Two samples of high school students (n = 315 and n = 395) completed the new Physical Self-Description Questionnaire (PSDQ). Confirmatory factor analysis (CFA) was used to demonstrate support for the 11 scales of physical self-concept (Strength, Body Fat, Activity, Endurance/Fitness, Sport Competence, Coordination, Health, Appearance, Flexibility, General Physical Self-Concept, and Self-Esteem) that the PSDQ is designed to measure, the replicability of its good psychometric properties in the two samples, and the replicability of the factor structure over gender. Subjects in Sample 1 also completed responses to the Physical Self-Perception Profile (Fox, 1990) and the Physical Self-Concept Scale (Richards, 1988). CFA models of this multitrait-multimethod data provided support for both the convergent and discriminant validity of responses to the three instruments. Comparisons of psychometric, theoretical, and pragmatic considerations of the three instruments led to the recommendation of the PSDQ in a wide variety of research and applied settings.
Herbert W. Marsh, Garry E. Richards, Steven Johnson, Lawrence Roche, and Patsy Tremayne
Herbert W. Marsh
The Physical Self-Description Questionnaire (PSDQ) is a multidimensional physical self-concept instrument with 11 scales: Strength, Body Fat, Activity, Endurance/Fitness, Sports Competence, Coordination, Health, Appearance, Flexibility, Global Physical, and Global Esteem. The purpose of this study is to evaluate the construct validity of PSDQ responses in relation to 23 external criteria, including measures of body composition, physical activity, endurance, strength, and flexibility for 192 (113 boys and 79 girls) high school students. Each external validity criterion was predicted a priori to be most highly correlated with one of the PSDQ scales. In support of the convergent validity of the PSDQ responses, every predicted correlation was statistically significant. In support of the discriminant validity of the PSDQ responses, most predicted correlations were larger than other correlations involving the same criterion. These results support the construct validity of PSDQ responses in relation to external criteria and their potential usefulness in a wide variety of sports and exercise settings.
Inés Tomás, Herbert W. Marsh, Vicente González-Romá, Víctor Valls, and Benjamin Nagengast
Test of measurement invariance across translated versions of questionnaires is a critical prerequisite to comparing scores on the different versions. In this study, we used exploratory structural equation modeling (ESEM) as an alternative approach to evaluate the measurement invariance of the Spanish version of the Physical Self-Description Questionnaire (PSDQ). The two versions were administered to large samples of Australian and Spanish adolescents. First, we compared the CFA and ESEM approaches and showed that ESEM fitted the data much better and resulted in substantially more differentiated factors. We then tested measurement invariance with a 13-model ESEM taxonomy. Results justified using the Spanish version of the PSDQ to carry out cross-cultural comparisons in sport and exercise psychology research. Overall, the study can stimulate research on physical self-concept across countries and foster better cross-cultural comparisons.
Herbert W. Marsh, Andrew J. Martin, and Susan Jackson
Based on the Physical Self Description Questionnaire (PSDQ) normative archive (n = 1,607 Australian adolescents), 40 of 70 items were selected to construct a new short form (PSDQ-S). The PSDQ-S was evaluated in a new cross-validation sample of 708 Australian adolescents and four additional samples: 349 Australian elite-athlete adolescents, 986 Spanish adolescents, 395 Israeli university students, 760 Australian older adults. Across these six groups, the 11 PSDQ-S factors had consistently high reliabilities and invariant factor structures. Study 1, using a missing-by-design variation of multigroup invariance tests, showed invariance across 40 PSDQ-S items and 70 PSDQ items. Study 2 demonstrated factorial invariance over a 1-year interval (test–retest correlations .57–.90; Mdn = .77), and good convergent and discriminant validity in relation to time. Study 3 showed good and nearly identical support for convergent and discriminant validity of PSDQ and PSDQ-S responses in relation to two other physical self-concept instruments.
Herbert W. Marsh
Theoretical models of relations between specific components of physical self-concept, global physical self-concept, and global esteem are evaluated. Self-concept models posit that the effect of a specific domain (e.g., strength, endurance, or appearance) on global components should vary with the importance an individual places on the specific domain, but empirical support for this prediction is weak. Fox (1990) incorporated a related assumption into his hierarchical model of physical self-concept, but did not test this assumption. In empirical tests based on responses to the newly developed Physical Self-Description Questionnaire, relations between specific and global components of physical self-concept did not vary with the perceived importance of the specific component, and unweighted averages of specific components were as highly related to global components as importance weighted averages. These results provide no support for the importance of importance in modifying relations between domain-specific and general components of self-concept.
Herbert W. Marsh
The similarity of the constructs measured by the Perceptions of Success Questionnaire (POS; Roberts, 1993) and the Sports Orientation Questionnaire (SOQ; Gill, 1993) were evaluated using (a) confirmatory factor analyses of responses by 395 high school students (217 males, 178 females, ages 12 to 18) to items adapted from the two instruments and (b) relations to external criteria. Although the POS Mastery and SOQ Goal scales were highly related and reflected task orientation, the SOQ Competitiveness scale was more highly correlated with the POS Mastery and SOQ Goal scales than with the POS Competitiveness scale. Apparently, competitiveness assessed by the SOQ reflects a task orientation, whereas the POS Competitiveness scale reflects primarily an ego orientation. Sport psychologists need to beware of jingle (scales with the same label reflect the same construct) and jangle (scales with different labels measure different construct) fallacies, and pursue construct validity studies more vigorously to test the interpretations of measures.
Florence Guérin, Herbert W. Marsh, and Jean-Pierre Famose
Two studies tested the generalizability of support for within- and between-construct validity based on responses to a French translation of the Physical Self-Description Questionnaire (PSDQ) by high school students. The PSDQ is a multidimensional physical self-concept instrument designed to measure 11 components: health, coordination, physical activity, body fat, sports competence, global physical, appearance, strength, flexibility, endurance, and esteem. In the first study (N = 752), preliminary reliability analysis revealed strong internal consistency and overall stability. Confirmatory factor analysis provided support for structural equivalence with the original instrument. In the second study (N = 288), PSDQ factors were related to 13 external criteria of physical fitness; each was predicted a priori to be most highly correlated with one of the PSDQ scales. Bivariate correlations and CFA models supported both the convergent and discriminant validity of the PSDQ responses. These overall results demonstrated good support for the generalizability of the PSDQ with French adolescents.
Genevieve Fridlund Dunton, Margaret Schneider, Dan J. Graham, and Dan M. Cooper
Cross-sectional research examined whether physical activity or physical fitness was more closely linked to physical self-concept in adolescent females ages 14 to 17 (N = 103, 63% Caucasian). Moderate physical activity and vigorous physical activity were measured through a 3-day physical activity recall. Physical fitness was assessed using highly accurate measures of peak oxygen consumption (via cycle ergometer) and percent body fat (via dual X-ray absorptiometer). The Physical Self-Description Questionnaire (PSDQ) assessed self-concept in 11 domains (e.g., health, endurance, appearance). Pearson’s correlations showed that vigorous physical activity was positively associated with scores on most of the PSDQ scales (p < .005). Peak oxygen consumption was positively related to all of the selfconcept domains (p < .001), and percent body fat was negatively related on most of the PSDQ scales (p < .005). Multiple-regression analyses found that physical fitness (i.e., peak oxygen consumption and percent body fat) was more closely related to physical self-concept than was physical activity. In addition to the possibility that genetically determined fitness levels may influence physical selfconcept, these findings suggest that programs designed to elevate self-perceptions may require physical activity levels sufficient to improve cardiovascular fitness and decrease body fat.
Deborah R. Shapiro and Jeffrey J. Martin
The purposes of this investigation were first to predict reported PA (physical activity) behavior and self-esteem using a multidimensional physical self-concept model and second to describe perceptions of multidimensional physical self-concept (e.g., strength, endurance, sport competence) among athletes with physical disabilities. Athletes (N = 36, M age = 16.11, SD age = 2.8) completed the Physical Self-Description Questionnaire. Participants reported mostly positive perceptions of self-esteem, global physical self-concept, endurance, body fat, sport competence, strength, flexibility, and physical activity (Ms ranging from 3.9 to 5.6 out of 6). Correlations indicated a number of significant relationships among self-esteem and reported PA and various dimensions of physical self-concept. Using physical self-concept, strength, endurance, and flexibility in the first regression equation and sport competence and endurance simultaneously in the second equation, 47 and 31% of the variance was accounted for in self-esteem and reported PA, respectively. The findings support the value of examining multidimensional physical self-concept as different aspects of the physical self appear to have different influences on reported PA engagement versus self-esteem.
Herbert W. Marsh, F. Hulya Asci, and Ines Marco Tomas
The present investigation demonstrated cross-cultural support for convergent and discriminant validity of the Physical Self-Description Questionnaire (PSDQ) in a multitrait-multimethod analysis of relations with responses to the Physical Self-Perception Profile (PSPP). The sample, 1,041 Turkish university students in elective physical education courses from 10 Turkish universities, provided a test of the cross-cultural generalizability of responses to these two widely used English language instruments. In support of construct validity interpretations, matching PSDQ and PSPP factors were highly correlated. However, support for the PSPP was undermined by extremely high correlations among several of its factors, due in part to a substantial method effect associated with its idiosyncratic response scale. Results based on this study with Turkish university students largely replicate and extend the findings of Marsh et al. (1994) with Australian high school students. Based on psychometric, theoretical, cross-cultural, and practical considerations, the results support the use of the PSDQ in a wide variety of research and applied settings.