I'm doing a paper and need to find a negative or positive relationship between attitudes (5 option likert scale) and level of education (5 options). Do I do chi-squared, spearman's rho or Kendal's tau. Please help.
Kendal's tau (or gamma would work). Chi-squared only tests significance of the association; it does not directly assess the strength or direction of association. Spearman's rho is most useful when the cases are all ranked from highest to lowest, with few ties. In your situation, there will be a large number of ties on both variables, as the cases are clumped into only 5 approximate ranks. Therefore, tau (which takes ties into account) or gamma (which ignores them) would be your better options.
Thank you for the replies, very much appreciated! The research question is, "are people with higher levels of education more accepting of LGBTQ?" and we testing education level with acceptance level.