sg$spread <- ifelse(is.na(sg$intSpread), sg$intSpreadCtrl, sg$intSpread)
sg$expSymptoms <- ifelse(is.na(sg$intExpSymp), sg$intExpSympCtrl, sg$intExpSymp)
sg$partyLikely <- ifelse(is.na(sg$partyLikelihood), sg$partyLikelihoodCtrl, sg$partyLikelihood)
sg$contLikely <- ifelse(is.na(sg$contLikelihood), sg$contLikelihoodCtrl, sg$contLikelihood)
#contrast code social gains: control = -.5; introspection = .5
sg$cond_cc <- ifelse(sg$Introspection_Condition == 1, .5, -.5)
sg$Introspection_Condition <- ifelse(sg$Introspection_Condition == 1, "introspect", "control")
sg$policy_condition_cc <-
(sg$policy_condition == 1)*(-1/2) +
(sg$policy_condition == 2)*(-3/8) +
(sg$policy_condition == 3)*(-1/4) +
(sg$policy_condition == 4)*(-1/8) +
(sg$policy_condition == 5)*(1/8) +
(sg$policy_condition == 6)*(1/4) +
(sg$policy_condition == 7)*(3/8) +
(sg$policy_condition == 8)*(1/2)
summary(model <- lmer(spread ~ cond_cc + policy_condition_cc + (1|country_names), data = sg))
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: spread ~ cond_cc + policy_condition_cc + (1 | country_names)
## Data: sg
##
## REML criterion at convergence: 21232.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.8936 -0.5265 0.3277 0.8623 1.4316
##
## Random effects:
## Groups Name Variance Std.Dev.
## country_names (Intercept) 0.2336 0.4833
## Residual 3.0825 1.7557
## Number of obs: 5349, groups: country_names, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 5.281e+00 1.989e-01 4.979e+00 26.554 1.48e-06 ***
## cond_cc 1.207e-01 4.803e-02 5.341e+03 2.514 0.012 *
## policy_condition_cc 3.527e-02 6.960e-02 5.341e+03 0.507 0.612
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cnd_cc
## cond_cc 0.000
## plcy_cndtn_ 0.002 -0.009
mean(sg$spread[sg$Introspection_Condition == "introspect"], na.rm = T) #introspection = 5.37
## [1] 5.366667
mean(sg$spread[sg$Introspection_Condition == "control"], na.rm = T) #control = 5.24
## [1] 5.243748
sg %>% ggplot(aes(Introspection_Condition, spread, fill = Introspection_Condition)) +
stat_summary(fun.data = mean_sdl, geom = "bar", position = "dodge") + theme_minimal() + coord_cartesian(ylim = c(4, 6))
## Warning: Removed 49 rows containing non-finite values (stat_summary).
sg %>% ggplot(aes(country_names, spread, fill = Introspection_Condition)) +
stat_summary(fun.data = mean_sdl, geom = "bar", position = "dodge") + theme_minimal() + coord_cartesian(ylim = c(4, 6))
## Warning: Removed 49 rows containing non-finite values (stat_summary).
summary(mExpSymp <- lmer(expSymptoms ~ cond_cc + policy_condition_cc + (1|country), data = sg))
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: expSymptoms ~ cond_cc + policy_condition_cc + (1 | country)
## Data: sg
##
## REML criterion at convergence: 21464.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.6825 -0.5190 0.2776 0.8410 1.3816
##
## Random effects:
## Groups Name Variance Std.Dev.
## country (Intercept) 0.1513 0.389
## Residual 3.1917 1.787
## Number of obs: 5361, groups: country, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 5.184e+00 1.608e-01 4.977e+00 32.236 5.68e-07 ***
## cond_cc 1.632e-01 4.882e-02 5.353e+03 3.343 0.000835 ***
## policy_condition_cc 2.624e-02 7.078e-02 5.353e+03 0.371 0.710879
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cnd_cc
## cond_cc 0.000
## plcy_cndtn_ 0.003 -0.008
mean(sg$expSymptoms[sg$Introspection_Condition == "introspect"], na.rm = T) #intro = 5.30
## [1] 5.297004
mean(sg$expSymptoms[sg$Introspection_Condition == "control"], na.rm = T) #control = 5.14
## [1] 5.137495
sg %>% ggplot(aes(Introspection_Condition, expSymptoms, fill = Introspection_Condition)) +
stat_summary(fun.data = mean_sdl, geom = "bar", position = "dodge") + theme_minimal() + coord_cartesian(ylim = c(4, 6))
## Warning: Removed 37 rows containing non-finite values (stat_summary).
sg %>% ggplot(aes(country_names, expSymptoms, fill = Introspection_Condition)) +
stat_summary(fun.data = mean_sdl, geom = "bar", position = "dodge") + theme_minimal() + coord_cartesian(ylim = c(4, 6))
## Warning: Removed 37 rows containing non-finite values (stat_summary).
summary(lmer(partyLikely ~ cond_cc + policy_condition_cc + (1|country), data = sg))
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: partyLikely ~ cond_cc + policy_condition_cc + (1 | country)
## Data: sg
##
## REML criterion at convergence: 21758.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.1481 -0.7779 -0.4076 0.6427 2.7424
##
## Random effects:
## Groups Name Variance Std.Dev.
## country (Intercept) 0.1021 0.3195
## Residual 3.3577 1.8324
## Number of obs: 5367, groups: country, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 2.57095 0.13297 4.96336 19.335 7.28e-06 ***
## cond_cc -0.23617 0.05004 5359.17342 -4.719 2.43e-06 ***
## policy_condition_cc 0.12071 0.07255 5359.59887 1.664 0.0962 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cnd_cc
## cond_cc 0.000
## plcy_cndtn_ 0.003 -0.009
mean(sg$partyLikely[sg$Introspection_Condition == "introspect"], na.rm = T) #intro = 2.44
## [1] 2.4394
mean(sg$partyLikely[sg$Introspection_Condition == "control"], na.rm = T) #control = 2.68
## [1] 2.678016
sg %>% ggplot(aes(Introspection_Condition, partyLikely, fill = Introspection_Condition)) +
stat_summary(fun.data = mean_sdl, geom = "bar", position = "dodge") + theme_minimal() + coord_cartesian(ylim = c(1, 3))
## Warning: Removed 31 rows containing non-finite values (stat_summary).
sg %>% ggplot(aes(country_names, partyLikely, fill = Introspection_Condition)) +
stat_summary(fun.data = mean_sdl, geom = "bar", position = "dodge") + theme_minimal() + coord_cartesian(ylim = c(1, 3))
## Warning: Removed 31 rows containing non-finite values (stat_summary).
summary(lmer(contLikely ~ cond_cc + policy_condition_cc + (1|country), data = sg))
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: contLikely ~ cond_cc + policy_condition_cc + (1 | country)
## Data: sg
##
## REML criterion at convergence: 21171.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.2398 -0.5099 0.1235 0.7393 1.7064
##
## Random effects:
## Groups Name Variance Std.Dev.
## country (Intercept) 0.1235 0.3514
## Residual 3.0088 1.7346
## Number of obs: 5367, groups: country, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.485e+00 1.455e-01 4.993e+00 30.816 6.86e-07 ***
## cond_cc -3.057e-02 4.737e-02 5.359e+03 -0.645 0.519
## policy_condition_cc 1.499e-03 6.871e-02 5.359e+03 0.022 0.983
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cnd_cc
## cond_cc 0.000
## plcy_cndtn_ 0.003 -0.010
mean(sg$contLikely[sg$Introspection_Condition == "introspect"], na.rm = T) #intro = 4.48
## [1] 4.484234
mean(sg$contLikely[sg$Introspection_Condition == "control"], na.rm = T) #control = 4.52
## [1] 4.523862
sg %>% ggplot(aes(Introspection_Condition, contLikely, fill = Introspection_Condition)) + stat_summary(fun.data = mean_sdl, geom = "bar", position = "dodge") + theme_minimal() + coord_cartesian(ylim = c(4, 5))
## Warning: Removed 31 rows containing non-finite values (stat_summary).
sg %>% ggplot(aes(country_names, contLikely, fill = Introspection_Condition)) + stat_summary(fun.data = mean_sdl, geom = "bar", position = "dodge") + theme_minimal() + coord_cartesian(ylim = c(4, 5))
## Warning: Removed 31 rows containing non-finite values (stat_summary).
summary(lmer(spreadLikelihood ~ cond_cc + policy_condition_cc + (1|country), data = sg))
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: spreadLikelihood ~ cond_cc + policy_condition_cc + (1 | country)
## Data: sg
##
## REML criterion at convergence: 21950.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.3366 -0.5581 0.2503 0.8889 1.3322
##
## Random effects:
## Groups Name Variance Std.Dev.
## country (Intercept) 0.06551 0.256
## Residual 3.48990 1.868
## Number of obs: 5364, groups: country, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.98245 0.10775 4.96966 46.242 9.67e-08 ***
## cond_cc 0.02127 0.05103 5356.25058 0.417 0.677
## policy_condition_cc -0.06833 0.07402 5356.94606 -0.923 0.356
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cnd_cc
## cond_cc 0.000
## plcy_cndtn_ 0.004 -0.009
mean(sg$spreadLikelihood[sg$Introspection_Condition == "introspect"], na.rm = T) #intro = 4.99
## [1] 4.993251
mean(sg$spreadLikelihood[sg$Introspection_Condition == "control"], na.rm = T) #control = 4.98
## [1] 4.977753
sg %>% ggplot(aes(Introspection_Condition, spreadLikelihood, fill = Introspection_Condition)) + stat_summary(fun.data = mean_sdl, geom = "bar", position = "dodge") + theme_minimal() + coord_cartesian(ylim = c(4, 6))
## Warning: Removed 34 rows containing non-finite values (stat_summary).
sg %>% ggplot(aes(country_names, spreadLikelihood, fill = Introspection_Condition)) + stat_summary(fun.data = mean_sdl, geom = "bar", position = "dodge") + theme_minimal() + coord_cartesian(ylim = c(4, 6))
## Warning: Removed 34 rows containing non-finite values (stat_summary).
social norms
e. groupLikelihood ~ condition + policy condition