I’m looking to test pre vs post differences in working memory (using R), and have a covariate (IQ) that was only measured once. My data frame looks something like this:

d <- tibble::tibble(time = c(rep("pre", 10), rep("post", 10)),
  wrking_mem = c(abs(rnorm(10, 50, 10.2)), abs(rnorm(10, 70, 13.5))),
  iq = c(rnorm(10, 100, 15), rep(NA, 10)),
  participant_id = as.factor(rep(1:10, 2)))

The R packages I’m using require that the # of outcome observations be equal to the number of predictor / covariate observations to fit the linear model or else you get an error.

Wondering if anyone else has encountered a similar situation, and what you did to solve the issue?

Any help is great - thanks!

  • $\begingroup$ How was the IQ measured? There are IQ test that involve the working memory. $\endgroup$
    – captcoma
    Oct 7, 2021 at 10:58
  • $\begingroup$ @captcoma Thanks for the comment. In this case, IQ is a hypothetical stand-in variable for any covariate that was measured once in a repeated measures experiment. Apologies, as I should have specified. $\endgroup$
    – giopico
    Oct 7, 2021 at 13:11

1 Answer 1


Assuming that your intervention is not intended to change IQ, then it might be a reasonable assumption that IQ is constant, so the IQ before can be used as the IQ after.

  • $\begingroup$ Thanks for this, @MaartenBuis. So simply repeating the same values a second time over sounds like the plan. Much appreciated! $\endgroup$
    – giopico
    Oct 7, 2021 at 13:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.