Advertisement

Surveys for senior Thesis

Started by November 25, 2014 09:34 PM
1 comment, last by slayemin 9 years, 11 months ago

Hello all. I don't know if this is the right place to put this, but I thought I might give it a try anyway. I am doing a few surveys for primary research for a senior paper I am working on and I was wondering if you all would help me out on this and participate. One is on Cormac McCarthy's The Road, one is on ALfonso Cauron's Children of Men and the last is on the game Fallout 3. The one on Fallout might interest you most, but feel free to participate in any and all. They are about 7 questions long, so not long, so it shouldn't take long. Anyway here are the links.
The Road
https://www.surveymonkey.com/r/8XDCRNR
Children of Men
https://www.surveymonkey.com/r/8XSFY7M
Fallout 3
https://www.surveymonkey.com/r/8753V6H

Thank you so much for reading this and participating. You have my eternal thanks.
gambit924

Hey all. This thread is probably going to come down soon, because I guess it's not the best way to ask people about books/movies/games, so just wanted to let you know. Anyway, no worries. Laters!!

Advertisement

Just a heads up: Your approach to gaining insight is very flawed.

Your first mistake is by creating a survey which only contains 7 questions each. If you don't have that many questions, it's probably more valuable to ask open ended questions in a different format. The purpose of a survey is to send it out to a defined sample size which would accurately represent a population. How many respondents do you need for your sample size to get a fix on the characteristics of a population? Usually its in the hundreds, but the sample size is a function of the population size. Surveys are efficient for getting a vague idea on what the population might be like, but to really get a good idea, you'll want to do a much more focused study... And, why does the opinions of a general population matter for this? (I suspect it doesn't)

Your second mistake is by using leading questions and pre-canned responses to those questions within your survey. You ask some very specific questions regarding the content of some media and I only get four choices to choose from? What if my response doesn't fit any of the four choices you gave me? This is the flaw of using multiple choice answers for open ended questions.

Your third mistake is going to be "selection bias". You will only get responses from people who feel strongly enough to take a survey on your topic. Those people won't necessarily be representative of the population as a whole, so you'll get skewed results. This will throw off your margin of error very much, if you care about that.

All in all, 90% of survey results are bullshit masquerading as "scientific". For the survey to be statistically valuable, you *really* need to be very careful about how it is conducted. A lot of time needs to be spent crafting the survey questions to minimize bias, and a lot of surveyors don't put in the right effort to be properly scientific about it (that is what you want for a senior thesis, right?)

What you're looking for are some deep insights into the design choices made to build a post apocalyptic world by the creators of those worlds. Surveys aren't going to give you that -- you'd only get useless speculation by a handful of irrelevant people. What you really want to do is conduct interviews with people who have designed those types of worlds and ask them questions which get at the questions you're trying to answer. Of course, you can also do a bit of internet research to see if those questions have already been asked and answered in some form or another.

This topic is closed to new replies.

Advertisement