Posts

In the UCLA Division of Social Sciences, we are dedicated to advancing research with real-world impact. As the #1 public university located in one of the most diverse cities in the world, we are ideally positioned to address critical issues facing our communities. Through the work of our world-class faculty – and our students who will become the leaders of tomorrow – we strive to be a leading agent for change across the nation and around the world.

Here in the Social Sciences at UCLA, we are very interested in the impact of technology on society. There are countless applications of big data that help us solve many of the problems that define life today in American society. LA Social Science is pleased to share this video highlighting two researchers, Dr. Till von Wachter and Dr. Safiya Noble, and the important, big data research they are leading in the social sciences.

As a public institution, our work is ultimately in service of you, our community. By engaging LA, we are changing the world.

UCLA Big Data and Politics Seminar Series

The Prevalence and Sharing Patterns of “Fake News” in the US in 2016 and 2020

David Lazer
University Distinguished Professor of Political Science and Computer Sciences

Co-Director of NULab for Texts, Maps, and Networks

Northeastern University

Friday, May 14, 12:00 PM – 1:30 PM PT

Zoom link: https://ucla.zoom.us/j/97899586814

Abstract: This presentation discusses the prevalence and sharing patterns of “fake news” in the United States in 2016 (regarding the election) and 2020 (regarding COVID-19). Substantively, the questions asked are: How common is fake news, as a specific genre of misinformation, been on Twitter? How concentrated are exposure and sharing patterns? And how does fake news fit into the broader information ecosystem on Twitter? Methodologically, the focus will in part be on the development of panels of accounts that are linked to administrative data as a method to measure aggregate behaviors on social media.

UCLA Latino Policy and Politics Initiative (LPPI) and the Alianza for Youth Justice have released a new “Call to Action” report titled “The Latinx DATA GAP in the Youth Justice System.” This report shows that inconsistent data collection methods complicate race and ethnicity tracking across different stages in the youth justice system.

“As our country undertakes a long overdue reckoning on race and justice, it is critical that Latinos be included in the conversation,” said Sonja Diaz, Founding Director of the UCLA Latino Policy & Politics Initiative and one of the report’s co-authors. “Far too often we are overlooked, but to effectively address inequities in the justice system, especially the egregious disparities facing Black Americans, policymakers and advocates need accurate data on Latinx youth. This new report details the challenges of collecting data on system-impacted youth and offers a way forward so that leaders can craft viable solutions based on facts as we reopen our economy and transform failed systems.”

Findings:

1. Today, Latinx youth represent 25%, or about 8.3 million, of the total U.S. youth population between ages 10-17.

2. State agencies involved in the criminal justice system collect data that identifies Latinx youth inconsistently if at all, creating an incomplete picture of Latinx ethnic data.

3. States report racial/ethnic data inconsistently across the 3 points of contact: arrests, detention, and probation.

  • 42% of states did not report racial or ethnic data for arrests
  • 30% of states did not report racial and ethnic data for detention
  • 52% of states did not have racial or ethnic data for probation

To read the full report, click HERE.

To learn more about UCLA Latino Policy and Politics Initiative, visit: https://latino.ucla.edu/

LRW Group, a local marketing services firm and UCLA Division of Social Sciences community partner, was a key contributor to the Los Angeles County COVID-19 study that sought to find the true spread of the virus in the county through serological testing. Led by UCLA alum, chairman and CEO Dave Sackman, LRW Group partnered with USC and the Los Angeles County Department of Public Health by identifying and recruiting a representative sample of LA residents to participate in the study. The findings released last week suggest infections from the new coronavirus are far more widespread, and the fatality rate is much lower here in LA County than previously thought.

LRW Group formed a partnership with the UCLA Division of Social Sciences in 2019. The partnership created a dedicated UCLA staff position tasked with recruiting undergraduates and graduate students for paid summer internships in data science. The initiative will support a proposed major in Data and Society in the division and feature support workshops and roundtables in social data science, as well as guest lectures by LRW leadership and employees.

According to Dean Darnell Hunt:

“LRW’s commitment will seed our efforts to establish a pipeline of underrepresented students into the new Data and Society major we are developing as part of the division’s larger big data initiative. Along with this generous and important contribution, LRW also has pledged to offer paid internships to some of our most promising students, creating an opportunity for them to envision future careers and gain workplace experience that enhances their academic pursuits at UCLA.”

 

Subscribe to LA Social Science and be the first to learn more insight and knowledge from UCLA social science experts in upcoming video/audio sessions and posts about current issues.

In the Department of Communication at UCLA, the Co-Mind Lab has rerouted some of its “big data” projects to create an informational resource about the coronavirus crisis. This resource offers members of the UCLA community and wider public a bird’s-eye view of news media coverage of the crisis, including a Los Angeles “dashboard” that maps positive cases in the county.

https://co-mind.org/cogmedia/browse/covid.php

Dr. Rick Dale, professor in the Department of Communication, who leads the Co-Mind Lab, further discusses this important and timely research in the following piece:

As social scientists, we ask questions that are of fundamental importance to the situation: How is the media covering the crisis? What are changing themes or trends in wider news coverage? How can we quantify and explore the public’s conception of the health crisis? It is now well established that public discourse and behavior—social and cognitive issues—are also of great importance to public health.

The CogMedia (“Cognition and Media”) project is a new endeavor of our group, and its goal is to find linkages between cognitive science and media activity. For example, our group is testing how research findings in the lab regarding language processing (how the mind processes language) can help to predict the spread of information in news media.

The project collects and analyzes many thousands of news items from major media each week. Despite its early stage of development, we decided to release information and resources to the public. One free resource is an application programming interface (API), so other researchers can use CogMedia’s data in a programming language called R. A second resource includes a news consumption platform for members of the public to track major news themes outside of “big tech” filtering.

At the onset of the crisis here, we reworked our project’s underlying code into a dashboard devoted to COVID-19. This dashboard is an “at-a-glance” tool, using automatic algorithms and data. We have hundreds of thousands of stories, and so we have a record of how news coverage has evolved, along with incoming coverage by the hundreds or thousands of stories per day. It seemed important to shape this information into something of direct relevance to this rapidly moving situation, and share it widely for all who may take interest in our approach.

For example, one resource we share is a kind of analysis on news stories referred to as the “newsgraph.” Stories that are related in their language, such as the words they use, can be interconnected into a diagram. This network diagram quickly reveals “clusters” of news stories and themes.

Consider the diagram below, from the morning of my writing (March 26th, 2020). We see the largest cluster of related news stories is the passage by the Senate of the coronavirus assistance bill. With this visualization, users can quickly see where much of the media attention is going. The network marks stories that are widely shared on social media (the size of the dot) and the relative recency (new stories are in red).

We also created a specific Los Angeles dashboard in the system. Here we have used Google’s API and case lists from LA Public Health to offer an animated map of positive test results across the region. In addition, we filtered results to prominent news items from the Los Angeles Times and a Twitter feed from LA Public Health itself. We will continue to expand this offering over the coming days and weeks as we organize our own information and integrate information from other sources.

More broadly, the Department of Communication at UCLA specializes in this “big data” research. Faculty in the department are developing cutting-edge tools and studies. They are creating an unprecedented TV News archive called NewsScape, political and social media analysis, and media-analysis applications using state-of-the-art machine learning and other quantitative methods. For example, the NewsScape contains over 15 years of wide ranging TV news, and faculty Tim Groeling and collaborators are continuously adding to the archive. The Department of Communication will be admitting its first graduate students in a new Ph.D. program this fall, and there will be great interest among our new students in using these tools to further our understanding of the lead up and evolution of the present crisis.

 

Subscribe to LA Social Science and be the first to learn more insight and knowledge from UCLA social science experts in upcoming video/audio sessions and posts dealing with the COVID-19 pandemic.