Developing media effects as a research programme: Reconsidering the meta-theoretical, conceptual, and methodological foundations of a paradigm in crisis

To what extent do media messages shape people’s opinions and behaviors? Despite decades of scientific research we still have no clear answer: we know that studies only report very small effects but the meaning of those findings is heavily debated. Some scholars argue that the lack of convincing results means that the effects model fails to account for reality and should be abandoned. Others have arrived at the opposite conclusion. To them, the lack of strong findings indicates that media effects are very complicated phenomena and require even more effort to be understood. In the current project I develop a philosophy for media effects studies that takes the middle ground between these arguments: I argue that effects research may still be valuable, but only if we reconsider the way it is typically approached. First, I reject the common assumption that media effects depend on a wide (and unidentified) range of moderators and mediators because it renders media effects hypotheses unfalsifiable. Second, I argue that, in order to make progress, we need to develop new and falsification-oriented assumptions. In particular, I put forward the notion of correspondence rules as a necessary addition to effects hypotheses. Such correspondence rules indicate under what conditions concepts from a hypothesis can(not) be translated to measures and, thus, they define falsification criteria allowing for a more systematic evaluation of the effects proposition.