Project Summary

Study Goals

We predict that moral diversity impacts how people evaluate norm violators. Specifically, morally diverse environments lead people to see norm violators has having higher status.

Hypotheses

Moral diversity (vs. homogeneity) leads norm violators to be ascribed more status.
Moral diversity (vs. homogeneity) is seen as culturally looser.
Moral diversity (vs. homogeneity) –> + –> Cultural Looseness –> + –> Ascribed Status.

Study Summary

Study	Context	Conditions	Goal
1.	Facebook - Affirmative Action	High (vs. Low) moral diversity.	Show initial support for hypotheses
2a.	Facebook - Affirmative Action	High (vs. Relative Vs. Low) moral diversity.	Replicate Study 1 w/ “relative” diversity
2b.	Facebook - Affirmative Action	High (vs. Relative Vs. Low) moral diversity.	Replicate Study 2 w/ larger sample
3	Facebook - Affirmative Action	High (vs. Relative Vs. Low) moral diversity. Tight vs. Loose	Manipulate mechanism

Measures

Each measure was presented on a scale from 1 (Strongly disagree) to 7 (Strongly agree).

Status Ascriptions

I think Fred has a great deal of prestige in this Facebook group.
I think Fred possesses high status on this Facebook group.
I think Fred occupies a respected position in this Facebook group.
I think Fred has a position of prestige on this Facebook group.

Cultural Tightness

There are many social norms that people should abide by in this facebook group.
In this facebook group, there are very clear expectations for how people should act in most situations.
In this facebook group, people agree upon what behaviors are appropriate versus inappropriate in most situations.
People in this facebook group have a great deal of freedom in deciding how they want to behave in most situations.
In this facebook group, if someone acts in an inappropriate way, others will strongly disapprove.
People in this facebook group almost always comply with social norms.

Manipulations

Moral Diversity

Study #	Low Moral Diversity	Relative Moral Diversity	High Moral Diversity
1.	You realize that your {Moral Values} are very similar in priority and belief to nearly everyone around you. In other words, the majority of people in this group share your particular moral values.	NA	You realize that your {Moral Values} are very different in priority and belief to nearly everyone around you. In other words, few people in this group share your particular moral values.
2a.	SAME AS ABOVE	You realize that your {Moral Values} are different in priority and belief to some people around you. In other words, some (but not all) people in this group share your particular moral values	SAME AS ABOVE
2a.	SAME AS ABOVE	You realize that your {Moral Values} are relatively different in priority and belief to some people around you. In other words, some (but not all) people in this group share your particular moral values	SAME AS ABOVE
3.	SAME AS ABOVE	SAME AS ABOVE	SAME AS ABOVE

Cultural Looseness

Study #	Tight Norms	Loose Norms
3	It is very important you adhere to these rules. If someone in the group doesn’t adhere to these rules, the admins will do the following: - Block violators from posting - Ban violators from the group - When violators return from their ban, the admins will severely limit (and monitor) their access to viewing, commenting, and liking posts -Report violators to Facebook	In general, adhering to these rules is recommended, but not required. If someone in the group doesn’t adhere to these rules, the admins may do the following (although they rarely see the need to): -Ask violators to post less -Ban violators from the group for a short amount of time -When violators return from the ban, the admins will grant them unmonitored access to viewing, liking, and commenting on posts -Acknowledge violator as having deviated, but will not report violator to higher authorities

In Paper

Study 1

Hypotheses

H1 is supported

## 
##  Two Sample t-test
## 
## data:  dj_stat by cond_num
## t = 4.5224, df = 312, p-value = 8.697e-06
## alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
## 95 percent confidence interval:
##  0.4454604 1.1316097
## sample estimates:
## mean in group 0 mean in group 1 
##        3.815287        3.026752

## 
## Cohen's d
## 
## d estimate: 0.5104269 (medium)
## 95 percent confidence interval:
##     lower     upper 
## 0.2847639 0.7360899

Other Measures

Main Effects

Moderation

Interaction Graphs

When there is high moral diversity, there is not a relationship between activism and status ascriptions. However, when there is low moral diversity, there is a positive relationship between activism and status ascriptions.

## SIMPLE SLOPES ANALYSIS 
## 
## Slope of al_act when cond = High Moral Diversity: 
## 
##   Est.   S.E.   t val.      p
## ------ ------ -------- ------
##   0.16   0.15     1.04   0.30
## 
## Slope of al_act when cond = Low Moral Diversity: 
## 
##   Est.   S.E.   t val.      p
## ------ ------ -------- ------
##   0.83   0.13     6.16   0.00

When there is high and low moral diversity, the relationship between perceived risk and moral diversity is negative. However, this is stronger in the low moral diversity condition compared to the high moral diversity condition.

## SIMPLE SLOPES ANALYSIS 
## 
## Slope of prc_risk when cond = High Moral Diversity: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.38   0.10    -3.83   0.00
## 
## Slope of prc_risk when cond = Low Moral Diversity: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.62   0.10    -6.52   0.00

There was a negative relationship between status ascriptions and kicking in both conditions. However, this effect was weaker in the “High” moral diversity condition than the “Low” moral diversity condition.

## SIMPLE SLOPES ANALYSIS 
## 
## Slope of kick_fred when cond = High Moral Diversity: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.23   0.07    -3.41   0.00
## 
## Slope of kick_fred when cond = Low Moral Diversity: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.39   0.07    -5.83   0.00

In the high moral diversity condition, there was not a difference between how much status people ascribed, regardless of whether they would (or would not) report Fred. However, in the low moral diversity condition, those who reported Fred ascribed significantly less status than those who did not report Fred.

## SIMPLE SLOPES ANALYSIS 
## 
## Slope of report when cond = High Moral Diversity: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.44   0.24    -1.78   0.08
## 
## Slope of report when cond = Low Moral Diversity: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -1.06   0.25    -4.26   0.00

Study 2a

Hypotheses

##              Df Sum Sq Mean Sq F value  Pr(>F)    
## cond          2   35.8  17.914   9.852 8.3e-05 ***
## Residuals   200  363.6   1.818                    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

##   Tukey multiple comparisons of means
##     95% family-wise confidence level
## 
## Fit: aov(formula = dj_stat ~ cond, data = status_sh_s2_clean)
## 
## $cond
##                                                       diff        lwr
## Low Moral Diversity-High Moral Diversity        -1.0134683 -1.5622800
## Relatively Morally Diverse-High Moral Diversity -0.6996324 -1.2541536
## Relatively Morally Diverse-Low Moral Diversity   0.3138360 -0.2264198
##                                                        upr     p adj
## Low Moral Diversity-High Moral Diversity        -0.4646566 0.0000615
## Relatively Morally Diverse-High Moral Diversity -0.1451111 0.0090725
## Relatively Morally Diverse-Low Moral Diversity   0.8540918 0.3577198

Other Measures

Main Effects

Moderation

Study 2b

Hypotheses

Hypothesis 1

##              Df Sum Sq Mean Sq F value   Pr(>F)    
## cond          2   43.7   21.86   10.82 2.35e-05 ***
## Residuals   696 1405.6    2.02                     
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

##   Tukey multiple comparisons of means
##     95% family-wise confidence level
## 
## Fit: aov(formula = dj_stat ~ cond, data = status_sh_s2b_clean)
## 
## $cond
##                                                             diff        lwr
## 2. Relatively Morally Diverse-1. High Moral Diversity -0.4283585 -0.7327449
## 3. Low Moral Diversity-1. High Moral Diversity        -0.5927591 -0.9044344
## 3. Low Moral Diversity-2. Relatively Morally Diverse  -0.1644006 -0.4770015
##                                                              upr     p adj
## 2. Relatively Morally Diverse-1. High Moral Diversity -0.1239721 0.0028631
## 3. Low Moral Diversity-1. High Moral Diversity        -0.2810839 0.0000275
## 3. Low Moral Diversity-2. Relatively Morally Diverse   0.1482003 0.4327737

Hypothesis 2

##              Df Sum Sq Mean Sq F value   Pr(>F)    
## cond          2   59.6  29.784    31.5 8.04e-14 ***
## Residuals   696  658.2   0.946                     
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

##   Tukey multiple comparisons of means
##     95% family-wise confidence level
## 
## Fit: aov(formula = tightness ~ cond, data = status_sh_s2b_clean)
## 
## $cond
##                                                            diff         lwr
## 2. Relatively Morally Diverse-1. High Moral Diversity 0.1557742 -0.05251472
## 3. Low Moral Diversity-1. High Moral Diversity        0.6919276  0.47865103
## 3. Low Moral Diversity-2. Relatively Morally Diverse  0.5361534  0.32224343
##                                                             upr     p adj
## 2. Relatively Morally Diverse-1. High Moral Diversity 0.3640631 0.1851853
## 3. Low Moral Diversity-1. High Moral Diversity        0.9052042 0.0000000
## 3. Low Moral Diversity-2. Relatively Morally Diverse  0.7500635 0.0000000

Hypothesis 3

## 
## Mediation/Moderation Analysis 
## Call: psych::mediate(y = dj_stat ~ cond_num + (tightness), data = status_sh_s2b_clean)
## 
## The DV (Y) was  dj_stat . The IV (X) was  cond_num . The mediating variable(s) =  tightness .
## 
## Total effect(c) of  cond_num  on  dj_stat  =  0.3   S.E. =  0.07  t  =  4.5  df=  697   with p =  7.8e-06
## Direct effect (c') of  cond_num  on  dj_stat  removing  tightness  =  0.35   S.E. =  0.07  t  =  5.14  df=  696   with p =  3.6e-07
## Indirect effect (ab) of  cond_num  on  dj_stat  through  tightness   =  -0.05 
## Mean bootstrapped indirect effect =  -0.05  with standard error =  0.02  Lower CI =  -0.1    Upper CI =  -0.01
## R = 0.2 R2 = 0.04   F = 14.35 on 2 and 696 DF   p-value:  4.43e-09 
## 
##  To see the longer output, specify short = FALSE in the print statement or ask for the summary

Looseness - w/ controls

## 
## Mediation/Moderation Analysis 
## Call: psych::mediate(y = dj_stat ~ cond_num + (tightness), data = status_sh_s2b_clean, 
##     z = c("issue", "age"))
## 
## The DV (Y) was  dj_stat . The IV (X) was  cond_num . The mediating variable(s) =  tightness .
## 
## Total effect(c) of  cond_num  on  dj_stat  =  0.3   S.E. =  0.07  t  =  4.5  df=  697   with p =  7.8e-06
## Direct effect (c') of  cond_num  on  dj_stat  removing  tightness  =  0.35   S.E. =  0.07  t  =  5.14  df=  696   with p =  3.6e-07
## Indirect effect (ab) of  cond_num  on  dj_stat  through  tightness   =  -0.05 
## Mean bootstrapped indirect effect =  -0.05  with standard error =  0.02  Lower CI =  -0.1    Upper CI =  -0.01
## R = 0.2 R2 = 0.04   F = 14.35 on 2 and 696 DF   p-value:  4.43e-09 
## 
##  To see the longer output, specify short = FALSE in the print statement or ask for the summary

Study 3

Hypotheses

Hypothesis 1

##              Df Sum Sq Mean Sq F value  Pr(>F)   
## cond_moral    2  14.14   7.072   4.898 0.00844 **
## Residuals   189 272.91   1.444                   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

##   Tukey multiple comparisons of means
##     95% family-wise confidence level
## 
## Fit: aov(formula = dj_stat ~ cond_moral, data = status_sh_s3_clean)
## 
## $cond_moral
##                                                      diff        lwr
## 2. Relatively Diverse-1. High Moral Diversity  -0.4454545 -0.9587413
## 3. Low Moral Diversity-1. High Moral Diversity -0.6338603 -1.1226109
## 3. Low Moral Diversity-2. Relatively Diverse   -0.1884058 -0.6964952
##                                                        upr     p adj
## 2. Relatively Diverse-1. High Moral Diversity   0.06783218 0.1032552
## 3. Low Moral Diversity-1. High Moral Diversity -0.14510981 0.0070427
## 3. Low Moral Diversity-2. Relatively Diverse    0.31968365 0.6561065

Hypothesis 2

##              Df Sum Sq Mean Sq F value Pr(>F)  
## cond_moral    2   5.21  2.6053   4.474 0.0126 *
## Residuals   189 110.05  0.5823                 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

##   Tukey multiple comparisons of means
##     95% family-wise confidence level
## 
## Fit: aov(formula = tightness ~ cond_moral, data = status_sh_s3_clean)
## 
## $cond_moral
##                                                     diff         lwr       upr
## 2. Relatively Diverse-1. High Moral Diversity  0.2074694 -0.11847494 0.5334138
## 3. Low Moral Diversity-1. High Moral Diversity 0.3929513  0.08258771 0.7033148
## 3. Low Moral Diversity-2. Relatively Diverse   0.1854818 -0.13716220 0.5081258
##                                                    p adj
## 2. Relatively Diverse-1. High Moral Diversity  0.2915016
## 3. Low Moral Diversity-1. High Moral Diversity 0.0088092
## 3. Low Moral Diversity-2. Relatively Diverse   0.3651073

Interaction

Status Ascriptions

## 
## Call:
## lm(formula = dj_stat ~ cond_moral_num * cond_norms_num, data = status_sh_s3_clean)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2.0074 -0.9556 -0.2225  0.7812  3.3636 
## 
## Coefficients:
##                               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                    2.63635    0.12876  20.475   <2e-16 ***
## cond_moral_num                -0.11921    0.09984  -1.194    0.234    
## cond_norms_num                 0.10944    0.18134   0.604    0.547    
## cond_moral_num:cond_norms_num -0.14242    0.13625  -1.045    0.297    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.206 on 188 degrees of freedom
## Multiple R-squared:  0.04796,    Adjusted R-squared:  0.03277 
## F-statistic: 3.157 on 3 and 188 DF,  p-value: 0.02598

## 
## Call:
## lm(formula = dj_stat ~ cond_moral_num * tightness, data = status_sh_s3_clean)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2.1987 -0.8977 -0.2367  0.7756  3.3240 
## 
## Coefficients:
##                          Estimate Std. Error t value Pr(>|t|)    
## (Intercept)               4.51355    0.55460   8.138 5.36e-14 ***
## cond_moral_num           -0.64714    0.48268  -1.341 0.181623    
## tightness                -0.37568    0.11161  -3.366 0.000924 ***
## cond_moral_num:tightness  0.09941    0.09506   1.046 0.297003    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.17 on 188 degrees of freedom
## Multiple R-squared:  0.1032, Adjusted R-squared:  0.08893 
## F-statistic: 7.215 on 3 and 188 DF,  p-value: 0.0001312

## SIMPLE SLOPES ANALYSIS 
## 
## Slope of tightness when cond_moral = 3. Low Moral Diversity: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.23   0.24    -0.97   0.33
## 
## Slope of tightness when cond_moral = 1. High Moral Diversity: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.49   0.16    -3.08   0.00
## 
## Slope of tightness when cond_moral = 2. Relatively Diverse: 
## 
##    Est.   S.E.   t val.      p
## ------- ------ -------- ------
##   -0.29   0.21    -1.40   0.16

Mediation

## 
## Mediation/Moderation Analysis 
## Call: psych::mediate(y = dj_stat ~ cond_moral_num + (tightness), data = status_sh_s3_clean, 
##     z = "cond_norms_num")
## 
## The DV (Y) was  dj_stat . The IV (X) was  cond_moral_num . The mediating variable(s) =  tightness .
## 
## Total effect(c) of  cond_moral_num  on  dj_stat  =  -0.2   S.E. =  0.07  t  =  -2.88  df=  190   with p =  0.0044
## Direct effect (c') of  cond_moral_num  on  dj_stat  removing  tightness  =  -0.15   S.E. =  0.07  t  =  -2.18  df=  189   with p =  0.03
## Indirect effect (ab) of  cond_moral_num  on  dj_stat  through  tightness   =  -0.05 
## Mean bootstrapped indirect effect =  -0.05  with standard error =  0.02  Lower CI =  -0.09    Upper CI =  -0.01
## R = 0.31 R2 = 0.1   F = 10.27 on 2 and 189 DF   p-value:  2.7e-06 
## 
##  To see the longer output, specify short = FALSE in the print statement or ask for the summary

## 
## Mediation/Moderation Analysis 
## Call: psych::mediate(y = dj_stat ~ cond_norms_num + (tightness), data = status_sh_s3_clean, 
##     z = "cond_moral_num")
## 
## The DV (Y) was  dj_stat . The IV (X) was  cond_norms_num . The mediating variable(s) =  tightness .
## 
## Total effect(c) of  cond_norms_num  on  dj_stat  =  0.05   S.E. =  0.18  t  =  0.28  df=  190   with p =  0.78
## Direct effect (c') of  cond_norms_num  on  dj_stat  removing  tightness  =  -0.3   S.E. =  0.19  t  =  -1.59  df=  189   with p =  0.11
## Indirect effect (ab) of  cond_norms_num  on  dj_stat  through  tightness   =  0.35 
## Mean bootstrapped indirect effect =  0.35  with standard error =  0.1  Lower CI =  0.17    Upper CI =  0.58
## R = 0.3 R2 = 0.09   F = 9.05 on 2 and 189 DF   p-value:  1.25e-05 
## 
##  To see the longer output, specify short = FALSE in the print statement or ask for the summary

Moderated Mediation

## 
## Mediation/Moderation Analysis 
## Call: psych::mediate(y = dj_stat ~ cond_norms_num * cond_moral_num + 
##     (tightness), data = status_sh_s3_clean)
## 
## The DV (Y) was  dj_stat . The IV (X) was  cond_norms_num cond_moral_num cond_norms_num*cond_moral_num . The mediating variable(s) =  tightness .
## 
## Total effect(c) of  cond_norms_num  on  dj_stat  =  0.06   S.E. =  0.17  t  =  0.32  df=  188   with p =  0.75
## Direct effect (c') of  cond_norms_num  on  dj_stat  removing  tightness  =  -0.25   S.E. =  0.19  t  =  -1.31  df=  187   with p =  0.19
## Indirect effect (ab) of  cond_norms_num  on  dj_stat  through  tightness   =  0.31 
## Mean bootstrapped indirect effect =  0.31  with standard error =  0.11  Lower CI =  0.12    Upper CI =  0.53
## 
## Total effect(c) of  cond_moral_num  on  dj_stat  =  -0.19   S.E. =  0.07  t  =  -2.81  df=  188   with p =  0.0055
## Direct effect (c') of  cond_moral_num  on  dj_stat  removing  tightness  =  -0.14   S.E. =  0.07  t  =  -2  df=  187   with p =  0.047
## Indirect effect (ab) of  cond_moral_num  on  dj_stat  through  tightness   =  -0.06 
## Mean bootstrapped indirect effect =  -0.06  with standard error =  0.02  Lower CI =  -0.11    Upper CI =  -0.02
## 
## Total effect(c) of  cond_norms_num*cond_moral_num  on  dj_stat  =  -0.14   S.E. =  0.14  t  =  -1.05  df=  188   with p =  0.3
## Direct effect (c') of  cond_norms_num*cond_moral_num  on  dj_stat  removing  tightness  =  -0.05   S.E. =  0.13  t  =  -0.38  df=  187   with p =  0.71
## Indirect effect (ab) of  cond_norms_num*cond_moral_num  on  dj_stat  through  tightness   =  -0.09 
## Mean bootstrapped indirect effect =  -0.09  with standard error =  0.04  Lower CI =  -0.19    Upper CI =  -0.02
## R = 0.33 R2 = 0.11   F = 5.62 on 4 and 187 DF   p-value:  7.44e-05 
## 
##  To see the longer output, specify short = FALSE in the print statement or ask for the summary

Status and Hate Study Summary

Project Summary

Study Goals

Hypotheses

Study Summary

Measures

Status Ascriptions

Cultural Tightness

Manipulations

Moral Diversity

Cultural Looseness

In Paper

Study 1

Hypotheses

Other Measures

Main Effects

Moderation

Interaction Graphs

Study 2a

Hypotheses

Other Measures

Main Effects

Moderation

Study 2b

Hypotheses

Hypothesis 1

Hypothesis 2

Hypothesis 3

Looseness - w/ controls

Study 3

Hypotheses

Hypothesis 1

Hypothesis 2

Interaction

Status Ascriptions

Mediation

Moderated Mediation