I want to test a model using discriminant function analysis. My question, as the title states, is very basic: What is a discriminant function? That is, how can I interpret the different discriminant functions?

I'll briefly explain this question. I have several continuous variables, and a categorical variable (with $N = 4$ groups) that I want to predict. Following a discriminant function analysis in SPSS, the output states $N-1$ discriminant functions, that is $3$ discriminant functions. My question is how I can interpret these functions, so that I can understand how my dependent variables predict the categorical variable.

  • 1
    $\begingroup$ Did you attempt to read anything on the topic? For example, Wikipedia article and Case Studies under SPSS Help menu? So far, it is not very clear what you mean under interpret. You want to know what a discriminant is or how to attach a substantial meaning to it (from the input variables)? $\endgroup$ – ttnphns Sep 2 '14 at 13:10
  • $\begingroup$ Yes, I have. Not the articles you're suggesting, but Chapter 16 from Field and chapter 5 in Hair's Multivariate Data Analysis. I want to know how to attach a substantial meaning to my model from the input variables. $\endgroup$ – TvBalkom Sep 2 '14 at 13:29
  • $\begingroup$ How many continuous variables do you have? Is it a small manageable number (say, around 5), or a huge number (say, around 1000)? $\endgroup$ – amoeba says Reinstate Monica Sep 2 '14 at 14:02
  • $\begingroup$ managable; it lies between 8 and 12, depending on my choice of variables $\endgroup$ – TvBalkom Sep 2 '14 at 14:52

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.