Measuring the radio

Someone had to come forward. And someone just did. Research agency AC Nielsen-ORG-Marg has launched a first-of-its kind study to measure radio listenership in India, the Radio Audience Measurement, which will track the various FM radio stations that have set up shop.

Right now, the project will be restricted to Mumbai, where five FM stations have been launched. A random sampling was conducted in Mumbai a fortnight ago, with a sample size of 1,000 people, with a fully structured questionnaire. In the pilot study, the target audience comprised individuals in the age group of 15 years and above, from all socio-economic categories.

In many ways the new project is also about testing waters. Several questions remain, including which system will be used finally - the diary system that asks respondents to jot down their preferences in a diary every week; the ‘recall' method to monitor listeners - one that involves face-to-face or telephone interviews; and the watch meter method, a technologically expensive method that involves recording the frequencies of stations listened to, on a small recorder in a wrist watch. Currently, the last method is used only in Switzerland, and is being contemplated by Australia. "The diary method involves a certain amount of dedication as it must be updated every week, and the watch meter method is too expensive. So, in the short run, we are using the recall method, but that does not mean that the others have been eliminated," says Pravin Tripathi, communications and media effects consultant, who is also chairman, Technical Committee for Radio Audience Measurement, ORG-Marg.

The committee is trying to look at comparative studies between the diary and recall methods, done anywhere in the world, by AC Nielsen, and which have the same sample target group. The idea is to figure out if there was a marked difference between the results obtained using the diary method and the recall method. Sources also say that in case the data is unavailable, the committee is considering holding an experiment with a small sample target group, to compare the effectiveness of the two methods.

One key input will be from the media agencies, and the broadcasters themselves. They will have to choose whether they want a measurement of only SEC ABC, or whether they want to include D and E also. If they chose to include D and E, many of whom are poor illiterate listeners, then the diary method, which calls for an amount of literacy, is effectively ruled out. In such a scenario, with watch meters being prohibitively expensive, the recall method will be chosen, say analysts.

ORG-Marg has not yet decided on a final method, and the final solution will also depend on many factors that are not technical. These include issues such as whether the broadcasters will foot the bill or if agencies will chip in. Right now, while radio stations will be charged Rs 18,000 by the research agency for buying the data, pricing for the advertising agencies will be on a differential basis based on their turnovers. However, selling it to clients will also be a problem. "Many clients on radio are small, such as the local eatery. In such cases, you would have questions like what percentage of the patrons of a restaurant are radio listeners, and have come to the eatery because of the radio advertisement. That will affect the marketing of such data by agencies who are seeking to woo clients," points out a senior media planner.

And then, there is the widely held perception that radio is too new a medium for viewers to have developed brand loyalties. Many analysts dispute this, saying that the media is already habit forming, and nascent brand loyalties have already emerged. However, there is no denying the need for such a study. Points out Nehal Medh, vice-president, ORG-Marg, "Right now, there is the need for an ‘immediate' measurement system. Potential problems such as top of mind brand recall will be addressed."

One way, that the team is considering, is a combination of two methods. "Right now, we plan to do it in waves every quarter. For two weeks, every three months, we will conduct face to face interviews with a sample size of 2,000 respondents. One way in which we are contemplating validating the data is to use, say, a hundred watch meters, but the costs are extremely high, and the idea is only under consideration," says Tripathi.

In the recall method that has been chosen now, the selection will be random, and avid listeners of radio will be quizzed on their favourite place of listening to radio, awareness level of different stations, frequency of listening to radio and listeners profile for different radio stations. Once the first part of the project in Mumbai is complete, ORG-Marg plans to set up a full-fledged system to monitor listenership. © 2002 agencyfaqs!