Study designs:
A. 2 (Gender: male vs. female) by 2 (Condition: Covid-19 Need vs. Aphorism)
B. 2 (Gender: male vs. female) by 3 (Condition: Personal Need vs. Abstract Need vs. Aphorism)
Reminder calculations:
gender_catscore <- gender_within_mistakes - gender_between_mistakes
state_catscore <- statement_within_mistakes - state_between_mistakes
Results quick summary:
For Covid-19 need vs. Aphorism: There is a significant difference between condition within vs. condition between categorization, prop_between = 3.60, prop_within = 3.90, t(1943) = -20.03, p < .001. This difference is competitive with gender categorization differences.
For Abstract Need vs. Aphorism: There is a significant difference between condition within vs. between categorization, prop_between = 3.34, prop_within = 3.91, t(78) = -2.01, p = .048.
## 1a. Gender-between mistakes vs. gender-within mistakes
t.test(d1$gender_between, d1$gender_within, paired = T)
##
## Paired t-test
##
## data: d1$gender_between and d1$gender_within
## t = -5.1036, df = 80, p-value = 2.201e-06
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -2.406638 -1.056325
## sample estimates:
## mean of the differences
## -1.731481
## 1b. Condition-between mistakes vs. Condition-within mistakes
t.test(d1$condition_between, d1$condition_within, paired = T)
##
## Paired t-test
##
## data: d1$condition_between and d1$condition_within
## t = -4.0652, df = 80, p-value = 0.0001116
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.9354830 -0.6632824
## sample estimates:
## mean of the differences
## -1.299383
##
## Paired t-test
##
## data: d2$gender_between and d2$gender_within
## t = -2.8299, df = 74, p-value = 0.005991
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.965405 -0.341262
## sample estimates:
## mean of the differences
## -1.153333
## Statement-between mistakes vs. Statement-within mistakes
t.test(d2$condition_between, d2$condition_within, mu = 0, paired = T)
##
## Paired t-test
##
## data: d2$condition_between and d2$condition_within
## t = -1.2437, df = 74, p-value = 0.2175
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.1189121 0.2589121
## sample estimates:
## mean of the differences
## -0.43
## 3a. Gender-between mistakes vs. gender-within mistakes
t.test(d3$gender_between, d3$gender_within, paired = T)
##
## Paired t-test
##
## data: d3$gender_between and d3$gender_within
## t = -3.4984, df = 75, p-value = 0.0007909
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.8895218 -0.5183729
## sample estimates:
## mean of the differences
## -1.203947
## 3b. Statement-between mistakes vs. Statement-within mistakes
t.test(d3$condition_between, d3$condition_within, paired = T)
##
## Paired t-test
##
## data: d3$condition_between and d3$condition_within
## t = -1.3495, df = 75, p-value = 0.1812
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.8715658 0.1676184
## sample estimates:
## mean of the differences
## -0.3519737
##
## Paired t-test
##
## data: d4$gender_between and d4$gender_within
## t = -4.7056, df = 78, p-value = 1.077e-05
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -2.1661443 -0.8781595
## sample estimates:
## mean of the differences
## -1.522152
## 4b. Statement-between mistakes vs. Statement-within mistakes
t.test(d4$condition_between, d4$condition_within, paired = T)
##
## Paired t-test
##
## data: d4$condition_between and d4$condition_within
## t = -2.0105, df = 78, p-value = 0.04783
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1.133662471 -0.005578035
## sample estimates:
## mean of the differences
## -0.5696203