Ten-Year Lookback

Having taken a three-year lookback at the demographics data in an earlier post to highlight relative data stability (see On Reliability of Data), I was asked to open up the time window and look a bit farther back into the Wargames, Soldiers, and Strategy's Great Wargaming Survey (GWS) annals.

Motivation

While recent changes over the last three years demonstrate relative stability in the demographic variables, what if we look back ten years instead of the most recent three years?  Past surveys have seen a lot of variation in both questions asked and data capture format so not all of the surveys are directly comparable.  Most of the survey years, however, do provide the basic demographic variables but some recoding and transformation are required to make direct comparisons.  Rather than transform all of the older datasets, three years will be selected from a ten-year period for comparison at roughly equal intervals.  This study looks at survey results from 2016, 2020, and 2025.  Will results show similar stability with a shift toward older gamers or will these data suggest more movement in the underlying times series?  If there are swings in demographic tendencies over these ten years, can any of these changes be explained?  

Total Counts

The graphic below illustrates the dramatic drop off in responses in 2025.  With only 5,005 responses, 2025's counts are roughly half of counts seen in 2016 and 2020.  Even in 2023 about 10,000 completed surveys were collected.  Was this reduction in surveys spread uniformly across all demographics or were certain groups targeted more than others?  Hopefully, examining the other demographic attributes can shed some light on answering this question. 

Prior Survey Indicator

While the percentages of those completing more than one survey remained similar for the 2023-2025 study, from 2016 to 2025 the percentages have practically reversed.  That is, in 2016 about 65% of respondents answered the call for the first time.  In the 2025 survey, about 63% of responses have taken the survey more than once.  Seeing the cadre of repeat respondents increase over the years suggests that completing the survey has become an annual event for many.  

Location

The mix of respondents' home countries has remained stable over the ten years from 2016 to 2025.  USA/Canada and UK/Ireland each make up about a third of responses.  Continental Europe/Scandinavia hovers around 20% with Australia/New Zealand coming in around 7-8%.

Primary Interest

Over the ten-year study period, the decrease in participation of those respondents claiming a primary interest in Fantasy/Sci-Fi grouping is undeniable.  Both Historical and Mixed categories picked up share of totals. In 2016, 31% of respondents listed Fantasy/Sci-Fi as their primary interest to only 22% for Historical.  By 2025, non-historicals as a primary interest fell to only 15% with Historicals coming in at 31% and Mixed increasing to nearly 54%.

Given past analyses have shown Fantasy/Sci-Fi interest is age-dependent (or at least correlated with age), will Age Group statistics confirm this shift away from gaming purely non-historicals as a primary interest?  Are those claiming a purely non-historical preference migrating to either the Mixed or Historical categories in subsequent years or dropping out of the survey altogether?  Is this tendency to shift away from non-historicals brought about by the survey sample or a shift in the interests of an aging wargaming population? 

Age Group

When examining age demographics, the ten-year statistics reinforce the tendency seen in the Primary Interest graphic.  That is, as sampled age increases, the proportion of gamers with a primary interest in non-historicals decreases.  

Survey samples show age increasing with each year's snapshot.  In 2016, less than 25% of the respondents were 61 and over.  By 2020, the percentage increased to about 35%.  In 2025, the 61+ cohorts made up almost 48% of total respondents.  Each survey reinforces the notion that older cohorts tend to reduce their exposure to non-historicals.  Now, we must ask if this is due to sampling bias or representing an actual greying of the wargaming hobby. 

Duration

After seeing Primary Interest and Age Group graphics, what would one expect to see from Duration or years in the hobby?  Given the earlier graphics, Duration results should come as no surprise.  The percentage of gamers having been in the hobby for 31 years or more showed nearly a 30% share in 2016, nearly 39% share in 2020, and a whopping 46% in 2025.  Again, are these shifts in longevity due to sampling or the result of the actual passage of time?

Final Thoughts

Well, final thoughts for this analysis.  This exercise raises a number of questions that are difficult to answer in a series of two-way summaries of select demographic attributes.  One question that pops into mind relates to the top wargaming period tallies.  Will a shift away from fantasy/sci-fi as a primary interest result in other historical periods climbing up the chart and dislodge some of the non-historical periods?  We will see.

Perhaps digging deeper into these data using multi-dimensional analyses or more longitudinal studies will shed light into some of these questions regarding possible shift of preferences and demographics.  Even without definitive answers to these questions, follow-on analyses with deeper exploration may provide interest and insight.  I, for one, find these studies both interesting and insightful.  This longitudinal study will stay in mind as I continue working through the 2025 survey results.  Many new and interesting topics remain to explore.