Women worry about family, men about the economy: Gender differences in emotional responses to COVID-19 (2004.08202v2)

Abstract: Among the critical challenges around the COVID-19 pandemic is dealing with the potentially detrimental effects on people's mental health. Designing appropriate interventions and identifying the concerns of those most at risk requires methods that can extract worries, concerns and emotional responses from text data. We examine gender differences and the effect of document length on worries about the ongoing COVID-19 situation. Our findings suggest that i) short texts do not offer as adequate insights into psychological processes as longer texts. We further find ii) marked gender differences in topics concerning emotional responses. Women worried more about their loved ones and severe health concerns while men were more occupied with effects on the economy and society. This paper adds to the understanding of general gender differences in language found elsewhere, and shows that the current unique circumstances likely amplified these effects. We close this paper with a call for more high-quality datasets due to the limitations of Tweet-sized data.

• The paper analyzes a text dataset using Bayesian methods and topic models to find significant gender differences in emotional responses during COVID-19.
• Findings showed women reported higher anxiety, sadness, and fear concerning family, while men expressed more desire, relaxation, and worry about the economy.
• These findings suggest that tailoring mental health interventions based on gender differences is important and that longer texts provide deeper emotional insights than short ones.

Gender Differences in Emotional Responses to COVID-19

Summary and Findings

The paper "Women worry about family, men about the economy: Gender differences in emotional responses to COVID-19" explores gender-specific differences in emotional reactions during the COVID-19 pandemic. By analyzing a "Real World Worry Dataset" consisting of long and short texts, the research identifies marked distinctions in concerns expressed by men and women regarding the crisis. Utilizing robust computational methodologies, the paper seeks to elucidate how gender modulates the expression of emotional responses through language.

Dataset Description and Analysis Methods

The Real World Worry Dataset (RWWD) consists of 5,000 texts written by participants enlisted via a crowdsourcing platform. Both long and short text samples were collected, allowing for an examination of variations in emotional expression based on text length. Using Bayesian hypothesis testing for emotional scoring, the paper assesses potential gender differences in the expressed emotions such as anger, anxiety, fear, happiness, relaxation, desire, disgust, and sadness.

A topic model approach linked specific concerns to text characteristics and gender differences in both long and short text formats. Utilizing n-gram, open vocabulary, and closed vocabulary approaches, the paper derives psycholinguistic constructs emphasizing gender-specific language usage about the pandemic.

Gender Differences in Emotional Concerns

Findings reveal substantial gender disparities. Women consistently express higher levels of anxiety, sadness, and fear and are more concerned about family and friends, translating to specific language usage related to emotional processes. Men show higher concern for economic and societal matters and utilize language more oriented toward analytical thinking, articles, and power-related constructs.

Bayesian hypothesis testing shows extreme evidence that women are more worried and anxious than men, whereas men report more desire and relaxation. Topic models highlight particular thematic focal points: women predominantly refer to sickness, death, and familial relationships, whereas men discuss national impact, health concerns, and future outlooks.

Implications and Future Directions

These gender differences have potential implications for tailoring targeted mental health interventions. Understanding the depth of emotional responses conveyed through longer text formats versus short tweets offers a nuanced perspective on emotional expression, suggesting that Twitter data may provide less comprehensive insight into deeper concerns related to crises. This prompts a call for more expansive, high-dimensional text datasets to capture the emotional landscape more fully.

Future research could explore the impact of different text lengths across diverse demographic contexts outside the UK to validate findings. Moreover, understanding how linguistic expression varies with different text types and emotional manifestations could refine computational techniques in emotion detection.

The paper underscores the importance of distinguishing general language differences from those amplified by the pandemic context, suggesting that COVID-19 may have intensified pre-existing gender differences in emotional expression through language. This contributes to the broader discussion on gender-specific communication and its impact on research in social sciences and mental health domains, providing valuable insights for developing contextualized intervention strategies.