The myth of an informed choice: Algorithms and information echo chambers on the rise in Indian politics

The digital revolution was supposed to bring the world, and the voters in a democracy together to discuss ideas and increase accountability and transparency. Has it ended in creating perpetual echo chambers?.—.THE 2024 general elections in India are going to be the largest exercise of electoral democracy in history— repeating the feat of the 2019 general elections in the country..Constitutionally speaking, Indian elections are premised on voters making an informed choice. However, the ideal of informed choices exists against the backdrop of extravagant practices of information manipulation, with most available sources of information remaining vulnerable to the holder of the strongest purse..It may then be the case that an informed choice is only as strong as the politics of the time will allow..The results of a study undertaken by this author indicate a similar trend— that partisanship among citizens may deeply influence what they perceive to be information about politics, and such biases among the electorate can be, and are, exploited by political elites to create a compromised and unequal ecosystem of political information..Recently, Prime Minister Narendra Modi presented the first-ever National Creator's Award, aimed at awarding digital creators (or quite simply, social media influencers) across various categories..“Indian elections are premised on voters making an informed choice..Such influencers have a mass following among the burgeoning youth population of India and also enjoy the full backing of social media algorithms that incentivise short-form content..For the wealthy politician, piggybacking off of such influencers is an easy way to appeal to the young crowd, without sounding preachy or political. Even as the crowd at the event cheered for Modi, the Prime Minister astutely stated, "The Lok Sabha elections will take place in the next few days. Don't think that this event is for that"..Also read: Black economy persists with or without electoral bonds.It may not be a coincidence then that the first such award was unveiled weeks before the general elections. Such events and influencer exposure, are likely to make their way rapidly into the social media algorithms of the most uninterested voter as well, allowing for a form of subtle, yet strong, political exposure..The above is a snapshot of how attention capture and opinion formation occurs in the digital era of contemporary politics— the ecosystem of political communication is one of algorithms, views, clickbait and content, all held together by expensive campaigns and events that are designed to target the citizen..This is different from the politicking of pre-digital days, when targeting citizens with specific, almost individually tailored propaganda, was extremely difficult. All that has now changed with the rise of online cookies and algorithm-based targeting of social media consumers..“Recently, Narendra Modi presented the first-ever National Creator's Award, aimed at awarding digital creators (or quite simply, social media influencers) across various categories..The Bharatiya Janata Party (BJP) alone has spent ₹37 crores on Meta ads, which is three times the money spent on Meta ads by the Congress party. Indian elections have become notorious for instances of fake news, content manipulation, targeted advertising and, more recently, social media censorship of non-Union-government-friendly voices..Given that Indian elections stress voting being an informed choice, what does it mean for Indian voters then, if information is selectively available and accessible?.The importance of free and fair elections that leverage an informed choice lies in the premise that any and all information regarding political issues is always and easily available for voters to engage with..Therefore, for there to be an informed choice, there cannot be barriers to information exposure. However, with the rise of algorithm-based targeting, digital marketing and social media-driven discourse, can it be possible that the voters are selectively exposed to information, creating possible echo chambers of discourse?.How voters believe they seek information.A national study was conducted along with C-Voter (which is an Indian international polling agency headquartered in Delhi) by this author to gauge how exposed citizens are to information regarding issues they deem important for national elections..Also read: A Hard Year For Federalism— Part 2 of the interview with P.D.T. Achary.Respondents were also classified as those who voted for the BJP/National Democratic Alliance (NDA) in the last general elections, and those who voted for any opposition party..The goal of the study was to check the self-perception and exposure of voters towards key issues, and any possible effects that partisanship may have on this exposure..“The Bharatiya Janata Party alone has spent ₹37 crores on Meta ads, which is three times the money spent on Meta ads by the Congress..Respondents were checked for baseline attitudes towards elections in general, and there appeared to be no significant effect of partisanship on the attitudes. Almost 40 percent of all respondents said they believed voting is a right and they always cast their vote, while almost 30 percent of all respondents believe it is a very important right and they make sure to cast their vote..The difference between opposition and BJP voters was small, with almost 30 percent of each saying it is a very important right and that they make sure to vote, while 37 percent of opposition and 43 percent of NDA voters felt that voting is a constitutional right and it is important to vote..For baseline attitudes, the study also checked the sources from which respondents believe they get most of their primary information regarding election issues..Here, social media platforms and television news channels emerged as the overwhelming source of information for most respondents regardless of partisanship, with almost 30 percent of respondents (NDA, opposition and total) saying they got their information primarily from social media, while 30 percent of all respondents saying they get their information primarily from news channels..“The results from this study point towards a potential for echo chambers, either self-created or manufactured, or both, regarding information that voters believe is key for Indian general elections..NDA voters were slightly more likely to get their information primarily from television (35.5 percent), compared to 26 percent of opposition voters..The baseline attitudes largely showed that regardless of partisanship, voters are invested in the political process, like to vote, and get their information predominantly from television or social media..Also read: The curious tale of missing migrant voters in India.It should be noted that as these figures are self-assessed by voters, they should not be confused with quantity, as a voter who consumes less information and a voter who consumes a lot of information, can both say they get the majority of their level of desired information from any given source..Thus, the baseline acts as an indicator of how voters perceive their own information-seeking behaviour and actions..Given that there have been no major influences of partisanship so far, one can, prime facie, assume that the information-seeking behaviour of voters actually are homogenous and non-selective..“NDA voters were slightly more likely to get their information primarily from television (35.5 percent), compared to 26 percent of opposition voters..Thus, based on just this, one could assume that the consumption and exposure to information would reflect this trend. However, significant differences emerged here..Unequal information exposure and echo chambers along partisan lines.The second part of the study tested the self-assessment of voters regarding five broad categories of information that are part of the general discourse between opposition–incumbent interactions during the years between general elections..Voters were asked two questions on a theme. First, to assess how important they felt an issue was for Lok Sabha elections. Second, regarding how much information they feel that they have received about that issue since the last general election took place..For the first question, voters were asked to choose any response between 'can't say', 'not important at all', 'not important to a certain extent', 'important' and 'very important'..Also read: 'One nation, one election' will test the limits of our democracy.For the second question, respondents were asked to choose between 'can't say', 'completely disagree', 'disagree', 'agree' and 'completely agree'. The survey was designed to gauge broad themes and should not be interpreted as conclusive opinions regarding more specific issues and causes..Unemployment and jobs.The first theme was regarding unemployment and the availability of jobs. Here, almost 59 percent of all respondents said they believed that issues of unemployment and job availability were a very important aspect of national elections..The fraction was similar for NDA and opposition voters, at 57 percent and 60 percent respectively. However, when asked whether voters felt they have regularly received information regarding what the national government has been doing about this issue since the last general election, a significant partisan difference emerged..While 23 percent of NDA voters completely agreed with the statement, more than double the fraction of opposition voters (49 percent) said they completely disagreed..Interestingly, 35 percent of NDA voters completely disagreed with the statement as well, putting them almost in the middle of opposition voters who completely disagreed and NDA voters who completely agreed..Of all respondents, 43 percent completely disagreed regarding receiving regular information on the actions of the national government on issues related to unemployment and jobs, thus leaning towards the opposition sentiment..China, Pakistan and national security.The second theme was regarding issues of national security and bilateral issues with China and Pakistan. This theme was included keeping in mind the media fanfare around surgical strikes during the pre-election season, which the opposition and anti-NDA voters had declared to be the politicisation of national security issues..“Of all respondents, 43 percent completely disagreed regarding receiving regular information on the actions of the national government on issues related to unemployment and jobs, thus leaning towards the opposition sentiment..Around 53 percent of all respondents said that national security and issues with China and Pakistan were a very important aspect of national elections, and there was only a small difference between NDA voters (56 percent), and opposition voters (50 percent) who believed the same..However, when asked whether they have regularly received information regarding this issue since the last elections, almost 48 percent of NDA voters completely agreed, compared to less than half of opposition voters who echoed the sentiment (23 percent)..Also read: Supreme Court directs ECI to respond in a plea seeking to tally EVM vote count with VVPAT.In stark contrast, almost a third of opposition voters completely disagreed regarding receiving information, which was almost double that of NDA voters (16 percent) who felt the same. Of all respondents, almost 24 percent completely disagreed while 34 percent completely agreed..Inflation and cost of living.The third theme looked at inflation and cost of living challenges as an issue to be discussed during national elections. Again, the trend of non-partisan opinion continued, with 59 percent of all respondents saying this was a very important aspect of national elections, with little difference between the opposition (60 percent) and NDA (56 percent) believing the same..Once again, however, when asked whether respondents regularly received information regarding how the national government has dealt with such issues, a whopping 48 percent of opposition voters completely disagreed..This was almost double that of NDA voters (27 percent) who completely agreed with the statement. Interestingly, an equal number of NDA voters completely disagreed as well, suggesting that even NDA voters may be divided regarding this issue..The overall trend leaned towards the opposition camp, with almost 39 percent of respondents completely disagreeing with receiving such information since the elections..Corruption and crime.The fourth theme looked at political corruption and crime as a key issue during national elections. This was included keeping in mind the corruption versus development rhetoric that is often at the heart of opposition–incumbent clashes, with corruption and crime also acting as levers of anti-incumbency in some cases..“Almost 60 percent of all respondents said that discussing issues of corruption by elected representatives was a very important aspect of national elections..As with the previous themes, almost 60 percent of all respondents said that discussing issues of corruption by elected representatives was a very important aspect of national elections, with opposition (57 percent) and NDA factions (63 percent) largely agreeing..However, responses for whether a respondent received enough information regarding political corruption during national elections were expectedly polarising, with 48 percent of NDA voters saying they completely agreed and 18 percent saying they agreed..Also read: Modi Magic Loses its Spell: Is it Just the Beginning?.A fraction of NDA voters (16 percent) also completely disagreed with the statement. For the opposition, only 29 percent completely disagreed, and 23 percent completely agreed, indicating a degree of information asymmetry within the opposition voters..Of the total respondents, the opinion was halfway between the NDA and opposition voters, with 34 percent saying they completely agreed with receiving information on political corruption and crime. In the same trend, 24 percent of all respondents completely disagreed with the statement..Religion and culture.The fifth theme centered on the protection of religion and cultural beliefs as a key aspect of national elections, an issue that has quickly come to the forefront of national politics since 2014..It should be noted that as key developments in this area, such as the Ram Mandir inauguration were done only recently, much after the survey, the response trend may have changed and become more polarising over time..This theme was the only one in which there were polarising responses for both questions, instead of just the second question regarding information exposure. On whether protection of religious and cultural beliefs was a key aspect of national elections, responses were almost equally divided, with 31 percent of respondents saying such issues are not important at all for national elections, while 33 percent saying they were very important, and 21 percent saying they were important..In terms of partisanship, NDA and opposition voters were almost equally split, with 37 percent of opposition respondents saying it was not important at all, while 41 percent of NDA voters saying it was very important..For the second question, when asked if they felt they have regularly received information regarding actions of the national government on issues of religion and culture, the polarisation between respondents remained similar..Almost 35 percent of opposition voters completely disagreed with receiving information, whereas 32 percent and 26 percent of NDA respondents completely agreed and agreed, respectively..Also read: UP Elections: Touching Stories of Migrant Workers' Suicides in Vast Interiors.Of all the respondents, 30 percent completely disagreed with receiving such information, while 23 percent and 22 percent completely agreed and agreed respectively..Preaching to the converted in Indian politics.The study preliminarily points towards significant silos of information exposure which are clearly split along partisan lines. Prima facie, except for the issue of religion and culture, voters are generally in agreement about whether or not something is a key issue, and exhibit strong and non-partisan investment in Indian politics..However, they are divided as to whether they receive or are exposed to enough information regarding what they believe to be key issues, along partisan lines..“One must ask if there is an incentive to deconstruct echo chambers or build controlled ones that are just expansive enough to win enough of the vote share required for a seat..To revisit the discussion on informed choice, the premise of that hinges on freely available widespread information. At the same time, social media and television news emerged as the predominant sources of information exposure, both of which are highly susceptible to manipulation and targeting by wealth and political power..Given this, the results from this study point towards a potential for echo chambers, either self-created or manufactured, or both, regarding information that voters believe is key for Indian general elections..Of course, as this is a public opinion study, there remains the issue of the chicken and the egg— do voters self-segregate information, or is it done for them, and then over time becomes organic?.While this can be debated, it may be the incorrect question to ask, as both scenarios are detrimental to the democratic process. In an environment where targeting attention through such means as algorithms, infotainment, content creation and influencers has become rather easy, the commitment towards creating homogenous and non-selective information access stands compromised..Any political power may want to perpetuate such echo chambers, ensuring that information flows enough only to satiate their aligned groups, and the rest is just noise that will not win them votes..With enough time, targeting and influence, these echo chambers and information asymmetries become the backdrop against which fake news, hate speech, censorship and manipulated information prevail..Also read: Uttarakhand HC asks ECI to consider issuing directions for virtual campaigning and voting in view of surge in COVID cases.This has become more of a concern in recent times as elected representatives have become exponentially more wealthy over time, with the 2019 Lok Sabha being one of the most expensive in history..The assumption for digital spaces was that they are, by nature, more democratic, as they break away from the limitations of geography and physical resources. Everyone has a phone, and most people are on social media, leaving them exposed to information on an almost permanent basis..Therefore, it is concerning that the limitations of the past in information exposure still exist, and are probably made worse. With enough time, information silos or echo chambers can become self-perpetuating, where voters actively just disengage or dissociate if any information goes against their beliefs or interest..Given that elections are deeply a function of beliefs, such trends make the quality of discourse poorer. What may be worse, is that if correctly leveraged, echo chambers may make it difficult for citizens to accurately critique their representatives, even those that they are aligned with..An informed choice, therefore, is premised not only on the voters' ability to make one but also on the voters' ability to access any information that is needed to do it..In information echo chambers, the information preaches to the converted, and may perpetuate the differences among voters rather than homogenise them on common issues. Such was the case in this study, where voters agree on issues, but remain divided on exposure..India has a first-past-the-poll system, where winning a fraction of the vote share can make a majority government. In such a case, one must ask if there is an incentive to deconstruct echo chambers or build controlled ones that are just expansive enough to win enough of the vote share required for a seat..Over time, such politics can be a race to the bottom, with under-prioritisation of key issues and over-prioritisation of others. An informed choice, then remains a lofty dream, and perhaps one that has become harder to achieve now than it was before.