Comparing medians

Parametric tests tend to compare means, while non-parametric tests tend to compare medians. This is because if your data is skewed, the mean may not be a good measure of central tendency, and you should work with the median instead.

Each of the university ranking organizations you just looked at also included a "total score" for each university. You'll start by determining if the scores are approximately normal by using a histogram. If they are not, you will perform a Mood's median test to compare their medians. In this exercise you'll work with the CW and ARW university rankings.

A DataFrame with three different these three sets of rankings in rankings_df. The packages pandas as pd, NumPy as np, and the stats package from SciPy have all been loaded for you.

Plot a histogram of the cw_score column.

Exercise

Comparing medians

Instructions 1/3