Explaining Radical Group Behaviour:

Developing Emotion and Efficacy Routes to Normative and Non-normative Collective Action

SUPPLEMENATRY MATERIALS

Nicole Tausch1, Julia C. Becker2, Russell Spears3, OliverChrist2, RimSaab3, Purnima Singh4 and Roomana N Siddiqui5

1University of St Andrews, Scotland

2University of Marburg, Germany

3CardiffUniversity, Wales

4Indian Institute of TechnologyDelhi, India

5Aligarh Muslim University, India

Table S1

Unstandardized Estimates, Standard Errors, 95% Bias-corrected Confidence Intervals, and Standardized Estimates of Direct, Indirect and Total Effects (Study 1)

B (SE) / 95% Bias-corrected Confidence Interval / β
Direct Effects
Anger on Injustice / .82 (.04)*** / .741/.902 / .75
Anger on Efficacy / .07 (.05) / -.028/.164 / .05
Contempt on Injustice / .37 (.05)*** / .277/.469 / .38
Contempt on Efficacy / .05 (.07) / -.084/.185 / .05
Normative Action on Injustice / .20 (.08)* / .051/.360 / .19
Normative Action on Efficacy / .36 (.06)*** / .249/.460 / .29
Normative Action on Anger / .38 (.08)*** / .220/.531 / .40
Normative Action on Contempt / -.04 (.05) / -.136/.057 / -.03
Non-violent Non-Normative Action on Injustice / .12 (.08) / -.022/.278 / .11
Non-violent Non-Normative Action on Efficacy / .12 (.07)+ / -.018/.242 / .10
Non-violentNon-Normative Action on Anger / .23 (.08)** / .060/.379 / .24
Non-violent Non-Normative Action on Contempt / .11 (.07) / -.029/.248 / .10
Violent Non-Normative Action on Injustice / .01 (.01) / -.010/.020 / .05
Violent Non-Normative Action on Efficacy / -.02 (.01)* / -.029/-.004 / -.14
Violent Non-Normative Action on Anger / -.00 (.01) / -.020/.012 / .03
Violent Non-Normative Action on Contempt / .02 (.01)* / .002/.034 / .19
Indirect Effects (via Anger)
Normative Action on Injustice / .32 (.06)*** / .186/.436 / .30
Normative Action on Efficacy / .03 (02) / -.008/.070 / .02
Non-violent Non-Normative Action on Injustice / .29 (.07)** / .050/.314 / .18
Non-violent Non-Normative Action on Efficacy / .02 (.01) / -.003/.050 / .01
Violent Non-Normative Action on Injustice / -.00 (.01) / -.016/.010 / -.03
Violent Non-Normative Action on Efficacy / .00 (.00) / -.003/.001 / -.00

Note. Standard errors and bias-corrected confidence intervals are based on bootstrapping with 5000 re-samples. * p < .05; ** p < .01; *** p < .001.

Table S1

Continued

B (SE) / 95% Bias-corrected Confidence Interval / β
Indirect Effects (via Contempt)
Normative Action on Injustice / -.01 (.02) / -.051/.021 / -.01
Normative Action on Efficacy / -.00 (.01) / -.023/.003 / -.00
Non-violent Non-Normative Action on Injustice / .04 (.03) / -.009/.101 / .04
Non-violent Non-Normative Action on Efficacy / .01 (.01) / -.005/.034 / .01
Violent Non-Normative Action on Injustice / .01 (.00)+ / .001/.014 / .07
Violent Non-Normative Action on Efficacy / .00 (.00) / -.001/.004 / .01
Total effects
Normative Action on Injustice / .50 (.06)*** / .404/.585 / .47
Normative Action on Efficacy / .38 (.06)*** / .274/.492 / .31
Non-violent Non-Normative Action on Injustice / .35 (.05)*** / .249/.447 / .33
Non-violent Non-Normative Action on Efficacy / .14 (.07)* / -.002/.268$ / .11
Violent Non-Normative Action on Injustice / .01 (.01) / -.004/.020 / .10
Violent Non-Normative Action on Efficacy / -.01 (.01)* / -.030/-.003 / -.13

Note. Standard errors and bias-corrected confidence intervals are based on bootstrapping with 5000 re-samples. +p < .10; * p < .05; ** p < .01; *** p < .001.$Note that slight inconsistencies between reported significance levels and confidence intervals are due to confidence intervals being bias-corrected and therefore not symmetrical.

Table S2

Unstandardized Estimates, Standard Errors, 95% Bias-corrected Confidence Intervals, and Standardized Estimates of Direct, Indirect and Total Effects (Study 2)

B (SE) / 95% Confidence Interval / Β
Direct Effects Anger on Injustice Disadvantage / .26 (.09)** / .077/.436 / .24
Anger on Efficacy / .00 (.13) / -.247/.243 / .00
Contempt on Disadvantage / 28 (.09)** / .106/.467 / .25
Contempt on Efficacy / .11 (.13) / -.169/.356 / .07
Policy Support on Disadvantage / .13 (.06)* / -.001/.249$ / .16
Policy Support on Efficacy / .08 (.10) / -.117/.274 / .07
Policy Support on Anger / .14 (.08) / -.030/.299 / .18
Policy Support on Contempt / -.16 (.08)* / -.323/-.012 / -.22
Violence Support on Disadvantage / .17 (.08)* / .018/.332 / .17
Violence Supporton Efficacy / -.16 (.12) / -.379/.084 / -.11
Violence Supporton Anger / -.09 (.11) / -.302/.121 / -.10
Violence Supporton Contempt / .26 (.10)* / .058/.458 / .29
Indirect Effects (via Anger)
Policy Support on Disadvantage / .04 (.03) / -.001/.108 / .04
Policy Support on Efficacy / .00 (.02) / -.042/.045 / .00
Violence Supporton Disadvantage / -.02 (.03) / -.102/.025 / -.02
Violence Supporton Efficacy / .00 (.02) / -.038/.036 / .00
Indirect Effects (via Contempt)
Policy Support on Disadvantage / -.05 (.03) / -.121/-.005$ / -.06
Policy Support on Efficacy / -.02 (.03) / -.090/.017 / -.02
Violence Supporton Disadvantage / .07 (.04)+ / .015/.176$ / .07
Violence Supporton Efficacy / .03 (.04) / -.031/.131 / .02
Total effects
Policy Support on Disadvantage / .12 (.06)+ / -.010/.241 / .15
Policy Support on Efficacy / .07 (.10) / -.139/.259 / .06
Violence Supporton Disadvantage / .22 (.08)** / .082/.379 / .22
Violence Supporton Efficacy / -.13 (.13) / -.365/.125 / -.09

Note. Standard errors and bias-corrected confidence intervals are based on bootstrapping with 5000 re-samples. +p < .10; * p < .05; ** p < .01; *** p < .001.$Note that slight inconsistencies between reported significance levels and confidence intervals are due to confidence intervals being bias-corrected and therefore not symmetrical.

Table S3

Unstandardized Estimates, Standard Errors, 95% Bias-corrected Confidence Intervals, and Standardized Estimates of Direct, Indirect and Total Effects (Study 3)

B (SE) / 95% Confidence Interval / β
Direct Effects Anger on Injustice / .43 (.05)*** / .332/.528 / .46
Anger on Efficacy / -.20 (.06)** / -.310/-.092 / -.16
Contempt on Injustice / .32 (.05)*** / .213/.422 / .30
Contempt on Efficacy / -.12 (.07) / -.265/.025 / -.08
Normative collective action on Injustice / .18 (.06)** / .051/.300 / .14
Normative collective action on Efficacy / .16 (.08)* / -.008/.013$ / .09
Normative collective action on Anger / .46 (.08)*** / .305/.609 / .35
Normative collective action on Contempt / -.01 (.06) / -.137/.118 / -.01
Voting intention on Injustice / .07 (.07) / -.071/.209 / .06
Voting intention on Efficacy / .25 (.08)** / .081/.406 / .14
Voting intention on Anger / .12 (.09) / -.047/.285 / .09
Voting intention on Contempt / -.13 (.07)+ / -.270/.009 / -.11
Violence support (military) on Injustice / .02 (.02) / -.025/.085 / .04
Violence support (military) on Efficacy / -.06 (.03)* / -.112/-.013 / -.11
Violence support (military) on Anger / .12 (.03)*** / .057/.173 / .27
Violence support (military) on Contempt / .06 (.02)* / .009/.100 / .14
Violence support (civilians) on Injustice / -.01 (.01) / -.026/.007 / -.07
Violence support (civilians) on Efficacy / -.01 (.01) / -.025/.014 / -.03
Violence support (civilians) on Anger / .01 (.01) / -.007/.033 / .09
Violence support (civilians) on Contempt / .02 (.01)* / .001/.034 / .13
Indirect Effects (via Anger)
Normative collective action on Injustice / .20 (.04)*** / .127/.286 / .16
Normative collective action on Efficacy / -.09 (.03)** / -.165/-.040 / -.05
Voting intention on Injustice / .05 (.04) / -.019/.130 / .04
Voting intention on Efficacy / -.02 (.02) / -.070/.007 / -.01

Note. Standard errors and bias-corrected confidence intervals are based on bootstrapping with 5000 re-samples. * p < .05; ** p < .01; *** p < .001.

Table S4

Continued

B (SE) / 95% Confidence Interval / β
Indirect Effects (via Anger)
Violence support (military) on Injustice / .05 (.01)*** / .025/.080 / .12
Violence support (military) on Efficacy / -.02 (.01)** / -.046/-.009 / -.04
Violence support (civilians) on Injustice / .01 (.01) / -.003/.015 / .04
Violence support (civilians) on Efficacy / -.00 (.00) / -.008/.001 / -.01
Indirect Effects (via Contempt)
Normative collective action on Injustice / -.00 (.02) / -.046/.037 / -.00
Normative collective action on Efficacy / .00 (.01) / -.015/.024 / .00
Voting intention on Injustice / -.04 (.02)+ / -.097/.000 / -.03
Voting intention on Efficacy / .02 (.02) / -.002/.053 / .01
Violence support (military) on Injustice / .02 (.01)* / .003/.036 / .04
Violence support (military) on Efficacy / -.01 (.01) / -.021/.000 / -.01
Violence support (civilians) on Injustice / .01 (.00)+ / .001/.012 / .04
Violence support (civilians) on Efficacy / -.00 (.00) / -.007/.000 / -.01
Total Effects
Normative collective action on Injustice / .37 (.07)*** / .242/.502 / .30
Normative collective action on Efficacy / .07 (.08) / -.104/.221 / .04
Voting intention on Injustice / .08 (.06) / -.041/.204 / .07
Voting intention on Efficacy / .24 (.08)** / .070/.396 / .14
Violence support (military) on Injustice / .08 (.02)*** / .044/.126 / .21
Violence support (military) on Efficacy / -.09 (.03)** / -.146/-.043 / -.16
Violence support (civilians) on Injustice / .00 (.01) / -.013/.016 / .01
Violence support (civilians) on Efficacy / -.01 (.01) / -.029/.010 / -.05

Note. Standard errors and bias-corrected confidence intervals are based on bootstrapping with 5000 re-samples. +p < .10; * p < .05; ** p < .01; *** p < .001.$Note that slight inconsistencies between reported significance levels and confidence intervals are due to confidence intervals being bias-corrected and therefore not symmetrical.