Need to run a regression over each subject in data frame in R

Question

I have a df in long format (resulting from cognitive psych experiments), with several subjects. The df has vectors representing reaction time (RT) and trial number, and a vector with subject number. I want to model the effect of trial number on reaction time (basically, how people get faster with practice). I wrote a simple loop that is supposed to run a regression: for (i in unique(red_incongruent$subject_nr)) { print(i) print(lm(red_incongruent$response_time_response ~ red_incongruent$trial_id)) } It returns a list with the regression results for each subject. However, I get the same results for each subject so something must be wrong but I don't understand what.

Answer 1

You are not selecting the subject in the loop. Perhaps something like this:

red_incongruent=data.frame(
    subject_nr=rep(1:100,each=20),
    trial_id=rep(1:20,100),
    response_time_response=rnorm(2000,100))


for (i in unique(red_incongruent$subject_nr))
{
 print(i)
 dat = red_incongruent[red_incongruent$subject_nr==unique(red_incongruent$subject_nr)[i],]
 print(lm(dat$response_time_response ~ dat$trial_id))
}