Back in March, we announced plans for a weekly series of FAWM 2008 summaries in the form of fun and interesting statistics about this year’s project. Perhaps it will be more of a bi-monthly thing at the rate we are going? At any rate, post #2 in this series focuses on some interesting data about us: the fawmers.
First of all, 2008 fawmers were 29.7% female and 70.3% male. By comparison, we were around 25% female in 2007, and roughly 20% during 2005-06, which means FAWM is gradually becoming less of a sausage fest each year. Also, the average age was 31 years (median was 29), up from about 28 last year, so we’re getting a little older. In spite of this, however, teenagers (fawmers aged 15-18) were actually one of the fastest growing demographics, increasing 393.8% since last year!
The pie chart above illustrates the top 5 ways fawmers learned about the project (according to the 2008 survey). As you can see, nearly half heard about it from a friend or acquaintance who has fawmed before. This is a large part of why (I think) we’ve been so successful: FAWM grows organically through word of mouth among interested parties, roughly doubling each year!
This chart is an interesting one: a scatterplot of FAWM 2008 song commenting behavior. Each dot represents a single fawmer, color coded by the number of songs s/he completed (see legend). The x-axis plots the total number of comments made, and the y-axis shows the average number of comments per song received (as of March 1, 2008). There is clearly a positive correlation… and in fact, the correlation coefficient among winners is 78.8%, which is very high! The obvious lesson? Those who give more feedback to the community tend to get more feedback as well.
Finally, this graph shows FAWM as a social network. Again, each dot represents a fawmer, and each thin grey line connects one fawmer to another, based on any song comments made between them (as of March 1, 2008). The more comments, the darker the line. The more connections a fawmer has, the more central s/he is to the network. Not surprisingly, there is a tight cluster of very active fawmers in the middle, most of whom were “winners” (i.e., wrote 14+ new songs). However, there are plenty of winners who are not connected to anyone, and plenty of fawmers who wrote fewer than 14 songs and are still in the “inner circles.” Even some fawmers with no songs are right there in the middle (mostly “FAWM widows” and other fans who create accounts to cheer the rest of us on). Of course, it also shows many people who signed up and did nothing, which isn’t unexpected.
There are several conclusions we can draw from this figure. First and most surprisingly: for as large as FAWM is (500+ active in 2008), there is really only one main “social cluster.” I’ve varied the minimum number of comments required for a connection, and the structure doesn’t change much at all, which means there aren’t a whole lot of cliques. (As FAWM continues to grow, this will likely change, which is probably fine.) Another interesting and, in my opinion, positive fact: a fawmer’s “success rate” seems to have little to do with how central they are to the network. I interpret this to mean that you are welcomed by and important to the FAWM community regardless of how prolific a songwriter you are coming in.
I bet, though, over time, you become more prolific. (And more confident, too!)
Burr, thanks for taking time to do this analysis — and for everything else you do to make FAWM the joy that it is.
Proud to be one of the dots in the top right corner of the comments per song / comments made axis,
phil
Dude.
DUDE.
Nice work. I’ve never a graph like the social network one before. Very interesting.
Keep hope alive! Data rocks!
Burr, what an interesting analysis of us FAWMers. It gives me hope, actually. I’ve recently finished reading a book by Lee Siegel called “Against the Machine: Being Human in the Age of the Electronic Mob”, which paints a horribly bleak view of the future of the Internet. My interaction in the FAWMunity, particularly with Mr. Guiles immediately above, has been very satisfying. I really wish I could get more local connections however. C’mon Detroit, represent. How can the city that is responsible for Motown and all of it’s hits, not have more FAWMtivity?
OK, I’m a nerd. I tried finding my spot on the comments vs. avg comments scatterplot and I can’t find my dot. I made 230 comments according to the site, and I added up my comments and divided by my number of songs, and got 9.9
Well, there’s a big white area in the graph where my red dot should be. I’m lost.
Hi Burr,
With my day job in research (to support my night job in rocking), I appreciate the graphs and data analysis. Excellent to see the average age is going up, but I must ask if you corrected for past fawming activity. That is to say, the average age would be expected to increase by one year for repeat fawmers, just by the fact that it’s an annual event. You can correct this figure by subtracting years of fawming from repeat fawmer ages to get a measure of “average age at first fawm,” or segregate out the first timers and report their average age separately. Semantics I suppose. Your number tells the truth, which is the average age of fawmers this year. But it’s always going to trend upwards unless a lot of fawmers quit to balance those who join, which is very unlikely, because its addictive and grows exponentially. I love it.
Thanks.
Joe
Scott — Note that the scatterplot and network figures are based on comments made on or before March 1 (i.e., during the actual FAWM event). So if you’ve made (or received) a lot of comments since then, your dot will be elsewhere.
Cactus — No, I didn’t do any correcting for anything, mostly because (1) FAWM is relatively young anyway and (2) I’m lazy.
That’s an interesting consideration for the future, though, thanks!
@Burr - Aha - that explains it. I did lots of commenting on other’s songs AFTER March 1 when I had more time on my hands.
The social network thing is a very cool way to look at the data.
I think the forums are a big reason that the social network is so welcoming. And although it’s not reflected in this data, I think the practice of “zonging” also broadens the connections among fawmers. It would be interesting to see how the scatterplot and network figures changed by April 1 or May 1 ….
Y’know, Burr, you just keep opening yourself up for more feature requests. I want these graphs for my profile page, updating in real time.
It’d be interesting to see the lines of the triangle between Nancy and Charlie and I get thicker and blacker all month long . . .
“You left a mark on me, it’s permanent, a tattoo”
-Lucinda Williams
I just think the social network graph is cool looking… like a sand portrait or a kaleidescope pointed at back-lit stained glass.
I Love The FAWM.