# SPSS: Which chi-square test should I use for one nominal and one ordinal variable?

#### Sandra22

##### New Member
My hypothesis is Gender and opinion on smoking is associated. I am using a likert scale (Strongly disagree..strongly agree) for the statement 'Smoking makes you look cool'.

In SPSS, I have coded gender as nominal and the opinion as ordinal. I have to use a chi-square test but dont which box to tick below that as i have one nominal and one ordinal variable.

The options are contingency coefficient, phi and cramer's v, lamba, uncertainty coefficient, gamma, somer's d, kendall's tau b/c, eta, kappa, risk, McNemar.

#### Karabiner

##### TS Contributor
Re: SPSS: Which chi-square test should I use for one nominal and one ordinal variable

Use the Mann-Whitney U-test, with gender as grouping variable.

BTW you do not have a Likert scale, but just 1 single Likert-type item.

With kind regards

K.