People often talk about how much better music used to be in the “old days.” Using examples of artists from a given decade to demonstrate this, a person may point out iconic acts from the 70’s and bad artists from today. It’s human nature to speak nostalgically and assume that the past was better. But is that really true?

First, this is a case of selection bias, something that can easily be applied to the past. Music that survives for a long time is usually of great quality, and thus not a true indicator of all music released in that period. With that said, how can we put this to a test?

If we truly want to compare eras, we need to set up a controlled experiment. We need to take all music released in the 1970’s and the 2000’s (assuming those are the decades we want to compare). Then we take a random sample of 100 songs from each era and randomly assign 10 songs to each subject. After the subjects rate their songs, we average out the ratings.  Statistically, I would guess that a two sample difference in means t-test would not return a significant result. It’s fun to discuss and debate eras, but beauty is in the eye of the beholder.