Feasibility and Validity of Geographically Explicit Ecological Momentary Assessment With Recall-Aided Space-Time Budgets

Abstract

We employ data from the Adolescent Health and Development in Context Study-a representative sample of urban youth ages 11-17 in and around the Columbus, OH area-to investigate the feasibility and validity of smartphone-based geographically explicit ecological momentary assessment (GEMA). Age, race, household income, familiarity with smartphones, and self-control were associated with missing global positioning systems (GPS) coverage, whereas school day was associated with discordance between percent of time at home based on GPS-only versus recall-aided space-time budget data. Fatigue from protocol compliance increases missing GPS across the week, which results in more discordance. Although some systematic differences were observed, these findings offer evidence that smartphone-based GEMA is a viable method for the collection of activity space data on urban youth.

Figure 1:

GPS tracks of pilot data from an AHDC staff member

Figure 2:

Convex Hull algorithm processing Note: The table on the left indicates the number of points in the convex hull calculation, and the increase in area by the addition of the last point. The pictures depict the addition of points in time from the upper left to the lower right. Gray points indicate the vertices/bounding points of the convex hull; yellow markers indicate GPS points falling within the convex hull. As points are added to the grouping, the area of the convex hull increases. The convex hull increases by 100% with the addition of a single point, marking the end of the stationary period and the beginning of a travel period at the last added point.

Figure 3:

Screenshot from the space-time budget application using pilot data