Question

I have a dataset that I need to sort by participant (RECORDING_SESSION_LABEL) and by trial_number. However, when I sort the data using R none of the sort functions I have tried put the variables in the correct numeric order that I want. The participant variable comes out ok but the trial ID variable comes out in the wrong order for what I need.

using:
fix_rep[order(as.numeric(RECORDING_SESSION_LABEL), as.numeric(trial_number)),]

Participant ID comes out as:

118 118 118 etc. 211 211 211 etc. 306 306 306 etc.(which is fine)

trial_number comes out as:

1 1 10 10 11 11 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19 2 2 20 20 …. (which is not what I want – it seems to be sorting lexically rather than numerically)

What I would like is trial_number to be order like this within each participant number:

1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 10 10 11 11 ….

I have checked that these variables are not factors and are numeric and also tried without the ‘as.numeric’, but with no joy. Looking around I saw suggestions that sort() and mixedsort() might do the trick in place of ‘order’, both come up with errors. I am slowly pulling my hair out over what I think should be a simple thing. Can anybody help shed some light on how to do this to get what I need?

Answer 1

Even though you claim it is not a factor, it does behave exactly as if it were a factor. Testing if something is a factor can be tricky since a factor is just an integer vector with a levels attribute and a class label. If it is a factor, your code needs to have a call to as.character() nested inside the as.numeric():

fix_rep[order(as.numeric(RECORDING_SESSION_LABEL), as.numeric(as.character(trial_number))),]

To be really sure if it’s a factor, I recommend the str() function:
str(trial_number)