Week 2: Assignment

Week 2 Assignment

Instructions

Download: Spotify Data Download Spotify Data

Submit: The text describing your story and the picture of your graph.

Assignment Details

See the attached CSV file (like a text-based spreadsheet), which contains data about the top 50 most-streamed songs on Spotify in 2019, along with machine learning-generated measures about the song from http://organizeyourmusic.playlistmachinery.com/# Links to an external site. (which you should scroll down and read the variable descriptions before starting! Always read the data documentation!)

1. Download the attached CSV file

2. Use R to look at the data. You won't be turning any of this part in but you should take this opportunity to get familiar with R. Look at the data (and probably do some preliminary calculations) and think of a story that you can tell with the data. Let me know if you're having difficulties.

You can load the data into R by downloading the file, setting your working directory to the folder where the file is stored (Session -> Set Working Directory), and then using the code:

library(tidyverse)
d <- read_csv('Spotify_top50.csv')

This will store the data in memory as a dataset called "d".

3. Write down, in a single sentence, the story you are trying to get across.

4. Draw, by hand, a visualization that helps get your story across. You can do this on paper or with MS Paint, draw.io, etc. Take a picture/screenshot of your sketch. Same idea as last time - doesn't have to be super exact but should reflect the data.

5. IMPORTANT: Keep some important data-communication and data-analysis notes in mind (1) the data includes some perhaps confusing terms that readers might not know, like "Valance", or even something understandable like "Popularity" but that is on an undefined scale- what's an 80 popularity mean? How much can your reader understand intuitively and how can you help them do it? (2) there are a lot of tiny little genres in here. Do we need all that little detail? Can we even say anything meaningful about these tiny little genres with one song? Be careful!

This data is from Kaggle datasets Links to an external site., which has a whole bunch of neat data sets, although it skews towards stuff for machine learning.

Grading will be based on fulfilling all the steps, having a reasonably accurate story, clearly reflecting that story in the diagram (I should be able to get the story just from the diagram without looking at your explanation, although annotations on the graph itself are fine), and following the principles we've learned in class.