Framing blame: toward a better understanding of the effects of populist communication on populist party preferences

Figures & data

Table 1. Mean scores of voting intentions for no blame attribution versus blame attribution.

	No blame attribution	Blame attribution
PVV (populist – right)	26.66a (34.88)	33.19b (37.11)
SP (opposition – left)	40.10a (34.51)	36.57a (32.35)
CDA (opposition – right)	33.30a (29.05)	29.51a (29.58)
VVD (government – right)	37.50a (34.06)	31.78b (32.85)
PvdA (government – left)	29.87a (28.94)	28.87a (28.84)

Notes: Propensities to vote for political parties were measured on a scale ranging from 0 to 100. Standard deviations are reported between brackets. Means with differing subscripts within rows differ significantly at the p < .05 level based on independent samples t-tests.

Table 2. Regression model predicting the effects of blame attribution and political distrust/cynicism on propensity to vote for a populist political party (PVV).

Variables	Model I	Model II
(Constant)	−4.84 (3.74)	−4.88 (3.71)
Blame	3.95 (2.93)^†	3.95 (2.98)^†
Cynicism	10.68 (1.27)***	10.46 (2.05)***
Blame × cynicism		0.35 (2.62)
Adjusted R^2	0.121	0.122
F for change in R^2		0.02

^†p < .10; *p < .05; **p < .01; ***p < .001.

Notes: Two-tailed tests. Unstandardized regression weights. Standard errors reported between brackets.

Figure 1. Interaction effect of blame attribution and political distrust/cynicism on propensity to vote for the governmental party VVD.

Table 3. Regression model predicting the effects of blame attribution and political distrust/cynicism on propensity to vote for the governmental party VVD.

Variables	Model I	Model II
(Constant)	−2.05 (3.50)	−0.60 (3.54)
Blame	−5.91 (3.13)*	−6.69 (3.15)*
Cynicism	−8.24 (1.22)***	−4.47 (2.00)*
Blame × cynicism		−6.19 (2.51)*
Adjusted R^2	0.076	0.086
F for change in R^2		6.06*

*p < .05; **p < .01; ***p < .001.

Notes: Two-tailed tests. Unstandardized regression weights. Standard errors reported between brackets.

Figure 2. Interaction effect of blame attribution and Dutch identity attachment on propensity to vote for the populist party PVV.

Figure 3. Interaction effect of blame attribution and European identity attachment on propensity to vote for the populist party PVV.

Table 4. Regression model predicting the effects of blame attribution and identity attachment on propensity to vote for a populist political party (PVV).

Variables	Model I	Model II	Model III	Model IV
(Constant)	−3.73 (2.22)^†	−3.56 (2.22)^†	−3.35 (2.23)	−3.34 (2.22)
Blame	4.86 (2.94)^†	4.59 (2.94)^†	4.48 (2.93)^†	5.10 (2.95)*
Dutch identity	4.06 (1.19)**	1.41 (1.83)	1.41 (1.83)	1.40 (1.82)
European identity	−4.20 (1.13)***	−4.26 (1.12)***	−4.26 (1.13)***	−4.20 (1.12)***
Political cynicism	5.99 (1.57)***	6.01 (1.56)***	6.16 (2.19)**	6.11 (2.18)**
Gov. approval	−1.21 (1.21)	−1.27 (1.21)	−1.27 (1.21)	−1.23 (1.21)
EU approval	−3.49 (1.33)***	−2.40 (1.33)***	−3.40 (1.33)*	−3.59 (1.33)**
Blame × Dutch identity		4.86 (2.47)*	3.81 (2.53)^†	3.89 (2.54)^†
Blame × European identity		−2.67 (2.08)	−2.62 (2.01)	−2.67 (2.06)
Blame × cynicism			−0.12 (2.59)	−0.14 (2.58)
Blame × European identity × Dutch identity				−1.86 (0.82)*
Adjusted R^2	0.180	0.185	0.183	0.190
F for change in R^2		3.62^†	0.02	5.13*

^†p < .10; *p < .05; **p < .01; ***p < .001.

Notes: Two-tailed tests. Unstandardized regression weights. Standard errors reported between brackets.

Figure 4. Three-way interaction effect of blame attribution and exclusionist identity attachment on propensity to vote for the populist party PVV.

Figure 5. Mediation model demonstrating the mediating role of blame perceptions on the effects of blame attributions on populist party preferences. **p < .01; ***p < .001.