Click here to see the Top 10's!Click here to see the complete results in spreadsheet format!First of all, thank you everyone who answered my poll. Like I said in the original post, I expected my post to be buried and get maybe a couple hundred people to notice it (let alone actually answer it), and by the end of it all I had roughly 2,000 responses! That being said, getting 100-200x as many answers as I had reasoned left me a lot more than I could chew. It took 2 weeks of doing barely anything else in my free time to sort through all of the answers and get them into a measurable format. I got through it, though, and I'm glad I did it. I think that it'll give me a lot of direction in what shows to watch to get content for my convention games next year, if nothing else.There are a few caveats I'd like to point out about these results.First and foremost, I sorted the responses through a Python script. It was pretty simple: I had a "replacement" file where I recorded what a "raw" answer said and what it was replaced with (i.e. all instances of "FMA:B" were turned into "Fullmetal Alchemist: Brotherhood"). Each answer was looked up in the list, and if it wasn't there, the script searched MAL, grabbed the top result, I checked to see if it was okay, then the script saved the new replacement to the list and wrote out the replacement to a new file of answers. (If you check the spreadsheet, this is the "replaced answers" page)That being said, when I got answers like just a first name (i.e. "Shinji"), I could only guess at what was the intended name (Ikari vs Matou vs every other Shinji). These guesses probably led to more than a few improper replacements. The replacement list didn't discriminate between questions either, so there will probably be a few characters in the voice actor responses or other similar situations. I tried to catch the errors when I saw them, but in no way shape or form was my system perfect.If, however, you followed my format, or at least something similar to it, I at least did a quick Google search to make sure your answer was correctly replaced. If you gave me a full name and a show title or the full show title, the fake statistician inside of me says you have about a 300x greater chance of your answer being correctly replaced.On a different note, It seemed most fair to me that larger series get consolidated under one name. My case-in point would be the "gone on too long" graph in the top 10's. "Naruto" covers the base series plus Shippuden plus any movies that may or may not have come through. If I had separated the base series from Shippuden, One Piece would have won by a landslide, since "Naruto" would have been dispersed into 2 peaks. Again, it's not a perfect system, but it seems the most fair in my head. Off the top of my head:All ~monogatari shows were replaced as "Monogatari Series"All Fate series shows were replaced as "Fate/*"Naruto and Naruto: Shippuden were replaced as "Naruto"Both seasons of Sword Art Online were replaced as "Sword Art Online"All Madoka series shows and movies were replaced with "Mahou Shoujou Madoka Magica"I'm sure there are more, these are just the bigger onesIf you feel like what I've done isn't good enough, the raw answers* I got from the poll are in the spreadsheet. I encourage you to prove to me that you can do a better job.* I did alter the data slightly, but only because a few people put in what appear to be phone numbers into the "Which voice actor could read a phone book..." question. Just in case they were real phone numbers, I erased them for the security and privacy of the number holders. I have not looked through the comments section completely yet, but if I find any sensitive information, I will be redacting it ASAP.On the note of the comments section, I will be reading all of them sometime soon. I definitely need a break from this project (I've only watched Luluco and the last few episodes of Galko-chan over the past two weeks; I've got a lot to catch up on). However, for sure by next Saturday I will have read all of the comments. If people feel like they need proof that I did, I'll set up a mailing list or something and I'll email everyone a link to a document that proves I read every one. However, if no-one is vocal about needing proof, I'll assume that you all trust me.===========================================As a final note: I'm releasing this data to the public, so you can use it however you'd like. I hope that all of this information will help people like me who enjoy putting on game show panels at cons to think up questions, or at least know which shows are good to pull ideas from.Thank you again everyone for helping me out with this poll! It really means a lot! http://ift.tt/269Vtw7
No comments:
Post a Comment