I'm using expss tables to calculate means values. I can't find a function to report confidence intervals around the means.
I know r has several packages that can calculate and report confidence intervals, but I'm focused on using the expss package. I tried to create a function using tab_stat_fun_df() with no success. I also reviewed the significance documentation for the expss package hoping to have an option that reports CI.
Looking to add CI for the means in this code:
df <- data.frame(Group=c(1,1,1,1,1,1,1,0,0,0,0,0,0,0,0),
Response=c(0,1,0,1,0,1,0,1,0,1,0,1,1,0,0),
Value=c(10,20,17,14,15,16,17,18,19,10,11,12,13,14,15))
library(expss)
df %>% tab_rows(Group) %>% tab_cols(Response) %>% tab_cells(Value) %>%
tab_stat_mean_sd_n() %>% tab_pivot()
What I propose is not an actual solution, but rather a starting point (based on the use of the tab_stat_fun_df command). Some further refinement is needed to achieve what you need.
library(expss)
df <- data.frame(Group=c(1,1,1,1,1,1,1,0,0,0,0,0,0,0,0),
Response=c(0,1,0,1,0,1,0,1,0,1,0,1,1,0,0),
Value=c(10,20,17,14,15,16,17,18,19,10,11,12,13,14,15))
df %>% tab_rows(Group) %>%
tab_cols(Response) %>%
tab_cells(Value) %>%
tab_stat_mean_sd_n() %>%
tab_stat_fun_df(function(x) {
out <- t.test(x)
mn <- round(out$estimate,2)
CI=round(out$conf.int,2)
c("Mean" = paste0(mn, " (", CI[1], " - ", CI[2], ")"))
}) %>%
tab_pivot()
The output is
| | | | | Response | |
| | | | | 0 | 1 |
| ----- | -- | ----- | ------------ | -------------------- | -------------------- |
| Group | 0 | Value | Mean | 14.75 | 13.25 |
| | | | Std. dev. | 3.30403793359983 | 3.40342964277702 |
| | | | Unw. valid N | 4 | 4 |
| | 1 | Value | Mean | 14.75 | 16.6666666666667 |
| | | | Std. dev. | 3.30403793359983 | 3.05505046330389 |
| | | | Unw. valid N | 4 | 3 |
| | 0 | Mean | | 14.75 (9.49 - 20.01) | 13.25 (7.83 - 18.67) |
| | 1 | Mean | | 14.75 (9.49 - 20.01) | 16.67 (9.08 - 24.26) |