There’s a pretty strong correlation between the population of a US state and the number of times it is mentioned on Jeopardy!
The data covers all episodes up until 2019 which is about 366,000 clues/solutions. I wasn’t sure how to best represent in a single visual so the dashboard is a little cheesy. I used Datawrapper for this particular project, though, so everything is interactive for those who want to look closer (for all the interactive things, you can see [https://shinycharts.substack.com/p/jeopardy](https://shinycharts.substack.com/p/jeopardy) — no sales pitch or need to subscribe, it’s just where I host some fun-for-me analyses).
What’s really interesting to me is seeing the deviations from the trend line; like California being so far above the line since it implies less trivia generated per person. The logical reason is because of history so I’d be curious to see something like cumulative population or a weighted over time population vs trivia since trivia does look backwards too.
Can you log-scale both axes?
does it filter out “new york city” related responses that are just given as “new york”?
![[OC] US State Names as Jeopardy! Clues or Responses and Population Correlation](https://www.europesays.com/wp-content/uploads/2024/08/vmed09nh92ld1.jpg)
4 comments
There’s a pretty strong correlation between the population of a US state and the number of times it is mentioned on Jeopardy!
The data covers all episodes up until 2019 which is about 366,000 clues/solutions. I wasn’t sure how to best represent in a single visual so the dashboard is a little cheesy. I used Datawrapper for this particular project, though, so everything is interactive for those who want to look closer (for all the interactive things, you can see [https://shinycharts.substack.com/p/jeopardy](https://shinycharts.substack.com/p/jeopardy) — no sales pitch or need to subscribe, it’s just where I host some fun-for-me analyses).
What’s really interesting to me is seeing the deviations from the trend line; like California being so far above the line since it implies less trivia generated per person. The logical reason is because of history so I’d be curious to see something like cumulative population or a weighted over time population vs trivia since trivia does look backwards too.
Can you log-scale both axes?
does it filter out “new york city” related responses that are just given as “new york”?