A study exploring predictors of cell phone use while walking among adolescents based on theory of planned behavior

Abstract

Background: Walking is a complex activity that requires high levels of perception and cognitive abilities. Healthy pedestrians who use mobile phones while walking will have their decision-making process affected to varying degrees and may be at greater risk of injury. Previous studies have shown that using mobile phones while walking is becoming increasingly common. Therefore, the objectives of this study are to solve the following problems: What factors influence the intention and behavior of teenagers to use mobile phones while walking? Do different factors play the same role in students of different genders, school types, mobile phone dependence, or mental health status? Based on the above results, what measures should we take to reduce the behaviors of teenagers using mobile phones while walking?

Method: This study used a cross-sectional online survey design. The study was conducted in six junior high schools and four senior high schools in Shanghai, China, from December 2019 to January 2020 (N = 4,082 students in Shanghai). The questionnaire was designed based on the theory of planned behavior and analyzed by structural equation model analysis and multi-group invariance analysis.

Results: Girls, junior middle school students, and students without mental health problems or mobile phone dependence can better understand the hazards of using mobile phones while walking. They are more willing to accept other people's suggestions about not using mobile phones while walking. Additionally, they exhibit better self-control and lower levels of intention and behavior when using mobile phones while walking.

Conclusions: The key points to preventing teenagers from using mobile phones while walking are to instill a correct attitude toward dangerous behavior, build a stronger sense of norm, help them form good habits in mobile phone use, and improve their ability to control their behavior.

Keywords: Adolescent; Mobile phone; Pedestrian; Structural equation model; Theory of planned behavior.

Fig. 1

Structural equation modeling of mobile phone use behavior while walking. Legend: λ measurement weight (loading); γ and b structural weight (path coefficient); δ and ε measurement residual; ζ structural residual; ϕ factor covariance; τ intercept (mean score of the item); ▭ observed variable; ⬭ latent variable

Fig. 2

Standardized measurement/structural weights in SEMs

Fig. 3

Comparison of item scores between different subgroups in sex. Legend: Bars represent the mean ± standard error of the mean (SEM) of each item score in each subgroup. Independent-sample t tests were performed to compare each item scores between different subgroups. *P<0.05.**P<0.01 (two-tailed test)

Fig. 4

Comparison of item scores between different subgroups in school type. Legend: Bars represent the mean ± standard error of the mean (SEM) of each item score in each subgroup. Independent-sample t tests were performed to compare each item scores between different subgroups. *P<0.05. **P<0.01 (two-tailed test)

Fig. 5

Comparison of item scores between different subgroups in mental health problem. Legend: Bars represent the mean ± standard error of the mean (SEM) of each item score in each subgroup. Independent-sample t tests were performed to compare each item scores between different subgroups. *P<0.05. **P<0.01 (two-tailed test)

Fig. 6

Comparison of item scores between different subgroups in mobile phone dependence. Legend:Bars represent the mean ± standard error of the mean (SEM) of each item score in each subgroup. Independent-sample t tests were performed to compare each item scores between different subgroups. *P<0.05.**P<0.01 (two-tailed test)