I would like to use Cronbach's Alpha to check the fiability or internal consistency of a cognitive task. The task has a certain number of trials which were coded with zeros (incorrect trial) and ones (correct trials). But it happens that not all participants responded all trials, because the task was terminated if participants did 3 incorrect trials in a row.

So I have some participants with 10 responded trials, other with 13, with 15, with 18 etc. The sample has 80 participants.

ID Trial10 trial12 trial13 Trial14 Trial15 Trial16 Trial17 Trial18
0        1       1       1       1       1       1       0       0
1        1       1       0       0       0      NA      NA      NA
3        1       1       1       0       1       0       0       0
4        1       0       0       0      NA      NA      NA      NA

What should I do? Should I consider the NA's as Zeros (assumming that participants would have responded incorrectly? Would this carry some problems for my analysis (overestimation of the statistic, for example).

For information, I am working with R.

  • $\begingroup$ Why are you using Cronbach's alpha at all? What are you trying to find out? $\endgroup$ – Peter Flom - Reinstate Monica Nov 9 '19 at 12:34
  • $\begingroup$ Hi! I'm trying to know if the different trials (levels) of my test are consistent. I want to know the fiability of my test. I will edit the question to make it clear. $\endgroup$ – fedegiov Nov 10 '19 at 20:20

Your Answer

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

Browse other questions tagged or ask your own question.