CRR Data

Comparing outcomes between control and treatment group

data.combined$treatment_group_i <- recode_factor(data.combined$treatment_group_i, ctrl_info = "control", trtm_info = "treatment", )
data.combined %>% group_by(treatment_group_i) %>% na.omit() %>% summarize(mean_homo = mean(homo), mean_info = mean(info), mean_wmb = mean(wmb), mean_bi = mean(bi))

Regression Analysis

Homo - Controlling for covariates

library(broom)

## Warning: package 'broom' was built under R version 4.0.2

fit <- lm(homo ~ factor(treatment_group_i) + age + gender + int_exp + work_exp + nat_sec_job + tech_efficacy + prior_training + prior_victimization + resources + prior_knowledge, data.combined)
tidy(fit) %>% select(term, estimate, p.value) %>% mutate(estimate = round(estimate, digits = 2), p.value = round(p.value, digits = 3))

Info - Controlling for covariates

fit <- lm(info ~ factor(treatment_group_i) + age + gender + int_exp + work_exp + nat_sec_job + tech_efficacy + prior_training + prior_victimization + resources + prior_knowledge, data.combined)
tidy(fit) %>% select(term, estimate, p.value) %>% mutate(estimate = round(estimate, digits = 2), p.value = round(p.value, digits = 3))

WMB - Controlling for covariates

fit <- lm(wmb ~ factor(treatment_group_i) + age + gender + int_exp + work_exp + nat_sec_job + tech_efficacy + prior_training + prior_victimization + resources + prior_knowledge, data.combined)
tidy(fit) %>% select(term, estimate, p.value) %>% mutate(estimate = round(estimate, digits = 2), p.value = round(p.value, digits = 3))

BI - Controlling for covariates

fit <- lm(bi ~ factor(treatment_group_i) + age + gender + int_exp + work_exp + nat_sec_job + tech_efficacy + prior_training + prior_victimization + resources + prior_knowledge, data.combined)
tidy(fit) %>% select(term, estimate, p.value) %>% mutate(estimate = round(estimate, digits = 2), p.value = round(p.value, digits = 3))

Descriptive Preferences (BUI)

For my cybersecurity needs, it makes sense to use the training that corresponds to:

data %>% group_by(BUI_1) %>% summarize(n = n())

Even if the videos have the same content, I would prefer to use the training that corresponds to:

data %>% group_by(BUI_2) %>% summarize(n = n())

If there is another form of cybersecurity training that is equally good, I would prefer to use the training that corresponds to:

data %>% group_by(BUI_3) %>% summarize(n = n())

If there is another form of cybersecurity training that is equally good, I would prefer to use the training that corresponds to:

data %>% group_by(BUI_4) %>% summarize(n = n())