A Longitudinal Analysis of Gaming- and Non-Gaming-Related Friendships and Social Support among Social Online Game Players

ABSTRACT

Research examining online games often focuses on their potential to negatively impact players. One of the most common concerns is that playing online with others can displace offline relationships and, consequently, detrimentally affect one’s level of “offline” social support. However, there has been little empirical evidence supporting these causal claims. The current study addresses this by outlining a longitudinal analysis between gaming- and non-gaming-related friendships and social support among a representative sample of social online players (i.e., people who play online video games with others). The results indicate that social online video game play with online or offline friends is not related to perceived social support, positively or negatively, cross-sectionally or longitudinally. Taken together, these results dispute the long-held claims of the social displacement hypothesis and instead suggest that social online video game play does not have negative real-world consequences on players’ offline friendships or levels of offline social support.

Funding

The research leading to these results has received funding from the European Union’s Seventh Framework Programme (FP7/2007–2013) under grant agreement number 240864 (SOFOGA).

Notes

1. The participants who dropped out from Wave 2 to Wave 3 (n = 1297) were slightly older than those who participated in both Wave 2 and 3 (n = 902): 43.12 versus 39.17 years, t(2197) = 5.927, p < .001. However, these subsamples did not differ in terms of gender distribution: dropouts 44.1% female/55.9% male versus stayers 55.7% female/44.3% male, χ2 (1) = 0.013, p = .910.

2. For the descriptive analyses we used the original values (see Table 1).

3. Because Little (2013) noted that the fit index SRMR has “not been well evaluated for longitudinal models in any systematic way” (p. 112), we refrained from using the SRMR for the evaluation of the cross-lagged models.

4. This decrease might, in parts, be explained by the procedure that was used to handle outliers (see data analysis section) and, hence, be a methodological artifact instead of an actual trend. Hence, we will refrain from interpreting this change here.

5. To control for the influence of the covariates, we correlated them with the focal variables and each other in both waves and added autoregressive paths for the covariates as well (but no cross-lagged paths).

Additional information

Funding

The research leading to these results has received funding from the European Union’s Seventh Framework Programme (FP7/2007–2013) under grant agreement number 240864 (SOFOGA).