[Paper Review] Race and Religion in Online Abuse towards UK Politicians: Working Paper

Genevieve Gorrell, Mehmet E. Bakir+3 more|arXiv (Cornell University)|Oct 2, 2019

Hate Speech and Cyberbullying Detection40 references4 citations

TL;DR

This study analyzes 8 million tweets to UK MPs in H1 2019, finding that ethnic minority politicians—particularly David Lammy and Diane Abbott—received disproportionately high levels of racist and Islamophobic abuse, with abuse patterns showing extreme burstiness. The research reveals that race and religion are central to online abuse, especially amid Brexit-related tensions, and that abuse volume and temporal patterns significantly shape MPs' online experiences beyond simple metrics.

ABSTRACT

Against a backdrop of tensions related to EU membership, we find levels of online abuse toward UK MPs reach a new high. Race and religion have become pressing topics globally, and in the UK this interacts with "Brexit" and the rise of social media to create a complex social climate in which much can be learned about evolving attitudes. In 8 million tweets by and to UK MPs in the first half of 2019, religious intolerance scandals in the UK's two main political parties attracted significant attention. Furthermore, high profile ethnic minority MPs started conversations on Twitter about race and religion, the responses to which provide a valuable source of insight. We found a significant presence for disturbing racial and religious abuse. We also explore metrics relating to abuse patterns, which may affect its impact. We find "burstiness" of abuse doesn't depend on race or gender, but individual factors may lead to politicians having very different experiences online.

Motivation & Objective

• To investigate the role of race and religion in online abuse directed at UK MPs during a period of heightened political tension linked to Brexit.
• To analyze how abuse patterns—particularly their burstiness and temporal distribution—affect the lived experiences of individual MPs.
• To examine the intersection of identity politics, party dynamics (e.g., antisemitism in Labour, Islamophobia in Conservatives), and online hate speech.
• To compare the volume, frequency, and temporal dynamics of abuse targeting MPs of different ethnicities, genders, and party affiliations.
• To explore how high-profile events and tweets by ethnic minority MPs triggered measurable spikes in abusive content, reflecting broader societal tensions.

Proposed method

• Collected a corpus of 8 million tweets authored by or replied to UK MPs from January to June 2019.
• Applied natural language processing (NLP) techniques to classify and detect hate speech, focusing on racial slurs, religious intolerance, and identity-based abuse.
• Used temporal analysis metrics such as 'churn' and 'Gini coefficient' to quantify burstiness of abuse, comparing it to general online attention patterns.
• Tracked topic modeling and keyword frequency to identify dominant themes, including Brexit, antisemitism, Islamophobia, and identity politics.
• Conducted case studies on high-profile MPs (e.g., David Lammy, Diane Abbott) to analyze responses to their tweets on race and religion.
• Employed statistical comparison of abuse volume and distribution across MPs, considering gender, party affiliation, and ethnic background.

Experimental results

Research questions

• RQ1 What abuse was received on Twitter by UK MPs in the first half of 2019? What topics were discussed and who was targeted? How are race and religion situated within this?
• RQ2 What evidence do we find for racial and religious hate speech and discrimination in tweets to UK MPs? What individuals, events and themes emerge? How does this compare with other forms of discrimination in the corpus?
• RQ3 How can we use temporal information to better understand the differing experience MPs have of receiving online abuse, in a way that goes beyond simple quantity?

Key findings

• David Lammy received the highest volume of abusive tweets of any MP in the study period, exceeding both Theresa May and Boris Johnson.
• Diane Abbott received a significant volume of racial abuse, particularly in response to her tweets on the repatriation of a young Muslim woman.
• Sajid Javid and James Cleverly were subjected to strong racial slurs, with Cleverly receiving the second-highest volume of general abuse.
• Ethnic minority MPs received more racist and Islamophobic abuse, while women received more sexist abuse, indicating intersectional patterns of online hostility.
• Boris Johnson received the highest proportion of abuse relative to overall engagement, though antisemitic undertones in some abuse were ambiguous.
• Abuse was significantly more bursty than general online attention, with churn and Gini metrics showing higher volatility, suggesting that sudden surges may amplify psychological impact.