Indian media need more ‘tracking’ surveys for richer public discourse
Until a few days ago, the ‘Khan market consensus’ was that India’s ruling Bhartiya Janta party was likely sweeping the February 2022 elections in Uttar Pradesh, an election as important as the midterms in the states. -United.
Then, last week, Prime Minister Narendra Modi made a shocking gesture, announcing that the next session of the national parliament will repeal three controversial laws that aimed to allow the privatization of agriculture.
Experts immediately began the post-facto explanations: the BJP was clearly losing enough voice in western Uttar Pradesh, one of the hotbeds of unrest among farmers, that they had to eat a rare humble pie. Of course, the main center of unrest is Punjab, but the BJP has no chance of winning the elections there. And then there is Haryana, where the elections are in a certain time.
As always, everyone said they knew there was enough trouble in the UP west. Everyone is wiser after the fact.
Call the voter
This disjunction is a good example of why the Indian media would do well to have more phone tracker surveys that give a weekly idea of public opinion, just like in the US or UK. United.
We regularly read the ups and downs of US presidents’ “ratings”, not just during the election campaign, but also in the ordinary course. We also have such surveys in India, but they are occasional and election-focused.
There are two types of surveys: one conducted in the field and the other by telephone. Telephone surveys are quicker to run and give a common idea of changing public moods. The news channels that commission surveys are still too hooked on a heavy and heavy field survey just before the start of the elections, then an exit poll. That’s it.
It is argued that field surveys are of better quality because people are less likely to make a joke of them. But someone who has supervised telephone surveys tells me that people are often honored to be asked for their opinion. It’s only a few elite people like us who are likely to give the wrong answers just for the fun of it.
Field surveys also have their problems and biases – the field interviewer is inclined to falsify the data to reduce legwork, or to investigate the asymmetric sample set around markets and storefronts. These problems have their solutions, but ultimately, the field survey is too long to conduct to be useful in measuring the evolution of public opinion. India Today publishes a “Mood of the Nation” survey only twice a year.
Political parties and consulting firms are ahead of the media in this regard. They directly or indirectly conduct phone surveys that ask people the same question over and over again over the course of months.
Trends and variations
The only telephone follow-up survey available in the public domain is that of Cvoter, although they do not yet publish it on a weekly basis. They have so far published three polls regarding Uttar Pradesh. Whether their “numbers” are correct or not is not the question.
It is more interesting to see the trend and the variation – if you look at the three surveys, there is a clear trend for the Samajwadi Party to accelerate and the BJP’s lead to decrease. Looking at this trend, it can be said that the Samajwadi party is picking up as it has finally started campaigning.
Earlier this year, Cvoter was kind enough to give me access to their daily monitoring dashboard for the legislative elections in West Bengal, Assam and Kerala. By seeing the daily trend and the weekly averages, we could understand why political parties were doing what they were doing.
In Kerala, I saw the congressional line go up after handing over his campaign to Oomen Chandy. This made sense as the investigation also showed Oomen Chandy to be by far the most popular congressional leader in Kerala. However, Congress fell too far behind, as Pinarayi Vijayan and the PMO were consistently ahead.
In other words, follow-up media surveys can help political discourse go beyond “who made the right prediction.” In West Bengal, Cvoter showed how the Trinamool Congress gained a percentage point or two after the Mamata Banerjee crash – clearly, the mass leader got the extra sympathy he needed.
The survey showed that the Trinamool Congress was consistently ahead, but the BJP was ahead in terms of the “perception gap”. This meant that many of those who did not vote for the BJP thought the BJP was winning anyway.
As Election Day approached, Trinamool also overtook BJP in the perception gap. If you look at the end result – the landslide victory for Trinamool Congress – the investigation made sense.
From prediction to understanding
The problem isn’t the cost, as it doesn’t cost much to set up a call center. It costs a lot less than hiring field researchers and having them travel.
The problem is one of a mindset: news channels need to move out of the prediction mindset and view polls rather as an ongoing process to help understand public opinion and its continuum, capturing even small changes.
The news media want big changes in every survey so they can make headlines. However, public opinion is gradually changing and it is important to closely monitor these gradual changes if we are to understand and reflect public opinion in real time.
Is inflation becoming a concern for people from week to week? Or, for example, what were the immediate consequences of the rollback of the BJP’s farm law on its popularity and on Prime Minister Modi’s approval ratings? Is there a significant increase in the percentage of women wishing to vote for the Congress Party in Uttar Pradesh after its pledge to give 40% tickets to women?
The right demographic mix
Another necessary change in mindset concerns the sample size. It is a myth that the larger the sample size, the more reliable the survey. A poll is like taking the first sip of coffee and deciding if it’s strong enough for you. You don’t have to swallow half the coffee to make up your mind.
It is more important to get the right demographic distribution, to reach voters that you cannot easily find on the side of the road, making sure that all castes, communities, age groups and geographic areas are proportionately represented in the final survey analysis.
This is why “online” polls in India are by definition absurd. They don’t measure demographics and exclude the digital poor to begin with.
Like all opinion polls, many daily tracking polls will also be bogus, as it’s a difficult science, and some will be rigged for propaganda purposes, but if we have at least 3-4 regular weekly polls, we can match them with each other, and our own sense of the terrain.