Masters Theses

Date of Award


Degree Type


Degree Name

Master of Science



Major Professor

Scott E. Crouter

Committee Members

Eugene C. Fitzhugh, Dawn P. Coe


BACKGROUND: Sedentary behavior (SB) is defined by an energy expenditure (EE) ≤1.5 METs while in a seated, lying, or reclining posture. Criterion measures for SB, direct observation [(DO); posture] and indirect calorimetry [(IC); EE], are used to validate objective estimates of SB. It is unclear whether the criterion measures provide different measurement for SB time and if this impacts validation studies. PURPOSE: To compare the measured SB time by DO, IC, and combined approach (DO+IC) and to examine how changing the criterion measure impacts the validity of estimated SB time in youth from accelerometer-based devices. METHODS: One-hundred youth (6-18 years) had their resting metabolic rate measured and completed 16 activities ranging from sedentary to vigorous intensities while wearing an ActiGraph GT9X (AG) on the right hip, and an activPAL 3C (AP) on each thigh. Criterion measures for SB time were the Cosmed K4b2 IC (EE), DO (posture), and DO+IC (EE+posture). Estimated SB by the AP and AG inclinometer and AG SB thresholds [vertical axis (VA) <25 counts/10 sec, vector magnitude (VM) <75 counts/10 sec; VA<100 counts/60 sec and VA<150 counts/ 60 sec] were compared against each criterion measure using repeated measures ANOVAs. RESULTS: Final sample was 65 youth. DO SB time was significantly greater than IC and DO+IC SB time by 7.0 min and 4.5 min, respectively (p<0.001). DO+IC SB time was significantly greater than IC by 2.5 min (p=0.014). All estimates, except APleft, VA<150 counts/60 sec and VM<75 counts/10 sec, were significantly different from DO SB time by 3.1 (VA<100 counts/60 sec) to 32.8 min (AG inclinometer) (p<0.05). All estimates significantly overestimated IC SB time by 3.9 min (VA<100 counts/60 sec) to 39.9 min (AG inclinometer; p<0.001). VA<100 counts/60 sec was the only estimate not significantly different from DO+IC SB time (mean difference=1.4min; p=0.212); all other estimates significantly overestimated SB time by 2.7 (VA<150 counts/60 sec) to 37.3 minutes (AG inclinometer; p<0.05). CONCLUSION: Validation studies using different criterion measures for SB should not be compared as DO, IC, and DO+IC provide different measured SB time. Standardization of criterion measures for SB time is needed to ensure consistent results going forward.

Available for download on Friday, May 15, 2026

Files over 3MB may be slow to open. For best results, right-click and select "save as..."