A Pearson correlation is a number between -1 and +1 that indicates how strongly two variables are *linearly* related.

This simple tutorial quickly explains the basics with outstanding illustrations and examples.

Read more...

This tutorial quickly walks you through the correct steps for a correlation analysis in SPSS.

We'll cover a quick data check, the assumptions, significance levels, APA reporting and more.

Read more...

A Spearman rank correlation is a number between -1 and +1 that indicates to what extent 2 variables are monotonously related.

This tutorial quickly walks you through the basics such as assumptions, significance levels, software and more.

Read more...

Kendall’s Tau is a number between -1 and +1 that indicates to what extent 2 variables are monotonously related.

This tutorial quickly walks you through some basics such as assumptions, significance and confidence intervals for Kendall’s Tau.

Read more...

Kendall’s Tau-B is found in SPSS under

Alternatively, both Kendall’s Tau-B and Tau-C can be obtained from

Be aware that this second option may report incorrect significance levels for small sample sizes.

Read more...

Cramér’s V is a number between 0 and 1 that indicates how strongly two nominal variables are correlated.

Because it's suitable for categorical variables, Cramér’s V is often used as an effect size measure for a chi-square independence test.

Read more...

Kendall’s Concordance Coefficient W is a number between 0 and 1 that indicates interrater agreement.

This tutorial explains the basic idea behind Kendall’s W and shows how to get it from SPSS.

Read more...

This tutorial presents a simple, menu based tool for computing confidence intervals for Pearson correlations given their sample sizes.

Since the tool does not require raw data, it can be used for correlations obtained from SPSS, other software or even journal articles.

Read more...

SPSS CORRELATIONS creates tables with Pearson correlations, sample sizes and significance levels.

Its syntax can be as simple as correlations q1 to q5. which creates a correlation matrix for variables q1 through q5. This simple tutorial quickly walks you through some other options as well.

Read more...

## THIS TUTORIAL HAS 4 COMMENTS:

## By Vaishali on October 4th, 2018

Hi

Thank you for such an easy and amazing way of reporting and making tables in SPSS in line with APA style.

If you can please can you share a similar kind of document for Partial correlation?

Waiting to hear soon.

-Vaishali

## By Ruben Geert van den Berg on October 4th, 2018

Hi Vaishali!

Thanks for the compliment. We haven't covered (semi)partial correlations yet. We aren't planning to do so any time soon either. It's a rather specialist topic for a small audience. We first like to expand our SPSS Beginners Tutorials.

Hope that helps!

SPSS tutorials

## By aya on September 2nd, 2020

thank you for the amazing and clear illustration. I have one question however, what test of correlation should I use to study association between one nominal variable and one continuous? For example, association between grade and occupation, marital status, smoking status... ?

thank you

## By Ruben Geert van den Berg on September 3rd, 2020

Great question!

You'd normally test if the mean scores of the quantitative variable are equal over all levels of the categorical variable with an ANOVA.

Now, the effect size for ANOVA is (partial) eta squared. This is the proportion of variance accounted for and thus has the same meaning as a (squared) Pearson correlation.

You'll also find this in the table presented in Which Statistical Test Should I Use? under "quantitative" and "nominal".

Hope that helps!

SPSS tutorials