The Grand Comics Database: User Survey Summer 2016
By Henry Andrews and Chuk Moran
The Grand Comics Database (GCD) has huge global reach and the best data on the history of comic publishing. We conducted a survey of those visiting the website and collected 416 responses in three days.
This survey reveals several things. First, the GCD is not reaching everyone it should. The GCD could do better reaching women, young people, those exclusively interested in modern comics, and regions outside North America and Europe. To a large extent, these populations overlap, with European visitors, for example, often being younger. Second, those talking on the GCD lists do not represent the current userbase, let alone the userbase that the organization could potentially have.
Therefore, we recommend changes to the site that would help rectify these problems without downgrading the experience of the large current userbase, without which the GCD would be nothing at all. In particular, we recommend improvements to the site's ease of use and further research.
The GCD could be attracting a wider range of visitors, which could increase site traffic, enhance quality of contributions, and serve the organization's mission of disseminating information on published comics of all eras and languages.
GCD visitors overwhelmingly read comics in English, though the site includes comics in many other languages. This chart uses data from account creation, rather than the survey, as it was similar in character but was just a larger set of data for the dozens of languages covered on the site.
While even the US comics market may have as much as a 46% female readership, the GCD's visitors are almost entirely men, with about 97% of survey respondents identifying as male. Unfortunately, this imbalance also makes it difficult to get representative data on the site's women users, as only about a dozen women completed a survey. Still, these respondents did tend to be noticeably younger, less concentrated in the US, more likely to use the site for academic work, and less likely to engage with the GCD's social media or mailing list.
Internationally, comics and graphic novels have readership across age groups, with a substantial portion under 30. While the GCD's mission is different from comic sales, site visitors are considerably older than the wider universe of comic afficianados.
Perhaps relatedly, almost all GCD visitors enjoy older comics, most also enjoy newer comics, but almost none read new comics exclusively. While we have not found comparative data for the broader comic markets, this extreme preference stands out.
The survey allowed respondents to give specific reasons for using the GCD and list specific forms of engagement with social channels. For our analysis, respondents were sorted into non-overlapping groups based on these questions.
| Group | Meaning | |
|---|---|---|
| Active | Posts to mailing list or Facebook | |
| Follows | Follows mailing list, Facebook, or other social media | |
| Not Social | Does not post or follow any channels |
| Group | Meaning | |
|---|---|---|
| Interactive | Indexes comics | |
| Research | Uses the GCD for research or writing but does not index | |
| Personal | Does not qualify for above groups |
While the GCD's volunteer membership comes from across the world, members of the board, mailing lists, and social media channels are not, of themselves, representative of all users. The GCD must work to represent the concerns and hopes of its current and prospective users. Doing this requires better understanding who uses the site.
The GCD's mailing lists are the soul of the organization, but only include a tiny fraction of all users. Recent votes on GCD policies are often determined by around 20 votes. In contrast, the GCD hosts approximately 100,000 unique visitors per month. The current number of members on GCD mailing lists is also a helpful referent, with each list numbering in the hundreds or less.
| Mailing List | Subscribers | |
|---|---|---|
| GCD Main | 420 | |
| GCD Chat | 204 | |
| GCD Tech | 151 | |
| GCD Policy | 91 | |
| GCD Board | 14 |
Those who are most active on the GCD's mailing lists are much more likely to index on the site than other groups.
Those active on lists and Facebook tend to be older and more concentrated in the middle of the site's broader age distribution.
Those active in the GCD's social forums also tend to over-represent the US and Canada. To put it another way, those from other countries tend not to engage socially with the GCD, and thus they are not proportionately represented in discussions on the GCD mailing lists or social media forums.
By asking a series of questions about posting to Facebook or GCD mailing lists, we were able to compare visitors by their level of social engagement around the GCD. Social channels aside from Facebook and mailing lists had very low numbers of followers from the group of respondents. Of all channels, Facebook lead in readership.
Researchers are essential to the GCD's mission to maintain and disseminate historical information on comics. Researchers in particular have the power to legitimize the organization for others and hold it to higher standards. However, outside of those who also index, a relatively small percentage of researchers actively engage socially with others about the GCD.
These researchers tended to read comics in more languages than others and were also more often interested in modern comics
The survey also included a free response question, "If there is anything else you'd like to tell us about the site and how you use it, please tell us here." This elicited a fair amount of praise for the site, as well as a number of requests that came up multiple times. Improving search was the most frequent request, with improvements for indexing and data interaction less common.
Of course these comments did not all ask for a generic improvement to search. Most were specific requests, including the following ideas:
The most common free form comment for the survey was to improve search, with 27 mentioning it. While many long-term uses have adapted to the existing search system, it is a ways off from state of the art best practices, and can easily be either too intimidating to learn or too complex to quickly yield a simple answer to a simple question, such as what was the most recent superman comic in the US.
GCD visitors speak many languages, but the website only offers versions in English, German, Dutch, and Swedish. A version in French, Spanish, Italian, German, and Portuguese might better serve existing visitors. Or, specially recruited volunteer teams could own language-specific variations of the site. Finally, a feature that would filter results to a user's preferred languages could simplify site navigation considerably.
Seven indexers commented that the indexing system is quite difficult to use, and this has been known by the GCD tech team for some time. This requires a substantial change to backend technologies, but is possible and worthwhile.
Followup interviews with researchers and other important user groups could greatly help the GCD understand where it is, and is not, fulfilling its mission. Our current research also did not reach those using the new “My Comics” site, who could be the subject of further work.
This survey measures visitors who felt motivated to complete a survey in English within a fairly short window during which it was open. This may over- or under-represent various populations. The language barrier is the most obvious, but for instance we are probably seeing fewer respondents who prefer other sites to the GCD simply because such visitors likely do not come here regularly.
Based on a series of questions about posting to mailing lists and the Facebook page, as well as following various social media channels, we grouped together those who said that they actively post as “Active.” Those who said they read the mailing list or followed some social media channel were grouped as “Follows.” Those who did not indicate any online social engagement were labeled “Not Social.”
A series of questions about reasons that the respondent uses the GCD were summarized in three non-overlapping categories of user type. Those who said that they indexed comics on the site were grouped together, regardless of other answers, as “Interactive” users. They interact with the GCD data set, and therefore should be regarded differently from others. Those who said they used the site for academic work, for research, or for writing were grouped together as “Research” users. Finally, those who did not give any of those reasons, but did use it for collecting or personal reasons were assigned to a final group of those whose use is exclusively “Personal.”
We created several regions to prevent a lot of tiny slices while grouping things geographically. Australia and New Zealand are grouped with the one respondent each from South Korea and the Philipines who both listed English as their only language. We've labeled this the “Pacific Anglophone” region.
The other regions are what you would expect, with the three largest country populations (U.S., U.K., and Canada) standing on their own. Note that there is only one respondent in the “North Africa” region, an active indexer in Libya who indexes comics in Arabic. Unlike with the Pacific Anglophone grouping, we decided not to put him in with Europe due to the language difference. There were no other respondents from Africa or Asia.
The data used for these charts is available as a sortable table.