The problem with this study and similar ones is that "use of social media" is measured as a homogeneous activity when in reality the various social media servic

The problem with this study and similar ones is that "use of social media" is measured as a homogeneous activity when in reality the various social media services have complex and important functional distinctions and usage patterns that have a meaningful impact on emotional investment. Spending many hours conversing in a private group chat on Facebook is completely different from lurking photos of attractive people on Instagram or taking photos of every meal for Snapchat or retweeting every celebrity you follow on Twitter. There are other factors to consider as well such as the prominence of and usage patterns of social media within the wider social heiarchy of the peer group. If the in-person social network doesn't have much online cohesion then it stands to reason that use of social media should have a smaller psychological impact compared to peer groups where social media participation is a significant part of socialization.