Time spent in sedentary behavior (SB) has increased during the last decades. Accurate assessments are of importance when studying health consequences of SB. This study aimed to assess concurrent validity between three different questions for self-reported sitting and thigh worn accelerometer data. In total, 86 participants wore the ActivPAL accelerometer during three separate weeks, assessing sitting time with different questions each week. The questions used were Katzmarzyk, GIH stationary single-item question (SED-GIH), and a modified version of the single-item from IPAQ short form. In total 64, 57, and 55 participants provided valid accelerometer and questionnaire data at each time-point, respectively, and were included for analysis. Spearman and Pearson correlation was used to assess the validity. The three questions, Katzmarzyk, SED-GIH, and a modified question from IPAQ all showed a weak non-significant correlation to ActivPAL with r-values of 0.26, 0.25, and 0.19 respectively. For Katzmarzyk and SED-GIH, 50% and 37% reported correctly, respectively. For the modified IPAQ, 53% over-reported and 47% under-reported their sitting time. In line with previous research, our study shows poor validity for self-reported sitting-time. For future research, the use of sensor-based data on SB are of high importance.
Keywords: questionnaire; sedentary behavior; self-reported; sensor-based measurements; sitting.