•  
  •  
 

TO TRACK OR NOT TO TRACK: VALIDITY AND RELIABILITY OF COMMONLY USED SELF-REPORT PHYSICAL ACTIVITY AND SEDENTARY BEHAVIOR INSTRUMENTS AMONG PA TRACKING DEVICE USERS AND NON-USERS

Abstract

BACKGROUND: Self-reported physical activity (PA) is higher among individuals who use PA tracking devices (PATs) compared to those who do not. Little is known about the relationship between PAT use and sedentary behavior (SED) or that of PAT-use and the validity and reliability of self-report instruments assessing these behaviors. The purpose of this study was to assess the validity and reliability of self-report PA and SED instruments among college student PAT users (PAT-U) and PAT non-users (PAT-NU). It was hypothesized that reliability and validity would be higher among PAT-U compared to PAT-NU because tracking may enhance recall accuracy. METHODS: PAT-U (n=55; 69.1% female) and PAT-NU (n=43; 72.1% female) completed the Global Physical Activity Questionnaire (GPAQ), International Physical Activity Questionnaire-Long Form (IPAQ-L) and Short Form (IPAQ-S), Sedentary Behavior Questionnaire (SBQ), and SIT-Q on 2 occasions separated by 7-8 days. Participants wore devices at the waist (ActiGraph GT9X Link) and on the thigh (activPAL3) to objectively assess PA and SED, respectively, between administrations of the self-report instruments. Intraclass correlation coefficients assessed test-retest reliability. Spearman’s rho assessed concurrent validity. Mean absolute error (MAE) was calculated for variables from self-report instruments using the devices as the comparison. RESULTS: For total PA (METmin/wk), reliability was consistently higher for PAT-NU compared to PAT-U (ICC: 0.59-0.75 & 0.49-0.56, respectively). For MVPA, reliability was not markedly different between the two groups but for SED, PAT-U showed higher reliability than PAT-NU (ICC: 0.40-0.83 & 0.34-82, respectively). For MVPA, validity was consistently higher for PAT-U whereas MAE was lower for PAT-NU. For vigorous-intensity PA (VPA), validity was higher and MAE was lower for PAT-U (ρ=0.265-0.295; MAE=80.8-103.0, respectively) compared to PAT-NU (ρ=-0.028-0.191; MAE=88.9-119.4, respectively). CONLUSIONS: Reliability was moderate to good among both PAT-U and PAT-NU. PAT-U may be more accurate in reporting VPA, but not overall MVPA or SED, compared to PAT-NU. Researchers seeking to assess PA and SED among college students should be aware of individual-level factors, such as PAT use, that could influence the validity of self-report instruments. Specifically, these factors may influence PA and SED outcomes (e.g., MVPA, VPA, SED) differently.

This document is currently not available here.

Share

COinS