Examining the Pipeline: Demographics of Undergraduate Students Studying Games

Monica M. McGill, Bradley University
Amber Settle, DePaul University
Adrienne Decker, Rochester Institute of Technology

Like it or not, diversity in the game industry is undeniably a current hot topic for a variety of reasons. Whether it’s about race, gender, or something less frequently discussed like political or religious leanings, strong opinions exist on all sides. As a social issue, calls for diversity may not be as well heard in the games industry where larger profits and higher returns on investments often take priority over larger social ideals.  But multiple industries have found that a key to broadening markets and market share is to include a variety of designers and developers in the process, a fact that remains irrefutable.  

The evidence for this keeps mounting. According to a 2008 Forbes Insights report, “companies with a strong commitment to diversity on average outperformed their peers with higher profit margins, and greater return on equity and assets.” A 2009 study on the effects of diversity on company performance showed that “companies with greater racial and gender diversity performed better in terms of sales, revenue, number of customers, and market share.” And a 2010 McKinsey report found that “companies with the highest share of women in their senior management teams outperformed those with no women by 41% in return on equity and by a 56% in operating results.”

Perhaps this is one of the reasons that, in a field where market share is becoming more and more competitive and exploring new avenues of growth is becoming increasingly important, this topic has become hot within the game industry. Even in 2005, the IGDA report on its diversity survey found that the typical game development professional is “white, male, heterosexual, not disabled, […] and agrees that workforce diversity is important to the future success of the game industry” (pp. 9-10). The report goes on to state that “… it is reasonable to believe that diversity does have an impact on the game industry and the products we create – either via broader markets and/or a means to attract future talent” (p. 22).

Ah, attracting future talent. That phrase certainly begs the question: what future talent are we attracting? And does the prospective talent pool differ in its composition than current game industry employees? Or are we attracting more of the same, trapped in a cycle like the one Anna Anthropy describes as "straight white developers [who] make games that straight white reviewers market to straight white players, who may eventually be recruited to become the new straight white developers and reviewers" (Anthropy 2012)?

Game degree programs at colleges and universities in the United States continue to grow and evolve. As competition and quality improves, these programs have started to become a viable pipeline for recruiting future industry talent. If the game industry as a whole becomes convinced that, like other companies across the world have recognized, that diversity matters and has an impact on market share, it’s important, then, to look at the composition of aspiring game developers at universities and colleges.

In 2012, we conducted a study to do just that—to find the characteristics of undergraduate students currently studying games. We looked at gender, race, disabilities, sexual orientation, and political and religious leanings and learned that, while some areas show marked improvement, other areas remain woefully lacking.

The Game Industry Employee Pipeline Survey

We created the Game Industry Employee Pipeline Survey in which many of the questions were taken directly from the 2005 IGDA survey “Game Developer Demographics: An Exploration of Workforce Diversity” and the 2011 IGDA Industry Survey with permission. The survey consisted of demographic questions, religious preferences, sexual preferences, political views, and disabilities. The survey also elicited information about student perceptions of diversity in the game industry.

Complete details about the methods used in this study are provided in our journal article, Demographics of undergraduates studying games in the United States: A comparison of Computer Science students and the General Population, recently published in the Computer Science Education journal. Data was collected from 261 students at four nationally-recognized universities in the U.S. The data was compared against the U.S. population as reported by the US Census Bureau in 2011 (United States Census Bureau, 2011a, 2011b, 2011c). When data from the U.S. population was not available from the Census Bureau, we reference data reported in peer-reviewed, published research.

Year of study, areas of study, and career aspirations

In the U.S., the majority of the participants (95.8%) are traditional students in the age range of 18-24, with the vast majority being full-time students (97.7%). Participants in their first year of studies made up 33.3% of the respondents, second year 26.8%, third year 19.5%, and fourth year 20.3%.

Respondents could choose from a broad selection of areas of study: Game Design, Game Software Development, Game Production, Game Art, Game Sound, Other, or Undecided. The top three areas of study were Game Design (41.0%), Game Software Development (31.8%), and Game Production (16.2%). Seven percent chose Game Art and fewer than 3% chose Game Sound. Several participants (12.3%) stated that they were undecided.

Sex, Race, and Ethnicity

The results of the study showed that 87.4% of respondents were male and 12.6% were female. According to the 2010 United States Census (United States Census Bureau, 2011a), the ratio of males to females in the general population is nearly 1 to 1, with males being 49.1% of the population and females 51.9%, clearly showing a gap in the representation of women.

Figure 1. Sex of game students in the US and the US population

Figure 1. Sex of game students in the US and the US population

95.8% of respondents were native English speakers. 259 respondents responded to the ethnicity question and 71.6% of participants identified as white, 10.2% as Asian, 6.5% as black, and 2.7% as Hispanic/Latino.

With respect to ethnicity, the data indicates that ethnicity is more equitably represented compared to the U.S. Population than the ratio is for gender (United States Census Bureau, 2011b). Blacks and Hispanic/Latinos are underrepresented, while Asians are overrepresented when compared with the US Population. 

Figure 2. Ethnicity of game students, computing students, and US population

Figure 2. Ethnicity of game students, computing students, and US population

Religious Leanings

255 participants responded to the religious preferences question, with 31 declining to specify. The majority (41.6%) of participants did not have a religious preference. 13.0% identified themselves as Roman Catholic, while 39.6% identified themselves as Christian (Other Christian, Church of Christ, Methodist, Episcopalian, Presbyterian, Baptist, Eastern Orthodox, or Lutheran). 1.5% identified themselves as Muslim, while 2.3% identified themselves as Jewish.

Compared to data reported by the Pew Forum on Religion and Public Life, the religious affiliations of the U.S. population differs significantly. Of significance, over 40% of game students claim no religious affiliation compared to only 16.1% of the population. 

Figure 3. Religious affiliation of game students and U.S. population

Figure 3. Religious affiliation of game students and U.S. population

Political Leanings

Questions about political preferences were also given. Participants mostly identified themselves as either liberal (27.4%) or did not care about political preferences (26.4%). Another large percentage (24.1%) identified with middle of the road ideologies. Only 4.5% identified with conservative ideologies. Seven participants stated other, with four (4.5%) self-identifying as Libertarian. The remaining three stated that their view depends on the issue.

Figure 4. Political preferences of game students.

Figure 4. Political preferences of game students.

Compared to the Gallup Politics poll reported in January 2012 (Gallup Politics, 2012), there are a number of extremely statistically significant differences between the political views of game students versus the U.S. population. 

Figure 5. Political preference comparison.

Figure 5. Political preference comparison.

Lesbian, Gay, Bisexual, and Transgender (LGBT)

255 participants responded to the LGBT questions, with 15 (5.9%) declining to specify (Table 6). Of those that did, 87.1% identified themselves as heterosexual. 5.5% identified themselves as bisexual with only four participants (1.6%) self-identifying as lesbian/gay. Though not shown in the table, two participants (0.8%) self-identified as transgendered.

According to the Williams Institute (Gates, 2011), 96.2% of the U.S. population identify as heterosexual, 1.7% as lesbian/gay, and 1.8% as bisexual. Compared to the population of game students, the lesbian/gay population is nearly identical; however, there appears to be a higher percentage of students who identify as bisexual rather than lesbian/gay, thereby decreasing the heterosexual percentage. 

Figure 6. Sexual orientation of game students and US population

Figure 6. Sexual orientation of game students and US population

Disabilities

Participants were asked to identify one or more medically diagnosed disabilities. This was a required field and 261 respondents selected at least one answer. The majority, 74.3%, of participants stated that they did not have a disability. Of the remaining responses, mental illness ranked highest with 8.4% (e.g. anxiety, obsessive compulsive disorder, post-traumatic stress disorder, bipolar, depressions, schizophrenia, etc.), with 7.3% of participants stating that they have a cognitive disorder (e.g. dyslexia, ADD/HD, specific learning disability, autism, Asperger’s, etc.). 1.1% reported they had a hearing disability and 0.4% reported being mobility impaired.

A second question was posed to those respondents who indicated that they had a cognitive disability to identify the one or more cognitive orders that they have. Of these, 5.7% stated that they had Attention Deficit/Hyperactivity Disorder (AD/HD) and 1.1% stated that they had a learning disability.

According to the U.S. Census Bureau (2011c), 78.7% of the US population do not have a disability, compared to 74.3% of game students, and 6.3% of the US has a mental illness, while 8.4% of participants do.

Blind and partially sighted constitute 3.1% of the participants and 3.3% of the population. Deaf/hard of hearing constitute 3.1% of the general population, but only 1.1% of the game students self-identified with being deaf or hard of hearing. Unfortunately, due to how the data is reported by the U.S. Census Bureau, categories do not match the categories in the survey, and thus data cannot be compared.

Summary

The results from this work provide new insights about the game industry pipeline. In many respects the results of our study align with the results from the 2005 IGDA survey of game industry employees and with information available on demographics in the science and engineering (S&E) workforce, and in the computing workforce in particular (S&E Indicators, IGDA 2005). In the S&E, computing, and game workforces, women, blacks, and Hispanics/Latinos are underrepresented and Asians have a higher presence than the general population.

What does this mean for the game industry? At this point, it means that in the short term, we are filling the pipeline with essentially the same types of people that are currently in the industry.  If we want to change the demographic makeup of the industry, it is incumbent on the industry (and the burgeoning academic institutions directly supporting them) to start the quest for change.  Without a concerted effort, the industry may not have a diverse group of skilled graduates from which to choose and the face of the game industry may not change. The path, after all, of least resistance is typically the path taken.

We leave our data for your consideration, and we invite the industry to decide whether a diverse workforce is important to increased market share and profits. For those industry leaders who believe it is, the next logical step is determine how to expand diversity in the pipeline.

Acknowledgements

We would like to acknowledge and extend our appreciation to the following for their time and resources in shepherding the IRB process at their respective institutions and recruiting students for participation: Briana Morrison, Southern Polytechnic State University; Jacques Carette, McMaster University; Mark Eyles, University of Portsmouth, United Kingdom; and Siobhan Thomas, London South Bank University.

References

Gallup Politics. (2012). Conservatives remain the largest ideological group in U.S. Retrieved October 9, 2012 from http://www.gallup.com/poll/152021/conservatives-remain-largest-ideological-group.aspx

Gates, G. (2011). How many people are lesbian, gay, bisexual, and transgender? Retrieved October 9, 2012 from http://williamsinstitute.law.ucla.edu/wp-content/uploads/Gates-How-Many-People-LGBT-Apr-2011.pdf

IGDA Curriculum Framework: The study of games and game development. (2003). Retrieved October 20, 2012 from http://www.igda.org/sites/default/files/IGDA_Curriculum_Framework_Feb03.pdf.

IGDA Curriculum Framework: The study of games and game development. (2008). Retrieved October 20, 2012 from http://www.igda.org/wiki/images/e/ee/Igda2008cf.pdf.

IGDA International Game Developers Association. (2005). Game Developer Demographics: An Exploration of Workforce Diversity. Retrieved October 15, 2012 from http://www.igda.org/game-developer-demographics-report

IGDA International Game Developers Association. (2011). Game Industry Survey 2011. Retrieved December, 2012 from http://www.research.net/s/IGDA_Industry_Survey_M2Research

McGill, M., Settle, A., and Decker., A. (2013). Demographics of undergraduates studying games in the United States: A comparison of Computer Science students and the General Population. Computer Science Education. Vol 23(2) 2013.  

Pew Forum on Religion and Public Life. (2008). U.S. Religious Landscape Survey. Retrieved October 9, 2012 from http://religions.pewforum.org/reports

Princeton Review. (2012). Top game design programs. Retrieved October 15, 2012 from http://www.princetonreview.com/game-design.aspx 

Science and Engineering Indicators. (2012). Chapter 3: Demographics of the S&E Workforce, National Science Foundation. Retrieved September 2012 from http://www.nsf.gov/statistics/seind12/c3/c3s4.htm

United States Census Bureau. (2011a). Age and Sex Composition: 2010. Retrieved October 9, 2012 from http://www.census.gov/prod/cen2010/briefs/c2010br-03.pdf

United States Census Bureau. (2011b). Overview of Race and Hispanic Origin: 2010. Retrieved October 9, 2012 from http://www.census.gov/prod/cen2010/briefs/c2010br-02.pdf

United States Census Bureau. (2011c). Prevalence of specific measures of disability among individuals 15 years and older: 2010. Retrieved October 9, 2012 from http://www.census.gov/hhes/www/disability/sipp/disab10/table_A1.pdf

United States Equal Employment Opportunity Commission. (2009). Annual Report on the Federal Work Force: Fiscal Year 2009. Retrieved October 19, 2012 from http://eeoc.gov/federal/reports/fsp2009/