Sometimes stories about games make their way into the media. Around the year 2000 they were usually about how games turns mild-mannered suburban kids into desensitized high school-shooters-in-training. But things have changed. Warnings about aggressive emotions, caricatured gender images, and detrimental effects of time spent in front of a screen now compete with claims about gamification as a magic key to business success and utopist visions of a better game-based tomorrow for education, citizenship, and science participation. The claims are many – but they all seem to agree on one thing: Games affect us. And we all want to prove our claims. But how do we test the impact of games? In this chapter, we discuss empirical logics and approaches known from large- to small-scale effect studies as traditionally found in educational, political, and biological sciences. There are two premises in this approach: We assume that causal effects of games can be specified and measured (often by proxy) in a statistically valid way, and that findings from studies allow us (at least with a few caveats) to generalize cause and effect to other players at other times. The balance between control and real-world relevance varies very much across methods, and is something we will return to repeatedly. This chapter provides an overview of methods, as well as a discussion of the dilemmas and limitations inherent to measuring anything in the lives of diverse groups like students or gamers. The chapter finishes by discussing the inherent problems in causal and probabilistic claims in media psychology, and argues that it is necessary to keep in mind that humans are interpretative beings. But first, a few words about evidence and (yes, unfortunately) math.
Game Research Methods: An Overview, 2015, p. 175-192
virtual learning; games; evidence based; quantitative methods; statistics; aggression; game based learning