Wer schaut sich die meisten Trans-Pornos an? (über Google Trends-Daten)

Note that some metro’s were missing search interest data for some search terms (particularly those in Alaska and Glendive, Montana). Missing data were treated as zero interest (and thus given the lowest z-score for that category) so their scores may be a bit distorted.

Each search term interest score for each metro was converted into z-scores. Those z-scores were then averaged to make an average z-score per metro. Those averages themselves were then converted to z-scores and plotted above. Red areas are above average, blue are below average.

The “average score” column in the table is actually the z-score of that metro’s average. So a score of +3 for instance would denote that that metro’s average score was 3 standard deviations higher than the mean average score. -3 would denote that that metro is 3 standard deviations below average.

Each search term column shows that metro’s z-score for that search term. So a score of 2.5 under “trans porn” would denote that that metro’s google trends search interest score was 2.5 standard deviations higher than the mean “trans porn” interest score.

All the data for search terms was taken from google trends. For instance, [here is the google trends page for “trans porn”](https://trends.google.com/trends/explore?date=today%205-y&geo=US&q=trans%20porn&hl=en)

Metro areas political lean was calculated by tallying votes of the counties within each metro. Note that the coefficient of correlation between Trump% and “average score” was 0.393 (this is really high for social science)

With 210 metros that’s a p-score of: 0.000000002

edit: There’s an error in the written formula on the bottom right for x bar. The data is correct but i forgot to update the label in the bottom right. I originally had used just the first three search terms along with weighting them by search volume. I later switched to a simple average because the weighted average would inflate scores of those missing data since they recorded 0 search volume for the terms they’re missing. I remembered to remove the weights from the forumla label but forgot to update it to 5 terms and denote that we’re dividing by 5 to get a simple average. So the formula for x bar should read:

x bar = 1/5 * ((s_1 – mu_s_1)/sigma_s_1 + (s_2 – mu_s_2)/sigma_s_2 + (s_3 – mu_s_3)/sigma_s_3 + (s_4 – mu_s_4)/sigma_s_4 + (s_5 – mu_s_5)/sigma_s_5)

[Here’s a picture of the correct formula for better clarity.](https://imgur.com/a/PaSNzGf)

Cajuns, huh?

Louisiana district 4 is looking pretty red. Y’all need to get Covenant Eyes downloaded on your phones right now.

imagine not watching trans porn. this map should look like the chinese flag

Mike Johnson is pumping the numbers up

Utah probably has more gay porn views than straight lol. I swear so many closeted guys there that just never figured it out. I remember one guy being totally serious looked me in the eyes and said “ I thought I was gay but I didn’t like ALL the positions“ and I was like ??? You tried them all?

Interesting. Rural black southerners and rural white northerners stand out


St Joseph, MO makes sense

well, well, well

Numbers thrown off by Transporns Georg who lives in a cave and watches 10,000 hours daily was an outlier and shouldn’t have been counted

I wonder how many shade tree mechanics working on transmissions have skewed this map?


I read that as Trains Porn…