Original reddit postGoogle Sheets raw resultsGoogle analyticsShort resultsFull ResultsGithub linkHey I finished this finally. There were 3000+ responses to the survey. Thanks for your help everybody! The results are posted above.* If you (for whatever reason) want to see all 3000+ responses click the raw results link.* If you are only interested in the salt-fest of best girl wars, see the analytics link.* If you want a short summary of the most significant correlations, see the short results.* If you want to see the full output click the full results link (fyi this is a 1.6MB txt file. If you want something from here, use ctrl+f).* If you want to see/download the source code (java), click the Github link. I realize that more can be done with all the information I got (especially the have you read the manga questions) so feel free to improve it if you want.The script calculates the percentage of people that voted for Character A that also voted for Character B. It then subtracts the percentage of the total vote Character B received.Using this"+39.024 percentage points between Tamura, Manami and Kushieda, Minori"as an example. If Minori got 14.2% of the vote, then 53.224% the people that voted for Manami also voted for Minori. (I don't know the actual percentage Minori got. I pulled the 14.2% from the analytics, which did not count people who didn't respond to the question in the total vote. It was merely for the sake of an example.)Thanks for taking the survey. I think the results are interesting. Hopefully you do too! http://ift.tt/1oFypoW
No comments:
Post a Comment