Problematic internet use and safety behavior: The moderating role of safety climate

climate were predictors and safety behavior was the dependent variable (Eq 1). The second model is called the interaction model in which PIU, safety climate, and the interaction term (i.e. PIU × safety climate) are predictors, and safety behavior is the dependent variable (Eq 2). The moderation effect is supported if the effect of the interaction term is significant [44]. Basic model : SB ¼ b0 þ SC:b1 þ PIU:b2 ð1Þ Interaction model : SB ¼ b0 þ SC:b1 þ PIU:b2 þ SC � PIU:b3 ð2Þ Where SB safety behavior, SC = safety climate, PIU = problematic internet use, and b1, b2, and b3 are regression weights. Visualization of the moderation effect can provide a deeper insight on how the moderation works. For this purpose, the procedure recommended by Hayes [45] was employed. In this regard, two new equations (simple slope lines) were generated from the interaction model for- mula by inserting certain values of the moderator into the equation, i.e. safety climate = mean + SD and safety climate = mean–SD. The effect of moderation variable would be visualized when these two slope lines are plotted for various values of PIU. Moreover, the size of moderation effect was calculated using Eq 3 [46]: f 2 ¼ R2 i R2 b 1 R2 i ð3Þ Where f2 = the size of moderation effect, R2 i = R2 of the interaction model, and R2 b = R2 of the basic model. The size of moderation effect is negligible when its value is lower than 0.002. Results A total number of 485 employees from a mining industry participated in this study (response

rate = 66%). All participants were male, most of them were married (84.1%), most of them were aged between 31 and 40 years old (61.4%), and most of them had 6–10 years of experi- ence (35.7%). Detailed demographic characteristics of the participants are demonstrated in Table 1. They were working in various units of the mining company, including extraction, pro- cessing mineral concentrates, and pelletizing units. Table 1. Demographic characteristics of participants (n = 485). Variable Number (%) Age (year) = < 30 115 (23.7%) 31–40 298 (61.4%) 41–50 63 (13.0%) > 50 9 (1.9%) Experience (years) <1 56 (11.5%) 1–5 159 (32.8%) 6–10 173 (35.7%) 11–15 70 (14.4%) >15 27 (5.6%) Marital status Single 77 (15.9%) Married 408 (84.1%) https://doi.org/10.1371/journal.pone.0279767.t001 PLOS ONE Problematic internet use and safety behavior PLOS ONE | https://doi.org/10.1371/journal.pone.0279767 December 30, 2022 5 / 13