Fig. 3: Forest plot of unstandardized beta coefficients and 95% confidence intervals for the multilevel model effects of all videos on confidence in detecting relevant manipulation for the three respective manipulation tactics.
From: Video inoculation against election misinformation across 12 EU nations
This shows that the long and short discrediting videos significantly increased confidence in detecting manipulative discrediting content compared to the control group. The long, but not the short scapegoating video significantly increased confidence in detecting manipulative scapegoating content compared to the control group. The long and short decontextualization videos significantly reduced confidence in detecting manipulative decontextualization content compared to the control group. Note. N = 16,156.
