Research talk:Master Thesis Cultural differences in motivations to contribute to Wikipedia

From Meta, a Wikimedia project coordination wiki

Yet another motivation study[edit]

Given the massive number of surveys given to Wikipedians, I'd like to know more about how this contributes beyond the current understanding. Can you give a list of related motivation related survey work and reasoning for why you can't use their results? I'm not trying to ask for a full literature review. A bulleted list of the form * <study reference> found <primary conclusion>, but we can't use their work because <reason> --EpochFail (talk) 18:42, 23 May 2012 (UTC)[reply]

Questions on scale and motivation[edit]

Hi Sjarlot and Nick. EpochFail is correct; there are quite a few motivation surveys out there already, even ones that compare motivations across languages. Here's a partial list (you can get most of these on the first page of Google results with the search string "Wikipedian motivation survey":
  1. http://wikipediasurvey.org/docs/Wikipedia_Overview_15March2010-FINAL.pdf (largescale, definitive, cross-cultural)
  2. http://dl.acm.org/citation.cfm?id=1297798 (pioneering, en-only)
  3. http://www.sciencedirect.com/science/article/pii/S0747563210000877 (theory-driven)
  4. http://mcdonough-dolmaya.ca/2011/08/24/wikipedia-survey-iv-motivations/ (just found this one; haven't read it)
  5. http://www.tandfonline.com/doi/abs/10.1080/15213260802669466#preview (haven't read this through, but the abstract looks promising)
Have you looked over these surveys? Is there any reason you couldn't use these data? I looked over your survey, and it doesn't seem to be closely tied to a particular theoretical framework. I imagine you could get rich comparative data by just reviewing the, first, second and fifth items on that list. The reason we want to know this is that surveys can be very disruptive (kind of like when you get a call from a telemarketer just as you're sitting down to dinner), so we want to know that there's something of value to science here, and that means a novel contribution of some sort. Surveys also have the potential to disrupt a lot of users, which puts a double onus on the researcher (whether they're researching Wikipedia or any other population) to make sure they're doing things right.
Regarding your methods, I think mailinglist distribution is a good idea (non-invasive). But I'm a little concerned about your plan to randomly spam the busiest Wikipedians, especially if you intend to deliver the survey to hundreds or thousands of users that way.
So, to review, my questions are:
  1. Have you read through the above publications?
  2. Is there any reason you could not use their data? (other than the inconvenience of tweaking your Master's thesis proposal :)
  3. How many editors are you planning to survey via userpage post? Per language, and in total? (this is perhaps the most critical question for me)
Also, a quick tip about this survey, and any future surveys of online communities that you run: I would strongly advise that you NOT make questions that elicit potentially sensitive demographic information mandatory. For instance, many people who would otherwise be happy to fill out your survey will not want to tell you their gender, which means you'll get a much lower response rate than you would otherwise. I learned this the hard way!

Cheers, Jtmorgan (talk) 16:53, 30 May 2012 (UTC)[reply]