Unlike the ranking methods from previous parts of the series, Fractional Ranking can take non-integer values. Compared to Standard and Modified competition ranking methods, which either has a gap after or before the equal measure, Fractional Ranking takes on the average value for equal rankings.
Instead of using Multi-row Formula, we can use Summarize and Join tool for calculation.
For each Decade, we will calculate the rank by averaging Row ID. If there is no repeats, then the average Row ID will be the Row ID itself, else we will take the average for each repeating Count.
Then we can inner join the summarize output back to the original data set on the key columns: Decade and Count.
Note 1: for record with Row ID 4-6, we take the average rank of 5 for each of them.
Note 2: fractional ranking involves non-integer rank, hence data type need to be changed.
The key benefit of using Fractional ranking method is that the sum of all ranks is equal to the sum of distinct(ordinal) rank, where it is a required condition for a lot of statistical tests on rank.
What can we do with the rank? Just for sorting the data?
See part 4 for Rank Test.
Looking for more guides, tips and tricks in Tableau or Alteryx? Go check out the other blog posts from the Data School.