Session Ready
Exercise

Comparing rankings

A number of organizations rank universities on factors such as research, teaching, and student satisfaction, among other factors. Three such organizations include the Times Higher Education World (THEW) University Ranking, the Academic Ranking (ARW) of World Universities, and the Center for World (CW) University Rankings. However, it's common for these organizations to give different ranks for the same university. How can you quantitatively measure the agreement or disagreement between rankings? One approach is to use a non-parametric rank correlation coefficient such as Kendall's tau.

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.

Instructions 1/2
undefined XP
  • 1
  • 2
  • Compute Kendall's tau rank correlation coefficient between columns thew_rank and arw_rank.
  • Compute Kendall's tau between the thew_rank and cw_rank columns.
  • Compute Kendall's tau between the arw_rank and cw_rank columns.
Loading