Different Scoring scale in quintile?

[soliverc]

I was comparing three different scoring methods, mean median and quintile.

I noticed that somebody who has a recency of 381.0 will get a score of 1 with scoring='Mean' and a score of 5 if you use scoring='Quintile'.

[zark-stark12]

This is with regards to the distribution of the dataset and how values cutoffs are different.

However it does seem interesting to see this happening. I haven't encountered this using this package as of lately.

When you try to manually do it for Quintile and Mean, do the results match up? e.g. building the cutoffs for the mean imply taking the mean of the variable (recency in this case) and anything below or above (above in this case) would be classified as a 1, the iterate on the next subset on the half below the mean and apply the same exercise however it would be a 2 this time etc.

[soliverc]

I'm sorry I completely forgot about this reply. I don't use this package too often. I am back using it again now, I will try to find this scoring discrepancy.

[soliverc]

Just focus on these three customers. The scoring system for recency is reversed for each method. Version 1.0.1

Quintile Scoring:

Median Scoring