Mid Term Report

It’s been five months since we shared an update with you. One of the reasons for that is that we have been extremely busy here at What Were Comics? headquarters as we have begun coding in earnest and generally moving the process forward. What have we been doing?

Acquiring comic books. In our last update I noted that had more than two-thirds of our corpus in hand from working with three different stores. As you can imagine, a lot of that involved picking the low-hanging fruit – comics that are pretty easily locatable in some of the larger stores or at conventions. Since that time we have continued to buy comics from literally dozens of dealers across North America, including a very successful purchasing trip to WonderCon in Anaheim in March. After six months of concerted efforts searching for 3,563 randomly selected comic books we were missing only 207 of them.

Those 207 comics fell into two broad categories: those that we can’t afford and those that we can’t find.

On the first category, we bought our books in stages. We have always been looking for reading copies and whenever we found multiple copies of the same book we would buy the least expensive (so long as it was complete). I joked to dealers that I was looking for Golden Age comic books that the dog had thrown up on. Initially we would not pay more than $25 for any book in the corpus. Over the months we raised that to $50, and then to $100. Nonetheless, I defy you find a copy of Superman #21 for $100. So, we have placed the purchase of some of our remaining comics on hold for the time being (mostly comics from 1934-1941) and we plan to come back to them a little further down the road when we are more clear about our budget. We also had some small troubles getting reimbursed for some of our purchases, but that has all been resolved (thankfully!).

On the second category, we have found that there are some comic books that we simply cannot find. Some of these would not be valuable at all (early-1990s black-and-whites from series that are barely recalled), but after six months of consistent searching online and in person, we’ve called it a day on those. Thus, we have re-randomized the Sampling Frame to account for some of the missing books. Having done that, we were able to find 170 of these comics, meaning that we are missing only 37 books from our corpus, or just slightly over one per cent. Many of the missing 37 we can buy online now, but generally in the range of several hundred dollars. We’re thinking about it.

The great Tom Gauld, in New Scientist. If this ratio of data collection and graphics making holds true, we are in serious trouble!

Coding comic books. With so many books in hand, we finally got down to work this summer. Specifically, we are using a custom-designed tool to enter data about our corpus (thank you Anissa!). This data is collected in several broad areas: at the level of the comic book (Title, Issue Number, Publisher, Price, Page Count, and so on), at the level of the story, at the level of the page, at the level of the panel, and also including all non-comics elements (ads, letters pages). Working with our tools as they have been designed, it is optimal to do each of these levels at once.

So, we began with the comic books themselves. We now have preliminary data on 99% of the books in our corpus, or about twenty per cent of our total data collection is now complete and we can begin making some very preliminary calculations.

A (very) small portion of the data for the early-1940s as it looks in Excel.

In the fall term, we will be moving on to coding of the paratexts. Why jump to the final element? Mostly because our first wave of data collection has given us solid leads for a number of articles and chapters that have almost nothing to do with the actual content of the comics themselves, and the information about the rise and fall of advertising and letters pages will be a necessary element to those analyses of economic shifts in the comics industry over time. We hope to have the paratextual elements coded by the end of 2018. We will then move on to the story coding in the first half of 2019, with the page and panel work being done in late 2019 and 2020.

So, to sum up, just past the midpoint of the first stage of our project we have:

- Finalized the Sampling Frame

- Generated the Randomized Corpus

- Acquired 99% of the corpus physically

- Constructed our coding tool

- Finished the first (of five) coding levels

We are extremely excited with what we’ve accomplished to date.

In May, I gave the keynote address at the Canadian Society for the Study of Comics annual conference in Toronto (and, once again, thanks to that fine organization for the invitation!). We were just about to start coding at that time, and now, three months later, I realize just how much more information I could have shared if we were then where we are today. We are learning things at an exponential rate now.

I should also note that the first publications from this project will be out soon, and we will share them when we can.

Ugly industrial shelving, wonderful spinner rack, so many comic books.

Finally, we bought a spinner rack! All of the 3, 563 comics almost fit onto two industrial shelving units that are so common on the University of Calgary campus, but we were going to need one additional shelf. So we bought a spinner rack, which will eventually be used to hold the “to be coded” part of the collection on a rotating (get it?) basis.

You can get your own spinner rack here (we think it’s great).