The purpose of this study was to determine the minimum number of days of accelerometry required to estimate accurately MVPA and total PA in 3- to 5-year-old children. The study examined these metrics for all days, weekdays, and in-school activities. Study participants were 204 children attending 22 preschools who wore accelerometers for at least 6 hr per day for up to 12 days during most waking hours. The primary analysis considered the intraclass correlation coefficient (ICC) for each metric to estimate the number of days required to attain a specified reliability. The ICC estimates are 0.81 for MVPA-all days, 0.78 for total PA-all days, 0.83 for MVPA weekdays, 0.80 for total PA-weekdays, 0.81 for in-school MVPA, and 0.84 for in-school total PA. We recommend a full seven days of measurement whenever possible, but researchers can achieve acceptable reliability with fewer days, as indicated by the Spearman-Brown prophecy: 3–4 days for any weekday measure and 5–6 days for the all-days measures.
Addy is with the Dept. of Epidemiology and Biostatistics, Arnold School of Public Health, University of South Carolina, Columbia, SC. Trilk is with the Dept. of Biomedical Sciences, University of South Carolina School of Medicine, Greenville, SC. Dowda and Pate are with the Dept. of Exercise Science, Arnold School of Public Health, University of South Carolina, Columbia, SC. Byun was with the School of Physical Education, Sport and Exercise Science, Ball State University, Muncie, IN at the time of this research and is now with North Dakota State University.