I'm trying to calculate row means on a dataset. I found a helpful function someone made here (dplyr - using mutate() like rowmeans()), and it works when I try out every column but not when I try to use a dplyr helper function.
Why does this work:
#The rowmeans function that works
my_rowmeans = function(..., na.rm=TRUE){
x =
if (na.rm) lapply(list(...), function(x) replace(x, is.na(x), as(0, class(x))))
else list(...)
d = Reduce(function(x,y) x+!is.na(y), list(...), init=0)
Reduce(`+`, x)/d
}
#The data
library(tidyverse)
data <- tibble(id = c(1:4),
turn_intent_1 = c(5, 1, 1, 4),
turn_intent_2 = c(5, 1, 1, 3),
turn_intent_3R = c(5, 5, 1, 3))
#The code that is cumbersome but works
data %>%
mutate(turn_intent_agg = my_rowmeans(turn_intent_1, turn_intent_2, turn_intent_3R))
#The output
# A tibble: 4 x 5
id turn_intent_1 turn_intent_2 turn_intent_3R turn_intent_agg
<int> <dbl> <dbl> <dbl> <dbl>
1 1 5 5 5 5
2 2 1 1 5 2.33
3 3 1 1 1 1
4 4 4 3 3 3.33
But this does not work:
#The code
data %>%
mutate(turn_intent_agg = select(., contains("turn")) %>%
my_rowmeans())
#The output
Error in class1Def@contains[[class2]] : no such index at level 1
Of course, I can type each column, but this dataset has many columns. It'd be much easier to use these wrappers.
I need the output to look like the correct one shown that contains all columns (such as id).
Thank you!
I think that you can simplify it to:
data %>%
mutate(turn_intent_agg = rowMeans(select(., contains("turn"))))
id turn_intent_1 turn_intent_2 turn_intent_3R turn_intent_agg
<int> <dbl> <dbl> <dbl> <dbl>
1 1 5 5 5 5
2 2 1 1 5 2.33
3 3 1 1 1 1
4 4 4 3 3 3.33
And you can indeed add also the na.rm = TRUE parameter:
data %>%
mutate(turn_intent_agg = rowMeans(select(., contains("turn")), na.rm = TRUE))