PT Mobile Logo

Search form


New Evidence Suggests Media Violence Effects May Be Minimal: Page 2 of 3

New Evidence Suggests Media Violence Effects May Be Minimal: Page 2 of 3

© grass-lifeisgood/shutterstock.com© grass-lifeisgood/shutterstock.com
Correlation of violent video game consumption to youth violenceFigure 1.
Surveys of scholars regarding endorsements of media violence effectsFigure 2.

Another common problem is the use of unstandardized outcome measures. Standardization is a base value of scientific measurement, but findings suggest that standardization was absent in many tests of media effects.8 This can lead to what is sometimes called the garden of forking paths, in which researchers may (even unconsciously, in good faith) select outcomes that best fit their hypotheses and ignore those that don’t.

Another problem commonly observed is citation bias in which researchers (or professional organizations such as the APA and AAP) cite only work that supports their personal views or organizational positions, which can make it seem that the evidence is more consistent than it actually is. Aside from arguably being an ethical issue, citation bias is linked to researcher expectancy effects that produce spurious results.9 Issues of citation bias have been documented as well for policy statements by professional organizations.10,11

What went right?

Science is inevitably self-correcting even if it takes time. Better methods are used, skeptics reexamine old beliefs, and new data comes to light—either reinforcing previous theories or rejecting them. We are entering an era of media effects research in which older theories of direct effects are beginning to crumble.

One innovation, borrowed from medical science, is to use preregistered studies. Preregistering methods and analyses before data collection limits the possibility of questionable researcher practices (taking advantage of unstandardized measures, garden of forking paths, etc) that can lead to spurious outcomes. Several preregistered studies of media effects have been done and, as occurred with social priming, the results have not supported previous theories.12,13

Other studies have examined for the potential of “vulnerable” populations of youth who may be particularly susceptible to media violence effects. Some of these studies have used either better standardized outcome measures and/or open science designs. Generally, no evidence was found for vulnerability to violent media effects, in connection with prior autism spectrum disorders or mental health symptoms such as depression or ADHD.11,14


Loading comments...

By clicking Accept, you agree to become a member of the UBM Medica Community.