2007 EP Survey Results

• About 33% of respondents report being outside the U.S. At first I thought this was another example of sample bias, because EP site stats indicate that about 20% of listeners are outside the U.S. But looking at the site stats more closely I realized that many visitors show up as "unknown" or with special HTTP status codes — as much as 20% of the bandwidth traffic. I'd always figured that the proportion of domestic to international for that tranche breaks down comparably to known visitor stats, but such may not be the case. If disproportionately more originate internationally the numbers could jive. And that would track with actual donors, of whom 35% are outside the U.S. Moreover, since (as discussed above) from the outset of the survey slightly over 30% of respondents reported being international, I'm inclined to think the actual number is much higher than 20%, though perhaps not quite so high as 35%. Higher, anyhow, than I'd previously believed.

• Median age of respondents is 41-55. This tracks with last year's survey and with what I know from correspondence. Only 4% are 25 or younger (2% last year) — I hope they learn something from the shows!

• Only 2% of respondents are female. I'm sure this understates female listeners, but it highlights an undeniable fact: by far most EP listeners are guys. To try to break it down, in last year's survey 5% of respondents were female and, this past year, 10% of actual donors are female. Even allowing, however, for a very large margin of error (and taking account of a much smaller donor population of only 29) I figure it's somewhat unlikely that female listeners would be so much more inclined to be donors. Conversely, I'd also be quite surprised if the gals make up more than 10% of listeners. To be honest, I don't know what this means. I would be very happy if more were listening, but how to make that happen? Sharon suggests guests who talk about relationships, or a more specific example, someone who went to live with arctic Caribou and wrote a book about it. I guess I'd like to get in touch with my inner Caribou as much as the next guy, but that's not necessarily so high on my priorities list. A vexing dilemma. Suggestions on fixes or whether just to leave well enough alone would be most welcome.

• Educational levels are up slightly from last year. The median is still a BA degree, with 34% of respondents (35% last year). 20% report an MA (up from 15%), and those with a professional degree, a Ph.D., and post-doc work are, amazingly, almost exactly the same: 9%, 5%, 2% versus last year's 8%, 5%, 2%. Going again by correspondence as a check I have no reason to doubt the rough accuracy of these numbers.

• 48% work in the private sector (56% last year); with a good chunk, 16%, in government or the military (8%). I should have broken out retired from unemployed, which seems high at 29% (26%).

• Median income is higher than I'd expected. Leave it at that.

• Politically, the largest category is "other" at 37%, followed by very liberal at 31%. More evidence that fusion politics is a not-unrealistic possibility.

• Regular listenership as a percentage of population appears to be either very slightly down or at least stable and changed around a bit, with 56% listening every week (65%), and 27% two or three times a month (15%).

• The "how often do you visit the EP website" question is very helpful to me in understanding traffic. 12% visit once a day (28%), and 51% a couple times a week (55%). I suspect more are using RSS feeds than last year, accounting for some part of the difference in the former. But virtually nobody, just 1%, is visiting EP more than once a day, which helps put into perspective the jumps I've seen in site traffic, e.g., large increases in visits per day are probably not an artifact of a constant population.

• The EP podcast rated on a scale of 1-10 gets a median score of 9 (again, 9). Very nice to know!

Which leaves the open-ended responses to do with the main political problems we face, which guests you all liked (or didn't), and the optional end comments. All this is quite fascinating and I'll need more time to pore over it and make sense of it. As I say, extremely helpful for programming.

To take just the guests that you liked part, though, for now, the most mentions are a tie between Al Bartlett and Paul Craig Roberts, with 13 each. Ray McGovern had the second most mentions, with 10, and Chalmers Johnson third, with 9. What's really interesting to me is that 48 individuals got mentioned, with a median of 2 mentions (yes, I worked this out). What this says — I think — is that there's something here for everybody. What's interesting also is that six names made both most-liked and most not-liked lists, Chalmers Johnson being one of them. Curious, huh? Then there were the two favorite guests who I haven't, in fact, ever interviewed, but the respondent gets a gold star for creativity... :)

Thanks very much for participating in the survey, if you did, and if you didn't I hope you find these results of interest, too.