Recording the Time Divide : A Comparative Study of Smartphone- and Recall-Based Approaches to Time Use Measurement

Based on a randomized survey experiment in Malawi, this study examines how innovative techniques in time use data collection could sidestep measurement concerns with traditional recall-based time use measurement. The experiment assigns random samples of households, and adult men and women within, to one of two treatment arms on time use measurement: a traditional 24-hour recall time use diary, and a self-administered smartphone-based pictorial time diary, known as the TimeTracker app, for real-time data collection. Compared to the recall arm, participation in employment and unpaid domestic and care work is shown to be higher in the smartphone arm for both men and women. The resulting estimates of gender gaps, while continuing to be large, are narrower in the smartphone arm, except for care work where the estimated gender gap increases. The recall treatment leads to substantial underreporting of activities after 6 pm, which otherwise accounts for nearly 30 percent of daily reported time in the smartphone arm. Likewise, the extent of simultaneous activities, particularly among women, is markedly lower in the recall arm. The overall reported time is, however, higher in the recall arm due to the minimum 15-minute duration that was used for recording activities the 24-hour recall diary, while over one-third of activities lasted less than 15 minutes in the smartphone arm. The analysis also shows that using stylized time use questions with a 7-day recall, as opposed to a 24-hour recall diary, results in an even greater overestimation of reported time in employment and unpaid work.