it's your job to make sure you are doingit."

Teachers were involved in many activities withstudents outside of the regular school day. Oneteacher runs an extracurricular sports program,another volunteers as the unofficial school pho-tographer. Another teacher mentioned that therewere some things he does for which he does notbelieve he should be responsible. He referred to twocrisis situations where he and the school gotheavily involved in a student's home life. Heexplained, "Yeah, there's a lot of stuff we do herethat we're not responsible for doing, but we do itjust because we love the kids."

Some teachers, however, expressed frustrationthat the students' needs led the community toexpect too much of teachers and the school. Oneteacher said: "I think we really need to gear moreon the academic areas than on the other areas,whether it's the violence prevention or whatever

areas are left to us, which is fine, but it's too badbecause we have got enough to do with the aca-demics. I mean it's expected that we providebreakfast, that we provide extra programs, weprovide homework centers, and it's not so muchwell . . . it's expected by some, appreciated bymany."

Despite these frustrations, teachers continued tofeel responsible for meeting the social and aca-demic needs of their students. They perceive theseneeds to be the driving force behind most teachers'decisions. Many teachers expressed a conviction thatin order to have success in their jobs, they had toassume responsibility for the many needs of theirstudents. One teacher explained:

I think kids are coming in with so much extrabaggage today, you can't get to the ABC'suntil you get past that point, and it would beignorant of me to think that you could becauseyou are not. If they don't come in ready tolearn, you have to get them to that point;that 's part of your job even if you have tospend all year on it.

Teachers' expectations of students also had astrong effect on the curriculum the teachers choseto introduce to students. Although there was acurriculum prescribed by the district, many teach-ers altered this curriculum based upon their percep-tions ofthe children's needs and abilities. Oneteacher called Pine Creek a "home-based school"meaning:

We plan our curriculum for the kids down herebased according to their needs because weknow . . . what the type of the population thereis down here. So we kind of gear the curricu-lum towards the needs of these kids becausewe know what the needs are and we know thatsometimes the curriculum that is set for theentire city doesn't meet the needs of the kids.When they're saying that the kids should bereading such and such and we know these kids



may come to us as non-readers . . . We do a lotof tutoring with the kids and we do a lot ofchanging the curriculum to meet their needsrather than to meet the needs of the city.

These expectations of students also appeared toshape the goals established by grade-level teams.For example, the third grade team chose to focustheir efforts on teaching children to write incomplete sentences. They felt that the activitiesprovided for students were too complex andtherefore chose to write their own set of questionsto accompany the basal reader. One teacherexplained that they "tried to write the questionsfrom a literal standpoint because we found that thebook was offering questions that were very evalu-ative and third graders could not do it. They werenot developmentally ready to do that, but wefound that they were ready to answer literal typequestions in a complete sentence." In this in-stance, the collective expectations of studentsworked within an internal accountability system tolower expectations of student performance.

At Pine Creek the expectations formed among theteaching staff regarding students' needs are muchstronger than any expectations from the district orprincipal. In the one area where the,principalpresented clear guidelinesstaff developmentthesecollective expectations shaped how the teachers metthose guidelines. In this school, these collectiveexpectations, closely aligned with the sense of per-sonal responsibility to act as foster parents that mostteachers feel, form the internal accountability system.As one teacher said, "You have to be willing to giveextra time on Saturdays to come and watch them playbasketball when they ask if you'll come to their gameand you know nobody else will be there . "

The consequences at Pine Creek operate not somuch on individual teachers, but for the entireschool. The possible loss of Title I status, or lossof recognition and discretion, appeared to be amotivating factor behind much of the work ofstaff development teams. Teachers collaborated to

create a plan that would help students achieve thatminimum level of progress. Sometimes thisincluded changing the curriculum, moving unitsout of the order they appear in the book, andstrategically choosing which material to cover andwhich to skip. Pine Creek teachers and the princi-pal felt pressure to achieve a five percent annualrise in test scores. This goal appeared to beconsistent with the teachers' collective expecta-tions of students, insofar as most teachers be-lieved the children could improve and learn.However, if the school were expected to meet acriterion-based standard level, that might be lessconsistent with the school's established internalaccountability systems.

Summary

The schools discussed in this section have man-aged to translate individual responsibility andcollective expectations into some kind of internalaccountability system. These internal accountabilitysystems operate in the context of external policy, andsometimes the internal and external are mutuallyreinforcing. Such is the case when the externalsystem, personal responsibility, and collective expec-tations are aligned within the school. At Turtle Haven,for example, the district's performance-based stan-dards, although accompanied by little external ac-countability, appear closely aligned with the school'scollective expectations. There is yet no defmedmechanism to hold teachers accountable for theimplementation ofthese standards, but this externalpolicy has been incorporated into the internal ac-countability system.

In schools where the external policy is not alignedwith the collective expectations, teachers tend tofollow the internal accountability system. Forex-ample, at Pine Creek, teachers collectively decidedthat the district's curriculum was not appropriate fortheir students, still the school pays careful attention tothe accountability system imposed by the federal TitleI program. In effect, Pine Creek teachers wrote thedistrict's curriculum based on their collective expecta-


4 637


tions of students with lower abilities than thoserepresented in the district's policy.

Internal accountability systems influence behaviorbecause they reflect an alignment within theschool of personal responsibility and collectiveexpectationsregardless of the external policy.This alignment of expectations and responsibility isalso accompanied by some sense that there will beconsequences if expectations are not met. At TurtleHaven, teachers were aware that they could lose theirjobs if they did not live up to the expectations set bythe principal and reinforced by the school community.Parents at St. B's may feel some sense of responsibil-ity to be involved in their children's education, butthey know they will be fmed if they do not follow theestablished guidelines.

It is the alignment between expectations andresponsibility, connected to certain consequences,that shapes the internal accountability within theseschools. All actors feel responsible for meetingexpectations, and there are consequences for notmeeting expectations. These expectations, how-ever, are for the most part generated internally, or,if generated externally, are modified to match pre-existing expectations within the school.

Conclusion

Our aim in this report has been to view the prob-lem of accountability primarily from the perspec-tive of schools, rather than from the perspective ofexternal policies that purport to influence schools.In taking this perspective, we have aligned our-selves with those who ask what conditions withinschools determine to whom, for what, and how theyare accountable (Wagner 1989; Newmann, King, andRigdon, 1997). In this sense, we have turned thetraditional formulation of educational accountability, asa problem of public policy, inside out. Instead ofasking how schools respond to policies designed tomake them accountable to external authorities, wehave asked how schools come to formulate theirown

conceptions of accountability and what role, ifany,external policies play in these conceptions. Ourworking theory of accountability, andour researchmethods in this study, were predicated on the belief,or expectation, that external accountability systemsoperate on the margins of powerful factors inside theschool, and that understanding these factors is amajor precondition to understanding howand whyschools respond the way they do to external pres-sures for accountability. In later phases ofourresearch, in light of what we have learned from thisstudy about how schools constnict accountability, wewill focus more explicitly on the design and implemen-tation of external accountability systems.

The first and most important finding of this studyis that our initial expectation about the power ofschool-level factors in shaping schools' concep-tions of accountability was correct. All theschools in our study had distinctive solutions to theproblem of to whom they were accountableforwhat. The relatively weak external accountabilityenvironment in which all of the schools operatedoffers some explanation for this lack of unifor-mity, but does not explain how or why schoolsarrive at the various configurations of accountabil-ity that we observed during our fieldwork. Inmost cases, solutions to the question of accountabilitywere tacit, unarticulated, informal, and grew morefrom the individual beliefs and values of teachers andadministrators as enacted in their daily practice, thanfrom formal or explicit agreements.

The baseline, or default solution to accountability,that we observed at several schools was character-ized by individual teacher responsibility, wherepersonal discretion appeared to be dominant overorganizational expectations or formalized ac-countability mechanisms. Phoenix CharterSchool typified this theme. The responsibility-driven formulation of accountability evident inthese schools was, in terms of our theoreticalmodel, rather simplistic because it reflected littleor no alignment with responsibility, expectationsand accountability, and equally little coherence



about teaching and learning. However, lack ofcomplexity should not be mistaken for lack ofinfluence on daily practice. Individual responsi-bility in these schools exerted a powerful influ-ence over day-to-day operations, although the netresult was a fragmented academic program.

Another group of schools, representing the mid-line of complexity in our sample, exhibited dis-cernible effects of collective expectations withinthe school on individual teachers' conceptions ofresponsibility. St. Aloysius typified this formula-tion. In these schools where group expectationsrelated to teaching and learning, the academicprogram was more coherent than in the firstcategory of schools. Where expectations andindividual responsibility were directed towardaffective needs, coherence was again evident.These schools were distinguishable from our thirdcategory of schools in that alignment and coher-ence were the incidental result of the schools'expectations, but there was little structure orconsequence associated with these expectations,and the object of these expectationswhetheracademic, affective, behavioralwas still largelydiscretionary.

The most complex formulation of accountabilityobserved in our sample was represented by ourthird category of schools, led by Turtle HavenPilot School. In these schools, collective expecta-tions gelled into highly interactive, relativelycoherent, informal and formal systems by whichteachers and administrators held each otheraccountable for their actions vis a vis students.Teachers and administrators in this category ofschool were able to describe and interpret theformal external accountability systems in whichtheir schools operated (such as testing systems,curriculum guidelines, charters, and the like), butin no case did these external systems seem toexercise the determining influence over theirindividual conceptions of responsibility, theircollective expectations of each other or their

students, or their internal accountability structures,where they existed.

Our findings did not accord with our initial expecta-tions on one dimension. We expected that differenttypes of schools would differ, if not systematically, atleast roughly, in their solutions to the accountabilityproblem. We deliberately designed the study toinclude parochial schools and charter schools, inaddition to types of mainstream public schools in thebelief that parochial schools and charter schools, asthe empirical literature and claims of policymakerssuggest, would present us with a stronger, clearer setof examples of internal accountability systems atwork. For our sample, at least, this expectationproved not to be true. The development ofaschool's collective expectations and its internalaccountability system in our sample seems to bemore a function of particular school-level charac-teristics than it is a function of the type of school.We should not over-generalize this finding, but itis interesting that parochial schools and chartersseemed to have the same problems as ordinarypublic schools in constructing a coherent concep-tion of accountability.

This study confirms widely-prevalent views insociological research that schools develop theirown internal normative structures that are rela-tively immune to external influences, and thatteaching is an essentially isolated occupation inwhich teachers are left largely to their own de-vices in deciding important issues of what andhow to teach (Lortie 1975). But the framework andfindings ofthis study advance this view in severalrespects. By distinguishing among individual concep-tions of responsibility, collective expectations, andinternal accountability structures, we have provided afiner-grained portrayal of the forces within schoolsthat affect solutions to the accountability problem.The persistent isolation of teaching as an occupation,in our framework and findings, means that theschool's conception of accountability collapses, bydefault, into individual teachers' conceptions of

CPRE Research Report Series, RR-42.' 39

48


responsibility. Schools operating in this mode, such asPhoenix Charter, Gateway Elementary, andHutchinson High, were typically characterized by anemphasis on order and control, possibly because thiswas the one collective expectation on which it ispossible to reach agreement in an essentially isolatedwork environment. Perhaps because we conductedour fieldwork in relatively weak external accountabil-ity districts, we found that big questions about thecollective purposes of the enterprise were oftenanswered by the accretion of individual teachers'decisions, based on their views of theirown and theirstudents' capacities, rather than by collective delib-eration or explicit management. The dominant patternwas not so much that schools developed their ownstrong internal normative structures that were inconflict with external influences, but rather that theyfailed to develop strong internal normative structures,and thereby defaulted to individual teachers on majorissues of collective expectations and accountability.In such circumstances, a school's incidental solutionto the accountability problemto whom, for what,and howbecame simply a collection of individual,often idiosyncratic, judgments by teachers, growingout of their backgrounds, capacities, and individualtheories about what students can do or need.

In some cases, these judgments were powerfullyinfluenced by teachers' preconceptions about theirstudents' characteristics. Where teachers andadministrators equated low socio-economic statuswith inevitably poor prospects for student learn-ing, they frequently wrapped their low expecta-tions in theories about the deprivation of students,their families, and their communities, uninformedby systematic knowledge of what students werecapable of learning under different conditions ofteaching. In other instances, teachers deliberatelygave affective needs precedence over teaching andlearning, but with the belief that physical, socialand emotional deficits must be addressed beforestudents could achieve at high levels. Hence, inseveral schools in our sample, teachers assignedthe most powerful causality, in their own concep-tions of responsibility, to factors over which they,

as teachers, had little or no control, and assumedthe least powerful causality to those over whichthey had the greatest control, the conditions ofteaching and learning in the school.

The exceptions to the baseline, responsibility-driven mode in our sample are instructive. Theyall challenge the isolation of teaching, often inhalting and tentative ways, sometimes moreaggressively and directly. And they do so usuallyby introducing the idea that collective expecta-tionsamong teachers, between teachers andstudents, between principals and teachers, andbetween families, communities, and schoolsshould influence individual teachers' conceptionsof responsibility. Sometimes these collectiveexpectations mirrored a culture of low expecta-tions for students, but often they challenged theselow expectations in important ways. At timescollective expectations were therapeutic in na-turethey cast the school in the role of substitutefor deficient families and communities; some-times they reflected the high academic expecta-tions that school people attributed to families; andother times they explicitly aimed to correct lowexpectations of students in the community or theschool. Usually, the development of more explicitcollective expectations was associated with thepresence of a principal whose model of leadershipembodied an explicit attempt to overcome theisolation of teaching, by shaping the normativeculture of the school through recruitment ofteachers and through direct involvement in theinstructional life of the school. And sometimesthe development of a stronger set of collectiveexpectationsthrough the active agency of aleader and the engagement of teachersled to thecreation of observable internal accountabilitystructures, informal and formal, that carried realstakes and consequences for members of theorganization.



In our sample, the relationships between the externalaccountability structures within which schools oper-ated and their internal solutions to the problems ofresponsibility, expectations, and accountability wereslippery, subtle, and often downright contradictory. Inthe default mode, teachers and principals often dealtwith the demands of formal external accountabilitystructures (curriculum guidance, testing, and the like)either by incorporating them in superficial waysclaiming, for example, that they were consistent withexisting practice when they clearly were notor byrejecting them as unrealistic for the type of studentsthey served. Without a way to address collectiveexpectations within the school, external accountabilitymeasures can only work through individual teachers'conceptions of responsibility. Some teachers seemedquite adept at deflecting external accountabilitymeasures or unable to translate the accountabilitymeasures into daily practice. In instances whereschools had developed some version of collectiveexpectations, sometimes these expectations werealigned with external accountability systems, andsometimes they were not. In a few cases, we wit-nessed principals and teachers engaged in some sortof collective deliberation about how to incorporateexternal accountability requirements into their internalconceptions of responsibility, expectations, andaccountability. But, in most cases, teachers andprincipals viewed external accountability systems likethe weathersomething that might affect their dailylives in some way, something they could protectthemselves against, but not something they could orshould do much about. In a few cases, the responsesof teachers and principals to external accountabilitysystems seemed to contradict in some fundamentalway the theory behind the external systems, such asthe charter schools in our study that seemed toexperience no special demands or requirementsstemming from the need for their charters to berenewed.

This finding about the slippery, subtle, and contradic-tory relationship between internal and external ac-countability may simply be an artifact of the design of

our study. We deliberately did not, at this stage, seekschools that were operating in strong and obtrusiveexternal accountability systems. Some schools in oursample were located in cities that are in the earlystages of developing stronger external accountabilitysystems. Some cities are located in states that are inthe early stages of implementing anew accountabilitysystem. The charter laws operating in the stateswhere we conducted our study were in the earlystages of implementation and these states had not yetdirectly confronted the issue of charter renewal. Theweakness of the effects of external accountabilitysystems may simply be attributable to the state ofpolicy. We will confront this issue more explicitly inlater stages of the study, when we will observeschools in more visible and powerful external ac-countability environments.

Taking the limitations of our design and sampleinto account, there are still important things to belearned from this initial study about the relation-ship between internal and external accountabilitysystems. It seems highly unlikely to us thatschools operating in the default modewhere allquestions of accountability related to student learningare essentially questions of individual teacher respon-sibility, will be capable of responding to strong,obtrusive external accountability systems in ways thatlead to systematic, deliberate improvement of instruc-tion and student learning. The idea that a school willimprove its instructional practice and, therefore, theoverall performance of its students implies a capacityfor collective deliberation and action that schools inour sample did not exhibit. Where virtually all deci-sions about accountability are decisions made byindividual teachers, based on their individual concep-tions of what they and their students can do, it seemsunlikely that these decisions will somehow aggregateinto overall improvement for the school. For schoolsoperating in the default mode, the question for futureresearch on the effect of external accountabilitysystems is whether these schools can, or will, respondby developing congruent internal expectations andaccountability systems. Perhaps more importantly, a


5 041


related question is how these schools will get thecapacity to develop these new internal norms andprocesses.

Schools that are not in the default modeschoolsthat have developed internal expectations, internalaccountability systems, or bothraise a differentset of issues about the relationship betweeninternal and external accountability. Our studysuggests that these schools answer the for-what-they-are-accountable question in very differentwayssome schools focus on students' affectiveneeds, others on high aspirations for students'academic performance. For these schools, theissues are: the degree of alignment between theirinternal expectations and accountability systemsand the demands of external systems; and thelevel of conflict and accommodation that arisesfrom the confrontation between internal andexternal accountability. Are schools that manifestsome capacity to deal collectively with the ac-countability problem internally more likely toadapt and align their internal norms and systemsto the requirements of external systems, or arethey likely to be more resistant to changing theirinternal norms and systems? Do these schoolshave the capacities necessary to do the work ofaccommodating and adapting new external re-quirements? The existence of internal expecta-tions and accountability structures, in other words,does not necessarily predict how a school willrespond to new external requirements regardingteaching and learning. We will pursue thesequestions in the next phase of our research.

In this report, we have tried to array schools in athree-fold typology showing the range of account-ability formulations that we observed in ourfieldwork: schools in the default mode where allquestions of accountability for student learningcollapse into questions of individual teacherresponsibility; schools that exhibit commonexpectations that influence and are influenced byindividual conceptions of responsibility; andschools where expectations and individual respon-

sibility are aligned to such an extent that thiscombination effectively functions as an internalaccountability systems with stakes and conse-quences for members of the organization. Theedges of these three types of schools are blurry ininteresting and informative ways, suggesting boththe possible limits of our working theory and thediversity of ways that schools have of coping withthe accountability problem.

One thing, however, seems quite clear from ourstudy to date. Conditions within schools arelogically and empirically prior to conditionsoutside schools when constructing a workingtheory of educational accountability. That is, wecannot know how an accountability system willwork, nor can we know how to design such asystem, unless we know how schools differ in theway they construct responsibility, expectations,and internal accountability. This finding is funda-mental to the study of educational accountabilityin all its forms. Schools will vary in their re-sponse to external accountability systems dependingon the level and type of solutions they have in place tothe problems of responsibility, expectations, andinternal accountability. Studies of accountability andattempts to design new accountability systems willsucceed to the degree that they consider the sourcesof variability and explain their impact on the wayschools respond to external demands. Accountabilitysystems are often constructed by policymakers andadministrators out of normative theories of howschools ought to act, uncorrupted by understandingsof why they act the way they do. Our study suggeststhat such systems should take their initial point ofdeparture not from normative theories about howschools ought to act, but from a finer-grainedunderstanding of why they act the way they do.

Our research also suggests that the attitudes,values, and beliefs of individual teachers andadministratorsabout what students can do,about what they can expect of each other, andabout the relative influence of student, family,community, and school on student learningare key


51


factors in determining the solutions that schoolsconstruct to the accountability problem. Put bluntly,many educators simply do not believe that they havethe capacity to influence student learning in the waysthat external accountability systems suggest theyshould. Hence, external accountability systems willbe relatively powerless in the absence of changedconceptions of individual responsibility and collectiveexpectations within schools. In our study,we havecome to call this problem: "when accountabilityknocks, is anyone home?" A strong normativeenvironment inside the school, based on a beliefin the capacity and efficacy of teachers and princi-pals to influence student learning, coupled withthe knowledge and skill necessary to act on thosebeliefs are prior conditions necessary to thesuccess of strong external accountability systems.


52


References

Elmore, Richard F. (1997). The politics of education reform. Issues in Science and Technology, XIV (1).

Ladd, Helen. (1996). Holding schools accountable: Performance based reform in education. Washington,DC: Brookings Institute.

Lortie, Daniel C. (1975). Schoolteacher: A Sociological Study. Chicago, IL: University of Chicago Press.

Metz, Mary Haywood. (1990). How social class differences shape teachers' work. In M. W. McLaughlin,J. E. Talbert, & N. Bascia (Eds.), The contexts of teaching in secondary schools: Teachers' reali-ties (pp. 40-107). New York, NY: Teachers College Press.

Metz, Mary Haywood. (1990). Real school: A universal drama amid disparate experience. In D. E. Mitchell &M. E. Goertz (Eds.), Education politics for the new century. New York, NY: Falmer Press.

Newmann, Fred, King, Bruce, & Rigdon, Mark. (1997). Accountability and school performance: Implicationsfrom restructuring schools. Harvard Educational Review, 67(1)41-74.

Schein, Edgar H. (1992). Organizational culture and leadership. San Francisco, CA: Jossey Bass.

Wagner, Robert B. (1989). Accountability in education: A philosophical inquiry. New York, NY:Routledge.


5 3


Appendix 1Interview Protocol

Interview Questions for Teachers

Direct Indirect Hybrid

1. As a teacher, what are you heldaccountable for?

1. As a teacher, what is your job?_

1. As a teacher, what are youexpected to do?

2. How do you know when you're incompliance?

2. How do you know when you're doingyour job well?

-,

2. To whom do you feel thegreatest sense of responsibility?

3. Do you think there's a commonperception in your school regardingwhat you're all accountable for?

3. Do you think most teachers in yourschool would answer the same way, ordifferently?

3. Do you think most of theteachers in your school have thesame point of view regardingtheir responsibilities?

,4. To whom are you accountable? 4. Who determines whether you're doing

your job well, and how is thatdetermined?

4. What makes a school year"good" or "bad" for you?

:5. What formal accountability measuresare in place at your school?

5. Are the formal assessments ofstudents used in your school an accuratemeasure of achievement and of yourteaching?

5. What does your schoolprincipal expect of you and doeshe/she expect the same ofeveryone?

6. Are those formal accountabilitymechanisms an accurate measure ofyour teaching and studentachievement?

6. How well does the content of thosemeasures map the content of what youdo in the classroom?

6. How would you alter theassessments your schoolcurrently uses to make themmore useful and/or informative?

7. How are the measures used? (i.e.,who sees the results, what happenswhen students/teachers do well/poorlyon these measures)?

7. Who or what has an influence (or thegreatest influence) over what you do inthe classroom?

7. How do you think the externalcommunity perceives yourschool?

8. What role do parents, and the largercommunity play in your school, and howdoes that compare to what you thinktheir role should be?

.-

8. What role do parents, and the largercommunity play in your school, and howdoes that compare to what you think theirrole should be?

8. What role do parents, and thelarger community play in yourschool, and how does thatcompare with what you think theirrole should be?

9. Who is accountable to you? 9. Are other teachers or your principalanswerable to you for anything?(Students?)

9. What do you expect of otherteachers? Of your principal? Arethey formally responsible to youin any way? (Students?)

10. Are there any conditions underwhich you believe you should not beheld accountable, or should be lessaccountable, for your student'slearning?

10. What level of schooling do you thinkmost of your students will achieve?

10. What factors do you believeinfluence students' achievement,and which of those are you in aposition to affect?

11. If you were hiring a teacher for yourposition, what characteristics would youlook for?

11. If you were hiring a teacher for yourposition, what characteristics would youlook for?

11. If you were hiring a teacherfor your position, whatcharacteristics would you lookfor?


47


Interview Questions for Parents


1. Who is accountable for your child'slearning?

1. Whose job is it to ensure that yourchildren are learning?

_

1. Who is responsible for yourchild's learning?

2. How do you know whether your childis learning?

2. How do you know whether your child islearning?

2. How do you know whetheryour child is learning?

3. For what information, and by whatmeans, is your child's school required togive an account of student learning andschool standing?

3. What information regarding your child'sprogress, and the school's scores, is theschool required to share with you, andhow is that info conveyed? (meetings,report cards, media, etc.)

3. What information regardingyour child's learning, and theschool's standing do you expectthe school to provide? How doyou expect that information to beconveyed?

4. If your child were having difficulty withhis/her work, who is accountable toaddress that problem?

4. If your child were having difficulty withhis/her work, what would you do, oradvise your child to do?

4. If your child were havingdifficulty with his/her work, whowould you expect to address thatproblem, and what would youexpect to happen in addressingit?

5. What do you feel accountable for inregard to your child's learning?

5. What is your role in your child'slearning?

5. What do you feel responsiblefor in regard to your child'slearning?

6. What do you think your child'steacher holds you accountable for, andwhy do you think that?

6. What do you think your child's teacherand principal believe your role is?

6. What do your child's teacherand principal expect of you inregard to your child's education?

7. Who is accountable for discipline inyour child's classroom and the schoolas a whole?

7. Whose job is it to maintain disciplinein your child's school?

.,

7. Who is responsible formaintaining discipline in theschool?

8. What type/degree of school-siteinvolvement should parents beaccountable for?

8. What type/degree of school-siteinvolvement is appropriate for parents?

8. What type/degree of school-site involvement should parentsbe responsible for?

9. How does this compare with yoursand other parents' involvement?

9. How does this compare with yours andother parents' involvement?

9. How does this compare withyour and other parents'involvement?

10. If you had (have) the option to sendyour child elsewhere, would you? Why?

10. If you had (have) the option to sendyour child elsewhere, would you? Why?

-10. If you had (have) the optionto send your child elsewhere,would you? Why?

11. What do you expect students toknow upon entering your class?Leaving?

11. What do you expect students toknow upon entering your class?Leaving?

11. What do you expect studentsto know upon entering yourclass? Leaving?


48


Interview Questions for Principals/Administrators


1. What are you accountable for? 1. As a principaVadministrator, what'syour job?

1. As principal/administrator,what are you responsible for?-

2. To whom are you accountable? 2. Who determines whether you're doingyour job well, and how is thatdetermined?

2. Who determines whetheryou're fulfilling your responsibilityand meeting expectations?

3. What do you hold teachersaccountable for'?

3. What is the job of teachers in yourschool?

3. What do you expect teacher'sto be responsible for in yourschool?

4. By what means, formal and informal,is that accountability enforced?

4. How do you know whether theyredoing their jobs well, and is there acommon understanding/procedure forwhat happens if they don't?

4. How do you know whetherthey're doing their jobs well, andwhat is the expectation aboutwhat happens if they don't?(Yours and theirs?)

5. Do you believe teachers in yourschool have a clear sense of whatthey're held accountable for?

5. Do you believe teachers have a clearsense of what their job is? Why or whynot?

5. Do teachers know what'sexpected of them, and how dothey know?

6. How well do the formal assessmentsused at your school reflect studentachievement and teaching?.

6. What happens if a teacher is notdoing his/her job?

6. What happens if teachers arenot doing what's expected ofthem?

7. How are results of assessmentsused in your school? (Who sees themor how well do they reflect what they'redoing in the classroom?)

7. How well do the formal assessmentsused at your school reflect your, andteachers' efforts at improving studentachievement and teaching?

7. How well do formalassessments used at your schoolreflect your expectations, andthose of teachers, for studentachievement and teachingeffectiveness?

8. What characteristics do you look forwhen hiring a new teacher?

8. What characteristics do you look forwhen hiring a new teacher?

-

8. What characteristics do youlook for when hiring a newteacher?

9. Since the beginning of this year, whathas been your most challenging internalissue?

9. Since the beginning of this year, whathas been your most challenging internalissue?

9. Since the beginning of thisyear, what has been your mostchallenging internal issue?

10. Since the beginning of the year,what was the most challengingexternally-generated issue?

10. Since the beginning of the year, whatwas the most challenging externally-generated issue?

10. Since the beginning of theyear, what was the mostchallenging externally-generatedissue?

11. Upon leaving your school, whatshould students know (and be able todo)?

11. Upon leaving your school, whatshould students know (and be able todo)?

11. Upon leaving your school,what should students know (andbe able to do)?


5649


Interview Questions for Students


1. As a student, what are youaccountable for?

1. What is your "job" as a student? 1. As a student, what are youexpected to do?

2. To whom are you accountable?How?

2. Who holds you responsible for doingyour "job" as a student? How?

2. Who holds you responsible forfulfilling those expectations?How?

3. How do you know when you're doingwell in school? (How do you know whenyou've learned a subject or lesson? )

3. How do you know when you're doingwell in school? (How do you know whenyou've learned a subject or lesson?)

3. How do you know when you'redoing well in school? (How doyou know when you've learned asubject or lesson?)

4. How is your school achievementmeasured?

4. How is your school achievementmeasured?

4. How is your schoolachievement measured?

5. Who is accountable for yourlearning?

5. Who's job is it to make sure youlearn?

5. Who is responsible for yourlearning?

6. As a student, is anyone accountableto you?

6. Is anyone in your school (adults orother students) answerable to you as astudent?

6. What do you expect of others,adults and students, in yourschool?

7. What makes a teacher a goodteacher?



8. What makes a school a goodschool?

8. What makes a school a good school? 8. What makes a school a goodschool?

9. How would you describe a goodschool year? (Or week?)

9. How would you describe a goodschool year? (Or week?)

9. How would you describe agood school year? (Or week?)

10. What are you expected to know, beable to do when you finish (x) grade?

10. What are you expected to know, beable to do when you finish (x) grade?

10. What are you expected toknow, be able to do when youfinish (x) grade?

11. What do you think you'll be doing 5-10 years from now?

11. What do you think you'll be doing 5-10 years from now?

11. What do you think you'll bedoing 5-10 years from now?



End Notes

' Newmann, et al 1997 identify a "complete school accountability system" as including: (1) Information aboutthe organization's performance (e.g., test scores); (2) Standards for judging the quality or degree of success oforganizational performance (e.g., a mean achievement score higher than other schools with comparable demo-graphic characteristics); (3) Significant consequences to the organization (i.e., rewards and sanctions such asbonuses to teachers in the school) for its success or failure in meeting specified standards; and (4) An agent orconstituency that receives information on organizational performance, judges the extent to which standards havebeen met, and distributes rewards and sanctions (e.g., the state department of instruction).

See Elmore, Education Policy and Practice in the Aftermath of TIMSS, 1997.

3 The research reported in this paper was part of the CPRE research project, "Accountability for Results,Capacity for Reform," jointly undertaken by Stanford and Harvard Universities.

4 Our working theory is informed by Robert Wagner's conception of accountability and responsibility, asdescribed in Accountability in Education, 1989.

See Newmann, et al. 1997

See Metz, in Mitchell and Goertz, eds., 1990 for more on "Real Schools," and standard constructions ofschooling.

' In some cases, the categorization of schools is a better fit than in others. Fullcase studies of each school areavailable upon request by contacting CPRE-Harvard.


Nondiscrimination Statement

The University of Pennsylvania values diversity and seeks talented students, faculty and staff fromdiverse backgrounds. The University of Pennsylvania does not discriminate on the basis of race, sex,sexual orientation, religion, color, national or ethnic origin, age, disability, or status as a Vietnam EraVeteran or disabled veteran in the administration of educational policies, programs or activities;admissions policies, scholarships or loan awards; athletic, or University administered programs oremployment. Questions or complaints regarding this policy should be directed to Executive Director,Office of Affirmative Action, 1133 Blockley Hall, Philadelphia, PA 19104-6021 or (215) 898-6993(Voice) or (215) 898-7803 (TDD).

5 9

U.S. Department of EducationOffice of Educational Research and Improvement (OERI)

National Library of Education (NLE)Educational Resources Information Center (ERIC)

NOTICE

REPRODUCTION BASIS

O.R 9 73c)

ERIC

This document is covered by a signed "Reproduction Release(Blanket) form (on file within the ERIC system), encompassing allor classes of documents from its source organization and, therefore,does not require a "Specific Document" Release form.

a/ This document is Federally-funded, or carries its own permission toreproduce, or is otherwise in the public domain and, therefore, maybe reproduced by ERIC without a signed Reproduction Release form(either "Specific Document" or "Blanket").

EFF-089 (9/97)

Elmore, Richard; Even, Johanna; Kenyon, Susan - US ...

Documents

Transcript of Elmore, Richard; Even, Johanna; Kenyon, Susan - US ...