Monday, October 13, 2014

Results from the videogame preferences and morals survey

Back in March I asked for volunteers for a survey about videogame preferences and morals. Then I had a baby and I kind of forgot about the results for a while. The blog link to the initial call for responses also has information on what questionnaires we used.


We solicited responses from Twitter and Facebook and a forum I frequent and received 75 responses. About 70% (53) of respondents were between the ages of 25-34. 29% (22) were women.

 Some people complained about how I switched the likert scale halfway through (i.e., 1 became "strongly agree" rather than "strongly disagree"), so after about 50 responses I added a heads-up about that. The reason for this switch was to preserve the wording of the original questionnaires. You can see everyone's responses here (yay for open source social science). There are a few different sheets on the results spreadsheet. We used the most conservative method of converting the videogame preferences to subscales (detailed in this master's thesis).

My friend Michael Davison helped with the data analysis (I have done similar analyses in SPSS, but it's been a long time I don't have access to that software anymore). He used the R project software to compute correlations between the videogame preference subscales and the morals subscales and within the subscales themselves. This was an exploratory study so we didn't try to predict the results, although I kind of thought a preference for shooting games would correlate negatively with harm/care, or possibly positively correlate with in-group loyalty.

Probably the most interesting correlation was that preferences for adventure and puzzle games were correlated with the fairness/reciprocity moral subscale. A preference for adventure games had a .33 correlation (p < .01) and a preference for puzzle games had a .31 correlation (p < .01).


I was wondering why there weren't any other interesting correlations between the moral subscales and the videogame preferences. One possible reason is that we had a better range of preferences for adventure and puzzle games (14-100 and 31-93 respectively), whereas the range for Action: shooting was 22-74. It's possible I didn't have enough participants who really liked shooting games, or that there is simply no correlation between liking shooting games and wanting other people to suffer.

As for why a preference for adventure and puzzle games correlates with wanting justice and things to be fair, I'm not sure. Perhaps a desire for fairness and a penchant for puzzle-solving stems from a desire for things to be logical or predictable?


I was hoping to be able to learn the R software and figure out how to compute Bayesian t-tests (this article inspired me) or a factor analysis and try it out with these data, but I'm not sure if I'll get around to it. If you would like to, go ahead!

No comments:

Post a Comment