I'm conducting research on eyewitness identification. More specifically, testing whether there's a difference between simultaneous line-up method (6 suspect photos viewed at the same time), and sequential line-up method (1 suspect photo viewed at a time), and whether angry compared to neutral suspect facial expression affects eyewitness identification.

The study is between-group design: enter image description here There will be 12 different conditions.


  • Line-Up Method (2): Simultaneous and Sequential
  • Target Status (2): Absent or Present
  • Line-Up Variation (3): 1 angry + 5 neutral suspects, 6 angry suspects, 6 neutral suspects


  • Line-up Selection (6): Correct Rejection, Correct Identification Angry, Correction Identification Neutral, Neutral Filler, Angry Filler, Miss

I've been advised to use a chi square test, and I'm wondering whether this would be the most appropriate analysis?

I thought I could cross-tabulate Line-Up Method X Line-Up Selection for the observed and expected frequencies between the Simultaneous and Sequential conditions, and then also use split file to view the output by each condition. Is this an appropriate way of analyzing? Is there a better analysis I could use?

Any help would be fantastic. Please let me know if I need to provide more info.


1 Answer 1


I think your DV is "correct identification". When you have a dependent variable, regression is often a good method. Here, since your DV is a dichotomy (correct, incorrect) you should look into logistic regression. You would have several IVs: Line-up method, target status and line up variation. Each of these would be dummy-coded.


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.