Factors Impacting the Quality of User Answers on Smartphones

TL;DR

This study tackles why user-provided smartphone context answers vary in quality, focusing on reaction time and completion time as key determinants. It builds a causal, hierarchical ML-SEM framework to relate these times to answer correctness, using data from a four-week SmartUnitn2 experiment with GPS verification and sensor data collected via iLog. The results show that longer reaction times increase the likelihood of errors and affect the Home–phone distance, with indirect effects mediated by pending notifications; the model fits well (RMSEA $= 0.022$, SRMR $= 0.013$, $R^2 = 0.127$) and explains about 13% of variance. The findings highlight the importance of aligning notification timing with user context and accounting for cognitive and psychosocial traits to improve data quality in experience-sampling on mobile devices.

Abstract

So far, most research investigating the predictability of human behavior, such as mobility and social interactions, has focused mainly on the exploitation of sensor data. However, sensor data can be difficult to capture the subjective motivations behind the individuals' behavior. Understanding personal context (e.g., where one is and what they are doing) can greatly increase predictability. The main limitation is that human input is often missing or inaccurate. The goal of this paper is to identify factors that influence the quality of responses when users are asked about their current context. We find that two key factors influence the quality of responses: user reaction time and completion time. These factors correlate with various exogenous causes (e.g., situational context, time of day) and endogenous causes (e.g., procrastination attitude, mood). In turn, we study how these two factors impact the quality of responses.