According to the documentation, when "ges" and "pes" are both used as a parameter to effect.size for anova_summary function, the table would have ges and pes values for each factor. However, only pes values are returned. The expected behavior is that the anova summary table has two values at the same time.
library(rstatix)
library(magrittr)
sample_data <- data.frame(
id = rep(1:5, each = 2),
dv = 1:10,
within = rep(c(0, 1), times = 5)
)
# anova_test uses anova_summary for table generation
res <- sample_data %>% anova_test(
dv = dv,
wid = id,
within = within,
effect.size = c("pes", "ges"),
detailed = TRUE
)
(actual)
ANOVA Table (type III tests)
Effect DFn DFd SSn SSd F p p<.05 pes
1 (Intercept) 1 4 302.5 80 15.125 0.018 * 0.791
2 within 1 4 2.5 0 Inf 0.000 * 1.000
(expected)
Effect DFn DFd SSn SSd F p p<.05 pes ges
1 (Intercept) 1 4 302.5 80 15.125 0.018 * 0.791 0.791
2 within 1 4 2.5 0 Inf 0.000 * 1.000 0.030
According to the documentation, when
"ges"
and"pes"
are both used as a parameter toeffect.size
foranova_summary
function, the table would haveges
andpes
values for each factor. However, onlypes
values are returned. The expected behavior is that the anova summary table has two values at the same time.