Journal Logo


Personality Correlates of Physical Activity in College Women


Author Information
Medicine & Science in Sports & Exercise: August 2015 - Volume 47 - Issue 8 - p 1691-1697
doi: 10.1249/MSS.0000000000000570


Physical activity in the United States is below recommended levels and is a target of public health interventions, which often have modest success (20). Evidence from observational studies (31,40) indicates that personality explains some of the natural variation in physical activity, might help explain genetic influence on physical activity (6,34), and could modify the success of interventions to increase physical activity (10).

Studies consistently report that extraversion and neuroticism, broad personality traits, are related to physical activity (31,40), but explanations for these relationship have only begun to be elaborated within motivational theory (24). Reinforcement sensitivity theory (17,18) conceptualizes two orthogonal personality dimensions representing differences in the intensity of functioning of the behavioral inhibition system (BIS) and the behavioral approach system (BAS), which offer trait descriptions that are reflective of reinforcement history and operate within the same conceptual space as extraversion and neuroticism (13). Trait level functioning of the BIS corresponds to trait anxiety, whereas trait level functioning of the BAS corresponds to a propensity for behavior motivated by positive reinforcement and to accompanying positive affective experience (17,18). Specific hypotheses regarding the relationship between these two sets of orthogonal dimensions indicate that BIS should correlate negatively with extraversion and positively with neuroticism, and BAS should correlate positively with extraversion and neuroticism (17,18).

The acute behavioral outputs of BIS and BAS suggest that their respective personality traits could plausibly be related to physical activity level (17,18). Nonetheless, there is a paucity of evidence regarding BIS/BAS and physical activity behavior. A cross-sectional study reported a small, positive relationship between BIS and a measure of inactivity, or sedentary behavior (38) and a small, negative relationship between inactivity and drive (a psychometric subfactor of BAS; (8) among 1014 undergraduates from four universities. Other authors have reported on the effects of BIS/BAS on subjective responses to acute exercise in a college-age sample (19) and adolescents (33), their relation to exercise dependence in adults with unhealthy body change behaviors (27), and the role of punishment sensitivity (i.e., BIS) in the personality profile of adults involved in high-risk sports (14). It is unclear whether relationships reported are confounded by broader personality dimensions, such as extraversion and neuroticism, as there are currently no studies testing their interrelations relative to physical activity using multivariate analyses.

The relationships reported between personality dimensions and physical activity are small and might depend on physical activity measurement method. Most of the literature on physical activity and personality has used self-report to measure physical activity level (31,40), and evidence for BIS/BAS is limited entirely to self-reports of physical activity. Thus, prior associations might have been biased by common-method artifact of self-report. A test of whether the associations between personality dimensions and physical activity depend on the method used to measure physical activity has not yet been reported. Some evidence indicates that self-reported physical activity is more congruent with accelerometer estimates of physical activity among women than men (12).

The main aims of the present investigation was to test direct, indirect, and interactive relations between extraversion, neuroticism, BIS and BAS, and physical activity and to examine whether those relations differed when physical activity was measured by self-report or accelerometry. We first used confirmatory factor analysis to test the factor validity and measurement equivalence/invariance of the personality measures between two samples of undergraduate women. We used fully latent structural models to test the expected relationships between the traits according to theory (17,18) and to simultaneously predict physical activity measured by self-report or accelerometry.



Two samples of women were recruited from the undergraduate population at a large university in the Southeast United States. All participants submitted informed consent. Inclusion in sample one (n = 409) extended to female undergraduates ages 18–25 (mean ± SD, 20.3 ±1.47 yr). Sample 2 (n = 298) consisted of female freshman students enrolled at the same university. Inclusion criteria for sample 2 required participants to be ages 18–20 yr (mean ± SD, 18.34 ± 0.49 yr), campus residents, and not intercollegiate athletic participants.



Extraversion and neuroticism were assessed by 10 items each, selected from the International Personality Item Pool (IPIP), using a 1–5 Likert-type response format (16). Behavioral inhibition system and BAS were measured with the BIS/BAS scales (8), which are the most widely used scales for the measurement of these constructs. A single seven-item scale measured BIS, and three scales labeled Drive (five items), Fun Seeking (four items), and Reward Responsiveness (four items) were used to measure BAS. All items have a four-point ordered response format. Items are shown in Appendix A (see document, Supplemental Digital Content 1, list of personality questionnaire items,

Physical activity

Physical activity level was assessed in sample 2 by convergent measures: the Global Physical Activity Questionnaire (GPAQ [3]) and the Godin Leisure Time Exercise Questionnaire (GLTEQ [15]), which asked about a typical week, and the International Physical Activity Questionnaire (IPAQ [5]), which asked participants to recall their physical activity during the past 7 d. All three self-report measures have demonstrated evidence of validity (7,9,11). A latent factor for self-reported physical activity was derived using the moderate and vigorous summary scores from the GLTEQ, the METs per week summary score from the GPAQ and the IPAQ. In addition to completing these self-report measures, participants wore an NL-1000 piezoelectric accelerometer (25,36) that assessed moderate to vigorous physical activity for 7 d concurrent with the week questionnaires were completed. Criteria for acceptable measurement were wear times of at least 10 h·d−1 on at least three weekdays and one weekend day. Daily estimates of moderate to vigorous physical activity provided by the accelerometers were used to derive a latent objective physical activity factor.


The Institutional Review Board approved the measurement protocols. Data collection was conducted separately for each sample. Participants in sample 1 gave informed consent, and were allowed to access the personality measures online (35). They were told to complete the measures in the order of their preference within the following 7 days. Data collection for sample 2 included two laboratory visits, 8 days apart, as well as online survey completion during the same week. After submitting informed consent, participants visited the laboratory for accelerometer assignment and instruction, as well as instructions for the online surveys. Participants were shown how to properly wear the accelerometer on their waistband and asked to wear it during all waking hours of the following 7 days, except during activities involving water, such as showering or swimming. Sample 2 was instructed to complete only one or two online surveys at a time in the order of their preference before returning to the lab at the end of the week to turn in their accelerometers.

Statistical Analysis

Full information robust maximum likelihood estimation was used in Mplus 7.11 (28). Critical z-scores (parameter estimate/SE) were used to test significance of relations (fully standardized β coefficients) between variables (P < 0.05). There was less than 2% missing data for sample 1 (156 of 11,919 questionnaire responses), and less than 1% missing self-report data (364 of 37,846 questionnaire responses) and less than 4% missing accelerometer data (79 of 2086 data entries) for sample 2.

Model fit

The chi-square (χ2) statistic, comparative fit index (CFI), root mean square error of approximation (RMSEA) and its 90% CI, and standardized root mean square residual (SRMR) were used to evaluate model fit (4,21). Values of CFI approximating 0.90 are commonly judged to be acceptable, whereas values greater than 0.95 indicate good fit. Values of the RMSEA 0.06 or less and 0.08 or less indicate close and acceptable fit. Although the number of indicators and nonnormal distributions affect statistical power, the available sample sizes were adequate for model tests (23). Concurrent values 0.95 or greater for CFI and 0.08 or less for SRMR provide optimal protection against type I and type II error rates (21). In the case that a significant interaction was revealed, model fit was judged using the Bayesian Information Criterion (BIC) (4).

Confirmatory factor analyses

The factor validity of each personality measure was examined in sample 1 by confirmatory factor analysis (CFA). Two primary factors were hypothesized for the BIS/BAS scales: one latent factor for BIS and a second-order BAS factor comprised of three first-order factors: drive, fun seeking, and reward responsiveness (8). A two-factor model was hypothesized for the measures of extraversion and neuroticism (13). If a hypothesized model was not supported, modification indices, standardized residuals, squared multiple correlations, and covariances between items were examined to determine the source of misfit (2). Problem items were trimmed until each model demonstrated acceptable fit (2). Trimmed measurement models were then specified for sample 2.

Measurement equivalence/invariance

Measurement invariance of the trimmed models for personality scales across samples was first tested with an omnibus test of equal covariances between groups. Had equivalence of covariance matrices not been supported, tests of decreasingly restrictive levels of equivalence would have proceeded to identify the highest level of measurement invariance reflected in these data (37).

Structural model equivalence

Equivalence of structural parameters for interrelationships of the four personality factors between groups was tested using the Wald statistic (39). First, all structural parameters were specified as equal, and the equivalence of the entire model was tested. In the case of a significant difference (P < 0.05), structural parameters were tested individually to identify the source(s) of significant difference in the structural model between the samples. Structural parameters found to be significantly different between samples were excluded from subsequent model specifications.

Structural model for physical activity measurement

The cross-validated structural model was retained for further analysis within sample 2 and expanded to simultaneously test the direct effects of personality factors on correlated physical activity factors representing moderate to vigorous physical activity measured by self-report or accelerometry.


Confirmatory factor analyses

Model fit statistics for confirmatory factor analyses are displayed in Table 1. Confirmatory factor analyses of the personality scales in sample 1 demonstrated poor fit. Modification indices specified collinearity or cross-loading of several items for the included scales. We trimmed a total of seven items from the IPIP (items 10, 12, and 14 from extraversion; and items 5, 7, and 9 and 15 from neuroticism) and four items from the BIS/BAS scales (items 2, 19, and 22 from BIS; and item 21 from the drive subscale of BAS) to obtain acceptable fit. Composite factor reliabilities were 0.881 for Extraversion, 0.814 for neuroticism, 0.683 for BIS, and 0.802 for BAS. The trimmed models for both personality measures had acceptable fit in sample 2 (Table 1). The model for objective physical activity within sample 2 also had acceptable fit. Factor loadings were significant (P < 0.001) and ranged from 0.502 to 0.724. Composite factor reliability was 0.818. The model for the latent variable representing self-report physical activity was fully saturated (i.e., identified by three indicators). Factor loadings were significant (P < 0.001) and ranged from 0.624 to 0.744. Composite factor reliability was 0.715. Full and trimmed measurement models for the personality measures are illustrated in Appendix B (see figures, Supplemental Digital Content 2, Figures for the full and trimmed measurement models of the personality measures in samples 1 and 2,

CFA model fit statistics for personality scales and objective physical activity.

Measurement equivalence/invariance

Equivalence of the covariance matrices between samples for the extraversion and neuroticism subscales (χ2 (91) = 119.446, CFI = 0.990, RMSEA = 0.030 (0.011, 0.044), SRMR = 0.051) and the BIS/BAS scales (χ2 (136) = 145.993, CFI = 0.995, RMSEA = 0.014 (<0.001, 0.030), SRMR = 0.082) was supported.

Structural equivalence/invariance

Structural parameters representing associations between personality factors for each sample, and Wald statistics comparing parameters between samples are listed in Table 2. Model fit was acceptable in both samples (sample 1: χ2 (368) = 772.909, CFI = 0.882, RMSEA = 0.052 (0.047, 0.057), SRMR = 0.066; sample 2: χ2 (368) = 611.655, CFI = 0.893, RMSEA = 0.047 (0.041, 0.054), SRMR = 0.068). Neuroticism significantly correlated with BIS and extraversion, which also significantly correlated with BAS in both samples. The correlation between BIS and extraversion was only significant in sample 1. Wald statistics confirmed that the only significant difference between the samples was the correlation between BIS and BAS, but consistent with theory, it did not differ from zero in either sample.

Structural parameters and equivalence statistics between samples.

Bivariate correlations between latent factors

Bivariate relationships between the latent factors measured in both samples are displayed in Table 3. Correlations between personality factors were significant, except for between BIS and BAS, and BAS and neuroticism. In sample 2, self-report physical activity was significantly correlated with extraversion, BAS, and objective physical activity, which was also significantly correlated with neuroticism.

Bivariate correlations between latent variables for personality and physical activity.

Structural model for physical activity measurement

The model testing prediction of physical activity by personality according to physical activity measurement method (χ2 (df ) = 990.275 (684); RMSEA (90% CI) = 0.039 (0.033–0.044); CFI = 0.900; SRMR = 0.062) is illustrated in Figure 1. Self-reported physical activity was independently predicted by BAS (R2 = 0.14). Neuroticism and BIS independently predicted objective physical activity (R2 = 0.18), indicating a suppression effect for the relationship between BIS factor and objective physical activity that resulted from an interaction between BIS and neuroticism. Decomposition of this interaction proceeded using standard procedures (1,22), yielding estimated scores for objective physical activity among individuals ±1 SE from the mean of each factor, expressed as T-scores (Fig. 2).

Structural model of the relationship between personality and physical activity measured objectively and subjectively.Broken lines indicate nonsignificant paths. Straight lines represent direct relationships indicated by fully standardized regression coefficients. Curved lines represent covariances indicated by correlation coefficients.
Relationship between objectively measured physical activity and BIS moderated by neuroticism. Values are estimated scores for objective physical activity among individuals ±1 SE from the mean of each factor, expressed asT-scores.


The primary finding is that the relationship between physical activity and personality depended on how physical activity is measured. Extraversion was related to self-reported physical activity, whereas neuroticism was related to physical activity measured objectively by an accelerometer. Moreover, the relationship between extraversion and self-reported physical activity was confounded by BAS, and neuroticism imposed a suppression effect on the relationship between BIS and objectively measured physical activity. Consistent with the cumulative evidence (30), self-reported physical activity had a low to moderate association with physical activity measured objectively. Poor agreement between recall and objective methods used to measure physical activity is common in population-based studies and represents a challenge for future studies of personality and physical activity. The relationships between personality dimensions and physical activity reported here and elsewhere (31,40) may be underestimates because the measures used do not assess all the true variance in physical activity. Nonetheless, when correlations of the sizes reported here and elsewhere (40) are viewed in population rates, the associations of physical activity with extraversion, neuroticism, and BAS represent a binomial difference of approximately 10%–30% in self-reported physical activity and 25% (BAS) in objectively measured physical activity between low and high personality scores in a normal distribution (32). This indicates that personality influences physical activity in as many as 1–3 people of 10 in the population. This is of sufficient size to recommend further study to clarify the role of personality in helping explain the genetic basis of physical activity and for better understanding of the effectiveness of interventions to increase physical activity behavior and its associated outcomes.

Strong evidence is provided here for these observations through extensive preliminary analyses that support the validity and equivalence of measurement models for the personality scales as well as the structural equivalence of the interrelationships between the personality factors. The use of convergent measures for self-reported physical activity allowed for the estimation of a latent physical activity factor covering a broader range of behavior than is typically captured in studies using self-reports of physical activity. In addition, to our knowledge, we report the first evidence for the relationship between BIS and BAS and objectively measured physical activity, concurrent with measures of extraversion and neuroticism as correlates of physical activity level. Evidence provided here supports the theoretical assumption that traits interact to influence behavior, and highlights the importance of understanding trait interactions to explain physical activity. Furthermore, it could be expected that certain traits may reflect a predisposition to socially desirable responding to physical activity, or other behavioral self-reports. Results herein support this hypothesis, drawing attention to a method artifact that may help explain the poor convergence commonly reported between physical activity measurement methods. More work will be necessary to strengthen the evidence for these effects.

The structural model for prediction of physical activity confirmed that traits did not uniformly predict physical activity measured by self-report or by accelerometer; nor did the traits correlate as strongly with objectively measured physical activity as with self-report physical activity, consistent with a possible common-method artifact (e.g., socially desirable responding to self-report measures of physical activity). The extent to which such an artifact influenced observations reported here is unclear however. Replication studies should include measures of other first-order personality factors as well as a lie scale to test the possibility of socially desirable responding or common-method artifacts for people with certain traits.

The bivariate relationship between extraversion and self-reported physical activity was accounted for by its shared variance with BAS in the multivariate model. There is strong support for a positive relationship between extraversion and physical activity in the literature (31,40), but prior studies did not examine indirect relations including BAS. More work is necessary to clarify this effect. People with high BAS functioning may be more physically active out of an appetitive drive for the benefits associated with physical activity or because they have a pleasant affective response to increased exertion. In addition, there is evidence that self-reported physical activity is susceptible to social desirability response bias among female undergraduates (26), although the role of extraversion and/or BAS has not yet been considered. It is plausible that individuals with high BAS functioning respond in a socially desirable way to self-report physical activity measures in anticipation of positive feedback or from a desire to be viewed in a more positive light by others and that the relationship between extraversion and physical activity is confounded by strong independent relationships with BAS. Because of limited evidence, it is unclear whether the strength of this relationship between BAS and self-reported physical activity is dependent on sample characteristics. It seems reasonable that a sample of female freshman students would place high importance on personal image, and increments in BAS would predispose them for socially desirable responding to items regarding health practices or body image. Furthermore, other social-cognitive constructs not measured here, such as self-efficacy and perceived social norms, might moderate the relationships between these personality dimensions and physical activity behavior. Replication among more diverse samples should clarify the confounding effect of BAS on the well-established relationship between extraversion and physical activity, as this effect may be sample specific.

Neuroticism had a negative relationship with objectively measured physical activity, which is consistent with the cumulative evidence (31,40). However, the relationship here failed to reach significance for self-reported physical activity. This inverse relationship disagrees with one report of a small, positive relationship observed in a small sample of obese women age 40–64 yr measured with accelerometry (29). Sample differences likely account for this discrepancy. Another explanation is that BIS confounded the observations in that study. We observed a moderate positive relationship between objectively measured physical activity and BIS, which was suppressed by the stronger negative effect between physical activity neuroticism and the strong positive relationship between BIS and neuroticism. Emotionally stable individuals scoring high in BIS were the most active, whereas those scoring highest in neuroticism and lowest in BIS were least active. Those who scored high in both neuroticism and BIS had a similar physical activity level as those who scored low for both neuroticism and BIS, effectively masking the effect for BIS and physical activity. It is possible that older obese samples have higher motivation to improve their health and reduce stigma often associated with obesity by participating in physical activity. If so, the positive relationship between physical activity and BIS we observed could explain the opposing effect previously reported (29). Just as the relationship between extraversion and self-reported physical activity vanished when we controlled for BAS in our sample, so might the inverse relationship between neuroticism and objectively measured physical activity vanish by controlling for BIS in a sample for whom the negative consequences of inactivity are more salient. More detailed studies are required to examine the extent to which personality dimensions interact with other sample characteristics to influence physical activity behavior.

The positive association between BIS and physical activity contradicts a previous report that BIS is positively related to self-reported inactivity (38). An explanation for our observations may come from the motivational aspects of Gray’s theory, which plausibly suggest that increments in BIS could encourage more physical activity in environments or populations that place high importance on health and/or physical appearance. For example, two individuals who both score high for neuroticism could differ in their sensitivity to aversive cues, or cues of nonreward (i.e., high vs low BIS). Although their level of neuroticism predisposes them to participate in less physical activity, the individual who is more sensitive to aversive cues (i.e., high BIS) may be more likely to act on negative reinforcers relative to health, physique, or athletic ability. The observed suppression effect suggests that BIS protects against risk for inactivity among those scoring high for neuroticism, although longitudinal analyses would be required to support this inference.


The cross-sectional design does not allow for causal inferences, which poses a problem in understanding the suppression effect of neuroticism on the relationship between BIS and physical activity. Furthermore, these observations are only generalizable to young women similar to those sampled here. This is particularly relevant to understanding cofounding effect of BAS on the relationship between extraversion and physical activity, which may be population specific. Nonetheless, these initial findings are sufficient to warrant replication studies in men and more diverse samples using a prospective design while stratifying or otherwise adjusting for potentially important participant characteristics such as ethnicity, percent body fat, and pre-existing conditions that might limit physical activity.


Results from this study suggest that the relationship between personality and physical activity depends on the method used to measure physical activity. Furthermore, motivational theories of personality such as BIS and BAS may provide insight to variation between samples in the relationship between physical activity levels and broad personality traits, such as extraversion and neuroticism.

This study was conducted as part of a parent project funded by the USDA (Grant Support: USDA 2008-55215-18825).

The authors would like to acknowledge additional members of the project team for their support and contribution to study design, data collection, and/or data management: Rachelle Acitelli, MS, Jeremy Dean, MS, Bill Evans, MS, Michael Fedewa, MS, Elizabeth Hathaway, MS, Marah Falle King, MS, Jill Lucas, PhD, Kelsey Lyon, MS, Whitni McConnell, MS, and Michael Schmidt, PhD.

The authors state that there are no conflicts of interest relative to the findings of this report. The results of the present study do not constitute endorsement by the American College of Sports Medicine.


1. Aiken LS, West SG. Multiple Regression: Testing and Interpreting Interactions. Thousand Oaks, CA: Sage; 1991. pp. 1–212.
2. Anderson JC, Gerbing DW. Structural equation modeling in practice: a review and recommended two-step approach. Psychol Bull. 1988; 103 (3): 411–23. doi: 10.1037/0033-2909.103.3.411.
3. Armstrong T, Bull F. Development of the World Health Organization Global Physical Activity Questionnaire (GPAQ). J Public Health. 2006; 14 (2): 66–70. doi: 10.1007/s10389-006-0024-x.
4. Bollen KA. A new incremental fit index for general structural equation models. Sociological Methods & Research. 1989; 17 (3): 303–16.
5. Booth M. Assessment of physical activity: an international perspective. Res Q Exerc Sport. 2000; 71 (2): 114–20. PubMed PMID: 218497972; 10925833.
6. Bray M, Hagberg J, Perusse L, et al. The human gene map for performance and health-related fitness phenotypes: the 2006–2007 update. Med Sci Sports Exerc. 2009; 41 (1): 34–73.
7. Bull FC, Maslin TS, Armstrong T. Global physical activity questionnaire (GPAQ): nine country reliability and validity study. J Phys Act Health. 2009; 6 (6): 790–804.
8. Carver CS, White TL. Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: The BIS/BAS Scales. J Pers Soc Psychol. 1994; 67 (2): 319–33. doi: 10.1037/0022-3514.67.2.319.
9. Craig C, Marshall A, Sjöström M, et al. International Physical Activity Questionnaire (IPAQ): 12-country reliability and validity. Med Sci Sports Exerc. 2003; 35: 1381–95.
10. Dishman RK. Gene-physical activity interactions in the etiology of obesity: behavioral considerations. Obesity (19307381). 2008; 16: S60–5. PubMed PMID: 79722024.
11. Dishman RK, Rooks CR, Thom NJ, Motl RW, Nigg CR. Meeting U.S. healthy people 2010 levels of physical activity: agreement of 2 measures across 2 years. Ann Epidemiol. 2010; 20 (7): 511–23. doi: 10.1016/j.annepidem.2010.04.004. PubMed PMID: 51443136.
12. Dyrstad S, Hansen B, Holme I, Anderssen S. Comparison of Self-reported versus accelerometer-measured physcial activity. Med Sci Sports Exerc. 2014; 46 (1): 99–106. doi: 10.1249/MSS.0b013e3182a0595f.
13. Eysenck HJ. The Structure of Human Personality. 3rd ed. London, UK: Methuen; 1970. pp. 1–470.
14. Freixanet MGI. Personality profile of subjects engaged in high physical risk sports. Personality and Individual Differences. 1991; 12 (10): 1087–93.doi: 10.1016/0191-8869(91)90038-d. PubMed PMID: 1992-14865-001.
15. Godin G, Shephard RJ. A simple method to assess exercise behavior in the community. Can J Appl Sport Sci. 1985; 10: 141–6.
16. Goldberg LR, Johnson JA, Eber HW, et al. The international personality item pool and the future of public-domain personality measures. J Res Pers. 2006; 40 (1): 84–96. doi: 10.1016/j.jrp.2005.08.007. PubMed PMID: 19334706.
17. Gray JA. The Neuropsychology of Anxiety: An Enquiry into the Functions of the Septo-Hippocampal System. New York: Oxford University Press; 1982. pp. 1–424.
18. Gray JA, McNaughton N. The Neuropsychology of Anxiety: An Enquiry into the Funtions of the Septo-Hippocampal System. 2nd ed. Oxford: Oxford University Press; 2000. pp. 1–440.
19. Hall EE, Ekkekakis P, Petruzzello SJ. Is the relationship of RPE to psychological factors intensity-dependent? Med Sci Sports Exerc. 2005; 337 (8): 1365–73.
20. Heath G, Parra D, Sarmiento O, et al. Evidence-based intervention in physical activity: lessons from around the world. Lancet. 2012; 380 (9838): 272–81.
21. Lt Hu, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal. 1999; 6 (1): 1–55.
22. Jaccard J, Wan CK. Measurement error in the analysis of interaction effects between continuous predictors using multiple regression: multiple indicator and structural equation approaches. Psychol Bull. 1995; 117 (2): 348–57. doi: 10.1037/0033-2909.117.2.348.
23. Kaplan D, George R. A study of the power associated with testing factor mean differences under violations of factorial invariance. Structural Equation Modeling: A Multidisciplinary Journal. 1995; 2 (2): 101–18.
24. Lochbaum MR, Litchfield K, Podlog L, Lutz R. Extraversion, emotional instability, and self-reported exercise: the mediating effects of approach-avoidance achievement goals. J Sport Health Sci. 2013; 2: 176–83. doi:
25. McMinn D, Rowe DA, Stark M, Nicol L. Validity of the new lifestyles NL-1000 accelerometer for measuring time spent in moderate-to-vigorous physical activity in school settings. Meas Phys Educ Exerc Sci. 2010; 14 (2): 67–78. doi: 10.1080/10913671003715516.
26. Motl RW, McAuley E, DiStefano C. Is social desirability associated with self-reported physical activity? Prev Med. 2005; 40 (6): 735–9.
27. Mussap AJ. Reinforcement sensitivity theory (RST) and body change behaviour in males. Personality and Individual Differences. 2006; 40 (4): 841–52. doi: 10.1016/j.paid.2005.08.013.
28. Muthén L. Mplus. Statistical analysis with latent variables. Version 7; 2012.
29. Ohmori Y, Suzuki N, Morita A, et al. Association of personality (neo-five factor inventory) with eating behaviors and physical activity levels in obese subjects in the Saku Control Obesity Program (SCOP). Anti-Aging Medicine. 2007; 4 (2): 43–50.
30. Prince S, Adamo K, Hamel M, Hardt J, Gorber S, Tremblay M. A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act. 2008; 5 (1): 56. PubMed PMID: doi:10.1186/1479-5868-5-56.
31. Rhodes RE, Smith NEI. Personality correlates of physical activity: a review and meta-analysis. Br J Sports Med. 2006; 40: 958–65.
32. Rosenthal R. Effect sizes: Pearson’s correlation, its display via the BESD, and alternative indices. Am Psychol. 1991; 46 (10): 1086–7. doi: 10.1037/0003-066X.46.10.1086. PubMed PMID: 1992-04003-001.
33. Schneider ML, Graham DJ. Personality, physical fitness, and affective response to exercise among adolescents. Med Sci Sports Exerc. 2009; 41 (4): 947–55. Epub 2009/03/12 doi: 10.1249/MSS.0b013e31818de009. PubMed PMID: 19276837; PubMed Central PMCID: PMC2761825.
34. Stubbe JH, Boomsma DI, Vink JM, et al. Genetic influences on exercise participation in 37,051 twin pairs from seven countries. PLoS One. 2006; 1: e22. Epub 2006/12/22. doi: 10.1371/journal.pone.0000022. PubMed PMID: 17183649; PubMed Central PMCID: PMC1762341.
35. SurveyMonkeyInc. Available from: Palo Alto, CA, USA2011-2013.
36. Tudor-Locke CE, Bassett DR, Shipe MF, McClain JJ. Pedometry methods for assessing free-living adults. J Phys Act Health. 2011; 8 (3): 445–54.
37. Vandenberg RJ, Lance CE. A review and synthesis of the measurement invariance literature: suggestions, practices, and recommendations for organizational research. Organizational Research Methods. 2000; 3 (1): 4–69.
38. Voigt DC, Dillard JP, Braddock KH, Anderson JW, Sopory P, Stephenson MT. BIS/BAS scales and their relationship to risky health behaviours. Personality & Individual Differences. 2009; 47 (2): 89–93. doi: 10.1016/j.paid.2009.02.003. PubMed PMID: 38910400.
39. Wald A. Tests of Statistical Hypotheses Concerning Several Parameters When the Number of Observations is Large. Transactions of the American Mathematical Society. 1943; 54: 426–82.
40. Wilson K, Dishman R. Personality and physical activity: a systematic review and meta-analysis. Personality & Individual Differences. 2015; 72: 230–42.


Supplemental Digital Content

© 2015 American College of Sports Medicine