53 research outputs found
Replication Data for: Educative Interventions to Combat Misinformation: Evidence From a Field Experiment in India
No full textMisinformation makes democratic governance harder, especially in developing countries. Despite its real-world import, little is known about how to combat misinformation outside of the U.S., particularly in places with low education, accelerating Internet access, and encrypted information sharing. This study uses a field experiment in India to test the efficacy of a pedagogical intervention on respondents' ability to identify misinformation during the 2019 elections (N=1224). Treated respondents received hour-long in-person media literacy training in which enumerators discussed inoculation strategies, corrections, and the importance of verifying misinformation, all in a coherent learning module. Receiving this hour-long media literacy intervention did not significantly increase respondents' ability to identify misinformation on average. However, treated respondents who support the ruling party became significantly less able to identify pro-attitudinal stories. These findings point to the resilience of misinformation in India and the presence of motivated reasoning in a traditionally non-ideological party system
Replication Data for: Misinformation and Support for Vigilantism: An Experiment in India and Pakistan
No full textThis repository includes data files and R scripts to replicate all original analysis from the paper and appendix
sj-pdf-1-hij-10.1177_19401612231158770 - Supplemental material for “I Don’t Think That’s True, Bro!” Social Corrections of Misinformation in India
No full textSupplemental material, sj-pdf-1-hij-10.1177_19401612231158770 for “I Don’t Think That’s True, Bro!” Social Corrections of Misinformation in India by Sumitra Badrinathan and Simon Chauchard in The International Journal of Press/Politics</p
La confianza en las noticias en América Latina ¿la misma historia de siempre?
No full textAl igual que en el resto del mundo, la confianza en las noticias en América Latina ha caído en años recientes. Sin embargo, datos de encuestas realizadas en Argentina, Brasil, Chile y México demuestran que los niveles de confianza varían significativamente de un contexto a otro y pueden estar relacionados con una variedad de fenómenos, desde las carencias del propio periodismo hasta el malestar social o el antagonismo entre políticos y periodistas, escriben Camila Mont’Alverne, Amy Ross Arguedas, Benjamin Toff y Sumitra Badrinathan (Reuters Institute for the Study of Journalism, University of Oxford)
The Politics Of Misinformation In India: Prevalence, Mechanisms, Solutions
No full textWhat is the scope of the misinformation problem in India, and what kinds of solutions can we implement to improve the quality of information processing? This dissertation explores this question by gathering original data collected over 3 years through three empirical projects that include a diversity of methods -- face-to-face surveys, online experiments, field interventions –- as well as a range of populations in the country, from highly-educated and English-speaking internet users, to those gaining access to mobile phones for the first time. These studies also cover different times centered around the pivotal 2019 general election in India: descriptive data was collected in the year before the election, the field intervention was rolled out during the election, and online experiments were conducted after the election. Overall, the dissertation sheds light on the role of new and developing media in India in fostering novel channels for the communication of information and misinformation, describes the prevalence of fake news and the mechanisms through which it spreads, and offers solutions through field interventions and experimental treatments aimed at reducing the uptake of political misinformation
Conspiracy Theories and Miracle Cures: Fighting COVID-19 Misinformation in India
No full textWe evaluate a correction strategy to reduce COVID-19 misinformation in India through an online experiment (N=1500). We present respondents with corrections from unlikely sources and evaluate the effect of these treatments on perceived accuracy of COVID-19 miracle cures and conspiracy theory headlines
Norms of Deference, Gender, and Belief in Misinformation
No full textMisinformation is now widely acknowledged as a leading social concern, especially since social media offers a fertile ground for its proliferation on a larger scale. As a result, scholars and policymakers have, over the past decade, sought to better understand the psychological mechanisms leading to the expression of belief in dubious claims. Beyond these psychological factors driving misinformation endorsement at the individual level, the endorsement of misinformed beliefs is also conditioned by societal and social norms, including norms around conformity to group-congruent beliefs and social pressure.
In the prevailing scholarship concerning susceptibility to misinformation, considerable attention has been directed towards societal divisions and affiliations based on demographic and social stratifications, notably encompassing factors such as race, religion, and party identity. Scholars posit that individuals exhibit a predisposition to consume information that aligns with the interests of their respective social groups. Conversely, scant consideration has been given to a very fundamental social unit within society, the family, and its potential role in fostering and perpetuating misinformed beliefs. Further, work in this area rarely discusses the subject of norms emphasized and enforced within familial frameworks. This study seeks to investigate the influence of familial norms on the endorsement of misinformation. In particular, we focus on a specific category of norms surrounding deference to adults and elders within familial contexts. We aim to highlight how adherence to such norms may engender a propensity to endorse misinformation, and whether experimental manipulations aimed at diminishing the perceived acceptance of these deferential norms can subsequently mitigate the endorsement of misinformation.
In societies with strong norms of deference for elders, such social dynamics should be especially relevant to the public endorsement of misperceptions grounded in traditional, religious and/or conservative belief systems, as such beliefs are likely to be perceived as dominant amongst those to whom deference is due. In more patriarchal contexts or those with traditional gender-based values, such attitudes of deference are especially likely among women. In societies in which misinformed beliefs about traditional medicine, home remedies, or superstitions are likely to be passed down from generation to generation within households, it stands to reason that stronger family norms around deference to elders might shape greater endorsement of misinformed beliefs. However, scholarship to date has not explored whether adherence to such norms affects vulnerability to misinformation grounded in traditional belief systems.
We accordingly ask two questions. Our first question is observational: what is the descriptive relationship between norms of deference to elders and endorsement to misinformation? Second, in order to better understand the relationship between these norms and endorsement of misinformation, we field an experiment in which we manipulate how prevalent such norms of deference to family elders are perceived to be. This leads us to ask: can reducing the perceived acceptance of norms of deference to elders decrease belief in misinformation and increase willingness to correct elders?
To answer these questions we field an in-person survey in Bihar, India with a sample of approximately 6,000 teenagers aged 13 to 18, drawn from 583 villages across the state, and representative of the rural school-going adolescent population in the state. We focus on adolescents, as young adults are vulnerable to norms of deference and respect for elders within the family. More generally, we deem adolescence a crucial time to study these questions. Misinformation studies consistently demonstrate that a factor limiting the effectiveness of treatments is motivated reasoning. This is especially crucial in the Indian case where strong partisans can be resistant to corrective information. We argue that targeting adolescents may overcome these issues: while they have not yet voted, they are actively developing social and political identities; thus, adolescence is a time when attitudes and behaviors may be less crystallized and hence more malleable.
To understand the correlation between norms of deference and endorsement of misinformation, our pre-test baseline survey measures attitudes towards collective decision making, deference to adults, and gender norms within the household. Crucially, these attitudes are measured in both adolescents and one parent/guardian present in the household. Consequently, we also measure the (potential) presence of elders during the adolescent's interview, which allows us to shed light on implicit deference and conformity to familial authority. Next, we implement a priming experiment where we assign half the sample to a treatment group and half to control. Treatment group respondents receive information that recent survey data from credible sources demonstrates that young Indians are increasingly engaging in individualistic decision-making and hence that they are decreasingly likely to defer to elders. Our goal with this prime is to provide information that relieves adolescent pressure to adhere to norms of deference towards elders and within the family, by reducing the perceived prevalence or acceptance of this norm. We measure how such manipulation impacts two related outcomes: endorsement of misinformation (as part of a discernment task) and self-reported willingness to correct misinformation when disseminated by an elder. Perceived accuracy of headlines (both true and false) is measured for both our primary respondents (adolescents) as well as one adult present in the household.
Respondent recruitment for this study commenced in September 2023 and baseline survey data collection took place October-December 2023. This pre-analysis plan was submitted during data collection (March-April 2024), but prior to any researcher access to the dataset. In addition to this pre-analysis plan, we plan to file additional pre-analyses plans using the same data to (1) study the effects of a conjoint experiment embedded in the same survey as the current study and (2) study the effects of a media literacy RCT of which this sample is part. There may also be additional analyses based on these data in the future
Misinformation and Anti-Minority Vigilantism
No full textActs of vigilante violence, most often targeted against religious and sectarian minorities, have become increasingly common in India and Pakistan in recent years. Journalistic accounts often link anti-minority vigilantism to misinformation on social media, leading politicians and activists to call on tech platforms to deal with such false information. Yet, empirically, the precise link between misinformation and support for vigilante violence remains largely unidentified. In this paper we fill this gap by asking what role, if any, misinformation plays in support for vigilantism. On the one hand, false information with a call to action could directly result in collective violence, especially in contexts where such information is readily believed. On the other hand, citizens could be motivated to engage in violence as a result of long-standing societal cleavages, with misinformation playing little to no causal role. To answer these questions we designed a jointly registered conjoint experiment in India and Pakistan using a novel audio medium treatment. Through realistic vignette scenarios describing vigilante acts in response to social media rumors, we manipulate corrections to misinformation, state positionality, and elite messaging and measure whether these variables affect support for vigilante violence
- …
