Skip to main content
. 2022 Apr 19;19(9):4937. doi: 10.3390/ijerph19094937

Table 6.

Between-group difference of scores on depressive symptoms as well as problematic social media use, perceived social support, social media violence, and loneliness among college students in Shaanxi, China (n = 60).

Variables Time Point Intervention Group (I) Control Group (C) Mean Difference (I-C) (95% CI) p-Values of ANOVA a,b Cohen’s d Effect Size c
Depressive symptoms T1 16.00 ± 10.77 13.23 ± 10.57 2.77 (−2.75, 8.28) 0.319
T2 3.00 ± 4.65 15.70 ± 9.73 −12.70 *** (−16.64, −8.76) <0.001 * 1.67
T3 3.43 ± 5.14 12.13 ± 9.36 −8.70 *** (−12.60, −4.80) <0.001 * 1.15
Problematic social media use T1 23.53 ± 7.12 20.50 ± 4.59 3.03 (−0.06, 6.13) 0.055
T2 12.87 ± 4.90 21.23 ± 4.92 −8.37 *** (−10.91, −5.83) <0.001 * 1.70
T3 12.67 ± 5.40 21.33 ± 5.37 −8.67 *** (−11.45, −5.88) <0.001 * 1.61
Perceived social support T1 67.63 ± 12.15 67.00 ± 14.38 0.63 (−6.25, 7.51) 0.854
T2 78.50 ± 9.05 65.93 ± 12.13 12.57 *** (7.04, 18.10) <0.001 * −1.17
T3 79.07 ± 9.71 65.50 ± 12.71 13.57 *** (7.72, 19.41) <0.001 * −1.20
Social media violence T1 4.87 ± 1.14 5.17 ± 1.76 −0.30 (−1.07, 0.47) 0.437
T2 4.30 ± 0.84 5.60 ± 2.06 −1.30 ** (−2.11, −0.49) 0.002 * 0.83
T3 4.43 ± 0.97 5.87 ± 2.32 −1.43 ** (−2.35, −0.52) 0.003 0.80
Loneliness T1 18.23 ± 3.87 17.70 ± 3.94 0.53 (−1.48, 2.55) 0.599
T2 13.17 ± 3.04 17.07 ± 3.52 −3.90 *** (−5.60, −2.20) <0.001 * 1.19
T3 11.0 ± 4.14 15.93 ± 4.56 −4.93 *** (−7.19, −2.68) <0.001 * 1.13

** p < 0.01; *** p < 0.001. a: We used Bonferroni Tests to adjust for the multiple comparison testing effect, p = 0.003 = 0.05/15 indicates the standard of statistical significance for each intervention effect. b: Baseline level of each variable was adjusted for in the ANOVA. c: Cohen’s d effect size was calculated with formula of (Mc-Mi)/SDpooled at T2 or T3. Cohen’s Rules of Thumb suggests that d values of 0.2, 0.5, and 0.8 represent small, medium, and large effect sizes, respectively. *: To indicate the significance after considering multiple comparison effect.