Validity of tests, psychometrics, perception, musculoskeletal diseases
Background: Dentists are exposed to occupational hazards, such as musculoskeletal disorders, in which symptoms can manifest early in clinical practice. Objectives: To estimate the reliability and validity of the Portuguese version of the “Questionnaire on work-related activities that may contribute to musculoskeletal symptoms” when applied to dental students. Methods: 553 Brazilian students participated in the study. The one-factor structure originally proposed was tested through confirmatory factor analysis (CFA) using the indexes c2/df, CFI, GFI, and RMSEA. After observing an inadequate good fit (c2/df=7.140, CFI=0.791, GFI=0.778, RMSEA=0.129), the sample was subdivided into 3 groups and an exploratory factor analysis was conducted (EFA) (n=255). A CFA was then conducted using a subsample (n=113). To test the invariance of the obtained factorial solution, a multi-group analysis was performed using a third sample, independent from the others (n=185). The convergent (AVE) and discriminant (r2) validity were assessed. The composite reliability (CR), Cronbach’s alpha coefficient (a), and intra-class correlation coefficient (r) were calculated. Results: In the EFA, 3 factors were extracted: “Repetitiveness,” “Working posture,” and “External factors” (c2=21895.154; p<0.001; KMO=0.905; l1=6.683; l2=1.464; l3=1.277) that explained 62.82% of the total variance. The three-factor model presented an adequate good fit (n=113 and n=185) (c2/df=2.259; CFI=0.885; GFI=0.833; RMSEA=0.066). The convergent validity was compromised only for “Repetitiveness” (AVE=0.412-0.653). The discriminant validity (r2=0.333-0.428), composite reliability, internal consistency, and reproducibility were adequate for all factors (CR=0.736–0.883; a=0.747–0.876; r=0.729-0.940). Conclusion: The questionnaire proved to be reliable and valid for the sample of dentistry students if the three-factor model is used.