I have two sets of data: DES scores (self-report) and whether dissociation was experienced during an experiment (yes or no). My sample size is 83 (71 no, 12 yes). I want to know if high DES scores are related to those 12 people experiencing dissociation.

To my understanding this means I have a discrete independent variable and a binary dependent variable. Is that right?

What would be the best way to compare these?

I have an feeling independent t-test or Pearson correlation would be wrong. Would I need to do a logistic regression?

  • $\begingroup$ You want to compare two means - why do you consider t-test inappropriate? $\endgroup$
    – Tim
    Apr 18, 2016 at 20:31
  • $\begingroup$ @Tim Because that would be treating the self-report score as the dependent variable $\endgroup$ Apr 20, 2016 at 17:09

1 Answer 1


You should try a logistic regression, although the number of cases is low, so do not expect to much. You should also show the data, visualize, plot it. A stripchart with two lines would be nice, an example can be seen here: How can I draw a boxplot without boxes in R?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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