Quality control The size of the sample and other components of it need to be overhauled. SHUTTERSTOCK/SCYTHER5
Indias size complicates media measurement efforts. Sample size and constituents need to reflect the diversity better, and more advertisers need to fund it with money and knowledge
Media measurement in India is an extremely complex practice due to the sheer size of the country and the heterogeneity of her population. Regional, ethnic and linguistic differences within the same region just add to the complexity. The obvious questions that arise before embarking on research of any sort are in relation to the adequate sample size for the research, whether there should be quotas for certain demographics, where should one conduct the research, the research methodology and the cost of the research.
It is a known fact that the sample sizes, especially in the TV measurement space, are inadequate. This results in high relative errors while reporting data and makes the data useless in certain cases. If we need robust data to help us make the right business decisions, this is the first issue that needs to be tackled. While the IRS sample size in the range of 2.5 lakh, there might be a case to examine the sample sizes by demographics by pop strata at an individual state level and in case of inadequacy, these samples should be boosted.
The issue of research methodology needs to be examined in relation to the advances in technology globally in research. Media reports suggest that technology/methodology in TV measurement in India has remained more or less static for almost a decade. While TAM, I am sure, would have improved its technology over the years, experts think it did not keep pace with the market developments. IRS, for the first time, has used Dual CAPI which is a much needed enhancement in the methodology. With this development and the advanced technology claims of BARC, I think we are moving in the right direction.
The most critical factor in conducting media research in India is the funding of a survey on such a large scale. The key issue is the fact that advertisers barring a few do not even pay for any media data. Most of the funding for this kind of research comes from media owners who earn advertising revenue based on this data and media planning and buying agencies who use this data to make decisions on behalf of advertisers who are their clients. What advertisers need to realise is that it is their money that is at stake here. Agencies have nothing to lose if the data is error-prone. But advertisers have a lot to lose as bad research data can lead to bad business decisions and it is in their interest to contribute to media research in India both financially and intellectually.
While it will be speculative to pinpoint any reason for the current situation that the media measurement practice in India is currently in, I do believe that the issues mentioned can address the situation effectively, if carefully thought through.
There are a few malpractices that have been reported in the past which can have serious implications for the research findings. Some of these might be hearsay and so should be looked at as potential ways of influencing the research.
For example, publications dumping copies in certain areas/cities where the survey is likely to happen can influence readership during the field work. Or publications could contact the research agency, offering people there bribes to report false data to show them in a better light.
In the television space, it has been alleged that channels were running promotions to influence viewership in select towns which are part of the TAM sample. Another charge was that TV channels were gifting TV sets to respondents in the panel with the diktat that their channel should play the maximum on the TV set attached to the peoplemeter.
Here is the original post:
Clearing the path to accurate data