Maartje Boer

THE COMPLEX ASSOCIATION BETWEEN SMU AND WELLBEING 243 8 internet gaming disorder as established in the appendix of theDiagnostic and Statistical Manual of Mental Disorders, including preoccupation, tolerance, withdrawal, persistence, displacement, problems, deception, escape, and conflict (American Psychiatric Association, 2013; Lemmens et al., 2015). Respondents were asked: ‘During the past year, have you (…)’, followed by, for example, ‘often felt bad when you could not use social media?’ (withdrawal). Response options were (1) yes and (0) no . A sum-score was computed that denoted adolescents’ number of present criteria. As appropriate for dichotomous items, internal consistency was calculated using the tetrachoric correlation matrix (Gadermann et al., 2012), which yielded an alpha of 0.85. Respondents’ gender was measured by asking whether they were (0) boy or (1) girl. Also, adolescents reported their educational level: (1) pre-vocational , (2) intermediate , or (3) pre-university . Adolescents’ educational level was defined as their most recent reported educational level. Immigrant background was determined based the reported country of origin of the parents. Table 8.1 shows the descriptive statistics of all study variables. Table 8.1 Descriptive Statistics Mean / proportion SD Min. Max. n Time variant variables Life satisfaction 4.664 0.842 1 6 5,676 SNS viewing 4.190 1.675 1 7 5,676 SNS posting 1.980 1.457 1 7 5,676 SNS liking 4.950 2.089 1 7 5,676 SNS responding 3.450 1.925 1 7 5,676 IM viewing 4.454 1.575 1 7 5,676 IM sending 4.259 1.732 1 7 5,676 Upward social comparison 1.841 0.815 1 5 5,676 SMU problems 1.186 1.519 0 9 5,676 Time invariant variables Girl 0.459 0 1 1,419 Pre-vocational education 0.578 0 1 1,419 Immigrant background 0.219 0 1 1,419 Notes. SNS = social network sites; IM = instant messenger; SMU = social media use; SD = standard deviation; Min. = minimum; Max. = maximum; n = sample size. Analytical Approach Missing Data Missing data ranged between 6.55% (SMU intensity T2) and 67.94% (upward social comparison T4). Little’s Chi-square test for missing data showed that the