Attempts to extend rpg theory into empirical research?

[matthijs]

A friend of mine is writing an article on RPG theory, but wants to make sure he's not just covering old ground, so he asked me to post this question:

"Have there been any attempts to extend rpg theory into empirical research, eg experiments designed to test crucial aspects of some theory, or questionnaires based on theoretical constructs?"

I can't think of any such research offhand - I know there have been some attempts in larp (see "Dissecting larp"), but in tabletop role-playing...?

[J. Tuomas Harviainen]

As far as academic research criteria are concerned, I'd say that any of the games designed through Forge discussion based processes and/or one of Edwards' models should be treated as /exemplary/ test run material.

Essentially, in them you have applied theories (i.e. the games themselves) that are undergoing continual test runs (i.e. playing) and have also been re-analyzed by other researchers (i.e. reviewed). In that, they're a definite exception from most tabletop games out there. And should thus be in some manner taken into account in your friend's research (even if it's just a nod in a footnote).

I'm not aware of any questionnaires within the tabletop field, though.

-Jiituomas

[Shreyas Sampat]

It's my impression that the RPG theory community has been specifically avoiding testing theory, to the point that theories are deliberately formulated such that testing is impossible.

I am only human, so my impression may be wrng in this regard.

But I do not think that the dearth of experimental testing is coincidental.

[J. Tuomas Harviainen]

But I do not think that the dearth of experimental testing is coincidental.

I agree with you this. In case the subject is of interest to anyone, here's an (intentionally provocative) article about this problem within larp theory I wrote for last year's Nordic larp conference proceedings book. A few people within the so-called Nordic theory circles (myself included) are currently trying to address the issue by organizing test runs on larp theory.

[matthijs]

How could I forget... Someone reminded me that in Sex & Sorcery, Ron mentions the use of player questionnaires. Don't know much about the context, though. Ron?

[Steve Verdon]

But I do not think that the dearth of experimental testing is coincidental.

I agree with you this. In case the subject is of interest to anyone, here's an (intentionally provocative) article about this problem within larp theory I wrote for last year's Nordic larp conference proceedings book. A few people within the so-called Nordic theory circles (myself included) are currently trying to address the issue by organizing test runs on larp theory.

Not to be a jerk or anything, but I think you've taken that quote a bit out of context.  The way I read it Shreyas is saying the theory is untestable not just that there hasn't been any testing.  For example, here are two behavioral hypotheses:

1.  People are loss averse.
2.  People do what they do.

The first is testable by offering people various lotteries and observing how they pick which lotteries to participate in.  For example, if people pick a lottery where the loss is $1,000 but with a probability of 20% vs. a lottery where the loss is $2000 but with a 10% probability (assuming benefits are the same) then they are loss averse.

The second one is untestable as you cannot derive anything testable from it.

If the above is true for some gaming "theory" then it really isn't a theory but a truism that is of little value in terms of developing testable hypotheses.

[J. Tuomas Harviainen]

Not to be a jerk or anything, but I think you've taken that quote a bit out of context.  The way I read it Shreyas is saying the theory is untestable not just that there hasn't been any testing.

Nope. I agree with Shreyas about the basic problem. But in my opinion the cause of untestability is a compound flaw created by a lack of initial outside testing. New material is written on an untested base, eventually leading to the entire theory being "ready" but no longer possible to validate. Not by intent, but by default.

The test process I was refering to is intentionally analyzing such theories with the idea of testing them despite their inherent untestability. By taking initially just parts of inaccessible theories into examination until the number of preliminary tests has provided enough data to support a full-scale test. And doing it all in connection with discussing the design intents of the theory and the results of the tests with the author of the said theory.