The impact of completion time on the reliability and validity of online electronic questionnaires: a preliminary study of the FABQ and ODI scales

This study uses secondary analysis of a health survey dataset to explore the impact of completion time on the reliability and validity of online questionnaires. A total of 1,758 chronic low back pain patients recruited online were included in the study. Using two widely used scales for assessing low back pain, the Fear-Avoidance Beliefs Questionnaire (FABQ) and the Oswestry Disability Index (ODI), we analyzed internal consistency, structural validity, and convergent validity across different completion time groups. Multi-group analyses based on time quartiles revealed that completion time exerted some influence on the internal consistency and structural validity of the FABQ, though slower completion times did not consistently yield better results. However, the internal consistency and structural validity of the ODI remained relatively stable across the different completion time groups. Longer completion times were associated with greater response variability in the FABQ but not in the ODI (in unadjusted analyses). Among participants who completed the questionnaires more quickly (the first 50% for FABQ and the first 25% for ODI), convergent validity was less ideal, with some results being contrary to expectations (when examining the correlations with self-reported happiness, sleep quality, and pain levels). In conclusion, our study supports the use of completion time as an important indicator for assessing the quality of online survey data.