Anova or correlation test?

#1
Hello everyone

Quick question. Participants of mine had to fill in likert scale questions of three sections: attitudes towards French English and Romanian. I have collected the median of each participant (so every participant has a number between 1 and 6). I want to find out if there is a correlation between background variables such as age and time spent abroad. The medians are continuous data and the background variables ordinal.

What test do I use? Kruskal-Wallis or Spearman (i have an uneven distribution so cannot opt for the normal variants).
 
Last edited: