Did you hear about Clarence Thomas? Measuring public attention toward the Supreme Court

Abstract

Research shows that U.S. Supreme Court events garner media attention, but the extent of public engagement remains uncertain. We examine online engagements with Supreme Court coverage, spanning the periods before, during, and after a scandal involving Justice Clarence Thomas drew broad media attention in April 2023. Analyzing internet browsing behavior of a nationally representative sample and Google Trends data, we discovered minimal interest in the Court prior to the scandal (3% seeking information). After the scandal, attention increased by 300% and stayed high for weeks. While Democratic participants were the most likely to seek out information about the Court following the scandal, we still observe strong effects among Republican participants, too. These results highlight how extra-judicial events draw attention to the Court and lead to lasting interest in it. Furthermore, our digital trace analysis of Supreme Court information seeking offers insights for scholars studying a variety of political topics.

On April 6, 2023, ProPublica published a three-thousand-word story detailing Supreme Court Justice Clarence Thomas’s decades of luxurious travel at the largess of a conservative megadonor named Harlan Crow (Kaplan et al., 2023). The story suggested the trips were gifts that should be disclosed on annual financial reports, and ethics experts counseled Thomas’s failure to do so violated federal law. The story itself was not new, as Court reporters wrote about Thomas and Crow’s relationship in the mid-aughts (Toobin, 2008). But this time, the report hit while Court support was at a record low (Gibson, 2024), and it struck a chord with a national media concerned about the Court’s impartiality. Would the public care about the story this time?

Decades of research shows that media discussion of a political scandal does not automatically translate to public awareness of it, let alone political consequences for it (see e.g., Nyhan, 2014). When a story creates a “media storm” in which outlets fixate on a particular issue with coverage that is both explosive and skewed, however, political elites feel pressure to address that issue with congressional hearings and legislative action (Boydstun, 2013; Thomas, 2017; Walgrave et al., 2017). Importantly, such coverage also draws popular attention toward an issue and compounds the pressure on elites to act (Boydstun et al., 2014). With scandals particularly, people’s engagement with coverage significantly affects their feelings toward the elected officials engaged in wrongdoing and the institutions in which they serve (Doherty et al., 2011; Pereira and Waterbury, 2019). Coverage of the Thomas scandal produced sustained political attention and some initial attempts at legislative action (de Vogue, 2023), but did the public notice or care about the story, given how little attention people afford the unelected Court in the first place?

In this study, we seek to understand how sudden and sustained coverage of a judicial scandal affects information seeking about the Court. The Thomas story offers a unique opportunity to study popular engagement with an institution that lacks an electoral connection with the public and thus receives little popular or media attention, so we use it to examine people’s attention toward the Supreme Court before, during, and after the ProPublica story went public. Using unobtrusive observations of the internet browsing behavior of a representative sample of 1,000 American adults provided by YouGov Pulse, paired with aggregated Google Trends data, we find that Americans did not consume much information about the Court before the story hit. Interest in the Court and its members spiked dramatically as scandal coverage unfolded, and that interest remained high in the weeks that followed. Notably, the results are strong regardless of partisanship; while Democratic participants were more likely to seek out information about the Court in the aftermath of the scandal, we observe strong effects among Republicans too.

By examining people’s engagement with a judicial scandal, we offer two contributions to literatures on political media and the Supreme Court’s role in society. First, we highlight how digital trace data like YouGov Pulse can help scholars examine behavioral indicators of information seeking and attention. Second, our work contributes to a growing literature on how media coverage of extra-judicial events can affect people’s perceptions of a low-coverage, unelected institution such as the Supreme Court (Armaly and Lane, 2023; Krewson, 2019). We hope scholars will exploit our relatively new approach, as such data are valuable for understanding dynamics between media reporting and public attention toward institutions in a shifting media landscape.

The Supreme Court, the media, and judicial scandal

Because the logistics of policymaking in the United States are opaque to the majority of Americans, most people need help interpreting and processing the government’s actions. The media fills that explanatory role but, for many reasons, covers only a small slice of the Court’s processes and outputs (Collins and Cooper, 2012; Zilis, 2015). Thus, while Americans understand basics about the Court’s duties and structure (Gibson et al., 2003), they only get selective information about the Court’s outputs and members (Hitt and Searles, 2018), much of which focuses on the Court’s legalistic nature and authority and is concentrated on the Court’s most salient cases each term (Vining and Marcin, 2014).

Disruptions to stable media coverage, such as the breaking of a scandal, draw new and different attention toward political actors and the institutions in which they work (Walgrave et al., 2017). While not all scandals receive public attention, those that do elicit reassessment for a leader’s job suitability, with financial scandals raising doubts about professional judgment and sex scandals leading to questions about moral integrity (Doherty et al., 2011). Accusations of misuse of office amplify the damage and diminish support for implicated politicians and their institutions (Basinger, 2013; Bowler and Karp, 2004). The media landscape surrounding a scandal influences its impact, with saturation levels shaping public awareness and coverage nuances shaping reactions (Pereira and Waterbury, 2019). Put simply, the sudden media attention associated with a scandal affects popular attention toward an institution and perceptions of it, as long as that attention lasts.

At the Supreme Court, a well-covered scandal would offer the public a view of the Court that conflicts with well-established popular belief in its authority while also drawing outsized attention toward it. Research suggests public response to judicial scandals is complicated, as they affect perceptions of individual justices but do not permanently affect the Court’s legitimacy (Boston et al., 2023; Krewson and Schroedel, 2023). But media coverage of a scandal also draws popular attention toward issues and the people involved in them (Boydstun et al., 2014), which, in this case, could lead to people seeking more information about the Court, its members, and its work.

Understanding how people’s Supreme Court information-seeking habits change based on news coverage is important for many reasons, though we highlight two here. First, demand for more information about the Court can lead to more media oversight of its activities (Boydstun, 2013). Supreme Court justices are appointed, few laws govern their conduct, and media outlets struggle to follow the justices’ daily activities (Longoria, 2023), which means the judiciary essentially polices itself. More public demand for stories lead to more resources which can lead to more coverage. Second, more attention on the Court’s work could change the way the justices make decisions. Limited media attention means the justices often make decisions without fear of public dissent (Zilis, 2015), but their decision calculus changes when people are paying attention (Badas and Justus, 2022), and it would change permanently if people demanded more coverage of the justices’ work.

The ProPublica report

ProPublica’s story about Justice Clarence Thomas’s failure to disclose expensive trips funded by a billionaire conservative donor came out at 5 a.m. Eastern Time on April 6, 2023 (Kaplan et al., 2023). The media latched onto the story quickly, ¹ and it stayed in the headlines for several days as outlets continued to pick up the story and Democrats in Congress demanded investigations and stronger ethics laws (Barnes and Marimow, 2023).

As a disruption of the regular news cycle, the report’s publication is useful for understanding changes in information-seeking behavior regarding the Court. First, the report was published during a period of relative calm at the Court; the Court was in the middle of a two-week recess, in which it did not hold oral argument or release any new decisions or orders, when ProPublica released the story. Thus, the period before the ProPublica article likely reflects “normal” coverage of the Supreme Court. Second, the unanticipated report provides us with an opportunity to see how a newsworthy scandal affects interest in the Court. ² Analyzing public interest and attention before and after the report offers information about the extent and durability of public attention to the Court.

We expect Supreme Court scandals will increase public attention toward the Court. Since the Court tends to be insulated from regular media attention and a majority of the public has only some baseline knowledge about it, Court coverage can have a seismic impact on the public’s impression of the institution. We believe broad media coverage of Court scandals are on par with events like judicial nominations (Armaly and Lane, 2023), confirmation hearings (Farganis and Wedeking, 2014), and controversial or salient decisions (Bailard, 2016), all of which research shows break the monotony of standard (mostly non-existent) Court coverage. These events should thus also have the power to alter individual-level behavior in ways that exhibit an increased interest in and a disposition towards seeking information about the Court (Boydstun et al., 2014).

We further expect that Democrats are more likely to engage in information seeking about scandals related to a conservative justice or majority conservative Court. Put simply, when a misfortune befalls an ideological or political opponent, individuals seek more information on the event to confirm their beliefs about the opposition or to develop an even wider chasm between them and the opposition (Nyhan, 2014). While an ideological ally might seek information to dismiss a misfortune, studies have shown that ideologically congruent messaging does not persuade as much as ideologically incongruent messaging polarizes (Nicholson, 2012). Thus, we believe that Democrats will disproportionately seek information about and engage in a judicial scandal on a political opponent like Thomas.

YouGov Pulse: measuring public attention to the Court

We contracted with YouGov to purchase information on the online behavior of a demographically representative sample of 1,000 American adults living in the United States for the entire month of April 2023. The total number of observed URL visits for our sample is approximately 4.2 million, with a median of 1,877 URL visits per unique respondent. In addition to URL links, the data contains title and truncated description information for approximately 400,000 views of media content on YouTube (the most frequently viewed), Netfix, Disney+, Amazon Prime, and BBC iPlayer.

With the data in hand, we examined the URLs visited, searches made, and media viewed for evidence they directly related to the Supreme Court, its members, or the scandal. We did this using a basic dictionary of search terms. ³ We then hand checked each unique URL, search term, or media title and truncated description to ensure the content reflected the topics of interest. Our efforts resulted in 833 unique URLs, 76 unique searches, and 44 unique media videos fitting our search criteria. While there could be articles and videos that reference the Court but do not have descriptive URLs, content directly referencing the Court or its members in its title or description reliably indicates that people are seeking out Court-related information. ⁴

YouGov also provided us with demographic information about respondents, including their gender (male or female), education (6-point scale), partisanship (7-point scale, Democrat to Republican), race (White or non-White), income (16-point scale), and year of birth. We use these characteristics to understand the types of people most likely to search for information about the Court and to respond to the judicial scandal.

Regression discontinuity design analysis

We treat the publication of the ProPublica report as a natural experiment, in which the disruption of the regular news cycle acts as an exogenous shock. We estimate the causal effect of the report on Court interest using a regression discontinuity design (RDD). We measure time in 24-hour units, and calculate the proportion of people in our dataset seeking information about the Supreme Court on the internet over each 24-hour period leading up to and beyond the publication of the scandal. We then estimate local linear regressions on both sides of this cutpoint and estimate the local average treatment effect utilizing the Imbens and Kalyanaraman (2012) optimal bandwidth calculation.

The results are displayed in Figure 1. At 0 on the x-axis, we observe an estimate of approximately 6% on the y-axis. This means that from the time of publication (5 a.m. EDT) until 4:59:59 a.m. the next day, we estimate that 6% of our sample demonstrated interest in or exposure to information about the Supreme Court. In contrast, in the 24 hours directly before 5 a.m. on April 6, slightly less than 1% showed similar interest. The estimated effect of the ProPublica publication is 4.7 percentage points and is statistically significant. ⁵ The impact of the publication on information seeking lingered through the second day, with nearly 5% of our sample showing interest. From the third day until the end of the month, attention remained elevated, hovering at approximately 3% of our sample, compared to a daily total interest between 0% and 1% before the scandal broke.

Figure 1.

Result of a regression discontinuity analysis showing the causal effect of the release of the ProPublica story on Justice Clarence Thomas on the daily percent of respondent seeking information about the Court on the internet.

Importantly, the consistently high percentage of people viewing the Court per day and the number of Court-related views per day, following the breaking of the news story, is not driven by a small number of individuals. On the left-side part of Figure 2, we plot the cumulative proportion of all respondents demonstrating interest in the Court over the month, which shows effects similar to those presented in the regression discontinuity analysis. The plot demonstrates that the strong causal estimates are driven by the interest of new persons showing interest in the Court, not the same people returning to websites about the Court. The right half of the figure contains a plot of the change in cumulative proportion each day. In the 24 hours following April 6, at 5 a.m., the percentage of persons with newly demonstrated interest in the Court increased dramatically by 4 percentage points. In the next 24 hours, another 2% of people demonstrated new interest. Outside of those 48 hours, the amount of new users per day was only an average of 0.3% points.

Figure 2.

Cumulative percentage of sample demonstrating attention to the court over each 24-hour period in April 2023 (left side) and the change in this cumulative proportion from each 24-hour period to the next (right side).

Information seeking or information avoidance?

Having established that the ProPublica report caused an increase in information seeking about the Supreme Court, we now test our expectation that the increase occurred disproportionately among Democrats. To test this theory, we estimate whether individuals sought out information about the Court using logistic regression models. For each model, the independent variables include two dummy variables—one indicating males and another indicate non-White individuals—and continuous variables for birth year, education level, party identification (strong Democrat to strong Republican), income level, and the total number of internet engagements during the month of April. For space considerations, we relegate the table of results to the appendix.

Notably, we find no partisan differences in innate interest. That is, analyzing information seeking about the Court before the ProPublica report only, partisanship does not predict information seeking. We next consider whether partisanship predicted change in information seeking following the report, especially among Democrats. (The dependent variable is 1 if they sought information after the report but not before it; otherwise it is 0). The data suggest that Democrats became most likely to seek out information about the Court whether we employ all of the data or smaller time frames surrounding the report.

The results demonstrate that Republicans did not avoid information altogether; they simply did not respond as strongly to the ProPublica report as Democrats. For example, holding all other variables at their modal values, the results predict that a strong Republican became 1.9 percentage points more likely to seek information about the Court as a result of the ProPublica report; a strong Democrat became 6.3 percentage points more likely.

This analysis portends some challenges for a society in which broad consensus is required to enact meaningful change because partisans became disproportionately interested in damaging information when it aligned with their policy preferences to do so. Beyond partisanship, people were most likely to become information seekers when they were more active on the internet generally and when they were older, though we did not have any expectations regarding these or other non-partisan independent variables.

Robustness check: Google Trends

Finally, additional RDD analysis shows that our findings generalize beyond our sample. We obtained data on all Google searches using Google Trends and Keywords Everywhere to estimate the historical monthly search volumes for terms similar to those we used in our YouGov data analysis throughout April 2023. ⁶ After collecting the estimated search volumes for each of these terms, we performed a regression discontinuity analysis.

Based on the analysis (see appendix), we find minimal searches related to the Supreme Court during the six days leading up to the publication (less than 100 thousand per day). In the 24 hours following the publication, that jumped to an estimated two million searches. The interest in the Court remained elevated for weeks, slowly having made its way back to the pre-report level of interest approximately two weeks later.

Conclusion

Understanding how the public engages with a sudden expansion of media coverage, like that of a political scandal, is crucial for understanding both public and elite reaction to issues that received little previous attention. This is particularly true when a scandalous story involves an unelected institution that only rarely receives coverage in the first place, such as the United States Supreme Court. In this paper, we utilize a unique data source—a representative sample of individual-level digital trace data from YouGov Pulse and more data from Google Trends—to gain insights into individual-level exposure to Court-related information, and we find that scandals draw attention toward unelected institutions. We also show that such data are valuable, not only to understand judicial scandal, but to unveil important dynamics between how the media report news, how the public consumes it, and what the public might ultimately do with it.

Supplemental Material

Supplemental Material - Did you hear about Clarence Thomas? Measuring public attention toward the supreme court

Supplemental Material for Did you hear about Clarence Thomas? Measuring public attention toward the supreme court by Christopher N Krewson, Jessica A Schoenherr and Marcy Shieh in Journal of Research & Politics.

Supplemental Material

Supplemental Material - Did you hear about Clarence Thomas? Measuring public attention toward the supreme court

Supplemental Material for Did you hear about Clarence Thomas? Measuring public attention toward the supreme court by Christopher N Krewson, Jessica A Schoenherr and Marcy Shieh in Journal of Research & Politics.