Health & Place 17 (2011) 519–524

Contents lists available at ScienceDirect

Health & Place

1353-82

doi:10.1

n Corr

E-m

journal homepage: www.elsevier.com/locate/healthplace

Mismatch between perceived and objectively assessed neighborhoodwalkability attributes: Prospective relationships with walking and weight gain

Klaus Gebel a, Adrian E. Bauman a,n, Takemi Sugiyama b, Neville Owen b

a Prevention Research Collaboration, School of Public Health, University of Sydney, Australiab Cancer Prevention Research Centre, School of Population Health, University of Queensland, Australia

a r t i c l e i n f o

Article history:

Received 30 April 2010

Received in revised form

25 November 2010

Accepted 9 December 2010Available online 24 December 2010

Keywords:

Body weight changes

Environment

Geographic information systems

Neighborhood

Physical activity

Prospective studies

92/$ - see front matter & 2010 Elsevier Ltd. A

016/j.healthplace.2010.12.008

esponding author. Tel.: +61 2 9036 3247; fax

ail address: adrian.bauman@sydney.edu.au (A

a b s t r a c t

We examined prospectively whether persons who perceive their objectively measured high walkable

environment as low walkable decrease their walking more and gain more weight than those with

matched perceptions. Walkability was measured objectively using GIS. Corresponding perceptions were

collected using the Neighborhood Environment Walkability Scale from 1027 urban Australian adults.

Objective and perceived measures were dichotomized and categories of match and mismatch were

created. Overall, walking levels decreased and BMI increased significantly over the four year follow-up

period. Those who perceived high walkability, dwelling density or land use mix as low decreased their

walking for transport significantly more than those with matched perceptions. Those who perceived high

walkability, land use mix or retail density as low increased their BMI significantly more than those with

concordant perceptions. These prospective findings corroborate recommendations from previous cross-

sectional studies. Interventions to improve negative perceptions of walkability among those living in high

walkable areas may be a relevant public health intervention to increase physical activity and support

weight maintenance.

& 2010 Elsevier Ltd. All rights reserved.

1. Introduction

There is a consistent body of evidence on the health benefits ofphysical activity (Pate et al., 1995; US Department of Health andHuman Services, 1996; Haskell et al., 2007). Over the last decadethere has been an increasing interest in the role of the physicalenvironment as a determinant of physical activity: neighborhoodbuilt-environment attributes have been shown to be associatedwith walking (Owen et al., 2004; Transportation Research Board,2005; Heath et al., 2006; Gebel et al., 2007; Saelens and Handy,2008), and with obesity (Frank et al., 2004; Papas et al., 2007; Blackand Macinko, 2008; Joshu et al., 2008; Mujahid et al., 2008).According to ecological models of health behavior, appropriateopportunities and settings that facilitate particular forms ofactivity, such as walking for recreation and exercise, or walkingto get to and from places, help adults to achieve sufficient levels ofphysical activity for health benefits (Owen et al., 2000; Sallis et al.,2008).

However, the provision of such built-environment facilitatorsalone might not automatically influence the behavior in theabsence of awareness of them, or without relevant awareness-raising initiatives (Kahn et al., 2002; Salmon et al., 2003; Timperio

ll rights reserved.

: +61 2 9036 3184.

.E. Bauman).

et al., 2004; Humpel et al., 2004; Mowen et al., 2007; Frohlich et al.,2007; McCormack et al., 2008; Ball et al., 2008). It has been arguedthat the relationships between the attributes of neighborhoodenvironments and physical activity may be mediated by environ-mental cognitions (Kremers et al., 2006). The outcomes of theprocessing, coding, and storing of information about environmen-tal attributes may be construed as cognitive maps. However, theliterature on spatial cognition and environmental psychologystates that these cognitive maps are unlikely to provide preciserepresentations of the actual environment, and may reflect sys-tematic cognitive distortions (Kaplan and Kaplan, 1982; Golledge,1987; Golledge, 1991; Lloyd, 1997). This is indicated by therelatively low rates of awareness of objectively present opportu-nities for physical activity that have been reported in some studies(Merom et al., 2003; Reed et al., 2004; Reed, 2007). As well, recentcross-sectional studies have found low agreement between objec-tive and perceived measures of environmental attributes known tobe associated with physical activity (McCormack et al., 2008;Ball et al., 2008; Reed et al., 2004; Kirtland et al., 2003; Boehmeret al., 2006; Kweon et al., 2006; McGinn et al., 2007; Macintyreet al., 2008; Gebel et al., 2009; Lackey and Kaczynski, 2009;Prins et al., 2009).

It has been hypothesized that the level of concordance ofperceptions with actual characteristics of the environment mightaffect physical activity behavior (Kirtland et al., 2003; Gebelet al., 2009; Napolitano et al., 2006). If the outcomes of such spatial

K. Gebel et al. / Health & Place 17 (2011) 519–524520

cognition processes mediate the relationship between environ-mental attributes and behavior, then the targeting of environ-mental cognitions (and particularly the awareness of settings forphysical activity) could be a promising public health communica-tion strategy (Mowen et al., 2007; van Stralen et al., 2008).

The potential mediating role of environmental cognitions on therelationships between environmental attributes and physicalactivity may be moderated by socio-demographic factors(Kremers et al., 2006). Such factors could either influence therelationships between the environmental attributes and cogni-tions, or the relationships between such cognitions and physicalactivity (Kremers et al., 2006). For example, in a cross-sectionalstudy (Gebel et al., 2009) we found that adults of lower socio-economic status, who had children in the household, or who wereoverweight were more likely to perceive a high-walkable neigh-borhood as low-walkable.

Few longitudinal studies have investigated either objectivelyassessed or perceived environmental attributes in relation toprospective changes in physical activity. For example, a prospectivestudy in the USA (Li et al., 2005) examined objectively assessedenvironmental correlates of change in neighborhood walkingamong older adults. One year after baseline measurement, walkinglevels had declined. Those living in neighborhoods with lowaccessibility to playgrounds, parks, or gyms, or with low perceivedsafety for walking showed greater decreases in walking thanthose living in more walkable neighborhoods. Another studyfrom the USA (Li et al., 2009) found that objectively assessedhigh neighborhood walkability positively moderated the relation-ship between people’s eating and physical activity habits andprospective changes in their weight and waist circumference.An Australian study (Humpel et al., 2004) examining perceptionsand behavior at two time points, found that improved perceptionsof accessibility to destinations and aesthetics were associated withincreased walking levels. These findings from prospective studiesprovide some evidence for the relevance of both objectivelyassessed and perceived environmental attributes in relation toprospective changes in physical activity; one study has shown aprospective association with weight gain (Li et al., 2009).

However, cross-sectional studies have found agreementbetween these objective measures of walkability and participant’sperceptions of the corresponding environmental attributes to below (McCormack et al., 2008; Ball et al., 2008; Reed et al., 2004;Kirtland et al., 2003; Boehmer et al., 2006; McGinn et al., 2007;Macintyre et al., 2008; Gebel et al., 2009; Lackey and Kaczynski,2009; Prins et al., 2009). As well, at baseline, we found that thosewith more positive perceptions are more likely to be physicallyactive (Gebel et al., 2009). In the baseline study, we examined thecorrelates of non-concordance between perceived and objectivemeasures of environmental attributes, and found that about a thirdof those living in objectively determined high walkable neighbor-hoods perceived their local environment as low in walkability.Similarly, a third of those living in neighborhoods with lowwalkability perceived them to be high walkable. This was the casefor a composite measure of walkability, and for the sub-domains ofdwelling density, street connectivity, land use mix, and retaildensity (Gebel et al., 2009). It has been suggested that strategiesto improve perceptions of the relevant environmental attributesparticularly among those who live in objectively determined highwalkable neighborhoods may have the potential to increasephysical activity levels (Reed et al., 2004; Gebel et al., 2009)and improve other health outcomes (Parra et al., 2010). Nostudy has examined whether mismatch between perceived andobjectively determined walkability influences change in walkingand weight.

Here, we examine prospective relationships of perceivingobjectively measured high walkability as low with changes in

adults’ walking, and weight gain. We hypothesized that over time,those who perceive their objectively measured high walkableenvironment as low walkable would decrease their walking moreand would gain more weight than would those with matchedperceptions.

2. Methods

2.1. Data collection procedure

Data were from the PLACE (Physical Activity in Localities andCommunity Environments) study conducted in 2003 (baseline) andin 2007 (follow-up) in the city of Adelaide, Australia. Ethicalapproval for the study was given by the Behavioural and SocialSciences Ethics Committee of the University of Queensland. Prior tobeing enrolled in the study, all participants provided informedconsent and were explicitly informed that they were free towithdraw from the study at any time.

An account of the design of the PLACE study (baseline) has beenprovided elsewhere (Owen et al., 2007; Leslie et al., 2007). Briefly,the aim of the PLACE study was to investigate how neighborhoodenvironmental characteristics, in particular walkability, influenceresidents’ physical activity patterns. Walkability was a compositemeasure of four objectively measured environmental attributes:dwelling density, street connectivity, land use mix, and net retailarea ratio. Several studies have found these environmental attributesto be consistently associated with walking behavior (Owen et al.,2004; Saelens et al., 2003b). Data were derived from street centerlinefiles, land use, zoning, shopping center locations, and census, andwere compiled in Geographic Information Systems (GIS). The fourobjective environmental attributes were calculated for each of theurban census collector districts (CCD) within the Adelaide StatisticalDivision; and they were then converted into deciles, with 1 repre-senting the lowest and 10 the highest level of walkability. A CCD isthe smallest spatial unit for census data collection defined by theAustralian Bureau of Statistics, and contains approximately 250households. The walkability index was the sum of these four decilescores with a possible score from 4 to 40. The resulting compositemeasure was divided into quartiles, with the first and fourthquartiles standing for the lowest and highest walkability. Thirty-two neighborhoods from these two quartiles were selected for thestudy, with a total of 156 CCDs represented.

For the baseline data collection, simple random samplingwithout replacement was used to select households within theselected neighborhoods. To be eligible for the study, people had tobe between the age of 20 and 65, English speaking, residing inprivate dwellings, and able to walk without any assistance. If therewas more than one eligible participant in a household, then themost recent birthday method was used. A total of 2650 respondentscompleted the baseline questionnaire (74.2% of those that werecontacted). The response rate did not differ significantly by socio-economic status or walkability of the identified study areas.Compared with the 2001 census data from the Australian Bureauof Statistics (Australian Bureau of Statistics, 2001), older adults,females, and people in paid work were overrepresented in thesample.

The follow-up survey was conducted four years after thebaseline survey (September 2007). A total of 1027 valid responseswere received. The relatively high attrition rate was partly due to alarge number of participants who had moved their place ofresidence after the baseline survey. Participants for whom fol-low-up data were available lived in neighborhoods with lowerobjectively assessed neighborhood walkability (p¼0.001) andwere older (po0.001) than those who dropped out after thebaseline survey.

Table 1Participants with matched/mismatched perceived and objective measurements of

neighborhood walkability.

Objective

walkability

High Low Total

340 176 516

High

Perceived 68.8% 33.0% 50.2%

walkability 154 357 511

Low

31.2% 67.0% 49.8%

494 533 1027

Total

100% 100% 100%

K. Gebel et al. / Health & Place 17 (2011) 519–524 521

2.2. Measures and instruments

The outcome variables for this study were the change in walking,and body mass index (BMI) between the baseline and the four-yearfollow-up. Data on physical activity were collected in both thebaseline and follow-up surveys using the long form of theInternational Physical Activity Questionnaire (IPAQ). This reliableand validated survey (Craig et al., 2003) distinguishes betweendomain-specific walking behavior. For this study, weekly minutesof walking for transport, and for recreation or exercise, wereexamined separately. Participants were also asked to reporttheir height and weight, from which BMI was calculated, in bothsurveys.

The independent variable was the mismatch between the objec-tively determined and perceived walkability at the baseline. Inaddition to the GIS measures of walkability, participants ratedaspects of their environment by completing the reliable and vali-dated Neighborhood Environment Walkability Scale (NEWS)(Saelens et al., 2003a; Brownson et al., 2004; Cerin et al., 2006;Cerin et al., 2008). The geographical scale for the self-reportedenvironmental measures was a 10–15 minute walk from partici-pants’ homes. Perceived measures corresponding to dwelling den-sity, street connectivity, land use mix, and net retail area ratio werecalculated, and categorized into deciles (or quartiles depending onvariability). In line with the procedure for the objective measures,perceived walkability was calculated as the sum of four sub-domainscores resulting in a possible range from 4 to 40. Moreover, in thefollow-up survey study participants had to rate perceived changes inneighborhood walkability across six categories (access to shops andservices or public transport, quality of places to walk, attractivenessof neighborhood and safety from traffic and crime).

2.3. Statistical analysis

Both the objective and perceived measures of walkability weredichotomized, using median splits. Then, participants were cate-gorized into the following four quadrants: high objective/highperceived; high objective/low perceived; low objective/high per-ceived; and low objective/low perceived. This applied to thecomposite walkability score, and to its sub-domains of dwellingdensity, street connectivity, land use mix, and net retail area ratio.Outliers in reported minutes of walking per day were truncated to120. Participants who had a change in walking for leisure (n¼3), orwalking for transport (n¼3) of 840 min per week were excluded.Subjects whose BMI changed by more than 20 (n¼2) were alsoexcluded. Multiple linear regression analyses were carried out toexamine the longitudinal effect of perceiving objectively measuredhigh walkability as low on changes in walking for transport andleisure, and BMI. The regression coefficients show the differences inchanges in walking, and in BMI, between participants whoseperceptions of walkability matched the objective measurements(high objective/high perceived) and those who perceived theirneighborhood as less walkable than objectively determined. In theregressions, we controlled for baseline levels in walking, or BMI,and adjusted for age, sex, and income. The duration of living in aresidential location is related to familiarity with the neighborhood(Liben, 1981). For instance, one study found an inverse relationshipbetween the time subjects had lived in their neighborhood andmismatch between objective and perceived measures of theenvironment (Ball et al., 2008). Therefore, we also adjusted forthe length of residency. The distributions for all variables related tochanges in perceptions of walkability were skewed, with almost allparticipants not reporting differences in their neighborhood. Asthere was hardly any change in perceptions, we did not control forthese variables. Statistical analyses were carried out using theStatistical Package for Social Sciences (SPSS 17).

3. Results

3.1. Agreement between objective and perceived measures of

neighborhood walkability

Consistent with the baseline study (Gebel et al., 2009), there wasonly a modest agreement between the objective and the perceivedenvironmental measures. One third of the subjects who lived inobjectively determined high walkable neighborhoods perceivedthem to be low walkable. Among those in low walkable environ-ments, a third perceived walkability to be high (Table 1). Theagreement between objective and perceived measures for the foursub domains of walkability was similar (data not provided here).

As mentioned above, there were hardly any sociodemographicdifferences between the baseline and the follow-up samples.Therefore, for descriptive data on sociodemographic attributesand for cognitive and walking variables by the four quadrants ofobjective and perceived walkability we refer to the baseline study(Gebel et al., 2009).

3.2. Changes in walking levels and BMI in total sample

For the total sample, walking for transport decreased by22.7 min/week (SD¼198.6; po0.0001) and walking for leisuredecreased by 9.8 min/week (SD¼165.1; p¼0.05) in the four yearsbetween the baseline and the follow-up survey. Mean BMIsincreased by 0.6 units (SD¼2.4; po0.001).

3.3. Changes in walking for transport by mismatch between perceived

and objective walkability

Those who perceived objectively measured high walkability aslow (see Table 2) decreased their walking for transport significantlymore than those whose perception matched the objective measure(b¼ �55.7, po0.001). The same applied to misperceptions ofdwelling density and land use mix. There was a trend towards aneffect of misperceiving retail density on changes in walking fortransport, but no difference for misperceiving street connectivity.

3.4. Changes in walking for leisure by mismatch between perceived

and objective walkability

Participants who perceived their neighborhood as less walkablethan objectively measured decreased their walking for leisure by23.6 min/week more than did those who perceived it as highwalkable. However, this difference was not statistically significant.Misperceiving street connectivity was the only significant factor

Table 2Prospective relationships of perceiving high walkable environmental attributes as low with changes in walking for transport, walking for leisure, and BMI.

Overall walkability Dwelling density Street connectivity Land use mix Retail density

b (95% CI) p b (95% CI) p b (95% CI) p b (95% CI) p b (95% CI) p

DWalking for transport �55.7 �86.8,

�24.5

o0.001 �49.9 �81.5,

�18.2

o0.01 �0.6 �28.3, 27.1 0.97 �29.4 �56.6, �2.3 0.03 �25.5 �56.5, 5.4 0.11

D Walking for leisure �23.6 �54.7, 7.5 0.14 2.4 �26.4, 31.3 0.87 �29.4 �52.9, 6.0 0.01 �14.8 �38.3, 8.7 0.22 �14.7 �46.0, 16.5 0.36

D BMI 0.54 0.06, 1.02 0.03 0.43 �0.14, 0.99 0.14 0.05 �0.4, 0.51 0.82 0.51 0.11, 0.92 0.01 0.63 0.19, 1.07 o 0.01

Analyses control for baseline levels of walking, BMI, and adjust for sex, age, income, and length of residency.

K. Gebel et al. / Health & Place 17 (2011) 519–524522

predicting changes in walking for leisure (b¼ �29.4, p¼0.01).There were trends in the expected direction for misperceiving landuse mix and retail density (non-significant) and there was nodifference for dwelling density.

3.5. Changes in BMI by mismatch between perceived and objective

walkability

Participants who perceived objectively measured high walk-ability as low gained significantly more weight than did thosewhose perception matched the objective measure (p¼0.03). Aswell, there were significant effects for misperceiving land use mixand retail density. Those who perceived the density of shops aslower than objectively determined increased their BMI by 0.6 morethan those whose perception matched the objective measure(p¼0.01). There was a trend for an effect for misperceiving highdwelling density as low. There was no effect of perceiving streetconnectivity as lower than objectively measured on the changes inBMI (Table 2).

4. Discussion

In the total sample, over the four years between the baseline andfollow-up surveys, there was a significant decrease in walkinglevels and a significant increase in BMI. These findings are con-sistent with those of other studies, which have found that withincreasing age, adults become physically less active and gain moreweight (US Department of Health and Human Services, 1996; Trostet al., 2002; Lewis et al., 2000; Jacobsen et al., 2001; Allman-Farinelli et al., 2008; Gordon-Larsen et al., 2009).

Misperceiving objectively measured high walkability as lowwalkable was significantly associated with a greater decrease inwalking for transport over four years compared to having percep-tions which were concordant with high walkability. The differencein the decrease in transport walking between those with mis-perceptions and those whose perceptions were concordant was450 min/week, which is substantial, given that public healthguidelines recommend 150 min/week of physical activity. If weadd the differences in walking for transport and walking for leisure,participants who misperceived walkability as lower than it wasobjectively determined to be decreased their walking levels by79.3 min/week more than those who had higher perceptions. Thisdifference is equivalent to more than half the recommended levelof physical activity. It is possible to argue that this large decline inwalking over time may be able to be prevented or reduced bymaking residents more aware of the opportunities for walkingwithin their neighborhoods.

Studies examining the associations of built-environment attri-butes with physical activity have found relatively consistentassociations between the four domains of walkability used in thisstudy (dwelling density, street connectivity, land use mix, andretail density) and walking for transport (Transportation ResearchBoard, 2005; Heath et al., 2006; Saelens et al., 2003b; Humpel et al.,

2002; Lee and Moudon, 2004). Those who misperceived objectivelymeasured high overall walkability, dwelling density, and land usemix as low had significantly greater decreases in walking fortransport than did those whose perceptions matched the objectivemeasure. As well, there was a non-significant trend in the relation-ship between the decrease in walking for transport and perceivingretail density as lower than it actually is. Overall, our findingsprovide some novel evidence that perceiving these domains ofwalkability as lower than they are objectively determined to be cannegatively influence changes over time in walking for transport.

There was no effect for misperceiving street connectivity onchanges in such utilitarian walking. This was somewhat surprising,as the literature relatively consistently states that street connec-tivity is a correlate of walking for transport.

With the exception of street connectivity, the influence ofmisperceiving high walkability as low was weaker for walkingfor leisure than for walking for transport. As stated above, researchhas found that the four domains of walkability used in this studyare more related to walking for transport than they are to walkingfor leisure. Therefore, this finding is consistent with the literature.Studies have mostly linked street connectivity with walking fortransport as more connected neighborhoods provide more directroutes and thereby higher accessibility to destinations (Saelenset al., 2003b). However, it has also been argued that connectivitycould influence walking by offering more choice of routes (Saelensand Handy, 2008). This variety of routes might explain why therewas a significant effect for misperceiving street connectivity onchanges in walking for leisure.

In the overall sample, there was a statistically significantincrease in body weight over four years. Misperceiving overallwalkability, land use mix, and retail density were significantpredictors of BMI change. The composite measure for perceivedretail density incorporated 11 items, including accessibility ofsupermarkets and greengrocers. The frequency of trips to these twodestinations can be expected to be higher than for most otheritems, such as post offices, hardware stores, or pharmacies.Furthermore, previous studies have shown that high accessibilityof supermarkets and greengrocers can not only affect the bodymass index by improving energy expenditure through facilitatingactive transport, but also by reducing energy intake through betteravailability of healthy food with lower energy density (Morlandet al., 2006; Morland and Evenson, 2009; Spence et al., 2009;Poortinga et al., in press). There was a significant effect ofmisperceiving land use mix as lower than it actually is. A highvariety of different land uses, such as residential, commercial,industrial, or educational, are associated with shorter traveldistances between the places of interest. Thereby, land use mixmight act to facilitate more physically active transport choices,which in turn would mediate weight maintenance.

A strength of our study is that participants were sampled fromneighborhoods with diverse urban form, for whom both objectiveand self-reported measures of four domains of walkability wereanalyzed and compared. Environmental cognitions and familiaritywith the neighborhood have been shown to be related to the

K. Gebel et al. / Health & Place 17 (2011) 519–524 523

duration of residence (Ball et al., 2008; Liben, 1981). Therefore,another strength of this study is that we could adjust our analysesfor the duration of living in respondents’ current residentiallocation. The study participants reported almost no differencesin perceived walkability between the baseline and follow-upsurveys. Thus, we can exclude that changes in perceptions overtime could explain differences in changes in walking and BMI overtime. A limitation is that older adults, women, and people in paidwork were overrepresented in the sample, which might limit thegeneralizability of our findings. Another limitation is that self-report measures were used for physical activity and BMI.

Some longitudinal studies have examined environmental attri-butes as predictors of change in physical activity (Humpel et al.,2004; Li et al., 2005; Sallis et al., 2007) or weight (Li et al., 2009).However, Li et al. (2005) identified a limitation of their study as theone-year period between their baseline and final surveys may havebeen too short to detect meaningful changes in walking levels. Inthe related studies carried out in the USA (Sallis et al., 2007) and inAustralia (Humpel et al., 2004), there were only six months and tenweeks, respectively, between the baseline and the follow-upsurveys. Therefore, it is a major strength of our study that thetime span of our follow-up was four years, which ought to haveallowed us to capture meaningful changes in walking and BMI forthe whole sample. We found a significant decrease in walking fortransport, a borderline significant decrease in walking for leisure,and a significant increase in BMI during the four-year period.

We found that for those who misperceived objectively deter-mined high-walkable attributes of their neighborhood to be lesswalkable, such misperceptions can negatively and significantlyaffect their walking for transport and for leisure, as well as theirweight compared with those whose perceptions match the objec-tive measures. Effects identified in this study are long term, andapplicable to a considerable portion of residents, thus may have asignificant, community-wide health impact. These prospectivefindings support recommendations from previous cross-sectionalstudies; interventions aiming at improving mismatched percep-tions among those who live in high-walkable neighborhoods maybe helpful in maintaining (or reducing declines in) physical activityand in facilitating weight maintenance.

References

Allman-Farinelli, M.A., Chey, T., Bauman, A.E., Gill, T., James, W.P., 2008. Age, periodand birth cohort effects on prevalence of overweight and obesity in Australianadults from 1990 to 2000. European Journal of Clinical Nutrition 62, 898–907.

Australian Bureau of Statistics, 2001. Census of population and housing: CDATA, CD-ROM. Final Release.

Ball, K., Jeffery, R.W., Crawford, D.A., Roberts, R.J., Salmon, J., Timperio, A.F., 2008.Mismatch between perceived and objective measures of physical activityenvironments. Preventive Medicine 47, 294–298.

Black, J.L., Macinko, J., 2008. Neighborhoods and obesity. Nutrition Reviews 66,2–20.

Boehmer, T., Hoehner, C., Wyrwich, K., Ramirez, L., Brownson, R., 2006. Correspon-dence between perceived and observed measures of neighborhood environ-mental supports for physical activity. Journal of Physical Activity and Health 3,22–36.

Brownson, R.C., Chang, J.J., Eyler, A.A., Ainsworth, B.E., Kirtland, K.A., Saelens, B.E.,Sallis, J.F., 2004. Measuring the environment for friendliness toward physicalactivity: a comparison of the reliability of 3 questionnaires. American Journal ofPublic Health 94, 473–483.

Cerin, E., Leslie, E., Owen, N., Bauman, A.E., 2008. An Australian version of theNeighborhood Environment Walkability Scale: validity evidence. Measurementin Physical Education and Exercise Science 12, 31–51.

Cerin, E., Saelens, B.E., Sallis, J.F., Frank, L.D., 2006. Neighborhood EnvironmentWalkability Scale: validity and development of a short form. Medicine andScience in Sports and Exercise 38, 1682–1691.

Craig, C.L., Marshall, A.L., Sjostrom, M., Bauman, A.E., Booth, M.L., Ainsworth, B.E.,Pratt, M., Ekelund, U., Yngve, A., Sallis, J.F., Oja, P., 2003. International physicalactivity questionnaire: 12-country reliability and validity. Medicine and Sciencein Sports and Exercise 35, 1381–1395.

Frank, L.D., Andresen, M.A., Schmid, T.L., 2004. Obesity relationships with commu-nity design, physical activity, and time spent in cars. American Journal ofPreventive Medicine 27, 87–96.

Frohlich, K.L., Bodnarchuk, J.L., Chateau, D., Roos, L., Forsyth, S., 2007. What you see iswhat you get? Questioning the relationship between objective and subjectiveappraisals of neighbourhood resources in relation to health. Canadian Journal ofPublic Health 98, 292–296.

Gebel, K., Bauman, A., Owen, N., 2009. Correlates of non-concordance betweenperceived and objective measures of walkability. Annals of Behavioral Medicine37, 228–238.

Gebel, K., Bauman, A.E., Petticrew, M., 2007. The physical environment and physicalactivity: a critical appraisal of review articles. American Journal of PreventiveMedicine 32, 361–369.

Golledge, R.G., 1987. Environmental cognition. In: Stokols, D., Altman, I. (Eds.),Handbook of Environmental Psychology. Wiley, New York, pp. 131–174.

Golledge, R.G., 1991. Cognition of physical and built environments. In: Garling, T.,Evans, G.W. (Eds.), Environment, Cognition, and Action–an Integrated Approach.Oxford University Press, New York, Oxford, pp. 35–62.

Gordon-Larsen, P., Hou, N., Sidney, S., Sternfeld, B., Lewis, C.E., Jacobs Jr., D.R., Popkin,B.M., 2009. Fifteen-year longitudinal trends in walking patterns and theirimpact on weight change. American Journal of Clinical Nutrition 89, 19–26.

Haskell, W.L., Lee, I.-M., Pate, R.R., Powell, K.E., Blair, S.N., Franklin, B.A., Macera, C.A.,Heath, G.W., Thompson, P.D., Bauman, A.E., 2007. Physical activity and publichealth: updated recommendation for adults from the American College of SportsMedicine and the American Heart Association. Medicine and Science in Sportsand Exercise 39, 1423–1434.

Heath, G., Brownson, R., Kruger, J., Miles, R., Powell, K., Ramsey, L., Task Force onCommunity Preventive Services, 2006. The effectiveness of urban design andland use and transport policies and practices to increase physical activity:a systematic review. Journal of Physical Activity and Health 3, S55–S71.

Humpel, N., Marshall, A.L., Leslie, E., Bauman, A., Owen, N., 2004. Changes inneighborhood walking are related to changes in perceptions of environmentalattributes. Annals of Behavioral Medicine 27, 60–67.

Humpel, N., Owen, N., Leslie, E., 2002. Environmental factors associated with adults’participation in physical activity: a review. American Journal of PreventiveMedicine 22, 188–199.

Jacobsen, B.K., Njølstad, I., Thune, I., Wilsgaard, T., Løchen, M.L., Schirmer, H., 2001.Increase in weight in all birth cohorts in a general population: the Tromsø Study,1974–1994. Archives of Internal Medicine 161, 466–472.

Joshu, C.E., Boehmer, T.K., Brownson, R.C., Ewing, R., 2008. Personal, neighbourhoodand urban factors associated with obesity in the United States. Journal ofEpidemiology and Community Health 62, 202–208.

Kahn, E.B., Ramsey, L.T., Brownson, R.C., Heath, G.W., Howze, E.H., Powell, K.E., Stone,E.J., Rajab, M.W., Corso, P., 2002. The effectiveness of interventions to increasephysical activity. A systematic review. American Journal of Preventive Medicine22, 73–107.

Kaplan, S., Kaplan, R., 1982. Cognition and Environment: Functioning in an UncertainWorld. Praeger, New York.

Kirtland, K.A., Porter, D.E., Addy, C.L., Neet, M.J., Williams, J.E., Sharpe, P.A., Neff, L.J.,Kimsey Jr., C.D., Ainsworth, B.E., 2003. Environmental measures of physicalactivity supports: perception versus reality. American Journal of PreventiveMedicine 24, 323–331.

Kremers, S.P., de Bruijn, G.J., Visscher, T.L., vanMechelen, W., de Vries, N.K., Brug, J., 2006.Environmental influences on energy balance-related behaviors: a dual-processview. International Journal of Behavioral Nutrition and Physical Activity 3, 9.

Kweon, B.S., Ellis, C.D., Lee, S.W., Rogers, G.O., 2006. Large-scale environmentalknowledge—investigating the relationship between self-reported and objec-tively measured physical environments. Environment and Behavior 38, 72–91.

Lackey, K.J., Kaczynski, A.T., 2009. Correspondence of perceived vs. objectiveproximity to parks and their relationship to park-based physical activity.International Journal of Behavioral Nutrition and Physical Activity 6, 53.

Lee, C., Moudon, A.V., 2004. Physical activity and environment research in the healthfield: implications for urban and transportation planning practice and research.Journal of Planning Literature 19, 147–181.

Leslie, E., Coffee, N., Frank, L., Owen, N., Bauman, A., Hugo, G., 2007. Walkability oflocal communities: using geographic information systems to objectively assessrelevant environmental attributes. Health & Place 13, 111–122.

Lewis, C.E., Jacobs Jr., D.R., McCreath, H., Kiefe, C.I., Schreiner, P.J., Smith, D.E.,Williams, O.D., 2000. Weight gain continues in the 1990s: 10-year trends inweight and overweight from the CARDIA study. Coronary artery risk develop-ment in young adults. American Journal of Epidemiology 151, 1172–1181.

Li, F., Fisher, J., Brownson, R.C., 2005. A multilevel analysis of change in neighborhoodwalking activity in older adults. Journal of Aging and Physical Activity 13,145–159.

Li, F., Harmer, P., Cardinal, B.J., Bosworth, M., Johnson-Shelton, D., Moore, J.M., Acock,A., Vongjaturapat, N., 2009. Built environment and 1-year change in weight andwaist circumference in middle-aged and older adults: Portland NeighborhoodEnvironment and Health Study. American Journal of Epidemiology 169, 401–408.

Liben, L.S., 1981. Spatial representation and behavior: multiple perspectives. In:Liben, L.S., Patterson, A.H., Newcombe, N. (Eds.), Spatial Representation andBehavior across the Life Span: Theory and Application. Academic Press, NewYork, pp. 3–36.

Lloyd, R., 1997. Spatial Cognition—Geographic Environments. Kluwer Academic,Boston.

Macintyre, S., Macdonald, L., Ellaway, A., 2008. Lack of agreement betweenmeasured and self-reported distance from public green parks in Glasgow,Scotland. International Journal of Behavioral Nutrition and PhysicalActivity 5, 26.

K. Gebel et al. / Health & Place 17 (2011) 519–524524

McCormack, G.R., Cerin, E., Leslie, E., Du Toit, L., Owen, N., 2008. Objective versusperceived walking distances to destinations: correspondence and predictivevalidity. Environment and Behavior 40, 401–425.

McGinn, A.P., Evenson, K.R., Herring, A.H., Huston, S.L., Rodriguez, D.A., 2007.Exploring associations between physical activity and perceived andobjective measures of the built environment. Journal of Urban Health 84,162–184.

Merom, D., Bauman, A.E., Vita, P., Close, G., 2003. An environmental intervention topromote walking and cycling—the impact of a newly constructed Rail Trail inWestern Sydney. Preventive Medicine 36, 235–242.

Morland, K., Diez Roux, A.V., Wing, S., 2006. Supermarkets, other food stores, andobesity: the atherosclerosis risk in communities study. American Journal ofPreventive Medicine 30, 333–339.

Morland, K.B., Evenson, K.R., 2009. Obesity prevalence and the local food environ-ment. Health & Place 15, 491–495.

Mowen, A., Orsega-Smith, E., Payne, L., Ainsworth, B., Godbey, G., 2007. The role ofpark proximity and social support in shaping park visitation, physical activity,and perceived health among older adults. Journal of Physical Activity and Health4, 167–179.

Mujahid, M., Diez Roux, A., Shen, M., 2008. Relation between neighborhoodenvironments and obesity in the Multi-Ethnic Study of Atherosclerosis. Amer-ican Journal of Epidemiology 167, 1349–1357.

Napolitano, M.A., Lerch, H., Papandonatos, G., Marcus, B.H., 2006. Worksite andcommunications-based promotion of a local walking path. Journal of Commu-nity Health 31, 326–342.

Owen, N., Cerin, E., Leslie, E., duToit, L., Coffee, N., Frank, L.D., Bauman, A.E., Hugo, G.,Saelens, B.E., Sallis, J.F., 2007. Neighborhood walkability and the walking behaviorof Australian adults. American Journal of Preventive Medicine 33, 387–395.

Owen, N., Humpel, N., Leslie, E., Bauman, A.E., Sallis, J.F., 2004. Understandingenvironmental influences on walking: review and research agenda. AmericanJournal of Preventive Medicine 27, 67–76.

Owen, N., Leslie, E., Salmon, J., Fotheringham, M., 2000. Environmental determinantsof physical activity and sedentary behavior. Exercise and Sport Sciences Reviews28, 153–158.

Papas, M.A., Alberg, A.J., Ewing, R., Helzlsouer, K.J., Gary, T.L., Klassen, A.C., 2007. Thebuilt environment and obesity. Epidemiologic Reviews 29, 129–143.

Parra, D.C., Gomez, L.F., Sarmiento, O.L., Buchner, D., Brownson, R., Schmid, T.,Gomez, V., Lobelo, F., 2010. Perceived and objective neighborhood environmentattributes and health related quality of life among the elderly in Bogota,Colombia. Social Science and Medicine 70, 1070–1076.

Pate, R.R., Pratt, M., Blair, S.N., Haskell, W.L., Macera, C.A., Bouchard, C., Buchner, D.,Ettinger, W., Heath, G.W., King, A.C., Kriska, A., Leon, A.S., Marcus, B.H., Morris, J.,Paffenbarger JR., R.S., Patrick, K., Pollock, M.L., Rippe, J.M., Sallis, J.F., Wilmore,J.H., 1995. Physical activity and public health: a recommendation from theCenters for Disease Control and Prevention and the American College of SportsMedicine. Journal of the American Medical Association 273, 402–407.

Poortinga, W., Gebel, K., Bauman, A.E., Vernez Moudon, A., in press. The neighbor-hood environment, physical activity and obesity. In: Nriagu, J. (Ed.), Encyclo-pedia of Environmental Health, Elsevier, Oxford.

Prins, R.G., Oenema, A., van der Horst, K., Brug, J., 2009. Objective and perceivedavailability of physical activity opportunities: differences in associations withphysical activity behavior among urban adolescents. International Journal ofBehavioral Nutrition and Physical Activity 6, 70.

Reed, J., 2007. Perceptions of the availability of recreational physical activity facilities ona university campus. Journal of American College Health 55, 189–194.

Reed, J.A., Ainsworth, B.E., Wilson, D.K., Mixon, G., Cook, A., 2004. Awareness and useof community walking trails. Preventive Medicine 39, 903–908.

Saelens, B.E., Handy, S.L., 2008. Built environment correlates of walking: a review.Medicine and Science in Sports and Exercise 40, S550–566.

Saelens, B.E., Sallis, J.F., Black, J.B., Chen, D., 2003a. Neighborhood-based differencesin physical activity: an environment scale evaluation. American Journal of PublicHealth 93, 1552–1558.

Saelens, B.E., Sallis, J.F., Frank, L.D., 2003b. Environmental correlates of walking andcycling: findings from the transportation, urban design, and planning litera-tures. Annals of Behavioral Medicine 25, 80–91.

Sallis, J.F., King, A.C., Sirard, J.R., Albright, C.L., 2007. Perceived environmentalpredictors of physical activity over 6 months in adults: activity counseling trial.Health Psychology 26, 701–709.

Sallis, J.F., Owen, N., Fisher, E.B., 2008. Ecological models of health behavior. In:Glanz, K., Rimer, B.K., Viswanath, K. (Eds.), Health Behavior and HealthEducation: Theory, Research and Practice, fourth ed. Jossey-Bass, San Francisco,pp. 465–486.

Salmon, J., Owen, N., Crawford, D., Bauman, A.E., Sallis, J.F., 2003. Physical activityand sedentary behavior: a population-based study of barriers, enjoyment, andpreference. Health Psychology 22, 178–188.

Spence, J.C., Cutumisu, N., Edwards, J., Raine, K.D., Smoyer-Tomic, K., 2009. Relationbetween local food environments and obesity among adults. BMC Public Health9, 192.

Timperio, A., Crawford, D., Telford, A., Salmon, J., 2004. Perceptions about thelocal neighborhood and walking and cycling among children. PreventiveMedicine 38, 39–47.

Transportation Research Board, 2005. Does the Built Environment InfluencePhysical Activity? Examining the Evidence. Transportation Research Board,Washington D.C.

Trost, S.G., Owen, N., Bauman, A.E., Sallis, J.F., Brown, W., 2002. Correlates of adults’participation in physical activity: review and update. Medicine and Science inSports and Exercise 34, 1996–2001.

US Department of Health and Human Services, 1996. Physical activity and health—

a report of the surgeon general. Centers for Disease Control and Prevention, Atlanta.van Stralen, M.M., Kok, G., de Vries, H., Mudde, A.N., Bolman, C., Lechner, L., 2008. The

Active plus protocol: systematic development of two theory- and evidence-basedtailored physical activity interventions for the over-fifties. BMC Public Health, 8.