Additional Analyses for Löb et al. (2024)

The following analysis complements the findings presented in the paper Löb, C., Rinke, E. M., Weinmann, C. & Wessler, H. (2024). Unpacking the determinants of outrage and recognition in public discourse: Insights across socio-cultural divides, political systems, and media types. The International Journal of Press/Politics, 29(1), 273–294. doi:10.1177/19401612221084206

Average Marginal Effects

The following table shows the corresponding average marginal effects of all independent variables, controlling for all other variables in the model. It makes plain that the largest differences in the expected probability of outrage occur in the lower two categories. The main difference across political systems, degrees of socio-cultural division, and especially media types occurs with respect to whether "no outrage" or "slight outrage" occurs. Average effects for political system, societal division and media type are essentially nil for the higher outrage categories (i.e., "moderate" to “very strong” outrage).

	No outrage	Slight outrage	Moderate outrage	Strong outrage	Very strong outrage
Political system
Mixed (vs Consensus)	-.05 [-.09, -.01]	.03 [.00, .06]	.01 [.00, .01]	.01 [.00, .01]	.00 [-.00, .00]
Majoritarian (vs Consensus)	.05 [.02, .09]	-.04 [-.06, -.01]	-.01 [-.01, -.00]	-.01 [-.01, -.00]	.00 [-.00, .00]
Societal division
Contested secularism (vs Stable secularism)	-.05 [-.08, -.02]	.04 [.02, .06]	.01 [.00, .01]	.01 [.00, .01]	.00 [-.00, .00]
Media type
News website (vs Blog)	.19 [.15, .23]	-.14 [-.17, -.11]	-.03 [-.04, -.02]	-.02 [-.03, -.01]	-.01 [-.01, -.00]
Printed newspaper (vs Blog)	.16 [.12, .20]	-.12 [-.15, -.08]	-.03 [-.04, -.01]	-.02 [-.02, -.01]	-.01 [-.01, -.00]

Note: Cell entries are average marginal effects on predicted probabilities of news items falling into each tone category, with 95% confidence intervals in brackets. Reference categories for categorical independent variables are given in parentheses.