Monitoring airwaves with a "portable people meter," Arbitron's methodology questioned
Are there more Hispanic country music fans out there than previously thought?
The Federal Communications Commission is investigating whether minority radio listeners are undercounted by the tracking firm Arbitron, according to this story in the Washington Post.
The investigation centers around the use of a new tracking system employed by Arbitron (and covered by Wired in 2007), which requires the listener being tracked to wear a device that constantly monitors the airwaves:
Arbitron has recently replaced its diary-based rating system in certain markets with the PPM [portable people meter] system. According to Arbitron, the PPM is a mobile-phone-sized device that consumers wear throughout the day. The PPM detects inaudible identification codes that are embedded in the audio of certain programming to which the consumer is exposed. An encoder at the programming or distribution source inserts the inaudible identification codes. In addition, a station monitor is installed at the programming source to ensure audio content is encoded properly. At the end of each day, each survey participant places the PPM device in a base station to recharge the battery and to send collected codes to a household collection device known as a “hub.” The household hub collects the codes from all the base stations in the survey household and transmits them to Arbitron.The concerns of the coalition, now made more formal by the FCC, are that Arbitron uses the PPMs in a way that underrepresents minorities, with the consequence that "undercounting could particularly affect the ratings of local, urban-formatted radio stations that broadcast programming of interest to African-American and Hispanic audiences."
Arbitron's response is that "samples effectively represent Blacks and Hispanics in the 18-34 age group, and across other factors such as geographic location and language preferences." Arbitron published this "Hispanic Radio Today" report in 2008.
Read the FCC's official Notice of Inquiry here.
From the comments of FCC officials at fcc.gov, it appears the investigation does not presume that the concerns raised are in fact correct, but are serious enough to flesh out until they are confirmed or rebutted.
Nashville's country music industry has been trying to attract Hispanic audiences to country music (see this story for background, or see the Country Music page on this site). The FCC investigation could reveal that the numbers of Hispanic listeners have been undercounted.
Here are the reasons why minority radio representatives think the Auditron PPM methods are flawed:
[O]nly five to six percent of the PPM sample is comprised of cell-phone-only households, while a significant and growing percentage of young adults and Hispanics and African-Americans live in cell-phone-only households.8 PPMC asserts that 19.3 percent of Hispanic households and 18.3 percent of African-American households are cell-phone-only, whereas 12.9 percent of non-Hispanic white households are cell-phone-only.9 Among other things, PPMC also complains that: (1) PPM has a 66 percent smaller sample size than the diary, often making it impossible to target age or gender subsets of minority audiences because standard industry metrics require at least 30 respondents in a cell to run ratings data; (2) PPM samples are not built using street addresses, and therefore fail to ensure statistically representative inclusion of cell-phone-only households; (3) young minorities are reluctant to carry visible PPMs; (4) Hispanic PPM recruitment methods skew toward English-dominant persons because potential panelists are identified by origin rather than by language; (5) PPM response and compliance rates fall below industry norms; (6) PPMs record exposure to radio signals, but they do not capture listener loyalty, which is high among minorities; (7) PPM reports provide less granular data in terms of geography; (8) PPM reports do not contain income data, country of origin data, or data that accounts sufficiently for language preferences;and (9) PPM panelists may be corrupted more easily by radio personnel because the PPM device often visibly identifies them and their expected participation is two years instead of the usual one-week participation in the diary system.Photo by Fabrizio Sciami. Licensed under Creative Commons.