I tried it with linear regression but it pops up it cant accept string variables. I tried with correlate......Bi-variate still same thing. Both methods work for continuous variables. Any ideas please?
Recode variables to numbers or better move to R (http://r-project.org) and define as unordered factor (see eg. https://swcarpentry.github.io/r-novice-inflammation/12-supp-factors ). You can then check with non-parametric correlation (eg Spearman's) or even move to a variable inflation factor analysis
Normally it is better to consult a local statistician first. However, as suggested by Alexander Herr, recode and try again.
Also check these previous RG threads out
https://www.google.com/amp/s/www.researchgate.net/post/How_to_measure_collinearity_categorical_variable_and_not_normal_distribution_in_SPSS/amp
https://www.google.com/amp/s/www.researchgate.net/post/How_to_test_multicollinearity_in_binary_logistic_logistic_regression/amp
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS