I was a bit suprised at not finding many really explicit dojinshi among my samples. Of course this was a very tiny little test, done just to see if the topic would be interesting to pursue further, so the margin for error is huge. I might just have grabbed the wrong dojinshi off the shelf. But I do get the impression that there's far more comedy and non-sexual stuff in dojinshi than you'd guess from the dojinshi news available in English.

There's more Harry Potter djs out there than I could ever examine, so number-wise, it would be perfectly doable to leave some out if the authors requested it. It's more of a methodological problem. Choosing samples so that they're pretty much representative of the entirety of dojinshi produced is a fuzzy enough science as it is. Dropping samples en masse for no reason other than that the authors requested it could really skew results, if a lot of authors do this...

Indeed, this probably should be one of the things I discuss in more detail. I can already see it's going to be an interesting and hugely annoying issue :) (Right now I'm also brainstorming on how to narrow down about a billion gazillion Harry Potter fanfics to a few hundred representative samples. In a few weeks I'll scream if you utter the phrase 'built-in sampling error' at me again, I'm sure.)

Aw, hell. Feeling very torn between fandom loyalty and academic rigor right now. I'm going to discuss this with my supervisors too... One of them is a Japanese professor who knows more dojinshi people than I do, maybe she can pull a rabbit out of a hat on this. I'll let you know if she has any good ideas as per communication with Japanese djka.