This the fourth in a series of essays on the use of Arbitron ratings as a tool for Pacifica programming decision-making. While the information is specific to station KPFA in Berkeley, Pacifica's general trend of dealing with their audiences in the same manner as commercial audiences is clear in these reports.

SUBJECT: KPFA DEMOGRAPHICS

In this message, we'll look at what the "Walrus Report" has to say about the demographics of KPFA listeners. Remember that these figures are fairly reliable for the overall KPFA audience, somewhat less so for the most popular dayparts (Morning Show and Evening News), but almost totally unreliable for most other specific timeslots (where values vary wildly from one survey to the next). However, the consultant and KPFA management apparently take all these figures as equally significant.

KPFA's Cume Persons (different persons (listeners) per week) are 52% male; the TSL runs 55% male. In other words, men slightly outnumber women in the audience, and they also tend to listen for slightly more time each week than women do. However, Walrus notes: "That ought to be considered the lack of an appeal by sex, because most stations skew strongly one way or the other. For example, a hard rock station will skew 75% male. A soft AC station will skew 75% female." KQED-FM has a distribution similar to KPFA's - just slightly more male listeners than female.

KPFA's Cume Persons are 6% African-American, 8% Hispanic, and 86% "other" (Arbitron has no separate category for Asians). The TSL has an almost identical distribution. I couldn't find corresponding figures for KQED in the Walrus report.

Ah, Middle Age
The median age of KPFA's Cume Persons is about 39; that for KQED is about 48. Walrus says: "The most interesting demographic skew in KPFA's appeal is age. One third of the audience is within 35-44. The next largest demo is 45-54, then 25-34. Hardly anyone over age 54 listens to KPFA, and few listen under age 30. ... KPFA's demographic range is identical with that for NPR. Any public radio station which carries Morning Edition and ATC generates an audience that looks like KPFA's audience by age." Well, KPFA's median is 9 years younger--a noticeable difference, I'd say--particularly since only KALW among stations used extensively by our listeners has a younger median age than we do (about 36 for KALW).

Here's the age breakdown (estimated from a bar graph):
          Cume Persons     TSL
12-24        8%                16%
25-34      19%                20%
35-44      32%                30%
45-54      22%                22%
55-64       8%                   8%
65-74       7%                   3%
75+         4%                    1%

Now, I'd look at these figures and say that we have a surprisingly healthy listenership among young adults. To me, the surprising thing is the small proportion of our listenership from among people my age (55) and older. Walrus, however, has dire predictions that our listenership (and NPR's) will dwindle over the years unless we do something to attract younger listeners. I think you'd find it hard to claim that KPFA has ever done much to appeal strongly to young listeners, and yet we have a sizeable audience of people who must have come to the station after age 25. Management bought Walrus's argument here; I don't.

Age and Gender Shifts
The Walrus report goes on to draw lots of conclusions about how these demographics change for different dayparts. The Morning Show audience, for example, has median age around 43 and is around 70% male. For the Morning Concert, the median age drops to around 33 and the audience shifts to around 55% female. The Evening News has a median age around 37 and an audience around 52% female. (The black & Hispanic segment of the audience varies between 10% and 20% through the day, with no clear pattern obvious - these look to me, and to Walrus, like random sampling errors rather than shifts with particular programming.)

The demographics for most other shows don't mean much because of the small audiences and hence small samples, plus the fact that in most cases you are averaging across five different shows during the week, so that the average values have little meaning.

Watching the Clock
Walrus asks: "Is the Morning Show a guy thing?" Interesting question. Also interesting to note that the Pacifica News audience runs about 65% male, with a sharp swing to a slightly female audience for the KPFA news. An unasked question (by Walrus or by KPFA, so far as I know): How do the demographics of People Using Radio change with the hours? Do some of these changes simply reflect overall shifts in age and sex of people able to listen to radio at different hours?

Walrus goes on to analyze Saturday and Sunday demographics, but remember that these samples are so small that the numbers are highly suspect. (Values estimated from graphs.)

Saturday
7am Median age 51, 75% male, 39% black & Hispanic
10am Median age 45, 60% male, 20% black & Hispanic
12n Median age 37, 54% female, 0% black & Hispanic
2pm Median age 37, 53% male, 10% black & Hispanic
4pm Median age 45, 62% male, 35% black & Hispanic
6pm Median age 52, 50% male, 35% black & Hispanic
8pm Median age 30, 100% female, 0% black & Hispanic
10pm No measurable audience

Sunday 7am Median age 40, 62% male, 0% black & Hispanic
10am Median age 44, 51% female, 11% black & Hispanic
12n Median age 35, 63% male, 15% black & Hispanic
2pm Median age 47, 60% male, 10% black & Hispanic
4pm Median age 48, 51% male, 6% black & Hispanic
6pm Median age 40, 72% male, 8% black & Hispanic
8pm Median age 33, 100% female, 32% black & Hispanic
10pm Median age 41, 100% female, 14% black & Hispanic
12m Median age 33, 75% male, 0% black & Hispanic

The main point that Walrus makes about these numbers is that the station is drawing a big audience of one sort for a show, then driving that audience away and attracting a different one for the next show. This, of course, is what one would expect with an eclectic station. However, the argument is that we would get much better ratings if we kept the same kind of show all day, or at least grouped similar shows together, so that we could hold the same audience longer. If you look at the new weekend schedules, you will see that KPFA has tried to cluster the shows as recommended.

A couple of interesting points about these numbers, though. Recall that a stated goal of the new schedule is to attract younger listeners. Yet, Ellwood/Strachwitz, who attracted a very young audience if these figures are to be believed, were dropped. It's also interesting that Bonnie Simmons drew a slightly female audience, but the audience shifted back to slightly male for Women's Magazine. Similarly, most of the women listeners tuned away during Focus on Women in Music on Sundays.

Of course, when looking at management's changes, you must combine these demographics with the ratings discussed earlier. A show may have desirable demographics but a very tiny audience.

Playing to the Opposition
Walrus tends to assume that many changes in KPFA's audience may simply be due to KQED's programming. That is, some program on KQED may draw away enough men or women, older or younger people to have a major effect on our demographics and ratings. In such cases, they recommend counter programming. That is, if KQED is appealing to older males at a given time, we should program a show with strong appeal to younger females, and so forth.

Next time: geographic distributions of KPFA listeners.

We believe decisions about our stations should be made by the representatives of all the communities that have a stake in Pacifica, and should occur in an open, accountable, community-oriented process . . . something now lacking at Pacifica.