S.I. 3 Effects on losers

Treatment effects on experimental subjects who find the decision outcome to align unfavorably with their own preferences.

3.1 Fairness

In the manuscript, we mention that the Generic good loser message for the respondents with an unfavorable outcome is on par with the outcome favorability effect (difference in fairness perceptions between winners and losers). The table below displays the outcome favorability effect, and table ‘tab:1055_post_fairness’ further below displays the treatment effect of the Generic good loser message.

(#tab:1050_int_outfav)Conditional effect of outcome favorability on Generic good loser message treatment, Study 1
Treatment value Estimate Std. Error t-statistic p value
Intercept 4.79 0.08 62.18 0
Unfavorable outcome -0.93 0.11 -8.28 0

3.1.1 Fairness I

Figure 2 in the manuscript:

(#tab:1055_post_fairness)Treatment effects on fairness perceptions of decision, Study 1
Treatment value Estimate Std. Error t-statistic p value
Not shown 3.46 0.16 22.32 0.00
Lamenting politician 0.27 0.21 1.26 0.21
Generic good loser message 0.88 0.21 4.15 0.00

The sample size of the fitted model is 481.

3.1.2 Fairness II

Lamenting politician as reference category

(#tab:105_post_fairness)Treatment effects on fairness perceptions of decision, Study 1
Treatment value Estimate Std. Error t-statistic p value
Intercept 3.73 0.15 25.41 0.00
Not shown -0.27 0.21 -1.26 0.21
Generic good loser message 0.61 0.21 2.96 0.00
Note:
Lamenting politician set as reference category.

3.2 Justice

(#tab:105_post_justice)Treatment effects on justice perceptions of decision, Study 1
Treatment value Estimate Std. Error t-statistic p value
Not shown 3.61 0.14 25.45 0.00
Lamenting politician 0.13 0.20 0.64 0.52
Generic good loser message 0.53 0.19 2.76 0.01

3.3 Decision evaluation

(#tab:105_post_eval)Treatment effects on decision evaluation, Study 1
Treatment value Estimate Std. Error t-statistic p value
Not shown 3.00 0.15 20.60 0.00
Lamenting politician -0.21 0.20 -1.03 0.30
Generic good loser message -0.09 0.20 -0.44 0.66

3.4 Willingness to accept

(#tab:105_post_accept)Treatment effects on willingness to accept decision, Study 1
Treatment value Estimate Std. Error t-statistic p value
Not shown 4.31 0.17 24.95 0.00
Lamenting politician -0.22 0.24 -0.93 0.35
Generic good loser message 0.13 0.24 0.57 0.57

3.5 Compliance

(#tab:105_post_comply)Treatment effects on decision compliance, Study 1
Treatment value Estimate Std. Error t-statistic p value
Not shown 4.46 0.16 28.58 0.00
Lamenting politician -0.14 0.22 -0.65 0.52
Generic good loser message 0.06 0.21 0.29 0.78

3.6 Moderating effects of issue importance on losers

Figure 5 in the manuscript:

(#tab:1045_importance_treatment_fairness)Moderating effects of issue importance on experimental treatment, Study 1
Factor Issue importance AME SE z-statistic p value Lower Upper
Generic good loser message 1 1.07 0.42 2.52 0.01 0.24 1.90
Generic good loser message 2 1.01 0.32 3.10 0.00 0.37 1.64
Generic good loser message 3 0.95 0.25 3.85 0.00 0.46 1.43
Generic good loser message 4 0.89 0.21 4.17 0.00 0.47 1.30
Generic good loser message 5 0.82 0.24 3.39 0.00 0.35 1.30
Generic good loser message 6 0.76 0.32 2.38 0.02 0.13 1.39
Generic good loser message 7 0.70 0.42 1.67 0.09 -0.12 1.52
Lamenting politician 1 0.38 0.43 0.88 0.38 -0.47 1.23
Lamenting politician 2 0.35 0.33 1.05 0.30 -0.30 0.99
Lamenting politician 3 0.31 0.25 1.24 0.21 -0.18 0.80
Lamenting politician 4 0.28 0.21 1.29 0.20 -0.14 0.69
Lamenting politician 5 0.24 0.25 0.98 0.33 -0.24 0.72
Lamenting politician 6 0.20 0.32 0.63 0.53 -0.43 0.84
Lamenting politician 7 0.17 0.42 0.40 0.69 -0.66 1.00
Note:
Sample: Respondents with unfavorable outcome.

The sample size of the fitted model is 481.

(#tab:1045_exp1_mods)Moderating effects of issue importance, Study 1
Treatment value Estimate Std. Error t-statistic p value
Intercept 3.42 0.25 13.73 0.00
Lamenting politician 0.03 0.34 0.10 0.92
Generic good loser message 0.81 0.34 2.41 0.02
Issue not important to respondent 0.07 0.32 0.21 0.83
Lamenting politician w/ issue not important to respondent 0.40 0.44 0.91 0.37
Generic good loser message w/ issue not important to respondent 0.12 0.43 0.29 0.78
Note:
Sample: Respondents with unfavorable outcome.