Organizing through stories: the role of emotions in increasing support for decarceration

Figures & data

Table 1. Completed canvass conversation demographics

Variable	Full sample
	n	%
Gender
Man	1321	51.95
Woman	1222	48.05
Transgender^a	0	0.00
Racial Identity
White	2091	82.22
Latinx	250	9.83
Asian	126	4.95
Black	76	3.00
Partisan Affiliation
Democrat	1205	47.39
Republican	768	30.20
Independent	479	18.83
Other	91	3.58

N = 2,543.

^aTransgender and additional gender identities were not offered in the voter registration data.

Table 2. Baseline policy opinion regression results

Outcome Variable: Baseline Policy Score	B	SE
Gender^a (Women)
Men	−0.326***	0.095
Race^a (White)
Black	1.031***	0.288
Asian	−0.138	0.224
Latinx	−0.520**	0.164
Political Party^a (Democrat)
Republican	−2.435***	0.113
Independent	−0.858***	0.130
Other	−1.143***	0.258
Age	−0.007*	0.003
Length into Campaign (Number of Canvass Days)	−0.032**	0.011
Canvasser Experience (Number of Times Canvased)	−0.001	0.012
Constant	8.399***	0.218
Observations	2543
R^2	0.191
Adjusted R^2	0.188

* p < .05; ** p < .01; ***p < .001

^aReference categories in parentheses.

Table 3. Change in policy opinion regression results

Outcome: Change in Policy Opinion	Model 1			Model 2		Model 3		Model 4		Model 5
	B	SE	B		SE	B	SE	B	SE	B	SE
Story-Sharing	0.545***	0.067	0.183*		0.074	−0.083	0.070	−0.048	0.070	−0.194*	0.095
Emotional Expression			0.859***		0.073	0.859***	0.067	0.841***	0.067	0.657***	0.105
Baseline Opinion						−0.239***	0.011	−0.271***	0.012	−0.272***	0.012
Race^a (White)
Black								0.147	0.176	0.155	0.176
Asian								0.139	0.137	0.127	0.137
Latinx								0.197	0.101	0.193	0.101
Age								0.003	0.002	0.003	0.002
Gender^a (Women)
Men								−0.114	0.059	−0.113	0.002
Party^a (Democrat)
Republican								−0.471***	0.076	−0.472***	0.076
Independent								−0.204*	0.080	−0.211**	0.080
Other								−0.048	0.159	−0.058	0.159
Time-point in Campaign								0.002	0.006	0.002	0.006
Canvasser Experience								0.001	0.007	0.001	0.007
Story x Emotion Interaction										0.309*	0.137
Constant	0.679***	0.052	0.468***		0.055	2.116***	0.093	2.312***	0.176	2.377***	0.178
Observations	2543			2543		2543		2543		2543
R^2	.026			.067		.208		.225		.227
Adjusted R^2	.026			.066		.207		.221		.222

* p < .05; ** p < .01; ***p < .001.

^aReference categories in parentheses.

Figure 1. Interaction effect of story-sharing and emotional expression on attitude change. Emotional expression moderates the effect of story-sharing on attitude change, with the combination of story-telling and emotional expression having a significantly larger effect than story-telling alone.

Table 4. Substantive changes in policy opinion logistic regression results

Outcome: Substantive Changein Policy Opinion	Log-odds Bk	Odds Ratio (OR)	95% CI for OR
Story-Sharing	−0.343	0.710	0.472–1.053
Emotional Expression	0.792***	2.207***	1.500–3.234
Baseline Opinion	−0.313***	0.731***	0.700–0.764
Race^a (White)
Black	−0.005	0.995	0.478–1.936
Asian	−0.295	1.343	0.845–2.088
Latinx	−0.053	0.948	0.672–1.321
Age	−0.002	0.998	0.992–1.005
Gender^a (Women)
Men	−0.244*	0.784*	0.638–0.963
Party^a (Democrat)
Republican	−0.660***	0.517***	0.396–0.672
Independent	−0.506***	0.603***	0.448–0.805
Other	−0.481	0.618	0.346–1.060
Time-point in Campaign	0.001	1.001	0.980–1.023
Canvasser Experience	−0.003	0.997	0.973–1.021
Story x Emotion Interaction	0.645*	1.906*	1.143–3.217
Constant	0.531	1.701	0.906–3.194

N = 2543.

* p < .05; ** p < .01; ***p < .001.

^aReference categories in parentheses.

Data availability statement

The data that support the findings of this study are openly available at the Center for Open Science (OSF) at DOI 10.17605/OSF.IO/7W3PG.