10-27-2015 11:35 AM
I want to run an analysis to test for significane between two variables that are both categorized by World Bank Income (low income, low-middle income, high middle income, and high income) groups. The two variables of interest are 1) per capita alocohol consumption and 2) prevalence of non-communiccable disease. Wanted to get some advice on the best type of analyses I should use. Thank you!
10-27-2015 02:19 PM
Well, without any more information, I would check non parametric correlations
proc corr data=WorldBankData Spearman; var lowIncome lowMiddleIncome highMiddleIncome highIncome; with alcoholConsumption nonCommunicableDiseasePrevalence; run;