Congressional Tweets on China: A Guide to Using the Data Portal

Written by Young Yang and Rachel Yu, University of California San Diego

In April 2019, the China Data Lab started the “Congress Tweets” project. In March 2022, we issued a four-part series of blogs exploring how Congress tweeted about China over a two-year period in 2019 and 2020. Using Twitter’s API to gather daily tweets, our new Congressional Tweets on China Portal (https://chinadatalab.ucsd.edu/tweets/) allows users to visualize and explore real-time tweets about China authored by members of Congress (MOC).

This portal showcases information about who tweets about China, when they tweet, and what topics they tend to cover.

Data Source

We use Twitter’s API to scrape daily tweets from current members of Congress as well as historical tweets published by MOCs during their tenure in office. The metadata and Twitter handle for each member are taken from an open source github repository. Our portal specifically focuses on the tweets that mention the words “China” or “Chinese”. Our dashboard is updated every day as new congressional tweets about China are collected. 

How to Use This Live Portal

  • The panel on the left hand side of the screen allows users to toggle the view of China tweets according to the Congress, session, chamber, state or territory, and member’s party identification.

  • Placards at the top of the dashboard allow users to view aggregate data about the number of China tweets and number of legislators tweeting about China based on what is selected in the left hand panel.

  • Graphics below these placards display information about the geographic spread, partisan differences, and topic variation in China tweets.

  • The right hand side of the dashboard also features a display of Congressional tweets about China.1

  • At the bottom of the dashboard, users can view and search for legislator-level information. 
Figure 1. A screenshot of our Congressional Tweets on China Portal

Topic Model

We train a topic model to predict the topic prevalence in each China tweet. This helps us to better understand the general themes and specific subjects members of Congress refer to when tweeting about China. Our initial model was trained on all congressional tweets about China from May 8, 2009 to February 8, 2023, with a total of 27,158 tweets. It yielded 89 detailed subtopics which were then hand-labeled based on the words and tweets most strongly associated with the subtopic. Upon further discussion, we consolidated subtopics with similar themes into a single subtopic.2 This resulted in 72 unique subtopics. Each subtopic was then manually coded into one of 9 broader topics.  Upon labeling each topic, we assessed the validity of the model on a new dataset of congressional tweets about China from February 19, 2023 to February 23, 2023.

For more details about the specific topics detected, see the section below titled Topic Details.

Interactive Features

  • Users can use the controls on the left to subset the tweets into posts by the 116th (2019-2021), 117th (2021-2023), or 118th (2023-Present) Congresses. This will be updated with the start of every new Congress. 

  • Users can also subset the data to the first, second, or both sessions of a particular congress.3 

  • Users can subset the viewable tweets by the members of Congress according to the chamber (House or Senate), state, and party identification (Democrat or Republican). 

  • The map shows the distribution of China-related tweets by state (for the Senate) and by congressional district (for the House). 

  • The boxplot on the right shows the ‘outlier’ politicians who sent the most tweets about China. Each dot represents a top tweeting politician and reveals their tweet count on China. You can click the dot to identify where the legislator’s constituency is on the map.

Preliminary Findings

As China and the U.S.-China relations become fraught, members of Congress have been active in communicating about China-related issues on Twitter. Early this year, the U.S. House of Representatives established the Select Committee on the Chinese Communist Party, formed by a bipartisan group of legislators,4 signaling increased congressional attention on China. It remains important to understand the opinions and positions of U.S. Congressional members in order to appreciate the overall U.S. attitudes and policy toward China.

Our real-time data confirms two key findings from the 2019 and 2020 data.

① Republicans remain far more active than Democrats in tweeting about China. In the 117th Congress, the median Republican tweeted 8 times more than the median Democrat about China (See Figure 2). This trend continues into the current Congress.

Figure 2. Count of Tweets Mentioning China by Top Tweeters from Both Parties in the 117th Congress

② Within congressional tweets about China, “instrumental tweets” are becoming more prevalent over time.  Instrumental tweets refer to tweets that do not target China as the focus of discussion but use China as a way to attack political opponents, signal legislative competence, or explain domestic crises.  Our data shows that members of Congress use instrumental themes more and more when talking about China on Twitter.

In the second session of the 116th Congress, for example, instrumental tweets accounted for 12.53% of all China tweets. In the 117th Congress, this number rose to 17.09% in the first session and to 21.86% in the second session. In the current Congress, instrumental tweets account for 22.64% of the China tweets thus far. All this is pointing to an increasing effort by Congressional members to use China to leverage gains in domestic politics.

Figure 3. Subtopics in the 118th Congress as of April 28th, 2023

Figure 3 displays the top 25 subtopics captured in the China tweets for the 118th Congress thus far. The teal boxes illustrate different themes in the instrumental tweets.

Figure 4 shows an example of a tweet that was identified as highly instrumental in reference to China as it is actually about the “Hunter Biden” subtopic.

Topic Details

Our topic model yielded 72 unique subtopics which are hand-labeled based on the words and tweets most strongly associated with the subtopic. Request for our full documentation for examples of the words and tweets used to label these topics. Each subtopic is also hand coded into one of 9 broader topics.

Click show more to see overviews of the broader topics.

ECONOMICS: Percentage of tweets pertaining to the economy, trade, tariffs, and supply chains.

Relevant subtopics include: Trade Negotiations, Trump & Trade Negotiations, Protecting American Workers, Agricultural Harms from Trade War, Trade War, Unfair Competition, Supply Chains & Chips Act, Economics (Other), Medical Supply Chains, Promoting U.S. Innovation, Leveling the Playing Field, Tariffs, Announcing Legislation on Trade with China

GEOPOLITICS: Percentage of tweets pertaining to geopolitical tensions, relations with China vis-a-vis other nations, and perceptions of Chinese international influence. Note that these tweets refer to Chinese activities, influence, or statements abroad or in relation to the international community5 

Relevant subtopics include: Constraining Chinese Influence, Diplomacy, Competition in Research & Science, Taiwan, Countering Chinese Military, Competition in Technology and Innovation, Positioning Against Adversaries, Countering China and Russia, China as a Global Threat, Climate Cooperation, Fighting Communism, Countering Chinese Influence.  

HUMAN RIGHTS: Percentage of tweets pertaining to human rights in China, China’s position on human rights, detainments, and democracy.

Relevant subtopics include: China and the Olympics, Detainments Abroad, Democracy in Hong Kong, Uyghurs, Support for Human Rights Advocates, Extradition Law in Hong Kong, Holding China Accountable, Tibet

INFLUENCE: Percentage of tweets pertaining to Chinese influence in the U.S. governance, business, security, education, research, entertainment, and U.S. society broadly. 

Relevant subtopics include: Tech and Privacy, IP Theft, Chinese Influence on the Military, Confucius Institutes, Covert Influence, Influence (Other), Urging Leaders to Prevent Chinese Influence, Hollywood, Research, Calls for Transparency, Chinese Propaganda

INSTRUMENTAL: Percentage of tweets that use China as a way to counter or criticize other legislators, signal legislative competence, or explain domestic crises. Note that these tweets are not necessarily about China per se; rather, they use China as a topic to position the legislators relative to other legislators or topics.

Relevant subtopics include: Statements on Presidential Action on China, Standing Firm Against China, Legislative Oversight, Announcing Legislation to Counter Chinese Influence, China used to Attack Domestic Policies, Criticizing Trump on China, Criticizing Biden on China, Announcing Committee Action on China, Hunter Biden, Criticize Biden on China, U.S. Energy Crisis, Instrumental (Other)

NATIONAL SECURITY: Percentage of tweets pertaining to U.S. national security interests and the state of national security in relation to China.

Relevant subtopics include: Natural Resources, Military, Chinese Tech Companies

ASIAN AMERICANS: Percentage of tweets pertaining to Asian Americans in the U.S.

Relevant subtopics include: Asian Hate in the U.S., Chinese American Veterans, Lunar New Year

PUBLIC HEALTH: Percentage of tweets pertaining to public health issues in relation to China, including the COVID-19 pandemic and fentanyl in the U.S.

Relevant subtopics include: COVID Accountability, COVID Origins, Fentanyl, Holding China Accountable for COVID-19

OTHER: Percentage of tweets not directly pertaining to China, informational in nature, or unable to be classified

Relevant subtopics include: Reporting on China, Q&A on China Related Issues, COVID-19 Alerts, Spanish Tweets, Other

Authors

Young Yang, Research Data Analyst, China Data Lab at the 21st Century China Center, UC San Diego School of Global Policy and Strategy

Rachel Yu,  Ph.D. student in political science and research assistant at China Data Lab, UC San Diego School of Global Policy and Strategy. Her research interests include diplomacy, U.S. foreign policy, foreign lobbying, and U.S.- China relations.

Read More

Footnotes

  1. The UI for this feature is adapted from items in the shinythings open source GitHub repository.
  2. The 89 subtopic list can be found in the methodology notes.
  3. Because our data collection began in April 2019, there is incomplete data and missing graphics for the first session of the 116th Congress.
  4. https://selectcommitteeontheccp.house.gov/
  5. This is different from the Influence topic which focuses on perceived Chinese influence within the United States.

Cover image generated by Young Yang using Midjourney.

Recommended Posts
Contact Us

We're not around right now. But you can send us an email and we'll get back to you, asap.

Not readable? Change text. captcha txt

Start typing and press Enter to search

Photo by Baim Hanif on Unsplash