Abstract: Explaining Socioeconomic Differences in Walking App Use with the Technology Acceptance Model in a Population-Based Sample

◆ Anne Vos, University of Amsterdam
◆ Edith Smit, University of Amsterdam
◆ Michel Klein, VU University Amsterdam
◆ Gert-Jan de Bruijn, University of Amsterdam

Low socioeconomic status (SES) is associated with a higher risk of non-communicable diseases due to physical inactivity. Lower SES individuals, for example, walk less than higher SES individuals. The wide availability of walking apps presents an opportunity for promoting walking behavior. However, low SES individuals have been shown to use walking apps to a lesser extent. We used the Technology Acceptance Model (TAM) to investigate whether perceived ease of use (PEOU) and perceived usefulness (PU) are related to SES differences in the use of walking apps.

A total of 1,346 respondents who were a representative reflection of Dutch adults completed an online survey in August 2019. They provided information on level of education as a proxy for SES. National guidelines were used to categorize respondents as low, middle, or high SES. Respondents’ walking app use status was assessed by asking if they (1) currently used walking apps, (2) previously used walking apps, but had discontinued using them, or (3) had never used walking apps. PEOU and PU were measured with 7 items on a 7-point scale (1 = strongly disagree, 7 = strongly agree) adapted from Davis, Bagozzi, and Warshaw (1989). Examples of items were “How the user environment of walking apps works is easy for me to learn” and “Walking apps are useful for me to keep track of my daily walking behavior”.

Analyses showed that low SES individuals were less likely to use walking apps and more likely to have never used apps, X2(4) = 18.64, p = .001. Walking app use positively contributed to PEOU (F(2, 1337) = 88.89, p < .001) and PU (F(2, 1337) = 226.72, p < .001). Non-users scored lowest on PEOU (M = 4.90, SD = 1.41) and PU (M = 3.15, SD = 1.58), whereas current walking app users scored highest (MPEOU = 5.93, SD = 0.86; MPU = 5.05, SD = 1.26). SES also had a significant effect on PEOU (F(2, 1337) = 10.99, p < .001), such that low SES individuals had the lowest PEOU of all SES groups (M = 4.83, SD = 1.44). Furthermore, there was a significant interaction effect between SES and walking app use status on PEOU, F(4, 1337) = 3.00, p = .018. Among the group of non-users, low SES respondents had significantly lower PEOU (M = 4.43, SD = 1.45) than middle (M = 5.01, SD = 1.37) and high SES respondents (M = 5.37, SD = 1.23). There were no significant SES differences in PEOU within the group of previous walking app users.

Results demonstrated SES differences in walking app usage to the disadvantage of low SES individuals. The findings indicate that PEOU can be a barrier that prevents low SES individuals from using walking apps, but that does not constitute a reason to discontinue app use once started. These findings broaden our understanding of differentiating walking app use patterns across SES populations and address the importance for practitioners to concentrate on making walking apps easier to use, rather than just more useful.