'Fake news is not a disease'

'It is more of a catalyst.'
'People bring in their own demons and they are not created by fake news'

A team of Dr Rob Jones and Dr Sayan Banerjee of the University of Essex and Dr Srinjoy Bose from the University of New South Wales was in India to conduct a five-week long research study funded by Facebook.

Their aim was to find out about the influence of fake news on WhatsApp, and on Indian voters.

"Fake news does not change your views," Dr Sayan Banerjee, below, tells Rediff.com's Shobha Warrier.

Is the study conducted by you and your team to find the influence of news on voters during the elections, or the influence of fake news on Indian voters?

We are looking at how fake news distributed on social media like WhatsApp affects the political behaviour in the ethnically diverse places of India and Afghanistan.

You conducted the study in Uttar Pradesh, Bihar, West Bengal and Jharkhand. Any particular reason why you chose these states?

We chose a mix of urban and rural areas, you can say mostly urban areas, Tier 1 and tier 2 towns because most of India's WhatsApp users are concentrated in the urban areas.

If you go to the rural areas, you will see that people there do not have proper data connection and they don't own smart phones.

Not only is our sample representative of India's population, but they also don't have long-term incumbents.

When you have a long-term incumbent MLA or MP, voters tend to have loyalty towards the person.

How did you conduct the study? 

Our study is a two-step survey experiment sandwiched by a field experiment using WhatsApp.

In the first step, we interviewed 3,500 people in these states across 18 Vidhan Sabha constituencies. None of these seats have any caste that has more than 60% representation.

We asked them hypothetical questions on candidates. We didn't use the names of any real candidates or political parties.

We then asked them the order of preference when they think about voting; for what is s/he voting for and who s/he is voting for.

Next is, what s/he wants from the candidates -- security from violence, or public goods like roads, electricity, water, etc, or free goods like TV, laptops, etc.

These are the three fundamental things that voters mostly want from their candidates.

When I talk about violence, I mean caste violence. And that is one thing that gets overlooked by the media and academics.

If you look at the data, you will see that caste violence including hate crimes and civil rights violations are the most dominant forms of security issues in India.

When you talk about violence, you always think, it is Hindus versus Muslims, but it is not though it is there in isolated places.

It is caste violence that is more widespread.

While West Bengal has very low percentage of caste violence, states like UP and Bihar are high on caste violence.

Does that mean you try to find out how fake news on these three fundamental things affect voters?

Yes. In the second stage, we shared verified news with a control group, and fake news regarding public goods and security issues to a treatment group.

Then, we try to find out whether people become more 'in-group oriented' when they get fake news.

For example, if you are a Brahmin and you see fake negative or positive news on WhatsApp against the Dalits, whether you will be more interested in voting for the Brahmin candidate.

We are trying to find out how caste relations play a role in voting after they get to read fake news.

Do you think fake news, in fact, influences people while voting?

We do not have the data right now on India. So, I cannot definitely say how it influences. But so far, we have seen how fake news affects American voters.

What academic studies have found is that fake news only reinforces people's views.

That means fake news is not a disease, it is more of a catalyst.

People bring in their own demons and they are not created by fake news.

For example, if you are a racist and bigoted against blacks and if you are a Brahmin and bigoted against Dalits, and when you see fake news about the other person, it only reinforces your negativity about the other person.

It means fake news does not change your views.

This is what other scholars have found in other countries, but we do not know how it is in India.

If people are divided along party lines in the US, it is based on caste lines in India.

Is fake news a new phenomenon? Has it not been there before?

Yes, fake news has always been there, the only thing that has changed is the medium. For example, you always had newspapers and television channels belonging to certain political parties.

The difference is, today, fake news is decentralised and nobody controls it.

Do you feel compared to the last US elections in which fake news played a major influence, will it be more in India?

As a by-product of our study, what we expect is, fake news is largely being used as a tool for mobilisation by political parties. Of course, this is not the focus of our study.

For example, the BJP used identity-based messages on social media to drive out their voters.

Say, if many of the voters who had voted for the BJP in 2014 are not happy with the party and unlikely to vote again, what the party did was, use messages to bring these voters to the polling booth.

Unless you make them interested, they will not go. That's where identity-based messages were used.

Are these messages to bring their voters to the polling booth, based on fake news?

It can be real news also. As that is not part of our study, we did not try to find out what kind of messages worked on social media to get voters more interested in their candidates.

Are the political parties of the developed world using social media like WhatsApp to mobilise people during the elections?

No, not much. They use Facebook more than WhatsApp.

WhatsApp is something that is unique to India and other developing countries because sharing messages on WhatsApp takes much less data than Facebook. For example, Jiophone, though not a full-fledged smart phone, has WhatsApp.

Do you feel your study will be helpful to political parties who want to use social media to mobilise people?

Yes, of course. This study will not only be useful to academics, but also to political parties, governments and practitioners.

That is because it shows fake news wouldn't change people's opinion. It also shows what kind of messages change people's opinion or reinforce them.

With this in hand, political parties can target their voters better. They will also know what voters want from them, and what kind of messages work.