Just today, I’ve been alerted from Johannes Breuer (University of Cologne) that his paper on the longitudinal relationship between videogame use and sexist attitudes has just been published online at Cyberpsychology, Behavior and Social Networking. Along with the mention of twitter exploding like crazy, so I dropped everything and reviewed it. The paper is co-authored by Rachel Kowert, Ruth Festl and Thorsten Quandt.

The good news is that the paper is published in Cyberpsych and thus is quite short for my review.

The authors’ theoretical angle for their longitudinal study is cultivation theory (see Wikipedia). The theory posits that long-term exposure to media content can affect our perception of the world. For example, a long history of seeing Black criminals shown on the news generate and slowly crystallize the perception that Blacks are mostly criminals. Most of cultivation theory research focused on television effects, but there are few cultivation research on videogames and from those studies they reveal not much support for the theory, that is some studies found partial support or weak effects.

Method

Participants: 824 German respondents who were part of a large nationally representative telephone survey ran by the Social Foundations of Online Gaming project (SOFOGA). The respondents were first surveyed in 2011 with 4500 respondents, 2012 with 2199 respondents and finally on 2013 with 902 respondents. These numbers differ due to attrition effects typically for a longitudinal study (see Wikipedia). The relevant data for this study was used for Time 1 (2011) and Time 3 (2013).

Measures

The study was part of a large nationally representative telephone survey, it also means the respondents were answering all sorts of questions. Some of the questions were related to other studies, such as Rachel Kowert and colleagues’ emotional sensitivity and friendship study (see blog post). Remember all questions were asked through the telephone, imagine how long each respondent had to listen and answer questions.

Sociodemographics: age, sex, and highest education within the German education system.

Videogame use: How often they play: every day, several times a week, several times a month or less often. How many hours per day, week or month they play on average. They also indicated how much they like each videogame genre on a 5-point scale, such as First-person shooters, action games, role-playing games, etc.

Sexist attitudes: 3 items answered on a 5-point scale. The items are: “The man should be responsible for all major decisions made in a family”, “In a group of male and female members, a man should take on the leadership”, “Even if both partners work, the woman should be responsible for taking care of the household”. Very clear sexist statements, little room for ambiguity.

Results

They analyzed the data by a cross-lagged structural equation model. Not going into the statistical details as those are interested would probably have access to the full paper. Here is the graphical representation of their findings.

The upper number represents the standardized coefficients for female respondents and the lower number represents the standardized coefficient for male respondents. The model accounted age and education as control variables and is not shown to keep the graph simple and clean.

So what does the graph mean? Sexist attitudes and videogame use is stable across time. A respondent’s sexist attitude in 2011 is relatively similar in 2013 (.60 for female and .74 for male), so is their level of videogame use (.64 for female and .57 for male). Sexist attitudes and videogame use are not correlated with each other in 2011 nor in 2013.

What are the effects of variables in 2011 to variables in 2013? That is does videogame use in 2011 affect sexist attitudes in 2013? Just a negligible effect for male respondents (-0.08) in that videogames use in 2011 is negligibly associated to lower sexist attitudes in 2013.

Discussion

The take home message is that the cultivation effects of sexist attitudes from a general use of videogames over a three year period from a German population has not resulted in any appreciable changes in sexist attitudes.

The study conflicts with previous experimental studies on sexist content and sexist outcomes. These experiments used very obvious sexist videogames with specific outcomes related to sexism (Dill et al., 2008, blog post), the survey did not ask any specific examples of sexist videogames, just videogames in general nor did the survey assess for the full spectrum of sexist attitudes. The experiments assessed for short-term effects whereas the survey assessed for long-term effects. Nevertheless, the study assessed for the cultivation of sexist attitudes from videogames as posited by cultivation theory.

The authors argued that factors, such as personal experience, peers and family would have a stronger effect on sexist attitude than video game content. Speaking of peers, IMO this can extend to peers in your videogame social network. Furthermore, cultivation theory cannot properly account for videogame effects because of videogames’ interactive nature that makes each play experience unique for each player. The authors remind you that while they failed to find support of cultivation effect of sexist beliefs from videogame exposure, they have not found a repudiation of these effects. They proposed instead to examine certain genres or individuals series (Dead or Alive for example) and specific aspects of sexism, instead of the three general items used in their telephone survey, something along the lines of gender roles, sexual harassment, etc.

The study has limitations to bear in mind. The study was funded for use in Germany, so limited generalizability to other countries. So, it would be great to see Americans ask for the same, and you’ve got researchers eager to do it like me. Second, respondents included adolescents as young as 14, so there are issues related to growing up and being more impressionable and other developmental issues. IMO, this study is analogous of taking photographs from a tall skyscraper down into the streets at three different time periods. You get a beautiful view of a lot of things, but not very clear if you try to focus on a single thing. This means we need a high resolution camera focusing on the most relevant aspects for sexist attitudes.

References

Breuer, J., Kowert, R., Festl, R., & Quandt, T. (2015). Sexist Games=Sexist gamers? a longitudinal study on the relationship between video game use and sexist attitudes. Cyberpsychology, Behavior, and Social Networking. doi: /10.1089/cyber.2014.0492