Experimental Determination of Effectiveness of an Individual Information-Centered Approach in Recovering Step-Count Missing Data

Missing values are a common phenomenon in physical activity research, which has a negative impact on the quality of the data collected. The purpose of this study was to determine empirically the effectiveness of an individual information-centered (II-centered) approach in recovering step-count missing values by comparing the performance of the II-centered approach with the traditional group information-centered approach. Using pedometers, the step counts of 117 participants were measured for 21 consecutive days. A semisimulation approach was used to create a missing data set and several conditions of weekdays, weekend days, or both, were compared under each approach. Two accuracy indexes, Root Mean Square Difference and Mean Signed Difference, were used to determine the effectiveness of the recovery methods, and paired t tests were used to examine the mean differences between the original and the replacement values. The II-centered approach produced a more accurate recovery method. The impact of the findings and future research directions in recovering physical activity data are discussed.