I have an Excel sheet with n columns, these columns contain info about the students. For admissions we have the score of test scores in multiple subject areas, scores from an interview, and scores of a written test and comprehension test. There is a column which contains student's academic level (High, M.High, Middle, M.Low, Low). I want to compare the last column with the others variables and see whether there are common features that passing students have in common.

Is there software for this? If this can be done with excel, how can I do it? Does SPSS provide this kind of analysis?


2 Answers 2


A good place to start would be to look at the Spearman's Rank Correlation between your dependent variable (academic level) and your individual independent variables (other columns). This should give you a basic indication of whether any of your columns are correlated with academic level. This should be straightforward to implement in both Excel and SPSS.

If you wanted to go a step further you could set the problem up as one of Multinomial Logistic Regression. This would allow you to build a model to directly attempt to predict academic level from your other variables. I'm certain there's probably some way to do this in Excel, but SPSS can definitely handle it.


You can try some form of correlation or you can try a dimensionality reduction in a prediction model. In this case what is left are the important features for the prediction problem.

See https://www.knime.org/files/knime_seventechniquesdatadimreduction.pdf

-- Rosaria


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.