Author Archives: Katherine Pearson

Shea Streeter examines the circumstances surrounding police violence and protest

Shea Streeter

Shea Streeter

Post developed by Katherine Pearson

Shea Streeter began her graduate work in political science as a comparativist interested in state repression around the world. When the protest movement in Ferguson, Missouri exploded after the killing of Michael Brown, Streeter turned her attention to police violence and protest in the United States. As a President’s Postdoctoral Fellow at the University of Michigan, she’s examining how race and gender shape the ways that people experience, perceive, and respond to incidents of violence.

“Racial animus is in the air we breathe,” Streeter says, “but when we look at police violence, we can get distracted by race and ignore other important factors.” Her dissertation included an experiment to examine how the race of victims of police violence determines whether the public sees the violence as just. Surprisingly, she finds that the race of the victims is less salient than expected. Instead, the social context strongly shaped the attitudes of the respondents. Those who were predisposed to consider societal and institutional forces were less likely to believe the victim deserved the outcome, compared to respondents who place sole responsibility on the individual. 

Racial differences in rates of protest 

Half of the people killed by police each year are white, and yet the rate of protest over white victims of police violence is very low. A dataset that Streeter is currently completing includes all publicly available information on police killings and any protests that happened in 2015-2016. For those two years, about a third of the police killings of African Americans led to some sort of protest, but when whites were killed by police, protests occurred only five percent of the time. “I argue that it’s the biggest racial gap related to policing,” Streeter says. “There are a lot of reasons we could point to why African Americans would be protesting. But why wouldn’t whites also be protesting when their community members are killed?” 

When conducting field research in several different cities in the United States, Streeter asked community organizers about protests for white victims of police violence. The organizers told her that they reach out to the families of white victims, but those families often do not want to be involved with protests. Instead, many white family members express understanding and forgiveness toward the police. Streeter makes sense of these reactions by tying them to the psychological concept of a belief in a just world. The idea is that people get what they deserve and they deserve what they get. Streeter observes that even when people who hold this belief lose a member of their own family, their trust in the police remains unchanged. “If you have these beliefs, it can be like a double loss,” Streeter notes, which may explain why there are fewer protests for white victims of police violence. 

The role of mentorship

Mentorship has played a large role in Streeter’s academic career. Christian Davenport became a mentor to her when she was a senior at Notre Dame. At that time, Streeter was thinking about her career but hadn’t considered pursuing research. While working as a research assistant for Davenport, he encouraged her to pursue graduate work in political science. Streeter cites this support as a key reason she decided to come to the University of Michigan. She also gives credit to David Laitin and Jeremy Weinstein at Stanford, who pushed her to study the United States when she was training as a comparativist. “I had confusion about what my identity as a scholar would be if I changed paths, but they put my fears to rest, so I give them a lot of credit for helping me pursue this research path,” Streeter says. 

Looking forward 

In addition to her ongoing research on police violence, Streeter is turning her attention to the ways interpersonal violence affects the way that people think and act politically. She sees connections between different types of violence, including mass shootings, domestic violence, and suicide. “We don’t often see these as political violence, but they affect how people operate in the world,” Streeter says. She’s especially interested in the ways violence affects people differently based on gender. Streeter’s work is innovative and varied, but united by a common theme, which she sums up as “How does violence affect our world, and what are the aggregate consequences of that? That’s the big picture.” 

Political Communication Meets Big Data

Post developed by Mike Traugott and Katherine Pearson. 

How do voters make sense of the information they hear about candidates in the news and through social media? This question was at the heart of a collaboration between researchers at the University of Michigan, Georgetown University, and Gallup to study political communication that took place during the 2016 U.S. presidential election. 

Mike Traugott, Ceren Budak, Lisa O. Singh, and Johnathan Ladd presented findings from the study at the Michigan Institute for Data Science (MIDAS) Seminar on November 14, 2019. The panel discussion, moderated by Rayid Ghani, covered results that will be published in a new book, Words That Matter, in May 2020. 

Rayid Ghani, Jonathan Lass, Lisa O. Singh, Ceren Budak, and Mike Traugott at the MIDAS symposium.

Rayid Ghani, Jonathan Lass, Lisa O. Singh, Ceren Budak, and Mike Traugott at the MIDAS symposium. 

Genesis of the project

The project began when Gallup contacted Mike Traugott, a scholar of political communication who works in the area of attention to media using survey methods. In the four months leading up to the 2016 presidential election, Gallup conducted 500 interviews per day, asking respondents whether they had heard, read, or seen anything in the last few days about each of the two major-party candidates. In addition, the research team analyzed a sample of tweets from the public and from journalists. Finally, they compiled a database of news articles about the election, and also conducted an analysis of fake news. 

Data visualizations were an important part of this work, said Traugott. As data was gathered and interpreted, the researchers created visualizations and analyses that were published on the Gallup website and in The Washington Post and other news outlets. Excellent graphics were essential to show complex data in an easily interpretable way. 

Interdisciplinary strengths

Researching political communication using big data and data from multiple sources was an exciting challenge for the members of the team. When survey respondents are asked to recall what they’ve heard, read, or seen, there is the potential for error stemming from everything from memory problems to social desirability bias. Working with an interdisciplinary team was an opportunity to use new methods to analyze big data and mitigate such errors. 

Closed-ended survey questions can be difficult to interpret; researchers sometimes try to find out what people actually mean by asking open-ended follow-ups. The surveys in this study only collected open-ended responses, allowing respondents to give more meaningful answers. With such a large sample of open-ended responses about what people remembered about the candidates, it was essential to find innovative ways to analyze the data.  

Lisa Singh and Ceren Budak, both computer scientists, contributed expertise in computational social science and experience working with social media data. A variety of techniques were used in the analyses contained in the book: frequent word analysis, topic analysis, network analysis, sentiment analysis, and more. The open-ended text from the survey responses was so noisy and short that the algorithms were not enough to interpret the results. It took a team effort to interpret the data through a semi-automated process. The team at Gallup and the political scientists sorted words into topics and created synonym dictionaries to clean the data and remove inconsistencies. Developing these tools to be applied in domains where the text is not as rich and complete will be a focus of future work. 

A long-lived narrative is worth more than many explosive stories

Ladd noted that by analyzing text data – tweets and open-ended survey responses – the research team found that people repeatedly remembered Hillary Clinton’s emails throughout the campaign. The fact that this one story dominated the narrative about Clinton seemed to have an effect on voters, and Ladd points out that Clinton echoed this finding in her book, What Happened, employing one of the project’s graphics in the text. On the other hand, people remembered many different news stories about Donald Trump over time. These stories appeared and disappeared quickly, and no one story made a big impression on respondents. 

This figure highlights the changing topics that Americans remember about Clinton since July. The x-axis shows the date and the y-axis the fraction of responses that fall into a particular topic.

This figure highlights the changing topics that Americans remember about Clinton since July. The x-axis shows the date and the y-axis the fraction of responses that fall into a particular topic.

Another major finding of the study is that there were differences between the news that survey respondents recalled hearing and the text analysis of media articles, and both of those were different from what journalists were tweeting about. By analyzing streams of data from multiple sources, the researchers were able to conclude that journalists’ tweets and the text of newspaper articles did not favor either candidate. 

Singh noted that Trump was masterful in keeping the issue of Clinton’s emails central to the campaign narrative. When the researchers analyzed new articles and tweets from journalists, email was not a dominant topic, as it was in the survey responses. She said that it was the Trump campaign that kept the narrative about the emails in the public’s awareness. 

Connecting media coverage and voting behavior

Members of the research team who were not available to participate in the panel discussion contributed further analyses to the book. Stuart Soroka conducted a sentiment analysis of the open-ended responses, and Josh Pasek did work on story life and length of time an item was in the news. One limitation of this study was that Gallup did not collect any direct measure of voter preference, although they did collect favorability ratings of the candidates every day, which gave the researchers an indirect measure to work with. There was a lagged relationship between the net sentiment of Trump and Clinton in the news and the relative favorability of the two candidates. 

We can’t know how fake news influenced votes, said Budak, who analyzed social media data in the 2016 election cycle. In a chapter on fake news in Words That Matter, she examined Clinton’s net favorability and found a strong relationship between fake news and her favorability rating. Specifically, Budak found that Clinton’s favorability would move first, and fake news responded to that. The creators of fake news were attuned to what was happening in the campaign and responded accordingly. 

When Budak analyzed retained information data according to political leanings, she found that Republicans retained fake news coverage about Clinton, but not for Trump. The conversation about Trump changed a lot over time, while the narrative about Clinton stayed focused on her emails. According to Budak, “we can’t say fake news caused the outcome of the election, but it shaped the agenda.”

Racial Dynamics in the American Context
: A Second Century of Civil Rights and Protest?

Post developed by Katherine Pearson and Dianne Pinderhughes 

Drawing from published work that will be compiled as a new book, Black Politics After the Civil Rights Revolution, Dianne Pinderhughes explored the arc of 20th-century civil rights reform and the growing political incorporation of African Americans into electoral politics when she delivered the 2019 Hanes Walton, Jr. lecture. A recording of the lecture is available below. 

Understanding the history of collective action is essential to tracing the development of 20th-century racial politics in the United States. Pinderhughes began by describing racial injustice in the U.S. starting with the Plessy v. Ferguson decision in 1896, which some consider the nadir of race relations in the U.S. Following this era, Pinderhughes described a period of innovation and institution building beginning in the early and mid 20th century, which saw the development of legal defense funds and an increase of racial diversity in academia.

Social and political scientists recognize the gradual increase in African American political participation and the increasing numbers of elected officials of color. As the political dynamics of the eras changed, Pinderhughes described how African Americans have pushed to enter, to change, and to reframe their status.

Pinderhughes posits that the election of Donald Trump in 2016 posed a direct challenge to that framing of the evolution of successful racial reform. In doing so, she asks whether the U.S. is entering a new nadir. “My own work around these issues of democracy, political participation and efforts to integrate on a stable basis, and to begin to address the economic and political dimensions of citizenship, was challenged by how they might be framed,” Pinderhughes said. “But most of that work began from and was conceptualized within a relatively stable set of policy values and expectations, and that racial and ethnic exclusion was no longer possible, or acceptable.”

In the end, Pinderhughes concludes that the state of politics in the 21st century is far more hopeful than the nadir of the 19th and early 20th centuries. Institutional reforms have substantially recreated the American electoral and political process. Race is central to American life, and it will continue to be a dynamic force in electoral politics.

The Hanes Walton, Jr. lecture series was launched in 2015, in honor of Hanes Walton, Jr. One of the most influential and productive political scientists to emerge from the civil rights era, Walton published numerous journal articles, several book chapters, and authored more than twenty books. Walton is remembered for his in-depth subject knowledge, sense of humor, and ability to connect with his students. He was a caring and supportive mentor to his countless graduate and undergraduate students, many of whom have gone on to distinguished careers in academia and industry. 

Computer simulations reveal partisan gerrymandering 

Post developed by Katherine Pearson 

How much does partisanship explain how legislative districts are drawn? Legislators commonly agree on neutral criteria for drawing district lines, but the extent to which partisan considerations overshadow these neutral criteria is often the subject of intense controversy.

Jowei Chen developed a new way to analyze legislative districts and determine whether they have been unfairly gerrymandered for partisan reasons. Chen, an Associate Professor of Political Science and a Research Associate at the Center for Political Studies, used computer simulations to produce thousands of non-partisan districting plans that follow traditional districting criteria. 

Simulated NC map

These simulated district maps formed the basis of Chen’s recent expert court testimony in Common Cause v. Lewis, a case in which plaintiffs argued that North Carolina state legislative district maps drawn in 2017 were unconstitutionally gerrymandered. By comparing the non-partisan simulated maps to the existing districts, Chen was able to show that the 2017 districts “cannot be explained by North Carolina’s political geography.” 

The simulated maps ignored all partisan and racial considerations. North Carolina’s General Assembly adopted several traditional districting criteria for drawing districts, and Chen’s simulations followed only these neutral criteria, including: equalizing population, maximizing geographic compactness, and preserving political subdivisions such as county, municipal, and precinct boundaries. By holding constant all of these traditional redistricting criteria, Chen determined that the 2017 district maps could not be explained by factors other than the intentional pursuit of partisan advantage. 

Specifically, when compared to the simulated maps, Chen found that the 2017 districts split far more precincts and municipalities than was reasonably necessary, and were significantly less geographically compact than the simulations. 

By disregarding these traditional standards, the 2017 House Plan was able to create 78 Republican-leaning districts out of 120 total; the Senate Plan created 32 Republican-leaning districts out of 50. 

Using data from 10 recent elections in North Carolina, Chen compared the partisan leanings of the simulated districts to the actual ones. Every one of the simulated maps based on traditional criteria created fewer Republican-leaning districts. In fact, the 2017 House and Senate plans were extreme statistical outliers, demonstrating that partisanship predominated over the traditional criteria in those plans. 

The judges agreed with Chen’s analysis that the 2017 maps displayed Republican bias, compared to the maps he generated by computer that left out partisan and racial considerations. On September 3, 2019, the state court struck down the maps as unconstitutional and enjoined their use in future elections. 

The North Carolina General Assembly rushed to adopt new district maps by the court’s deadline of September 19, 2019. To simplify the process, legislators agreed to use Chen’s computer-simulated maps as a starting point for the new districts. The legislature even selected randomly from among Chen’s simulated maps in an effort to avoid possible accusations of political bias in its new redistricting process.

Determining whether legislative maps are fair will be an ongoing process involving courts and voters across different states. But in recent years, the simulation techniques developed by Chen have been repeatedly cited and relied upon by state and federal courts in Pennsylvania, Michigan, and elsewhere as a more scientific method for measuring how much districting maps are gerrymandered for partisan gain. 

Rousing the Sleeping Giant? Emotions and Latino Mobilization in an Anti-Immigration Era

Post developed by Nicholas Valentino, Ali Valenzuela, Omar Wasow, and Katherine Pearson 

ICYMI (In Case You Missed It), the following work was presented at the 2019 Annual Meeting of the American Political Science Association (APSA).  The presentation, titled “Rousing the Sleeping Giant? Emotions and Latino Mobilization in an Anti-Immigration Era” was a part of the session “The Rhetoric of Race” on Friday, August 30, 2019.

Since the 2016 presidential campaign anti-immigration policies have been very popular among President Trump’s strongest supporters, though they do not present obvious benefits to the economy or national security. Strategists suppose that the intent of the anti-immigration rhetoric and policies is to energize the president’s base. 

But what about people who identify with the targets of these policies, specifically Latinos? Are they mobilized against anti-immigration proposals, or are they further deterred from political participation? 

New research by Nicholas A. Valentino, Ali Valenzuela, and Omar Wasow finds that anger was associated with higher voter turnout among Latinos, but the Latinos who expressed more fear had lower voting rates.

voting rates by race and emotion

The role of emotions in politics is complex. The research team begins with the observation that negative emotions do not always have negative consequences for politics. Indeed, negative emotions may promote attention and interest, and drive people to vote. They draw a distinction between different negative emotions: while anger may spur political action, fear can suppress it. 

The research team fielded a nationally-representative panel survey of white and Latino registered voters before and after the 2018 midterm elections. Respondents were asked about their experience with Immigration and Customs Enforcement (ICE) officials and their exposure to campaign ads focused on immigration. Participants were also asked to rate their emotional reactions to the current direction of the country. 

The results showed that Latinos interacted with ICE more frequently than whites did, but both groups had the same level of exposure to campaign ads. Latinos reported more anger than whites, and also more fear. In fact, among the negative emotions in the survey, fear among Latinos was highest.  

In the sample the validated voting rate among Latinos was 39%; among whites in the sample it was 72%, demonstrating the under-mobilization of Latino voters. Whether Latinos vote in greater numbers in 2020 may depend on whether they are mobilized by anger against anti-immigration rhetoric, or whether they are deterred by fear stemming from policies like ICE detention and deportation. 

Accuracy in Reporting on Public Policy

Post developed by Katherine Pearson and Stuart Soroka

ICYMI (In Case You Missed It), the following work was presented at the 2019 Annual Meeting of the American Political Science Association (APSA).  The presentation, titled “Media (In)accuracy on Public Policy, 1980-2018” was a part of the session “Truth and/or Consequences” on Sunday, September 1, 2019.

Citizens can be well-informed about public policy only if the media accurately present information on the issues. Today’s media environment is faced with valid concerns about misinformation and biased reporting, but inaccurate reporting is nothing new. In their latest paper, Stuart Soroka and Christopher Wlezien analyze historical data on media coverage of defense spending to measure the accuracy of the reporting when compared to actual spending. 

In order to measure reporting on defense spending, Soroka and Wlezien compiled text of media reports between 1980 and 2018 from three corpuses: newspapers, television transcripts, and public affairs-focused Facebook posts. Using the Lexis-Nexis Web Services Kit, they developed a database of sentences focused on defense spending from the 17 newspapers with the highest circulation in the United States. Similar data were compiled with transcripts from the three major television broadcasters (ABC, CBS, NBC) and cable news networks (CNN, MSNBC, and Fox). Although more difficult to gather, data from the 500 top public affairs-oriented public pages on Facebook were compiled from the years 2010 through 2017. 

Soroka and Wlezien estimated the policy signal conveyed by the media sources by measuring the extent to which the text suggests that defense spending has increased, decreased, or stayed the same. Comparing this directly to actual defense spending over the same time period reveals the accuracy of year-to-year changes in the media coverage. For example, if media coverage were perfectly accurate, the signal would be exactly the same as actual changes in spending. 

As the figure below shows, the signal is not perfect. While there are some years when the media coverage tracks very closely to actual spending, there are other years when there is a large gap between the signal that news reports send and the defense budget. The gap may not entirely represent misinformation, however. In some of these cases, the media may be reporting on anticipated future changes in spending. 

media signal

For most years, the gap representing misinformation is fairly small. Soroka and Wlezien note that this “serves as a warning against taking too seriously arguments focused entirely on the failure of mass media.” This analysis shows evidence that media coverage can inform citizens about policy change. 

The authors conclude that there are both optimistic and pessimistic interpretations of the results of this study. On one hand, for all of the contemporary concerns about fake news, it is still possible to get an accurate sense of changes in defense spending from the media, which is good news for democratic citizenship. However, they observed a wide variation in accuracy among individual news outlets, which is a cause for concern. Since long before the rise of social media, citizens have been at risk of consuming misinformation based on the sources they select. 

Toward a Typology of Populists

Post developed by Pauline Jones, Anil Menon, and Katherine Pearson 

ICYMI (In Case You Missed It), the following work was presented at the 2019 Annual Meeting of the American Political Science Association (APSA).  The presentation, titled “Putin’s Pivot to Populism” was a part of the session “Russia and Populism” on Sunday, September 1, 2019. 

The rise in populism around the world has received much attention, but not all populists are the same. In a new paper, Pauline Jones and Anil Menon present an original typology of populists that goes beyond typical left-wing versus right-wing classifications. 

To better understand the different types of populists and how they operate, Jones and Menon examine two key dimensions: position within the political landscape (outsider versus insider), and level of ideological commitment (true believer versus opportunist). 

Populists tend to frame their criticism of political elites differently depending on whether they are political outsiders or government insiders. While outsiders are free to criticize those in power broadly, populists who hold political power are more likely to tailor their criticisms to their political opponents. Insiders are also more careful not to attack members of the elite with whom they will need to build political coalitions. 

Many populists evoke the past, but outsiders and insiders tend to do so differently. Whereas outsiders focus on the near past as a critique of a corrupt elite, political insiders instead focus on the distant past to evoke better days of shared national values. 

Jones and Menon also draw distinctions between true believers in populism and those who embrace populism for purely strategic reasons. True believers will remain strongly committed to enacting their populist agenda once in office; opportunists will use populist rhetoric to gain power, but won’t support their platform strongly if elected.  

The intersection of these two dimensions leads to the classification of populists into four types, illustrated in the table below: Oppositional, Classical, Strategic, and Pivot. 

Classification of populists

The most common variety of populist is the oppositional populist, who are outsiders and true believers. Oppositional populists put their agenda before all else and distance themselves from the mainstream elite. 

Classical populists sometimes start out as outsiders who become insiders once they are elected to office. Like oppositional populists, they are strongly committed to enacting their agenda; unlike oppositional populists, classical populists can enact their agenda from a position of power. Because they are insiders, classical populists are more selective about criticizing elites. 

Pivot populists are a rare group of political insiders who adopt populist rhetoric with little or no commitment to the populist ideology. Jones and Menon point to Russia’s Vladimir Putin as an example of a pivot populist who has adopted populism to bolster support for his regime while deflecting blame for the country’s problems. 

The final category is strategic populists. Like Donald Trump in the United States, strategic populists are outsiders with a weak commitment to the populist agenda. Strategic populists are broadly anti-elite, and also use their rhetoric to create divisions among the people. Once in power, they are unlikely to alienate elites by pursuing populist policy goals. 

Incidental Exposure to Political News Increases Political Knowledge

Post developed by Brian Weeks and Katherine Pearson 

ICYMI (In Case You Missed It), the following work was presented at the 2019 Annual Meeting of the American Political Science Association (APSA). The presentation, titled “Can Incidental Exposure to News Close the Political Knowledge Gap?” was a part of the session “News in the Digital Age” on Friday, August 30, 2019.

We’re immersed in a media landscape full of choices. News, information, and entertainment are all at our fingertips. But does this mean that people are better informed about important issues? Is it is possible for people who aren’t interested in seeking out political news to learn about candidates and issues through the information they’re exposed to casually? Brian Weeks, Daniel S. Lane, Lauren B. Potts, and Nojin Kwak conducted two surveys to answer this question. 

Motivation and opportunity play a big role in the amount of news we’re exposed to. People who are deeply interested in politics are motivated to seek out information, and as a result, they are better informed about candidates and policies. 

The nature of the media environment makes it hard to avoid news and political information; many people consume news without trying. As we have more access to all types of media, we are incidentally exposed to political information. Does increased accidental exposure make up for a lack of motivation to seek out news, or does all of that information rush past us without making us more knowledgeable? 

To test whether this incidental exposure to news translates into an increase in political knowledge, Weeks and his co-authors conducted a series of surveys. They collected panel survey data two waves during the 2012 presidential election and conducted another two waves of surveys during the 2016 presidential election. The surveys asked participants about whether they were exposed to political information they didn’t seek out, their level of political interest, and measured their knowledge of candidates’ policy positions. 

The surveys showed strong evidence that people who had incidental exposure to news about presidential candidates knew more about the candidates’ policy positions. 

Incidental exposure to media

The biggest benefit of incidental exposure was seen in the group of people who rated themselves least politically interested, which suggests that greater exposure can make up for a lack of motivation to seek out news. 

Knowledge of candidates and their policy positions is still essential for well-informed citizens, and the growth of opportunities to be exposed to news from many sources may reduce gaps in knowledge. 

Using Text and Images to Examine 2016 Election Tweets

Post developed by Dory Knight-Ingram 

ICYMI (In Case You Missed It), the following work was presented at the 2019 Annual Meeting of the American Political Science Association (APSA).  The presentation, titled “Using Neural Networks to Classify Based on Combined  Text and Image Content: An Application to Election Incident Observation” was a part of the session “Deep Learning in Political Science” on Friday, August 30, 2019.

A new election forensics process developed by Walter Mebane and Alejandro Pineda uses machine-learning to examine not just text, but images, too, for Twitter posts that are considered reports of “incidents” from the 2016 US Presidential Election. 

Mebane and Pineda show how to combine text and images into a single supervised learner for prediction in US politics using a multi-layer perceptron. The paper notes that in election forensics, polls are useful, but social media data may offer more extensive and granular coverage. 

The research team gathered individual observation data from Twitter in the months leading up to the 2016 US Presidential Election. Between Oct. 1-Nov. 8, 2016, the team used Twitter APIs to collect millions of tweets, arriving at more than 315,180 tweets that apparently reported one or more election “incidents” – an individual’s report of their personal experience with some aspect of the election process. 

At first, the research team used only text associated with tweets. But the researchers note that sometimes, images in a tweet are informative, while the text is not. It’s possible for the text alone to not make a tweet a report of an election incident, while the image may indeed show an incident. 

To solve this problem, the research team implemented some “deep neural network classifier methods that use both text and images associated with tweets. The network is constructed such that its text-focused parts learn from the image inputs, and its image-focused parts learn from the text inputs. Using such a dual-mode classifier ought to improve performance. In principle our architecture should improve performance classifying tweets that do not include images as well as tweets that do,” they wrote.

“Automating analysis for digital content proves difficult because the form of data takes so many different shapes. This paper offers a solution: a method for the automated classification of multi-modal content.” The research team’s model “takes image and text as input and outputs a single classification decision for each tweet – two inputs, one output.” 

The paper describes in detail how the research team processed and analyzed tweet-images, which included loading image files in batches, restricting image types to .jpeg or .png., and using small image sizes for better data processing results. 

The results were mixed.

The researchers trained two models using a sample of 1,278 tweets. One model combined text and images, the other focused only on text. In the text-only model, accuracy steadily increases until it achieves top accuracy at 99%. “Such high performance is testimony to the power of transfer learning,” the authors wrote. 

However, the team was surprised that including the images substantially worsened performance. “Our proof-of-concept combined classifier works. But the model structure and hyperparameter details need to be adjusted to enhance performance. And it’s time to mobilize hardware superior to what we’ve used for this paper. New issues will arise as we do that.” 

The Politicization of Policies to Address Climate Change

Post developed by Erin Cikanek, Nicholas Valentino, and Katherine Pearson 

ICYMI (In Case You Missed It), the following work was presented at the 2019 Annual Meeting of the American Political Science Association (APSA).  The presentation, titled “The Politicization of Policies to Address Climate Change” was a part of the session “The Dynamics of Climate Policy Support in the US” on Friday, August 30, 2019.

Climate change is a truly polarizing issue. Partisans on either side of the issue have such deeply entrenched beliefs that there is little that can change minds. But this wasn’t always the case. For example, in 1988 Democrats and Republicans were in close agreement about the amount of money the government should spend on environmental protection. More recently, partisans have become more polarized in their level of concern about climate change. 

How do scientific policy issues become so polarized, and how quickly does this happen? New research by Nicholas Valentino and Erin Cikanek measures public awareness of Carbon Dioxide Removal (CDR) polices, and explores whether attitudes toward these policies are as politicized as climate change overall. 

Valentino and Cikanek conducted two studies to examine political polarization of CDR. First, they surveyed a large, nationally representative sample of people to measure how much they knew about climate change. The questions covered a broad set of issues and strategies for dealing with the problem. This survey revealed that the public has a high level of knowledge about climate change. 

The study also demonstrates that partisanship is highly predictive of knowledge about climate change. Democrats responded with significantly more accuracy than Republicans did. When respondents were asked specifically about technologies to remove carbon dioxide from the atmosphere, overall knowledge was lower, but Democrats and Republicans answered questions with the same level of accuracy, as shown in the figure below.  Valentino and Cikanek note that “this pattern is consistent with the possibility that elite rhetoric has come to suppress accuracy on general climate change knowledge among Republicans, but this has not yet occurred for knowledge in this newer domain (CDR).” 

climate change

The second study experimentally tested whether CDR policies are sensitive to partisan cues. CDR policies have not been debated as much or as publicly as climate change in general. Are these policies as susceptible to political polarization?  

Survey respondents were randomly assigned to one of three groups. A control group was asked about current climate change policies, as well as carbon reduction policies. The first treatment group received information that applied partisan stereotypes to the CDR policies: Republican hesitation about CDR because it might hinder business, and Democratic encouragement to save the environment. The second treatment group received information that ran counter to those stereotypes: Republican support for a pro-business solution to climate change, and Democratic concern that the solution may encourage businesses to pollute. 

The partisan cues had very little effect on the response to CDR policies. Interestingly, the counter-intuitive partisan cues backfired: when Republican respondents read the treatment showing Republican support for CDR, they opposed it slightly more. The very weak effect of partisan cues on support for CDR may show that CDR policies may be more resistant to polarization. 

The more politicians discuss scientific policy issues, the more polarized the discussion tends to become. However, Valentino and Cikanek see reason to hope that compromises remain possible for issues like carbon removal, which have not yet been subjected to partisan rhetoric.