Do celebrity endorsements matter? And if so, how can celebrities communicate effectively? We conduct a nationwide Twitter experiment in Indonesia promoting vaccination. Celebrity messages are 72% more likely to be passed on or liked than similar messages without a celebrity’s imprimatur. In total, 66% of the celebrity effect comes from authorship, compared to passing on messages. Citing external medical sources decreases retweets by 27%. Phone surveys show that those randomly exposed to messaging have fewer incorrect beliefs and report more vaccination among friends and neighbours. The results can inform public health campaigns and celebrity public service more generally.

Partisan segregation within the news audience buffers many Americans from countervailing political views, posing a risk to democracy. Empirical studies of the online media ecosystem suggest that only a small minority of Americans, driven by a mix of demand and algorithms, are siloed according to their political ideology. However, such research omits the comparatively larger television audience and often ignores temporal dynamics underlying news consumption. By analyzing billions of browsing and viewing events between 2016 and 2019, with a novel framework for measuring partisan audiences, we first estimate that 17% of Americans are partisan-segregated through television versus roughly 4% online. Second, television news consumers are several times more likely to maintain their partisan news diets month-over-month. Third, TV viewers’ news diets are far more concentrated on preferred sources. Last, partisan news channels’ audiences are growing even as the TV news audience is shrinking. Our results suggest that television is the top driver of partisan audience segregation among Americans.

We use a series of experiments to understand whether and how people’s beliefs about their own abilities are biased relative to the Bayesian benchmark and how these beliefs then affect behavior. We find that subjects systematically and substantially overweight positive feedback relative to negative (asymmetry) and also update too little overall (conservatism). These biases are substantially less pronounced in an ego-free control experiment. Updating does retain enough of the structure of Bayes’ rule to let us model it coherently in an optimizing framework, in which, interestingly, asymmetry and conservatism emerge as complementary biases. We also find that exogenous changes in beliefs affect subjects’ decisions to enter into a competition and do so similarly for more and less biased subjects, suggesting that people cannot “undo” their biases when the time comes to decide.

Although it is under-studied relative to other social media platforms, YouTube is arguably the largest and most engaging online media consumption platform in the world. Recently, YouTube’s scale has fueled concerns that YouTube users are being radicalized via a combination of biased recommendations and ostensibly apolitical “anti-woke” channels, both of which have been claimed to direct attention to radical political content. Here we test this hypothesis using a representative panel of more than 300,000 Americans and their individual-level browsing behavior, on and off YouTube, from January 2016 through December 2019. Using a labeled set of political news channels, we find that news consumption on YouTube is dominated by mainstream and largely centrist sources. Consumers of far-right content, while more engaged than average, represent a small and stable percentage of news consumers. However, consumption of “anti-woke” content, defined in terms of its opposition to progressive intellectual and political agendas, grew steadily in popularity and is correlated with consumption of far-right content off-platform. We find no evidence that engagement with far-right content is caused by YouTube recommendations systematically, nor do we find clear evidence that anti-woke channels serve as a gateway to the far right. Rather, consumption of political content on YouTube appears to reflect individual preferences that extend across the web as a whole.

Since the 2016 US presidential election, the deliberate spread of misinformation online, and on social media in particular, has generated extraordinary concern, in large part because of its potential effects on public opinion, political polarization, and ultimately democratic decision making. Recently, however, a handful of papers have argued that both the prevalence and consumption of “fake news” per se is extremely low compared with other types of news and news-relevant content. Although neither prevalence nor consumption is a direct measure of influence, this work suggests that proper understanding of misinformation and its effects requires a much broader view of the problem, encompassing biased and misleading—but not necessarily actually incorrect—information that is routinely produced or amplified by mainstream news organizations. In this paper, we propose an ambitious collective research agenda to measure the origins, nature, and prevalence of misinformation, broadly construed,
as well as its impact on democracy. We also sketch out some illustrative examples of completed, ongoing, or planned research projects that contribute to this agenda.

The DeGroot model has emerged as a credible alternative to the standard Bayesian model for studying learning on networks, offering a natural way to model naïve learning in a complex setting. One unattractive aspect of this model is the assumption that the process starts with every node in the network having a signal. We study a natural extension of the DeGroot model that can deal with sparse initial signals. We show that an agent’s social influence in this generalized DeGroot model is essentially proportional to the degree-weighted share of uninformed nodes who will hear about an event for the first time via this agent. This characterization result then allows us to relate network geometry to information aggregation. We show information aggregation preserves “wisdom” in the sense that initial signals are weighed approximately equally in a model of network formation that captures the sparsity, clustering, and small-world properties of real-world networks. We also identify an example of a network structure where essentially only the signal of a single agent is aggregated, which helps us pinpoint a condition on the network structure necessary for almost full aggregation. Simulating the modeled learning process on a set of real-world networks, we find that there is on average 22.4 percent information loss in these networks. We also explore how correlation in the location of seeds can exacerbate aggregation failure. Simulations with real-world network data show that with clustered seeding, information loss climbs to 34.4 percent

“Fake news,” broadly defined as false or misleading information masquerading as legitimate news, is frequently asserted to be pervasive online with serious consequences for democracy. Using a unique multimode dataset that comprises a nationally representative sample of mobile, desktop, and television consumption, we refute this conventional wisdom on three levels. First, news consumption of any sort is heavily outweighed by other forms of media consumption, comprising at most 14.2% of Americans’ daily media diets. Second, to the extent that Americans do consume news, it is overwhelmingly from television, which accounts for roughly five times as much as news consumption as online. Third, fake news comprises only 0.15% of Americans’ daily media diet. Our results suggest that the origins of public misinformedness and polarization are more likely to lie in the content of ordinary news or the avoidance of news altogether as they are in overt fakery.

The application of modern NLP and ML techniques to large-scale datasets can generate implicit graphs that are so densely connected as to be unusable when rendered as node-link diagrams. We present a two-stage approach to extracting usable, map-like layouts from large, dense input graphs. This approach uses edge-cutting strategies based on node and edge metrics to reduce a graph to a skeletal structure showing only essential relationships, before filling in the resulting communities to create dense but usable layouts. Through a case study on a 145k-document adversarial health communication dataset, we show that each edge-cutting strategy has advantages and disadvantages, and that the appropriate choice of strategy depends on the data, user, and task.

Social networks can facilitate informal lending and risk-sharing in situations where for-
mal institutions such as banks and insurance companies do not exist. The social collateral approach provides an analytically tractable framework that can be used to analyze a wide range of informal transfers. Moreover, the approach is easily amenable to empirical analysis.

We study the role of ethnicity in experimental labor markets where “employers” determine wages of “workers” who perform a real effort task. This task requires a true skill which we show is not affected by minority status. In some treatments, we provide subtle priming to employers about minority status of workers as commonly depicted on Chinese “Hukou” identification system. We conduct our experiments at two sites located in provinces that differ by their historical shares of ethnic groups in the population. We find that: (1) Han and minority workers are equally productive in both provinces; (2) in the diverse province, there is no difference in the wages between Han and minority workers; (3) in the non-diverse province, minority workers receive 4%-7% lower wages than Han workers.

Increasingly detailed consumer information makes sophisticated price discrimination possible. At fine levels of aggregation, demand may not obey standard regularity conditions. We propose a new randomized sales mechanism for such environments. Bidders can "buy-it-now" at a posted price, or "take-a-chance" in an auction where the top d > 1 bidders are equally likely to win. The randomized allocation incentivizes high valuation bidders to buy-it-now. We analyze equilibrium behavior, and apply our analysis to advertiser bidding data from Microsoft Advertising Exchange. In counterfactual simulations, our mechanism increases revenue by 4.4% and consumer surplus by 14.5% compared to an optimal second-price auction.

We develop a model in which connections between individuals serve as social collateral to enforce informal insurance payments. We show that: (1) The degree of insurance is governed by the expansiveness of the network, measured with the per capita number of connections that groups have with the rest of the community. Two-dimensional networks---like real-world networks in Peruvian villages---are sufficiently expansive to allow very good risk-sharing. (2) In second-best arrangements, insurance is local: agents fully share shocks within, but imperfectly between endogenously emerging risk-sharing groups. We also discuss how endogenous social collateral affects our results. (JEL D02, D31, D70)

Social learning is a rapidly growing field for empirical and theoretical research in economics. We encounter social learning in many economically important phenomena, such as the adoption of new products and technologies or job search in labor markets. We review the existing empirical and theoretical literatures and argue that they have evolved largely independently of each other. This suggests several directions for future research that can help bridge the gap between both literatures. For example, the theory literature has come up with several models of social learning, ranging from naïve DeGroot models to sophisticated Bayesian models whose assumptions and predictions need to be empirically tested. Alternatively, empiricists have often observed that social learning is more localized than existing theory models assume, and that information can decay along a transmission path. Incorporating these findings into our models might require theorists to look beyond asymptotic convergence in social learning.

We present a simple auction mechanism which extends the second-price auction with reserve and is truthful in expectation. This mechanism is particularly effective in private value environments where the distribution of valuations are irregular. Bidders can “buy-it-now”, or alternatively “takea- chance” where the top d bidders are equally likely to win. The randomized take-a-chance allocation incentivizes high valuation bidders to buy-it-now. We show that for a large class of valuations, this mechanism achieves similar allocations and revenues as Myerson’s optimal mechanism, and outperforms the second-price auction with reserve. In addition, we present an evaluation of bid data from Microsoft’s AdECN platform. We find the valuations are irregular, and counterfactual experiments suggest our BINTAC mechanism would improve revenue by 11% relative to an optimal second-price mechanism with reserve.

We conduct a field experiment in a large real-world social network to examine how subjects expect to be treated by their friends and by strangers who make allocation decisions in modified dictator games. While recipients’ beliefs accurately account for the extent to which friends will choose more generous allocations than strangers (i.e. directed altruism), recipients are not able to anticipate individual differences in the baseline altruism of allocators (measured by giving to an unnamed recipient, which is predictive of generosity towards named recipients). Recipients who are direct friends with the allocator, or even recipients with many common friends, are no more accurate in recognizing intrinsically altruistic allocators. Recipient beliefs are significantly less accurate than the predictions of an econometrician who knows the allocator’s demographic characteristics and social distance, suggesting recipients do not have information on unobservable characteristics of the allocator.

We conduct online field experiments in large real-world social networks in order to decompose prosocial giving into three components: (1) baseline altruism toward randomly selected strangers, (2) directed altruism that favors friends over random strangers, and (3) giving motivated by the prospect of future interaction. Directed altruism increases giving to friends by 52 percent relative to random strangers, while future interaction effects increase giving by an additional 24 percent when giving is socially efficient. This finding suggests that future interaction affects giving through a repeated game mechanism where agents can be rewarded for granting efficiency-enhancing favors. We also find that subjects with higher baseline altruism have friends with higher baseline altruism.

This paper builds a theory of trust based on informal contract enforcement in social networks. In our model, network connections between individuals can be used as social collateral to secure informal borrowing. We de…ne network-based trust as the highest amount one agent can borrow from another agent, and derive a reduced-form expression for this quantity which we then use in three applications. (1) We predict that dense networks generate bonding social capital that allows transacting valuable assets, while loose networks create bridging social capital that improves access to cheap favors like information. (2) For job recommendation networks, we show that strong ties between employers and trusted recommenders reduce asymmetric information about the qualityof job candidates. (3) Using data from Peru, we show empirically that network-based trust predicts informal borrowing, and we structurally estimate and test our model.