Academic Study of Tabletop Wargamers: The Results are in! Part I: the Survey

Particular stereotypes I wanted to test in this study include the following:

American tabletop wargamers are…
… male
… white
… upper middle-class
… politically conservative
… politically disinterested / apolitical
… “geeky” or “nerdy”
… unshowered, smelly, or exhibiting poor personal hygiene
… etc.

These stereotypes listed above were some of the most commonly referenced ones in the case study interviews, but there were many more that could be listed – and many more that can be tested on the collected data. Some of the stereotypes were more readily testable than others; I will discuss this as it comes up later (e.g., how do you test personal hygiene on an online survey?).

The Survey

I imagine many reading this article remember taking the survey (thanks for doing so!). The survey was completely online, built with the Qualtrics survey design program. The survey itself was based on the information gleaned from the case study interviews, as well as from a literature review. Both the online survey and the case study interviews were approved by the College of William & Mary’s Institutional Review Board Protection of Human Subjects Committee. The survey included questions about demographic information, hobby participation, and political and policy preferences. When the survey went online, it consisted of 58 different questions (not every participant was asked every question) and could be taken by each respondent once during its seven week accessibility (mid-November to early January). A large percentage of the survey responses were taken on the opening day of the survey period; the most common time for participants to take the survey was between 6:00 PM and 8:00 PM.

Within this seven weeks, the survey collected an impressive 2,724 responses! Of these, 2,321 were usable surveys: the several hundred unusable responses were those who did not accept the confidentiality agreement, were not at least 18 years old, or were not citizens of the United States. After taking out the unusable responses, the completion rate on the survey was very high (greater than 90%).

Before we continue, a quick word about some limitations of the survey. As I discuss the results of the survey below, my most important assumption is of a random sample. If we assume the survey was a random sample of the population (American tabletop wargamers), because of the high number of responses (greater than 1000) we can reasonably infer the data are representative of the population as a whole. The biggest flaw of this assumption in my mind is that the survey was only online, and although it was posted on many blogs, forums and websites – some aimed at fantastical wargaming and some oriented towards historical wargaming – it may not have reached a significant portion of the tabletop wargamer population if any significant portion does not visit those sites (or for a variety of reasons chose not to take the survey). Similarly, the survey was entirely in English, and thus might not have reached any American wargamers who were not comfortable taking the survey in English. We should keep these limitations in mind, but while they serve as a grain of salt, I think the results are still reasonably accurate to the population. As always, more studies in the future will help reinforce, explore or correct these findings.

Of Race, Gender, and Wealth

Before I wrap up this first part of the series, let’s investigate the first few stereotypes listed above. Assuming the sample is representative, American tabletop wargamers are both overwhelmingly male and overwhelmingly white. Out of survey respondents, 97% were male (only 38 out of 2,321 were female). Eighty-six percent of respondents indicated they were white; the next-most chosen category (“multi-racial”) was selected by only 4%. Based on these responses, we could characterize the stereotype of wargamers as white men as reasonably accurate. The stereotype that tabletop wargamers in the U.S. are in the upper middle class is less clear. Although 65% of respondents indicated income over $32,000 a year (and 12% indicated annual income over $105,000), income levels were very spread out in the data, and a significant portion of respondents indicated annual income under $32,000. The “upper-middle class” stereotype is thus less well-supported by the survey data, but it is not clearly debunked either. Asking about income levels does not capture levels of family wealth; many college students, for example, would have low income but still be considered middle or higher socio-economic class.

Next time we’ll look more at the demographic information collected by the survey as well as more on the stereotypes! In particular we will look at some surprising results from the survey concerning demographics like religious affiliation and educational attainment.

If you would like to contact me, please feel free to send me a message in the BOLS Lounge; my username is MajorMcNicol.

Part 2 coming soon.  Have at it.

A quick but important note of thanks – the successes of this project relied heavily on many other people, including: my faculty advisor, my family and friends, and the several wargamers who helped out in different ways to make the survey better. The failures, by comparison, can only be attributed to myself. Also, it is important to acknowledge everyone who took the online survey, providing me with enough data to reasonably investigate! And thanks must go to BOLS for publicizing the survey and posting these results, as well. Thank you to everyone who participated in the survey, and many thanks to those who supported and assisted me in its design and implementation.