I have a normally distributed SPSS dataset with 1400+ cases. I want to create sub-groups based on "age in months" that are distributed evenly. The dataset ranges from 12 to 60 "age in months". Weighted average percentiles are 20 mos (5%), 23 mos (10%), 30 mos (25%), 36 mos (50%), 44 mos (75%), 52 mos (90%), and 56 mos (95%). SD is 10.4. What's the best approach to splitting this dataset into age subgroups for pairwise analysis?

  • 1
    $\begingroup$ The "Visual Binning" option in the transform menu does this. $\endgroup$ Nov 20, 2014 at 19:32
  • $\begingroup$ It seems worth asking if discretization of age is the best approach. I realize that's not your question, but it's hard to know the experience level of new users. Can you not do an analysis on the continuous version of the age variable? It seems like you're asking for inferential trouble by discretizing and doing pair-wise comparison. $\endgroup$
    – ndoogan
    Nov 20, 2014 at 20:15


Your Answer

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

Browse other questions tagged or ask your own question.