An Investigation of Song Popularity

Preregistration of analyses

Analysis #1

We want to investigate the relationship between a song’s popularity and its artists’ popularity . Specifically, we are interested in how the hotness or popularity of the artist relates to a song’s level of popularity.

We hypothesize that more popular artists are going to have songs that are more popular as well.

Null: The true proportion of songs that are popular and were created by a popular artist is no different from the true proportion of songs that are not popular and/or not created by a popular artist.

\[ H_0: p_1 - p_2 = 0 \]

Alternative: The true proportion of songs that are popular and were created by a popular artist is different from the true proportion of songs that are not popular and/or not created by a popular artist.

\[ H_A: p_1 - p_2 \neq 0 \]

Plan: We plan to make both song hotness and artist hotness into categorical variables based on a specific threshold. For example, we could say that a song is popular if it has a song hotness score of 0.7 and a artist is popular if they have an artist hotness of 0.7 as well. Once we do this, we will find the proportion of songs that have a yes in both categories and make them a group (popular song + popular artist), while all the other songs fall into the other group (either or neither). We’re looking to see if there’s a difference in the true proportions of these groups.

Analysis #2

We want to investigate the relationship between a song’s popularity and whether it was released before or after the creation of iTunes. Specifically, we are interested in how the creation of a streaming platform, such as iTunes, increased the popularity of songs. We would like to see if the creation of streaming platforms marks a change in the distribution of music in the industry, since after the creation of streaming platforms people don’t need to buy a whole album and instead can buy individual songs. In addition, songs are more easily-accessible.

We hypothesize that creation of streaming platforms marks a relevant change in increasing the popularity of songs.

Null: The true median popularity of songs after the creation of iTunes in 2003 is no different from the true median popularity of songs before the creation of iTunes.

\[ H_0: median_{before} - median_{after} = 0 \]

Alternative: The true median popularity of songs after the creation of iTunes in 2003 is different from the true median popularity of songs before the creation of iTunes.

\[ H_A: median_{before} - median_{after} \neq 0 \]

Plan: We plan to filter the songs by if they were released before 2003 and turn it into a categorical variable that has values of “before” and “after”. Once we do this, we will take the medians of the songs popularity before and after the creation of iTunes in 2003 and compare them against each other. We’re looking to see if there’s a difference in the true medians of popularity before and after the creation of streaming platforms.