Field work has been a whole lot better this time, aided by automation: Vikram Sakhuja

MRUC on Friday published the IRS Q1 2019 data. We caught up with Vikram Sakhuja, Group CEO, Madison Media & OOH & IRS Technical Committee Chairman, to talk about the changes that have been made in the latest IRS compared to the last one in 2017, the impact of increased internet penetration on the survey and more.

Excerpts 

What changes or corrections have you made in this IRS technically as compared to last year?

We brought in a lot of tech to increase controls. The clear measures that we took were to make sure that the reliability and validity of the data is unbeatable, automation of the contact sheet, and ensuring quality in the process of fieldwork.

Also the way we treated upscale interviews on the basis of feedback received. We tried to figure out what could possibly go wrong in the field, and then tried to address it systematically.

Did any aberrations come to your notice? And if yes, how did you deal with such aberrations?

In any field work there are controls; Nielsen has its own field controls. So, there is a system of back-checks and audio checks. In this dual CAPI, if any of the readers are being interviewed, it’s all being recorded by a machine also. Now, we are able to go back and listen to each and every interview. So, if it turns out that you read newspaper ‘x’ and newspaper ‘y’ has been recorded, obviously that cannot be taken. These kind of errors happen. Some of them are just normal human errors. Some of them could be motivated. But, regardless of the origin, we are now able to check that.

Did accounting for variants in the last IRS put an end to the controversies around variants in the Print industry?

I think the variant issue is not a research issue. It’s basically a management issue of the newspaper owner. So, if in a survey we show mastheads to people and say ‘Have you read this or not?’ misattribution of a main masthead to the variant or vice versa is likely. Now, if you by design made your masthead so difficult to differentiate, then you are trying to create confusion. And research will just tell you that confusion has been created. It’s not a field issue. We do have some controls; some follow-up questions to make sure there is no misattribution. I don’t think we can do much better in the field in terms of actually detecting a variant versus main edition.

The last IRS created an opportunity to compare apple to apple numbers of total readership, total viewership, total listenership. How has this worked out, and what is the kind of feedback that you received?

I think it’s been favourable. People like this apple to apple piece. It gives you a better idea of the general media consumption landscape which is arguably different frequencies across different mediums… I would like to see that. Print continues to be a very close number 2 to TV. So that’s nice, in terms of just number of people exposed to it. Digital now seems to be coming up too. So, it’s a good metric.

Internet penetration has increased from 19% to 24%; how has this impacted IRS?

If you look at the reported four quarters, and three quarters are of 2017, Internet penetration has gone from 19% to 24%. If you look at it just for the Q1 2019, it is 36% because of the post Jio boom - which means close to 390 million people who are actually exposed to Internet in one form or the other. In urban India, the latest quarter number is 51%. These are very sizeable numbers. I don’t like it when people say that the Tech Com is not doing a very good job of measuring the entire digital boom. I think we have captured a pretty good picture; quite representative of the reality right now. I am actually very pleased with the way we captured this. In surveys like these, 80% is field work. In this case, we put in a lot of effort to make sure that the quality of the field work is good. There was criticism that we weren’t capturing upscale homes very well. We put a lot of effort into that. I think field work has been a whole lot better this time.