## Metadata - Author: [[karen-grace-martin|Karen Grace Martin]] - Full Title:: Five Ways to Analyze Ordinal Variables - Category:: #🗞️Articles - Document Tags:: [[statistics|Statistics]], - URL:: https://substack.com/redirect/38a2896d-fe3b-4d24-99ae-8422f30bfff7?j=eyJ1IjoiNDRpMmEifQ.txKr3BEB06jM7pp-5wphmyXof7jFdPvpfRX5kIjhK8g - Finished date:: [[2024-01-28]] ## Highlights > 1. Analyze ordinal variables as if they’re nominal ([View Highlight](https://read.readwise.io/read/01hn7xpxbp66zdcppmrrvqj83p)) > The biggest advantage to this approach is you won’t violate any [assumptions](https://www.theanalysisfactor.com/assumptions-of-linear-models/). ([View Highlight](https://read.readwise.io/read/01hn7xq84hdxszvjsa89jctb1r)) > Analyze ordinal variables as if they’re numeric ([View Highlight](https://read.readwise.io/read/01hn7xpzs3sh218attjgxe5tn3)) > This approach requires the assumption that the distance between each set of subsequent categories is equal. And that can be very difficult to justify. ([View Highlight](https://read.readwise.io/read/01hn7xq287ky0x027hd7f6k63n)) > 3. Non-parametric tests ([View Highlight](https://read.readwise.io/read/01hn7xr35gy1808arja5bxfjzv)) > Many non-parametric descriptive statistics are based on ranking numerical values. Ranks are themselves ordinal — they tell you information about the order, but no distance between values ([View Highlight](https://read.readwise.io/read/01hn7xrckvfd809n8k3zjjnnc9)) > Common rank-based non-parametric tests include [Kruskal-Wallis](https://www.theanalysisfactor.com/non-parametric-anova-in-spss/), [Spearman correlation](https://www.theanalysisfactor.com/r-tutorial-pearson-spearman-correlation/), [Wilcoxon-Mann-Whitney,](https://www.theanalysisfactor.com/what-is-mann-whitney-u-test/) and Friedman ([View Highlight](https://read.readwise.io/read/01hn7xxq0h9hsq462dxsjdj0kr)) > 4. Ordinal logistic & probit regression ([View Highlight](https://read.readwise.io/read/01hn7xyze4vzhm5nm1da6y6agq)) > These models are complex, have their own assumptions, and can take some practice to interpret. But they are also sometimes exactly what you need. ([View Highlight](https://read.readwise.io/read/01hn7xze06k2ct99mpzdsyk522)) > The basic idea is a rank transformation: transform each ordinal outcome score into the rank of that score and run your regression, two-way ANOVA, or other model on those ranks. > The thing to remember though, is that all results need to be interpreted in terms of the ranks. Just as a log [transformation](https://www.theanalysisfactor.com/the-difference-between-link-functions-and-data-transformations/) on a dependent variable puts all the means and coefficients on a log(DV) scale, the rank transformation puts everything on a rank scale. Your interpretations are going to be about mean ranks, not means ([View Highlight](https://read.readwise.io/read/01hn7y08xvvd1gtt2hpbt0fsx8))