I'm investigating the prevalence of a specific disease in 4 cohorts; symptomatic male, symptomatic female, asymptomatic male, asymptomatic females.

I've tested the samples and have positive/negative results for each sample:

I would like to check if age, gender, and symptomatic status has any influence on the detection of disease.

I'm not great at SPSS, how should I input the data and what test would I best use?


  • 1
    $\begingroup$ How to input data into SPSS is off-topic here. Please see advice in the Help Center on software-specific questions. Otherwise, have you looked at previous questions under your tags? This kind of question is routine for epidemiology or medical statistics/biostatistics texts and courses. $\endgroup$ – Nick Cox Jul 10 '18 at 11:27
  • $\begingroup$ Hi Nick thanks for your reply. Yes i have a rough idea of what to do but I was looking for a more experienced opinion to make sure I was not missing anything :) $\endgroup$ – Andrew Jul 10 '18 at 11:57
  • 1
    $\begingroup$ If your response is binary you would presumably looking at something like logistic regression. $\endgroup$ – Glen_b Jul 10 '18 at 12:24
  • $\begingroup$ As a start, can you provide four counts in a $2 \times 2$ table for Men; Symptomatic; Yes/No; Postive Test: Yes/No? There are various tests for such tables, binomial, chi-squared, and Fisher exact. (To check 'predictability'.) // Maybe you know how to use SPSS to get the counts. Once you have the table, software is not really necessary. // Then a similar table for Women, and another for everyone. That may not be the end of the analysis, but could give you a good idea what to explore next. $\endgroup$ – BruceET Jul 26 '18 at 5:09

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.