Wednesday, 28 September, 2022
HomePageTurnerBook ExcerptsIndian voters care about ideas. But they are not telling pollsters that

Indian voters care about ideas. But they are not telling pollsters that

In ‘Whole Numbers and Half Truths’, Rukmini S writes why journalists can’t capture motivation of voters.

Text Size:

Why aren’t journalists and pollsters able to capture the multiple motivations of voters? In part, it is because they are asking the wrong questions and are too quick to use one answer as the answer. But part of the issue also is that people’s beliefs colour what they think of as their reality.

First, there are information asymmetries. Indians might not always believe that government programmes have benefited them to the extent that they credit a state or national government. Despite a majority across states saying that they use government schools and hospitals and a large majority saying that they use the PDS, 45 per cent said immediately following the 2014 election that no government programme had benefited their family in the last year, or they did not know if any had.

Sometimes, the blame or credit for a scheme is laid at the wrong door. Following the 2014 elections, 20 per cent of people said that they had benefited from the UPA’s Mahatma Gandhi National Employment Guarantee Scheme, but over two out of three credited the local or state government for it. Over 20 per cent benefited from the National Rural Health Mission, but over half of them credited the state government for it.

Post-election analyses abound with assertions that the presence or absence of ‘development’ makes all the difference. One argument is that an individual’s perception of economic progress is associated with a more favourable opinion of politicians: in one survey, among those who believed that their households were in a better economic condition than in the past, 56 per cent had an unfavourable opinion of the government; among those whose view of their own economic condition did not change, 58 per cent had an unfavourable opinion of the government; and among those who felt they were doing badly, 65 per cent had an unfavourable opinion of the government.

Perhaps the problem is with the very notion that voter motivations can be objectively quantified and measured, something that recent evidence contradicts.

Before the 2014 election, a majority of urban respondents said that economic growth would be the biggest issue influencing their voting choice. That should have meant that if the economy was doing well, the government in charge would be re-elected. Yet, after an election in which the incumbent was dislodged, more than half said that their financial situation was better than it was five years before.

An eventual BJP voter telling a journalist or pollster that unemployment mattered to her was largely meaningless then—it did ‘matter’, but her ideological commitment to the BJP was strong enough to colour her reality of under whom it had worsened.

Also read: Lockdown-stressed Indians would have defeated Modi if polls were held in June 2020: Study

How India (Probably) Votes

A more honest assessment of Indian voting is probably this: Indian voters care about ideas, they are more ideologically committed than ever before, they care about people who embody some of these ideas. There is no doubt that they want material improvements in their lives, but the ideological signalling—whether overt or implied—by the leaders and parties they vote for and the ones they hope to defeat matters and is something they factor in. These ideas they might

arrive at through their life experiences and through targeted messaging. And they’re not telling surveyors all of this—they might not even know it themselves.

Opinion polls don’t just produce electoral predictions: the same sample that produces  erroneous results is also the one that produces ‘research’ about what voters voted for. While a bad call can be apologised for and retracted, the narrative that the faulty data was based on can outlast the wrong call.

Make no mistake—this is not an easy electorate to survey and draw conclusions from. India’s politics are unique in myriad ways; the social heterogeneity of voters, the high turnover of incumbents, the frequent changes in alliances, all add up to a pollster’s nightmare. However, and partly for that very reason, information on who votes and why is invaluable.

As political parties increasingly turn to data to understand their electorates, ensuring that voter motivations are not misrepresented and the big ideas that voters care about are communicated to leadership becomes an important project of democracy. So how can we do a better job of capturing the Indian voter’s decision-making process?

One, opinion polling that is not transparent about its methodology and sampling deserves to be excluded from the public record, else the assumptions that stem from this data live on, unchallenged. Problematic opinion polling helps strengthen a growing narrative that all electoral polling is ‘fake’ or ‘paid’. It is only by separating the wheat from the chaff that we can build public trust in credible polling.

Opinion polling that is transparent needs to be held to better standards as well; if a pre-election poll or an exit survey significantly deviates from the final vote-share outcome, there’s a good chance that it got its sampling wrong. Instead of allowing these surveyors to merely recalibrate their findings about voter motivations to the final numbers (the current accepted practice), there’s good reason to consider discarding, or at the very least caveating, such surveys.

Two, there needs to be much more high-frequency, easily accessible opinion polling. Greater public scrutiny of opinion polling can only happen if the data is made public, which, in the case of a number of pollsters, is not currently the mode of operation. For this to change, the mode of funding of opinion polls—currently paid for either by media organisations or political parties—will need to be reimagined. In the US, for instance, Pew is a non-profit organisation supported by charitable funds.

Three, surveyors, journalists and analysts need to get more creative about the questions that they ask, knowing as we do now, that voters do not state their motivations upfront. Ahead of the 2014 general elections, a group of researchers conducted a survey where they tried out an experiment: first, they asked voters directly whether they would be willing to support a candidate who could deliver benefits but faced serious criminal cases—26 per cent of respondents said yes. But, since this sort of question is subject to a social desirability bias, they also asked the question indirectly through a list experiment. They randomly assigned all respondents to one of two equal-sized groups: a control group and a treatment group.

The control group received a list of three types of candidates (a candidate who was wealthy, a candidate who was poor, and a candidate who did social service but was not affiliated with any party) and were asked how many trouble them. The treatment group was provided the same three options but also a fourth option, a candidate who delivered benefits but faced serious criminal cases. Since the respondents were being asked how many—and not which—types troubled them, any difference in the average responses between the treatment and control groups would be due to the inclusion of the fourth option. They found that unlike the first time around, when only 26 per cent said that they were untroubled by a candidate facing criminal charges, the indirect question elicited this response from 48 per cent of respondents. This meant that the direct question got an answer that was an underestimate by twenty-two percentage points. Thus, questions that seek to better understand the voter’s world view might be more effective than those that ask her to simply choose what she is voting for from a list of options, a strategy that almost demands virtue signalling.

Also read: Most SC/STs would vote for Opposition unity in elections, survey shows

Random response techniques, endorsement experiments and survey experiments are other methods that surveyors can employ to reduce biases associated with the misrepresentation of data, the researchers suggested.

Four, there could and should be more triangulation of information between what seasoned reporters see on the ground and what the data says, instead of the current atmosphere of mutual distaste and suspicion.

Finally, we might not be able to get analysts, whose job it is to spin quick and pithy narratives on-air, to be more circumspect, but there’s no reason to allow these narratives

to dominate our understanding of how India really votes. Perhaps their narratives confirm our own biases and play to our own desire to understand the voter as an ‘aspirational’ creature. The more complicated truth is harder to fashion into six-word headlines, and is messier, a bit more unexpected and a bit less saccharine—much like Indian democracy.

But this is a worthy endeavour. Indian voters are inspiring in their idealism. Indian politicians are moving in the odds they have overcome and the changes that their pioneering paths have produced in broader society. The way a voter is able to engage with the world of ideas, with real tangible benefit and change, with self-respect, with an entitlement to be heard is transformational—few other interactions with the Indian state offer her that true egalitarianism, that power. It could be that even the best data is never able to capture this. But we needn’t let our notions about the universe of democracy be defined by the much smaller world of numbers.

Extracted with permission from Whole Numbers and Half Truths: What Data Can and Cannot Tell Us About Modern India by Rukmini S., published by Context, an imprint of Westland Publications, December 2021.

Subscribe to our channels on YouTube & Telegram

Support Our Journalism

India needs fair, non-hyphenated and questioning journalism, packed with on-ground reporting. ThePrint – with exceptional reporters, columnists and editors – is doing just that.

Sustaining this needs support from wonderful readers like you.

Whether you live in India or overseas, you can take a paid subscription by clicking here.

Support Our Journalism

Most Popular