Slashiness vs Writing Quality Scatterplot #1

[cimorene]

(Big version here)

Please click for the bigger version. This is based on data taken from two polls performed by thefourthvine, with the data in my tables taken out on 22 Feb 2009.

This plot follows the same format as my earlier graph here, which, of course, did not have any Science behind it, only me squinting at the screen and placing the dots by the seat of my pants. I've reversed the axes in this one, however, so that better-written canons are at the top of the graph and gayer ones to the right this time. I prefer placing slashiness and quality on the two axes so that there's one data point per show and one show per data point, which lets the interaction be more easily seen.

(ETA: The clearest example of skewing I see on the plot is Highlander. LOTR might be skewed, too, by people's childhood memories or something, I guess. But Highlander? Look at it! IT HAS THE SAME QUALITY SCORE AS SGA. I'm pretty sure that it's not just my personal... weirdness, or whatever, that wants to ROFL at that. SGA is bad, but it's not that bad.)

It should be noted that mean is only one measure of central tendency and, at these sample sizes - which is to say they're all rather small but they vary from fandom to fandom - it's important to look at the others as well. To that end, here is the table I used to generate the graph:


(big version here)

Median and mode (ie, the most popular rating) are just as interesting as mean in their own ways. Scanning down the standard deviation columns, a few of those numbers will jump out as extremely large, representing a much higher degree of disagreement on the rating in question: the slashiness of Supernatural (no surprise where the primary pairing is incest) and the quality of Merlin (I'm kind of at a loss for this one but there it is), for example.

It's well worth checking out the results in the two posts of thefourthvine's where I took the data because you can easily see the bell curves for some questions, and distinguish others where the distribution shows much higher or lower consensus. (The Science of Slashiness & Moar Science!)

Comments

[mecurtin.livejournal.com]

I LOVE YOU AND SCIENCE. Now I have to go wash dishes, but I shall return to worship at your feet. Scientifically.

[cimness.livejournal.com]

Aren't statistics beautiful? :D

[emmuzka.livejournal.com]

I love it! By the way, I'm one of those making fan fiction program in Finncon 09, one of the programs being "Ten years of slash" Would you give your permission for using this in it?

[cimness.livejournal.com]

Sure, but I think you ought to wait! I'm thinking TFV is probably going to make some more polls and these represent kinda preliminary results!

[emmuzka.livejournal.com]

Sure thing! and of course I will credit the appropriate sources etc.

you wouldn't happen to know who made the old "fandom timeline" picture that is circulating the net? I would like to use that as well.

[cimness.livejournal.com]

Hmm, no. Maybe they have some info on that at FanLore. http://fanlore.org/wiki/Main_Page

[brownbetty]

Which timeline is this? Is it this one? Because if it is, that was me, and if it wasn't, I'd really like to see it, as a collector of such things.

[emmuzka.livejournal.com]

I was thinking about this:
(http://www.aijaa.com/v.php?i=3660320.gif)

but the one that you linked looks very interesting, too. And more up to date.

[brownbetty]

Oh. I made that one, too. But it's mostly guesswork.

[emmuzka.livejournal.com]

But it was something like the first of its kind, and for that, I applaud you :)

[copracat]

Hi, hope you don't mind me butting in. What do the red lines through the timeline mean?

[cimness.livejournal.com]

You probably didn't notice that you addressed this comment to emmuzka, and if you want it answered you need to bring it to the attention of brown_betty.

[hicaniplay.livejournal.com]

I think it had to do with fandom migration patterns. The red lines was how long the fandom lasted (whether fans stuck aound till the end or got disheartened/bored and moved on). In general. Maybe...

[copracat]

Oh, yeah. That makes sense. Thanks!

[thesewordselope.livejournal.com]

You've put more thought and effort into this than I did into my entire undergraduate thesis. Now I feel like a bad scientist!

[cimness.livejournal.com]

Thought and effort??? Hahaha, yeah right. I mean, physical effort, sure, in that I had to sit there typing in the values and figure out the Create A Graph Wizard in Open Office. But otherwise, no.

[mecurtin.livejournal.com]

Is HnG Hikaru no Go?

[cimness.livejournal.com]

Yep.

[mecurtin.livejournal.com]

Having only one anime in there is kind of odd, because they are such a different genre and with different expectations (especially re: slashiness).

[cimness.livejournal.com]

True. I saw that brought up in TFV's comments. I think her policy was to include everything that anyone suggested.

[brownbetty]

I strongly believe some people are sniffing glue: Merlin is not that much better written than Smallville, this is a defamatory statement, and on behalf of Merlin fandom, I protest!

[cimness.livejournal.com]

Wow. I didn't even notice that. That's on crack! But wax_jism points out that the difference could be caused by Merlin's having less of what she calls "ickiness" and other people might call skanky issues. Her intuition says that that is part of the "good writing" composite, which seems like a fair assumption. It's a big part of why SGA is so low, I'm sure, not that SGA has great plots or dialogue or anything ahahha.

[ariadne83]

Oh come on, SGA has better plots than Hercules!! :-P

[seperis.livejournal.com]

Merlin has not yet used the excuse of junk mail when confronted by Arthur about a courier bringing him mail. MERLIN WINS.

(No, I don't still seethe over that or anything. *headdesk*)

[nicocoer.livejournal.com]

This is very hot. OMG.

[copracat]

I love you, I love thefourthvine, I love science.

*hearts*

[dreamnnightmare.livejournal.com]

you could improve it by adding error bars. Either plus or minus one standard deviation or maximum and minimum or something like that. That would allow people to see graphically which ones are problematic and which result from something like consensus.

[cimness.livejournal.com]

We call those whiskers in the modern social sciences' statistical forays. I wanted to do so, but I couldn't figure out how to make the program obey me. It's not made for statistics. Open Office Calc - the userbase is responsible for the FAQs and Help guidelines, and I couldn't find out how to do that.

[cimness.livejournal.com]

However, since the sample size is quite small, size of σ doesn't really show that the results are reliable anyway(statistically speaking).

[ariadne83]

ROFL SGA, Highlander and Hercules??? I have to agree with you, SGA isn't *that* bad.