In the poll above: it should say "kennen ur a r-tard" and then "sorry for being such a r-tard, lol".
Here's the original thread:
http://na.leagueoflegends.com/board/....php?t=2934959
The initial results are attached here as a PDF.
Wow, Reds! I'm still on my weekend adventure, but I wanted to thank you for responding and reply to your comments.
|
This is a very interesting analysis. Thanks for sharing!
Keep in mind that some of your results might be surprising because you are working with a biased sample (participating bias). Your sample consists of LoL players that read the forums, and that found the interest to click on you survey. Also, you should not be shy about tossing out entire surveys from respondents that are obvious trolls. |
And yes, one thing I plan to do when I do a proper analysis is go through and delete duplicate results or the most obvious troll responses.
(on the "are you rethinking your use of any of these words" question)
|
This is great! I found it interesting that people reconsidered their use of the word "******" but not some of the more inflammatory words.
|
Anyway, I was quite pleased to see that so many respondents are apparently willing to rethink their use of language.
|
If you did not do so already, you should probably throw out the responses on this question for anyone that said "None of these" for the previous question.
|
|
You should also throw out any responses that answer "None of these" for this question AND mark one of the other answers, as they are invalid.
|
|
The responses for "unacceptable/acceptable" words is kind of startling. Though, there could be a troll bias, and age may be a factor (the most common age in the sample is 16 and younger).
|
|
51% of respondents do not autopunish for any word. This is powerful because to me that means that a player-supported system like that Tribunal is better than a computer program that might automatically punish someone based on their choice of words.
|
|
Note that many people that chose "I do not autopunish for any words" probably also selected "I do not autopunish for any of these words." Additionally, there is troll potential on this question as well.
|
|
As you mentioned, they are most likely trolls and their responses can be tossed... but it is true that some people do not identify with either ***.
|
|
Would be cool to see the differences between male/female and age (such as "teenager", "college aged", "adult" etc.) as that might explain some of the more startling results.
|
|
Interesting! Poll data is often tricky--beyond the non-random sample considerations that need to be made when interpreting data like these (as RiotDerivative noted), you might be interested in doing things like covarying out social desirability bias! There's some fun controls you can apply to help filter out troll responses and get even more 'clean' responses from non-trolls.
|
|
With a large enough set of statements, you could begin to randomize presentation and code based on statement characteristics (e.g., antagonistic vs. apologetic), then perform statistical analyses. For example, there seems to be a trend in these data toward apologetic statements containing hate speech being rated as less problematic. Though if you ask me, I don't like people throwing around 'gay' and 'rape' in those contexts.
Psychometrics! |
|
EDIT: Oh hey, I just noticed you're the same person who compiled the arachnophobia solutions for people. I'm noticing a Psychology theme
![]() |
I don't have much of a psych background beyond AP Psych in high school, but I did major in the social sciences in college, so I have some rudimentary survey/statistical competence which I've found quite handy. I do find psychology a very interesting field with a lot to say of relevance to fields I've spent more time studying, such as political science, economics, and law. Like with political science in particular, my hunch is that doing good psych work is tricky because you have to think very hard about how you operationalize concepts, and how you're willing to trade accuracy for simplicity (because "a theory that explains everything explains nothing", in the sense that it doesn't make the world any easier to understand).



