HDI intervales for specific interaction between gender and leadership
# HDI
m1_int_l_hdi <- bayestestR::hdi(m1_int_l, effects = "fixed", component = "conditional", ci = .95)
kable(m1_int_l_hdi[sign(m1_int_l_hdi$CI_low) == sign(m1_int_l_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
5
|
b_Intercept
|
0.95
|
1.45
|
5.98
|
|
2
|
b_dominanceSum
|
0.95
|
0.86
|
4.72
|
|
1
|
b_Age
|
0.95
|
-4.73
|
-0.86
|
|
4
|
b_Gender1
|
0.95
|
-5.05
|
-1.19
|
# Interaction between gender and prestige
m1_int_p <- brm(generalRiskPreference ~ dominanceSum+ prestigeSum*Gender + leadershipSum + Age, data=experiment_dataset_analysis,
prior=c(prior(normal(0,1), class="Intercept"),
prior(normal(3,1), class="b", coef="dominanceSum"),
prior(normal(0,1), class="b", coef="prestigeSum"),
prior(normal(-2,1), class="b", coef="leadershipSum"),
prior(normal(-3,1), class="b", coef="Gender1"),
prior(normal(-3,1), class="b", coef="Age"),
prior(normal(0,1), class="b", coef="prestigeSum:Gender1")),
save_all_pars = T, iter = 4000)
summary(m1_int_p)
# HDI
m1_int_p_hdi <- bayestestR::hdi(m1_int_p, effects = "fixed", component = "conditional", ci = .95)
kable(m1_int_p_hdi[sign(m1_int_p_hdi$CI_low) == sign(m1_int_p_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')], format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
4
|
b_Intercept
|
0.95
|
1.40
|
5.92
|
|
2
|
b_dominanceSum
|
0.95
|
1.05
|
4.97
|
|
3
|
b_Gender1
|
0.95
|
-4.85
|
-1.05
|
|
1
|
b_Age
|
0.95
|
-4.81
|
-1.04
|
# Domain specific model
m2 <- brm(mvbind(ethicalPreference, financialPreference, socialPreference, healthAndSafetyPreference, recreationalPreference) ~ dominanceSum + prestigeSum + leadershipSum + Gender + Age, data = experiment_dataset_analysis, cores = 6
prior = c(prior(normal(0, 1), coef = "Age", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "Gender1", resp = "ethicalPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "ethicalPreference"),
prior(normal(0, 1), class = "Intercept", resp = "ethicalPreference"),
prior(normal(0, 1), class = "sigma", resp = "ethicalPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "financialPreference"),
prior(normal(0, 1), coef = "Gender1", resp = "financialPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "financialPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "financialPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "financialPreference"),
prior(normal(0, 1), class = "Intercept", resp = "financialPreference"),
prior(normal(0, 1), class = "sigma", resp = "financialPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "Gender1", resp = "healthAndSafetyPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "sigma", resp = "healthAndSafetyPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "Gender1", resp = "recreationalPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "recreationalPreference"),
prior(normal(0, 1), class = "Intercept", resp = "recreationalPreference"),
prior(normal(0, 1), class = "sigma", resp = "recreationalPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "socialPreference"),
prior(normal(0, 1), coef = "Gender1", resp = "socialPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "socialPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "socialPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "socialPreference"),
prior(normal(0, 1), class = "Intercept", resp = "socialPreference"),
prior(normal(0, 1), class = "sigma", resp = "socialPreference")),
save_all_pars = T, iter = 6000, warmup = 500)
summary(m2)
m2_hdi <- bayestestR::hdi(m2, effects = "fixed", component = "conditional", ci = .95)
kable(m2_hdi[sign(m2_hdi$CI_low) == sign(m2_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
4
|
b_ethicalPreference_Intercept
|
0.95
|
2.86
|
4.40
|
|
2
|
b_ethicalPreference_dominanceSum
|
0.95
|
0.58
|
1.48
|
|
10
|
b_financialPreference_Intercept
|
0.95
|
7.47
|
9.70
|
|
8
|
b_financialPreference_dominanceSum
|
0.95
|
0.12
|
1.37
|
|
28
|
b_socialPreference_Intercept
|
0.95
|
8.27
|
11.64
|
|
26
|
b_socialPreference_dominanceSum
|
0.95
|
0.48
|
2.54
|
|
16
|
b_healthAndSafetyPreference_Intercept
|
0.95
|
4.68
|
6.60
|
|
14
|
b_healthAndSafetyPreference_dominanceSum
|
0.95
|
0.52
|
1.63
|
|
22
|
b_recreationalPreference_Intercept
|
0.95
|
0.85
|
2.40
|
|
20
|
b_recreationalPreference_dominanceSum
|
0.95
|
0.76
|
1.66
|
|
21
|
b_recreationalPreference_Gender1
|
0.95
|
-1.85
|
-0.51
|
|
19
|
b_recreationalPreference_Age
|
0.95
|
0.04
|
0.85
|
# Domain specific model with Gender-DOPL interactions
m3 <- brm(mvbind(ethicalPreference, financialPreference, socialPreference, healthAndSafetyPreference, recreationalPreference) ~ dominanceSum*Gender + prestigeSum*Gender + leadershipSum*Gender + Age, data = experiment_dataset_analysis, cores = 6,
prior = c(prior(normal(0, 1), coef = "Age", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "Gender1", resp = "ethicalPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "ethicalPreference"),
prior(normal(0, 1), class = "Intercept", resp = "ethicalPreference"),
prior(normal(0, 1), class = "sigma", resp = "ethicalPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "financialPreference"),
prior(normal(0, 1), coef = "Gender1", resp = "financialPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "financialPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "financialPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "financialPreference"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "financialPreference"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "financialPreference"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "financialPreference"),
prior(normal(0, 1), class = "Intercept", resp = "financialPreference"),
prior(normal(0, 1), class = "sigma", resp = "financialPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "Gender1", resp = "healthAndSafetyPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "sigma", resp = "healthAndSafetyPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "Gender1", resp = "recreationalPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "recreationalPreference"),
prior(normal(0, 1), class = "Intercept", resp = "recreationalPreference"),
prior(normal(0, 1), class = "sigma", resp = "recreationalPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "socialPreference"),
prior(normal(0, 1), coef = "Gender1", resp = "socialPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "socialPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "socialPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "socialPreference"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "socialPreference"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "socialPreference"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "socialPreference"),
prior(normal(0, 1), class = "Intercept", resp = "socialPreference"),
prior(normal(0, 1), class = "sigma", resp = "socialPreference")),
save_all_pars = T, iter = 6000, warmup = 500)
summary(m3)
# HDI
m3_hdi <- bayestestR::hdi(m3, effects = "fixed", component = "conditional", ci = .95)
kable(m3_hdi[sign(m3_hdi$CI_low) == sign(m3_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
7
|
b_ethicalPreference_Intercept
|
0.95
|
2.85
|
4.42
|
|
2
|
b_ethicalPreference_dominanceSum
|
0.95
|
0.61
|
1.71
|
|
16
|
b_financialPreference_Intercept
|
0.95
|
7.50
|
9.67
|
|
11
|
b_financialPreference_dominanceSum
|
0.95
|
0.14
|
1.59
|
|
43
|
b_socialPreference_Intercept
|
0.95
|
8.34
|
11.67
|
|
38
|
b_socialPreference_dominanceSum
|
0.95
|
0.60
|
2.87
|
|
25
|
b_healthAndSafetyPreference_Intercept
|
0.95
|
4.65
|
6.59
|
|
20
|
b_healthAndSafetyPreference_dominanceSum
|
0.95
|
0.41
|
1.77
|
|
34
|
b_recreationalPreference_Intercept
|
0.95
|
0.95
|
2.48
|
|
29
|
b_recreationalPreference_dominanceSum
|
0.95
|
0.66
|
1.74
|
|
31
|
b_recreationalPreference_Gender1
|
0.95
|
-1.83
|
-0.47
|
|
28
|
b_recreationalPreference_Age
|
0.95
|
0.06
|
0.87
|
# Model Comparison: M2 and M3
#co <- loo(m2,m3, moment_match = T)
#co$loos$m2$looic
#co$loos$m2$elpd_loo
#co$loos$m3$looic
#co$loos$m3$elpd_loo
# Model Comparison (m2 and m3)
co <- loo(m2,m3)
looic_ <- c(co$loos$m2$looic, co$loos$m3$looic)
looic_se <- c(co$loos$m2$se_looic, co$loos$m3$se_looic)
looic_elpd <- c(co$loos$m2$elpd_loo, co$loos$m3$elpd_loo)
looic_elpd_se <- c(co$loos$m2$se_elpd_loo, co$loos$m3$se_elpd_loo)
loo_table <- data.frame(looic_, looic_se, looic_elpd,looic_elpd_se, row.names = c("m2", "m3"))
kable(loo_table, format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>%
kable_styling(full_width = F)
|
|
looic_
|
looic_se
|
looic_elpd
|
looic_elpd_se
|
|
m2
|
2407.76
|
48.03
|
-1203.88
|
24.02
|
|
m3
|
2422.23
|
47.55
|
-1211.11
|
23.77
|
apa_table(bayes_R2(m2))
(#tab:unnamed-chunk-15)
**
| R2ethicalPreference |
0.21 |
0.06 |
0.10 |
0.32 |
| R2financialPreference |
0.12 |
0.05 |
0.04 |
0.23 |
| R2socialPreference |
0.16 |
0.06 |
0.05 |
0.27 |
| R2healthAndSafetyPreference |
0.17 |
0.05 |
0.07 |
0.28 |
| R2recreationalPreference |
0.28 |
0.05 |
0.17 |
0.38 |
apa_table(bayes_R2(m3))
(#tab:unnamed-chunk-15)
**
| R2ethicalPreference |
0.21 |
0.05 |
0.11 |
0.32 |
| R2financialPreference |
0.13 |
0.05 |
0.05 |
0.24 |
| R2socialPreference |
0.17 |
0.05 |
0.07 |
0.28 |
| R2healthAndSafetyPreference |
0.17 |
0.05 |
0.07 |
0.28 |
| R2recreationalPreference |
0.29 |
0.05 |
0.18 |
0.39 |
m6 <- brm(mvbind(ethicalPreference, financialPreference, socialPreference, healthAndSafetyPreference, recreationalPreference) ~ dominanceSum*Age + prestigeSum*Age + leadershipSum*Age + Age, data = experiment_dataset_analysis, cores = 6,
prior = c(prior(normal(0, 1), coef = "Age", resp = "ethicalPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "dominanceSum:Age", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "ethicalPreference"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "ethicalPreference"),
prior(normal(0, 1), class = "Intercept", resp = "ethicalPreference"),
prior(normal(0, 1), class = "sigma", resp = "ethicalPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "financialPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "financialPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "financialPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "financialPreference"),
prior(normal(0, 1), coef = "dominanceSum:Age", resp = "financialPreference"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "financialPreference"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "financialPreference"),
prior(normal(0, 1), class = "Intercept", resp = "financialPreference"),
prior(normal(0, 1), class = "sigma", resp = "financialPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "healthAndSafetyPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "dominanceSum:Age", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "sigma", resp = "healthAndSafetyPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "recreationalPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "dominanceSum:Age", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "recreationalPreference"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "recreationalPreference"),
prior(normal(0, 1), class = "Intercept", resp = "recreationalPreference"),
prior(normal(0, 1), class = "sigma", resp = "recreationalPreference"),
#----
prior(normal(0, 1), coef = "Age", resp = "socialPreference"),
prior(normal(2, 1), coef = "dominanceSum", resp = "socialPreference"),
prior(normal(0, 1), coef = "leadershipSum", resp = "socialPreference"),
prior(normal(0, 1), coef = "prestigeSum", resp = "socialPreference"),
prior(normal(0, 1), coef = "dominanceSum:Age", resp = "socialPreference"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "socialPreference"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "socialPreference"),
prior(normal(0, 1), class = "Intercept", resp = "socialPreference"),
prior(normal(0, 1), class = "sigma", resp = "socialPreference")),
save_all_pars = T, iter = 6000, warmup = 500)
# summary(m6)
m6_hdi <- bayestestR::hdi(m6, effects = "fixed", component = "conditional", ci = .95)
kable(m6_hdi[sign(m6_hdi$CI_low) == sign(m6_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
6
|
b_ethicalPreference_Intercept
|
0.95
|
2.93
|
4.27
|
|
4
|
b_ethicalPreference_dominanceSum
|
0.95
|
0.52
|
1.42
|
|
14
|
b_financialPreference_Intercept
|
0.95
|
7.63
|
9.54
|
|
12
|
b_financialPreference_dominanceSum
|
0.95
|
0.09
|
1.27
|
|
38
|
b_socialPreference_Intercept
|
0.95
|
8.23
|
11.31
|
|
36
|
b_socialPreference_dominanceSum
|
0.95
|
0.44
|
2.47
|
|
22
|
b_healthAndSafetyPreference_Intercept
|
0.95
|
4.89
|
6.56
|
|
20
|
b_healthAndSafetyPreference_dominanceSum
|
0.95
|
0.51
|
1.60
|
|
30
|
b_recreationalPreference_Intercept
|
0.95
|
0.26
|
1.65
|
|
28
|
b_recreationalPreference_dominanceSum
|
0.95
|
0.79
|
1.73
|
# HDI
m6_hdi <- hdi(m6, effects = "fixed", component = "conditional", ci = .95)
m6_hdi[sign(m6_hdi$CI_low) == sign(m6_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')]
# Benefit, perception and risk taking across subdomains for DOPL motives
m4 <- brm(mvbind(riskSum, riskPerceptionSum, riskBenefitSum) ~ dominanceSum + prestigeSum + leadershipSum + Age + Gender, data= experiment_dataset_analysis,
prior = c(prior(normal(0, 1), class = "Intercept", resp = "riskSum"),
prior(normal(0, 1), class = "sigma", resp = "riskSum"),
prior(normal(-3, 1), coef = "Age", resp = "riskSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "riskSum"),
prior(normal(3, 1), coef = "dominanceSum", resp = "riskSum"),
prior(normal(-2, 1), coef = "leadershipSum", resp = "riskSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "riskSum"),
#----
prior(normal(0, 1), class = "Intercept", resp = "riskPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "riskPerceptionSum"),
prior(normal(3, 1), coef = "Age", resp = "riskPerceptionSum"),
prior(normal(3, 1), coef = "Gender1", resp = "riskPerceptionSum"),
prior(normal(-3, 1), coef = "dominanceSum", resp = "riskPerceptionSum"),
prior(normal(2, 1), coef = "leadershipSum", resp = "riskPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "riskPerceptionSum"),
#----
prior(normal(0, 1), class = "Intercept", resp = "riskBenefitSum"),
prior(normal(0, 1), class = "sigma", resp = "riskBenefitSum"),
prior(normal(-3, 1), coef = "Age", resp = "riskBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "riskBenefitSum"),
prior(normal(3, 1), coef = "dominanceSum", resp = "riskBenefitSum"),
prior(normal(-2, 1), coef = "leadershipSum", resp = "riskBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "riskBenefitSum")), save_all_pars = T, iter = 6500, cores = 6, warmup = 500)
summary(m4)
# HDI
m4_hdi <- bayestestR::hdi(m4, effects = "fixed", component = "conditional", ci = .95)
kable(m4_hdi[sign(m4_hdi$CI_low) == sign(m4_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
16
|
b_riskSum_Intercept
|
0.95
|
0.02
|
0.50
|
|
14
|
b_riskSum_dominanceSum
|
0.95
|
0.18
|
0.57
|
|
15
|
b_riskSum_Gender1
|
0.95
|
-0.86
|
-0.16
|
|
8
|
b_riskPerceptionSum_dominanceSum
|
0.95
|
-0.46
|
-0.04
|
|
9
|
b_riskPerceptionSum_Gender1
|
0.95
|
0.05
|
0.82
|
|
4
|
b_riskBenefitSum_Intercept
|
0.95
|
0.01
|
0.52
|
|
2
|
b_riskBenefitSum_dominanceSum
|
0.95
|
0.01
|
0.42
|
|
3
|
b_riskBenefitSum_Gender1
|
0.95
|
-0.98
|
-0.23
|
# Benefit, perception and risk taking across subdomains for DOPL motives and interactions
m5 <- brm(mvbind(riskSum, riskPerceptionSum, riskBenefitSum) ~ dominanceSum*Gender + prestigeSum*Gender + leadershipSum*Gender + Age, data= experiment_dataset_analysis, iter = 6000, cores = 6, warmup = 500,
prior = c(prior(normal(0, 1), class = "Intercept", resp = "riskSum"),
prior(normal(0, 1), class = "sigma", resp = "riskSum"),
prior(normal(-3, 1), coef = "Age", resp = "riskSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "riskSum"),
prior(normal(3, 1), coef = "dominanceSum", resp = "riskSum"),
prior(normal(-2, 1), coef = "leadershipSum", resp = "riskSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "riskSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "riskSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "riskSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "riskSum"),
#----
prior(normal(0, 1), class = "Intercept", resp = "riskPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "riskPerceptionSum"),
prior(normal(3, 1), coef = "Age", resp = "riskPerceptionSum"),
prior(normal(3, 1), coef = "Gender1", resp = "riskPerceptionSum"),
prior(normal(-3, 1), coef = "dominanceSum", resp = "riskPerceptionSum"),
prior(normal(2, 1), coef = "leadershipSum", resp = "riskPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "riskPerceptionSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "riskPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "riskPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "riskPerceptionSum"),
#----
prior(normal(0, 1), class = "Intercept", resp = "riskBenefitSum"),
prior(normal(0, 1), class = "sigma", resp = "riskBenefitSum"),
prior(normal(-3, 1), coef = "Age", resp = "riskBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "riskBenefitSum"),
prior(normal(3, 1), coef = "dominanceSum", resp = "riskBenefitSum"),
prior(normal(-2, 1), coef = "leadershipSum", resp = "riskBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "riskBenefitSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "riskBenefitSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "riskBenefitSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "riskBenefitSum")), save_all_pars = T)
summary(m5)
# HDI
m5_hdi <- bayestestR::hdi(m5, effects = "fixed", component = "conditional", ci = .95)
kable(m5_hdi[sign(m5_hdi$CI_low) == sign(m5_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
20
|
b_riskSum_dominanceSum
|
0.95
|
0.35
|
0.94
|
|
22
|
b_riskSum_Gender1
|
0.95
|
-0.86
|
-0.16
|
|
21
|
b_riskSum_dominanceSum.Gender1
|
0.95
|
-0.84
|
-0.10
|
|
13
|
b_riskPerceptionSum_Gender1
|
0.95
|
0.05
|
0.80
|
|
18
|
b_riskPerceptionSum_prestigeSum
|
0.95
|
0.01
|
0.61
|
|
14
|
b_riskPerceptionSum_Gender1.leadershipSum
|
0.95
|
0.03
|
0.81
|
|
2
|
b_riskBenefitSum_dominanceSum
|
0.95
|
0.06
|
0.70
|
|
4
|
b_riskBenefitSum_Gender1
|
0.95
|
-0.96
|
-0.21
|
m5_hdi <- hdi(m5, effects = "fixed", component = "conditional", ci = .95)
m5_hdi[sign(m5_hdi$CI_low) == sign(m5_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')]
## Highest Density Interval
##
## Parameter | 95% HDI
## ----------------------------------------------------------
## b_riskSum_dominanceSum | [ 0.35, 0.94]
## b_riskSum_Gender1 | [-0.86, -0.16]
## b_riskSum_dominanceSum.Gender1 | [-0.84, -0.10]
## b_riskPerceptionSum_Gender1 | [ 0.05, 0.80]
## b_riskPerceptionSum_prestigeSum | [ 0.01, 0.61]
## b_riskPerceptionSum_Gender1.leadershipSum | [ 0.03, 0.81]
## b_riskBenefitSum_dominanceSum | [ 0.06, 0.70]
## b_riskBenefitSum_Gender1 | [-0.96, -0.21]
# checking interaction slopes
# plot(conditional_effects(m5), points=T)
# Model comparison
# co2 <- loo(m4,m5)
# co2
# comparisonm4 <- loo(m4)
# comparisonm5 <- loo(m5)
# loocomparison4_5 <- loo_compare(comparisonm4, comparisonm5)
# loocomparison4_5
#
# co2$loos$m4$looic
# co2$loos$m4$se_looic
# co2$loos$m5$looic
# co2$loos$m5$se_looic
#
# co2$loos$m4$elpd_loo
# co2$loos$m4$se_elpd_loo
# co2$loos$m5$elpd_loo
# co2$loos$m5$se_elpd_loo
looic_2 <- c(co2$loos$m4$looic, co2$loos$m5$looic)
looic_2_se <- c(co2$loos$m4$se_looic, co2$loos$m5$se_looic)
looic_elpd_2 <- c(co2$loos$m4$elpd_loo, co2$loos$m5$elpd_loo)
looic_elpd_2_se <- c(co2$loos$m4$se_elpd_loo, co2$loos$m5$se_elpd_loo)
loo_table2 <- data.frame(looic_2, looic_2_se, looic_elpd_2,looic_elpd_2_se, row.names = c("m4", "m5"))
kable(loo_table2, format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>%
kable_styling(full_width = F)
|
|
looic_2
|
looic_2_se
|
looic_elpd_2
|
looic_elpd_2_se
|
|
m4
|
862.59
|
32.33
|
-431.30
|
16.16
|
|
m5
|
867.63
|
31.28
|
-433.82
|
15.64
|
# bayes_R2(m4)
# bayes_R2(m5)
#
# pp_check(m5, resp="riskSum", nsamples = 1000)
# pp_check(m5, resp="riskPerceptionSum", nsamples = 1000)
# pp_check(m5, resp="riskBenefitSum", nsamples = 1000)
m7 <- brm(mvbind(riskSum, riskPerceptionSum, riskBenefitSum) ~ dominanceSum*Age + prestigeSum*Age + leadershipSum*Age + Gender, data= experiment_dataset_analysis, iter = 4000, cores = 6,
prior = c(prior(normal(0, 1), class = "Intercept", resp = "riskSum"),
prior(normal(0, 1), class = "sigma", resp = "riskSum"),
prior(normal(-3, 1), coef = "Age", resp = "riskSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "riskSum"),
prior(normal(3, 1), coef = "dominanceSum", resp = "riskSum"),
prior(normal(-2, 1), coef = "leadershipSum", resp = "riskSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "riskSum"),
prior(normal(0, 1), coef = "dominanceSum:Age", resp = "riskSum"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "riskSum"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "riskSum"),
#----
prior(normal(0, 1), class = "Intercept", resp = "riskPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "riskPerceptionSum"),
prior(normal(3, 1), coef = "Age", resp = "riskPerceptionSum"),
prior(normal(3, 1), coef = "Gender1", resp = "riskPerceptionSum"),
prior(normal(-3, 1), coef = "dominanceSum", resp = "riskPerceptionSum"),
prior(normal(2, 1), coef = "leadershipSum", resp = "riskPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "riskPerceptionSum"),
prior(normal(0, 1), coef = "dominanceSum:Age", resp = "riskPerceptionSum"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "riskPerceptionSum"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "riskPerceptionSum"),
#----
prior(normal(0, 1), class = "Intercept", resp = "riskBenefitSum"),
prior(normal(0, 1), class = "sigma", resp = "riskBenefitSum"),
prior(normal(-3, 1), coef = "Age", resp = "riskBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "riskBenefitSum"),
prior(normal(3, 1), coef = "dominanceSum", resp = "riskBenefitSum"),
prior(normal(-2, 1), coef = "leadershipSum", resp = "riskBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "riskBenefitSum"),
prior(normal(0, 1), coef = "dominanceSum:Age", resp = "riskBenefitSum"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "riskBenefitSum"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "riskBenefitSum")), save_all_pars = T)
summary(m7)
# HDI
m7_hdi <- bayestestR::hdi(m7, effects = "fixed", component = "conditional", ci = .95)
kable(m7_hdi[sign(m7_hdi$CI_low) == sign(m7_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')], format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
25
|
b_riskSum_Intercept
|
0.95
|
0.01
|
0.52
|
|
22
|
b_riskSum_dominanceSum
|
0.95
|
0.18
|
0.57
|
|
24
|
b_riskSum_Gender1
|
0.95
|
-0.88
|
-0.17
|
|
13
|
b_riskPerceptionSum_dominanceSum
|
0.95
|
-0.47
|
-0.05
|
|
18
|
b_riskPerceptionSum_prestigeSum
|
0.95
|
0.01
|
0.44
|
|
15
|
b_riskPerceptionSum_Gender1
|
0.95
|
0.05
|
0.81
|
|
4
|
b_riskBenefitSum_dominanceSum
|
0.95
|
0.02
|
0.43
|
|
6
|
b_riskBenefitSum_Gender1
|
0.95
|
-0.97
|
-0.21
|
library(brms)
demo_m1 <- brm(mvbind(dominanceSum, prestigeSum, leadershipSum) ~ Gender + Age + Ethnicity + Education,
data = experiment_dataset_analysis, iter = 4000, cores = 6, warmup = 500,
prior = c(prior(normal(0,1), class="Intercept"),
prior(normal(3,1), coef="Gender1", resp = "dominanceSum"),
prior(normal(0,1), coef="Gender1", resp = "leadershipSum"),
prior(normal(-2,1), coef="Gender1", resp = "prestigeSum")))
summary(demo_m1)
# demo_m1_hdi <- bayestestR::hdi(demo_m1, effects = "fixed", component = "conditional", ci = .95)
# demo_m1_hdi[sign(demo_m1_hdi$CI_low) == sign(demo_m1_hdi$CI_high),
# c('Parameter', 'CI','CI_low', 'CI_high')]
```{ visualization, echo = T, results = ‘hide’, message=FALSE, warning=FALSE}
experiment_dataset_analysis %>% data_grid(generalRiskPreference = seq_range(generalRiskPreference, n = 111)) %>% add_predicted_draws(m2) %>% ggplot(aes(x = generalRiskPreference, y = dominanceSum)) + stat_lineribbon(aes(y = .prediction), .width = c(.99, .95, .8, .5), color = “#08519C”) + geom_point(data = experiment_1_Dataset, size = 2) + scale_fill_brewer()
```r
priorTest <- c(prior(normal(0, 1), class = "b",resp = "ethicalPreference"),
prior(normal(0, 1), class = "b", coef = "Age", resp = "ethicalPreference"),
prior(normal(0, 1), class = "b", coef = "Gender1", resp = "ethicalPreference"),
prior(normal(2, 1), class = "b", coef = "dominanceSum", resp = "ethicalPreference"),
prior(normal(0, 1), class = "b", coef = "leadershipSum", resp = "ethicalPreference"),
prior(normal(0, 1), class = "b", coef = "prestigeSum", resp = "ethicalPreference"),
prior(normal(0, 1), class= "b", coef = "dominanceSum:Gender1", resp = "ethicalPreference"),
prior(normal(0, 1), class= "b", coef = "Gender1:prestigeSum", resp = "ethicalPreference"),
prior(normal(0, 1), class= "b", coef = "Gender1:leadershipSum", resp = "ethicalPreference"),
prior(normal(0, 1), class = "Intercept", resp = "ethicalPreference"),
prior(normal(0, 1), class = "sigma", resp = "ethicalPreference"),
prior(normal(0, 1), class = "b",resp = "financialPreference"),
prior(normal(0, 1), class = "b", coef = "Age", resp = "financialPreference"),
prior(normal(0, 1), class = "b", coef = "Gender1", resp = "financialPreference"),
prior(normal(2, 1), class = "b", coef = "dominanceSum", resp = "financialPreference"),
prior(normal(0, 1), class = "b", coef = "leadershipSum", resp = "financialPreference"),
prior(normal(0, 1), class = "b", coef = "prestigeSum", resp = "financialPreference"),
prior(normal(0, 1), class= "b", coef = "dominanceSum:Gender1", resp = "financialPreference"),
prior(normal(0, 1), class= "b", coef = "Gender1:prestigeSum", resp = "financialPreference"),
prior(normal(0, 1), class= "b", coef = "Gender1:leadershipSum", resp = "financialPreference"),
prior(normal(0, 1), class = "Intercept", resp = "financialPreference"),
prior(normal(0, 1), class = "sigma", resp = "financialPreference"),
prior(normal(0, 1), class = "b",resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "b", coef = "Age", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "b", coef = "Gender1", resp = "healthAndSafetyPreference"),
prior(normal(2, 1), class = "b", coef = "dominanceSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "b", coef = "leadershipSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "b", coef = "prestigeSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class= "b", coef = "dominanceSum:Gender1", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class= "b", coef = "Gender1:prestigeSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class= "b", coef = "Gender1:leadershipSum", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "sigma", resp = "healthAndSafetyPreference"),
prior(normal(0, 1), class = "b",resp = "recreationalPreference"),
prior(normal(0, 1), class = "b", coef = "Age", resp = "recreationalPreference"),
prior(normal(0, 1), class = "b", coef = "Gender1", resp = "recreationalPreference"),
prior(normal(2, 1), class = "b", coef = "dominanceSum", resp = "recreationalPreference"),
prior(normal(0, 1), class = "b", coef = "leadershipSum", resp = "recreationalPreference"),
prior(normal(0, 1), class = "b", coef = "prestigeSum", resp = "recreationalPreference"),
prior(normal(0, 1), class= "b", coef = "dominanceSum:Gender1", resp = "recreationalPreference"),
prior(normal(0, 1), class= "b", coef = "Gender1:prestigeSum", resp = "recreationalPreference"),
prior(normal(0, 1), class= "b", coef = "Gender1:leadershipSum", resp = "recreationalPreference"),
prior(normal(0, 1), class = "Intercept", resp = "recreationalPreference"),
prior(normal(0, 1), class = "sigma", resp = "recreationalPreference"),
prior(normal(0, 1), class = "b",resp = "socialPreference"),
prior(normal(0, 1), class = "b", coef = "Age", resp = "socialPreference"),
prior(normal(0, 1), class = "b", coef = "Gender1", resp = "socialPreference"),
prior(normal(2, 1), class = "b", coef = "dominanceSum", resp = "socialPreference"),
prior(normal(0, 1), class = "b", coef = "leadershipSum", resp = "socialPreference"),
prior(normal(0, 1), class = "b", coef = "prestigeSum", resp = "socialPreference"),
prior(normal(0, 1), class= "b", coef = "dominanceSum:Gender1", resp = "socialPreference"),
prior(normal(0, 1), class= "b", coef = "Gender1:prestigeSum", resp = "socialPreference"),
prior(normal(0, 1), class= "b", coef = "Gender1:leadershipSum", resp = "socialPreference"),
prior(normal(0, 1), class = "Intercept", resp = "socialPreference"),
prior(normal(0, 1), class = "sigma", resp = "socialPreference"))
m7_DoPL_DOSPERT <- brm(mvbind(dominanceSum, prestigeSum, leadershipSum) ~ ethicalPreference + financialPreference + socialPreference + healthAndSafetyPreference + recreationalPreference, data = experiment_dataset_analysis,
iter = 6500, warmup = 500,
prior = c(prior(normal(0, 1), class = "Intercept"),
prior(normal(0, 1), class = "sigma", resp = "dominanceSum"),
prior(normal(0, 1), class = "sigma", resp = "prestigeSum"),
prior(normal(0, 1), class = "sigma", resp = "leadershipSum"),
prior(normal(0, 1), coef = "ethicalPreference", resp = "dominanceSum"),
prior(normal(0, 1), coef = "financialPreference", resp = "dominanceSum"),
prior(normal(0, 1), coef = "healthAndSafetyPreference", resp = "dominanceSum"),
prior(normal(0, 1), coef = "recreationalPreference", resp = "dominanceSum"),
prior(normal(0, 1), coef = "socialPreference", resp = "dominanceSum"),
prior(normal(0, 1), coef = "ethicalPreference", resp = "prestigeSum"),
prior(normal(0, 1), coef = "financialPreference", resp = "prestigeSum"),
prior(normal(0, 1), coef = "healthAndSafetyPreference", resp = "prestigeSum"),
prior(normal(0, 1), coef = "recreationalPreference", resp = "prestigeSum"),
prior(normal(0, 1), coef = "socialPreference", resp = "prestigeSum"),
prior(normal(0, 1), coef = "ethicalPreference", resp = "leadershipSum"),
prior(normal(0, 1), coef = "financialPreference", resp = "leadershipSum"),
prior(normal(0, 1), coef = "healthAndSafetyPreference", resp = "leadershipSum"),
prior(normal(0, 1), coef = "recreationalPreference", resp = "leadershipSum"),
prior(normal(0, 1), coef = "socialPreference", resp = "leadershipSum")),
save_all_pars = T)
summary(m7_DoPL_DOSPERT)
m7_DoPL_DOSPERT_hdi <- bayestestR::hdi(m7_DoPL_DOSPERT, effects = "fixed", component = "conditional", ci = .95)
kable(m7_DoPL_DOSPERT_hdi[sign(m7_DoPL_DOSPERT_hdi$CI_low) == sign(m7_DoPL_DOSPERT_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
5
|
b_dominanceSum_recreationalPreference
|
0.95
|
0.00
|
0.26
|
|
16
|
b_prestigeSum_Intercept
|
0.95
|
-2.41
|
-0.46
|
bayes_R2(m1)
m4_perceivedRisk_Age <- brm(mvbind(ethicalQuestionsPerceptionSum, financialQuestionsPerceptionSum, socialQuestionsPerceptionSum, recreationalQuestionsPerceptionSum, healthAndSafetyQuestionsPerceptionSum) ~ dominanceSum*Age + prestigeSum*Age + leadershipSum*Age + Gender + Age, iter = 4000, warmup = 500,
data = experiment_dataset_analysis,
prior =
c(prior(normal(0, 1), class = "Intercept",resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Age", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(-3, 1), coef = "dominanceSum:Age", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(-2, 1), coef = "Age:leadershipSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(-2, 1), coef = "Age:prestigeSum", resp = "ethicalQuestionsPerceptionSum"),
#---- Perception of financial risk taking
prior(normal(0, 1), coef = "Age", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1", resp = "financialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum:Age", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), class = "Intercept", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "financialQuestionsPerceptionSum"),
#---- Perception of risk in health and safety
prior(normal(0, 1), coef = "Age", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum:Age", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "healthAndSafetyQuestionsPerceptionSum"),
#---- Perception of risky recreational settings
prior(normal(0, 1), coef = "Age", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum:Age", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), class = "Intercept", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "recreationalQuestionsPerceptionSum"),
#---- Perception of social risk
prior(normal(0, 1), coef = "Age", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1", resp = "socialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum:Age", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Age:leadershipSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Age:prestigeSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), class = "Intercept", resp = "socialQuestionsPerceptionSum")),
save_all_pars = T)
summary(m4_perceivedRisk_Age)
m4_perceivedRisk_Age_hdi <- bayestestR::hdi(m4_perceivedRisk_Age, effects = "fixed", component = "conditional", ci = .95)
kable(m4_perceivedRisk_Age_hdi[sign(m4_perceivedRisk_Age_hdi$CI_low) == sign(m4_perceivedRisk_Age_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
4
|
b_ethicalQuestionsPerceptionSum_dominanceSum
|
0.95
|
-0.46
|
-0.07
|
|
9
|
b_ethicalQuestionsPerceptionSum_prestigeSum
|
0.95
|
0.04
|
0.43
|
|
5
|
b_ethicalQuestionsPerceptionSum_dominanceSum.Age
|
0.95
|
0.00
|
0.39
|
|
28
|
b_recreationalQuestionsPerceptionSum_Age
|
0.95
|
-0.42
|
-0.02
|
|
36
|
b_recreationalQuestionsPerceptionSum_prestigeSum
|
0.95
|
0.04
|
0.44
|
m5_generalRiskPreference <- brm(generalRiskPreference ~ ethicalPreference + socialPreference + financialPreference + healthAndSafetyPreference + recreationalPreference, data = experiment_dataset_analysis,iter = 4000, warmup = 500,
prior = c(prior(normal(0, 1), class = "Intercept"),
prior(normal(0, 1), coef = "ethicalPreference"),
prior(normal(0, 1), coef = "financialPreference"),
prior(normal(0, 1), coef = "healthAndSafetyPreference"),
prior(normal(0, 1), coef = "recreationalPreference"),
prior(normal(0, 1), coef = "socialPreference")),
save_all_pars = T)
summary(m5_generalRiskPreference)
m5_benefitRisk_Age <- brm(mvbind(ethicalQuestionsBenefitSum, financialQuestionsBenefitSum, socialQuestionsBenefitSum, recreationalQuestionsBenefitSum, healthAndSafetyQuestionsBenefitSum) ~ dominanceSum*Age + prestigeSum*Age + leadershipSum*Age + Gender + Age,
data = experiment_dataset_analysis, iter = 5000, warmup = 500,
prior = c(prior(normal(0, 1), class = "Intercept", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Age", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1", resp = "ethicalQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum:Age", resp = "ethicalQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Age:leadershipSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Age:prestigeSum", resp = "ethicalQuestionsBenefitSum"),
#---- Believed benefit of financial risk
prior(normal(0, 1), class = "Intercept", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Age", resp = "financialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "financialQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "financialQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum:Age", resp = "financialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Age:leadershipSum", resp = "financialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Age:prestigeSum", resp = "financialQuestionsBenefitSum"),
#---- Believed benefit of health and safety risk
prior(normal(0, 1), coef = "Age", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(3, 1), coef = "dominanceSum:Age", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Age:leadershipSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Age:prestigeSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyQuestionsBenefitSum"),
#---- Believed benefit of recreational risk
prior(normal(0, 1), coef = "Age", resp = "recreationalQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "recreationalQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum:Age", resp = "recreationalQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Age:leadershipSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Age:prestigeSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), class = "Intercept", resp = "recreationalQuestionsBenefitSum"),
#---- Believed benefit of social risk
prior(normal(0, 1), coef = "Age", resp = "socialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "socialQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "socialQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum:Age", resp = "socialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Age:leadershipSum", resp = "socialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Age:prestigeSum", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), class = "Intercept", resp = "socialQuestionsBenefitSum")),
save_all_pars = T)summary(m5_benefitRisk_Age)
m5_generalRiskPreference_hdi <- bayestestR::hdi(m5_generalRiskPreference, effects = "fixed", component = "conditional", ci = .95)
m5_benefitRisk_Age_hdi <- bayestestR::hdi(m5_benefitRisk_Age, effects = "fixed", component = "conditional", ci = .95)
kable(m5_benefitRisk_Age_hdi[sign(m5_benefitRisk_Age_hdi$CI_low) == sign(m5_benefitRisk_Age_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
4
|
b_ethicalQuestionsBenefitSum_dominanceSum
|
0.95
|
0.10
|
0.51
|
|
6
|
b_ethicalQuestionsBenefitSum_Gender1
|
0.95
|
-0.90
|
-0.16
|
|
16
|
b_financialQuestionsBenefitSum_Intercept
|
0.95
|
0.06
|
0.60
|
|
15
|
b_financialQuestionsBenefitSum_Gender1
|
0.95
|
-1.07
|
-0.30
|
|
43
|
b_socialQuestionsBenefitSum_Intercept
|
0.95
|
0.07
|
0.58
|
|
45
|
b_socialQuestionsBenefitSum_prestigeSum
|
0.95
|
0.02
|
0.43
|
|
42
|
b_socialQuestionsBenefitSum_Gender1
|
0.95
|
-1.04
|
-0.30
|
|
34
|
b_recreationalQuestionsBenefitSum_Intercept
|
0.95
|
0.20
|
0.69
|
|
31
|
b_recreationalQuestionsBenefitSum_dominanceSum
|
0.95
|
0.16
|
0.56
|
|
33
|
b_recreationalQuestionsBenefitSum_Gender1
|
0.95
|
-1.29
|
-0.59
|
|
22
|
b_healthAndSafetyQuestionsBenefitSum_dominanceSum
|
0.95
|
0.05
|
0.47
|
|
24
|
b_healthAndSafetyQuestionsBenefitSum_Gender1
|
0.95
|
-0.86
|
-0.10
|
m4_perceivedRisk_Gender <- brm(mvbind(ethicalQuestionsPerceptionSum, financialQuestionsPerceptionSum, socialQuestionsPerceptionSum, recreationalQuestionsPerceptionSum, healthAndSafetyQuestionsPerceptionSum) ~ dominanceSum*Gender + prestigeSum*Gender + leadershipSum*Gender + Gender + Age, iter = 6000, warmup = 500,
data = experiment_dataset_analysis,
prior =
c(prior(normal(0, 1), class = "Intercept",resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Age", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(-3, 1), coef = "dominanceSum:Gender1", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(-2, 1), coef = "Gender1:leadershipSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(-2, 1), coef = "Gender1:prestigeSum", resp = "ethicalQuestionsPerceptionSum"),
#----
prior(normal(0, 1), coef = "Age", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1", resp = "financialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum:Gender1", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), class = "Intercept", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "financialQuestionsPerceptionSum"),
#----
prior(normal(0, 1), coef = "Age", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum:Gender1", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "healthAndSafetyQuestionsPerceptionSum"),
#----
prior(normal(0, 1), coef = "Age", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum:Gender1", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), class = "Intercept", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "recreationalQuestionsPerceptionSum"),
#----
prior(normal(0, 1), coef = "Age", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1", resp = "socialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "dominanceSum:Gender1", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), class = "Intercept", resp = "socialQuestionsPerceptionSum")),
save_all_pars = T)summary(m4_perceivedRisk_Gender)
m4_perceivedRisk_Gender_hdi <- bayestestR::hdi(m4_perceivedRisk_Gender, effects = "fixed", component = "conditional", ci = .95)
kable(m4_perceivedRisk_Gender_hdi[sign(m4_perceivedRisk_Gender_hdi$CI_low) == sign(m4_perceivedRisk_Gender_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
9
|
b_ethicalQuestionsPerceptionSum_prestigeSum
|
0.95
|
0.13
|
0.67
|
|
5
|
b_ethicalQuestionsPerceptionSum_Gender1.leadershipSum
|
0.95
|
0.00
|
0.71
|
|
36
|
b_recreationalQuestionsPerceptionSum_prestigeSum
|
0.95
|
0.06
|
0.60
|
|
28
|
b_recreationalQuestionsPerceptionSum_Age
|
0.95
|
-0.39
|
-0.03
|
|
30
|
b_recreationalQuestionsPerceptionSum_dominanceSum.Gender1
|
0.95
|
-0.76
|
-0.04
|
|
23
|
b_healthAndSafetyQuestionsPerceptionSum_Gender1.leadershipSum
|
0.95
|
0.07
|
0.80
|
m5_generalRiskPreference <- brm(generalRiskPreference ~ ethicalPreference + socialPreference + financialPreference + healthAndSafetyPreference + recreationalPreference, data = experiment_dataset_analysis,iter = 4000, warmup = 500,
prior = c(prior(normal(0, 1), class = "Intercept"),
prior(normal(0, 1) ,class = "b", coef = "ethicalPreference"),
prior(normal(0, 1) ,class = "b", coef = "financialPreference"),
prior(normal(0, 1) ,class = "b", coef = "healthAndSafetyPreference"),
prior(normal(0, 1) ,class = "b", coef = "recreationalPreference"),
prior(normal(0, 1) ,class = "b", coef = "socialPreference")),
save_all_pars = T)
summary(m5_generalRiskPreference)
m5_benefitRisk_Gender <- brm(mvbind(ethicalQuestionsBenefitSum, financialQuestionsBenefitSum, socialQuestionsBenefitSum, recreationalQuestionsBenefitSum, healthAndSafetyQuestionsBenefitSum) ~ dominanceSum*Gender + prestigeSum*Gender + leadershipSum*Gender + Gender + Age,
data = experiment_dataset_analysis, iter = 5000, warmup = 500, cores = 6,
prior = c(prior(normal(0, 1), class = "Intercept", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Age", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1", resp = "ethicalQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum:Gender1", resp = "ethicalQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1:leadershipSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1:prestigeSum", resp = "ethicalQuestionsBenefitSum"),
#----
prior(normal(0, 1), class = "Intercept", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Age", resp = "financialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "financialQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "financialQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum:Gender1", resp = "financialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1:leadershipSum", resp = "financialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1:prestigeSum", resp = "financialQuestionsBenefitSum"),
#----
prior(normal(0, 1), coef = "Age", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(3, 1), coef = "dominanceSum:Gender1", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1:leadershipSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1:prestigeSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyQuestionsBenefitSum"),
#----
prior(normal(0, 1), coef = "Age", resp = "recreationalQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "recreationalQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum:Gender1", resp = "recreationalQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1:leadershipSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1:prestigeSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), class = "Intercept", resp = "recreationalQuestionsBenefitSum"),
#----
prior(normal(0, 1), coef = "Age", resp = "socialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1", resp = "socialQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "socialQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum:Gender1", resp = "socialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1:leadershipSum", resp = "socialQuestionsBenefitSum"),
prior(normal(-3, 1), coef = "Gender1:prestigeSum", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), class = "Intercept", resp = "socialQuestionsBenefitSum")),
save_all_pars = T)summary(m5_benefitRisk_Gender)
m5_benefitRisk_Gender_hdi <- bayestestR::hdi(m5_benefitRisk_Gender, effects = "fixed", component = "conditional", ci = .95)
kable(m5_benefitRisk_Gender_hdi[sign(m5_benefitRisk_Gender_hdi$CI_low) == sign(m5_benefitRisk_Gender_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')] , format = "html", booktabs = T, escape = F, longtable = F, digits = 2) %>% kable_styling(full_width = T)
|
|
Parameter
|
CI
|
CI_low
|
CI_high
|
|
4
|
b_ethicalQuestionsBenefitSum_Gender1
|
0.95
|
-0.93
|
-0.18
|
|
16
|
b_financialQuestionsBenefitSum_Intercept
|
0.95
|
0.04
|
0.58
|
|
13
|
b_financialQuestionsBenefitSum_Gender1
|
0.95
|
-1.06
|
-0.30
|
|
14
|
b_financialQuestionsBenefitSum_Gender1.leadershipSum
|
0.95
|
-0.90
|
-0.09
|
|
43
|
b_socialQuestionsBenefitSum_Intercept
|
0.95
|
0.04
|
0.57
|
|
40
|
b_socialQuestionsBenefitSum_Gender1
|
0.95
|
-1.06
|
-0.32
|
|
45
|
b_socialQuestionsBenefitSum_prestigeSum
|
0.95
|
0.04
|
0.63
|
|
34
|
b_recreationalQuestionsBenefitSum_Intercept
|
0.95
|
0.19
|
0.69
|
|
31
|
b_recreationalQuestionsBenefitSum_Gender1
|
0.95
|
-1.31
|
-0.60
|
|
28
|
b_recreationalQuestionsBenefitSum_Age
|
0.95
|
0.01
|
0.38
|
|
22
|
b_healthAndSafetyQuestionsBenefitSum_Gender1
|
0.95
|
-0.92
|
-0.14
|
|
27
|
b_healthAndSafetyQuestionsBenefitSum_prestigeSum
|
0.95
|
0.01
|
0.61
|
|
23
|
b_healthAndSafetyQuestionsBenefitSum_Gender1.leadershipSum
|
0.95
|
-0.81
|
-0.01
|
m8 <- brm(mvbind(ethicalQuestionsRiskSum, ethicalQuestionsPerceptionSum, ethicalQuestionsBenefitSum,
financialQuestionsRiskSum, financialQuestionsPerceptionSum, financialQuestionsBenefitSum,
healthAndSafetyQuestionsRiskSum, healthAndSafetyQuestionsPerceptionSum, healthAndSafetyQuestionsBenefitSum,
recreationalQuestionsRiskSum, recreationalQuestionsPerceptionSum, recreationalQuestionsBenefitSum,
socialQuestionsRiskSum, socialQuestionsPerceptionSum, socialQuestionsBenefitSum) ~ dominanceSum*Gender + prestigeSum*Gender + leadershipSum*Gender + Age, data= experiment_dataset_analysis, iter = 8000, warmup = 500, save_all_pars = T, cores = 6,
#----
prior = c(prior(normal(0, 1), class = "Intercept", resp = "ethicalQuestionsRiskSum"),
prior(normal(0, 1), class = "sigma", resp = "ethicalQuestionsRiskSum"),
prior(normal(-2, 1), coef = "Age", resp = "ethicalQuestionsRiskSum"),
prior(normal(-2, 1), coef = "Gender1", resp = "ethicalQuestionsRiskSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "ethicalQuestionsRiskSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "ethicalQuestionsRiskSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "ethicalQuestionsRiskSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "ethicalQuestionsRiskSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "ethicalQuestionsRiskSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "ethicalQuestionsRiskSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "Age", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "Gender1", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(-2, 1), coef = "dominanceSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "ethicalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "ethicalQuestionsPerceptionSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), class = "sigma", resp = "ethicalQuestionsBenefitSum"),
prior(normal(-2, 1), coef = "Age", resp = "ethicalQuestionsBenefitSum"),
prior(normal(-2, 1), coef = "Gender1", resp = "ethicalQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "ethicalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "ethicalQuestionsBenefitSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "financialQuestionsRiskSum"),
prior(normal(0, 1), class = "sigma", resp = "financialQuestionsRiskSum"),
prior(normal(-2, 1), coef = "Age", resp = "financialQuestionsRiskSum"),
prior(normal(-2, 1), coef = "Gender1", resp = "financialQuestionsRiskSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "financialQuestionsRiskSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "financialQuestionsRiskSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "financialQuestionsRiskSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "financialQuestionsRiskSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "financialQuestionsRiskSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "financialQuestionsRiskSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "financialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "Age", resp = "financialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "Gender1", resp = "financialQuestionsPerceptionSum"),
prior(normal(-2, 1), coef = "dominanceSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "financialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "financialQuestionsPerceptionSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), class = "sigma", resp = "financialQuestionsBenefitSum"),
prior(normal(-2, 1), coef = "Age", resp = "financialQuestionsBenefitSum"),
prior(normal(-2, 1), coef = "Gender1", resp = "financialQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "financialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "financialQuestionsBenefitSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyQuestionsRiskSum"),
prior(normal(0, 1), class = "sigma", resp = "healthAndSafetyQuestionsRiskSum"),
prior(normal(-2, 1), coef = "Age", resp = "healthAndSafetyQuestionsRiskSum"),
prior(normal(-2, 1), coef = "Gender1", resp = "healthAndSafetyQuestionsRiskSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "healthAndSafetyQuestionsRiskSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "healthAndSafetyQuestionsRiskSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "healthAndSafetyQuestionsRiskSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "healthAndSafetyQuestionsRiskSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "healthAndSafetyQuestionsRiskSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "healthAndSafetyQuestionsRiskSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "Age", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "Gender1", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(-2, 1), coef = "dominanceSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "healthAndSafetyQuestionsPerceptionSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), class = "sigma", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(-2, 1), coef = "Age", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(-2, 1), coef = "Gender1", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "healthAndSafetyQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "healthAndSafetyQuestionsBenefitSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "recreationalQuestionsRiskSum"),
prior(normal(0, 1), class = "sigma", resp = "recreationalQuestionsRiskSum"),
prior(normal(-2, 1), coef = "Age", resp = "recreationalQuestionsRiskSum"),
prior(normal(-2, 1), coef = "Gender1", resp = "recreationalQuestionsRiskSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "recreationalQuestionsRiskSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "recreationalQuestionsRiskSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "recreationalQuestionsRiskSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "recreationalQuestionsRiskSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "recreationalQuestionsRiskSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "recreationalQuestionsRiskSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "Age", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "Gender1", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(-2, 1), coef = "dominanceSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "recreationalQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "recreationalQuestionsPerceptionSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), class = "sigma", resp = "recreationalQuestionsBenefitSum"),
prior(normal(-2, 1), coef = "Age", resp = "recreationalQuestionsBenefitSum"),
prior(normal(-2, 1), coef = "Gender1", resp = "recreationalQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "recreationalQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "recreationalQuestionsBenefitSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "socialQuestionsRiskSum"),
prior(normal(0, 1), class = "sigma", resp = "socialQuestionsRiskSum"),
prior(normal(-2, 1), coef = "Age", resp = "socialQuestionsRiskSum"),
prior(normal(-2, 1), coef = "Gender1", resp = "socialQuestionsRiskSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "socialQuestionsRiskSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "socialQuestionsRiskSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "socialQuestionsRiskSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "socialQuestionsRiskSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "socialQuestionsRiskSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "socialQuestionsRiskSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), class = "sigma", resp = "socialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "Age", resp = "socialQuestionsPerceptionSum"),
prior(normal(2, 1), coef = "Gender1", resp = "socialQuestionsPerceptionSum"),
prior(normal(-2, 1), coef = "dominanceSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "socialQuestionsPerceptionSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "socialQuestionsPerceptionSum"),
#
prior(normal(0, 1), class = "Intercept", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), class = "sigma", resp = "socialQuestionsBenefitSum"),
prior(normal(-2, 1), coef = "Age", resp = "socialQuestionsBenefitSum"),
prior(normal(-2, 1), coef = "Gender1", resp = "socialQuestionsBenefitSum"),
prior(normal(2, 1), coef = "dominanceSum", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "leadershipSum", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "prestigeSum", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "dominanceSum:Gender1", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1:prestigeSum", resp = "socialQuestionsBenefitSum"),
prior(normal(0, 1), coef = "Gender1:leadershipSum", resp = "socialQuestionsBenefitSum")))
summary(m8)
m8_hdi <- hdi(m8, effects = "fixed", component = "conditional", ci = .95)
m8_hdi[sign(m8_hdi$CI_low) == sign(m8_hdi$CI_high),
c('Parameter', 'CI','CI_low', 'CI_high')]
mcmc_areas_m1 <- mcmc_areas(m1, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
mcmc_dens_m1 <- mcmc_dens_overlay(m1, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
m1_plot <- mcmc_plot(m1, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
mcmc_areas_m2 <- mcmc_areas(m2, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
mcmc_dens_m2 <- mcmc_dens_overlay(m2, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
m2_plot <- mcmc_plot(m2, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
mcmc_areas_m3 <- mcmc_areas(m3, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
mcmc_dens_m3 <- mcmc_dens_overlay(m3, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
m3_plot <- mcmc_plot(m3, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
mcmc_areas_m4 <- mcmc_areas(m4, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
mcmc_dens_m4 <- mcmc_dens_overlay(m4, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))
m4_plot <- mcmc_plot(m4, pars = c("b_dominanceSum", "b_prestigeSum", "b_leadershipSum", "b_Gender1"))