1

I would like to test if there is a significant difference between the pre-test and post-test in a dichotomous variable (0 = "False", 1 = "True"). I'm using only one sample of 30 subjects.

So, my 2 variables are:

  • accuracy_pre_test (0 = "False", 1 ="True")
  • accuracy_post_test (0 = "False", 1 ="True")

Also, I would appreciate if you can give me a general idea of how to perform this test in SPSS.

gung - Reinstate Monica
  • 132,789
  • 81
  • 357
  • 650
mat
  • 639
  • 1
  • 4
  • 19
  • Could you tell us a little more about your sample? How many items are in your variable, and how many individuals are in the sample? – doug.numbers Jun 03 '13 at 16:35
  • 1
    @doug.numbers additional information provided in the OP. – mat Jun 03 '13 at 16:41
  • @Glen_b Thanks for your answer, however do you think there is another way so I dont have to rearrange the data? – mat Jun 03 '13 at 17:00
  • 2
    Actually, mat, sorry, it just occurred to me that your data are paired. You don't want that you want a McNemar test. – Glen_b Jun 03 '13 at 17:03
  • 1
    Here are some relevant answers relating to that - [1](http://stats.stackexchange.com/a/31578/805) [2](http://stats.stackexchange.com/a/55527/805) [3](http://stats.stackexchange.com/a/14453/805) – Glen_b Jun 03 '13 at 17:13
  • @Glen_b I think I'll perform a McNemar test. Thank you for your help! – mat Jun 03 '13 at 17:21

0 Answers0