ZScores Screen 8: Using Standardization for Comparisons

Hi there

I’m just confused why a negative z-score is considered better than in terms of a positive one in this specific example.

The description before the code assignment states:
One thing the two systems have in common is that the index is directly proportional with the house
quality — a lower index means lower quality, and a higher index means higher quality.


Average houses will have z-scores around 0 in both distributions, good-quality houses will have z-scores significantly greater than 0, and low-quality houses will have z-scores significantly lower than 0

So this leads me to believe that positive scores will be considered better than negative scores. But the solution states the opposite?

print(houses[['z_1', 'z_2']].head(2))
       z_1       z_2
0      NaN  0.429742
1 -0.93592       NaN

better = 'first'

Doesn’t the negative z-score reflect that the house is ~1 std below the mean showing that is below average in rating?

Hi @cole.regnier, the solution says that the first house is better, not the first z-score.

print(houses[['z_1', 'z_2']].head(2)) outputs two rows: the first row corresponds to the first house, and the second row to the second house.

Let me know if that clarifies things for you.