Using Large Language Models to Examine Family Stresses During Pandemic

<a href="https://embed.stanford.edu/iframe/?url=https%3A%2F%2Fpurl.stanford.edu%2Fyx926fy9172" class="su-underline">Show Content</a>

Abstract/Contents

Abstract: The COVID-19 pandemic has exposed families with young children to various difficulties and challenges, significantly affecting children’s early experiences and overall development. Despite numerous studies investigating the pandemic's impact through quantitative measures, they potentially overlook the dynamic, evolving experiences, perspectives, and challenges families confront. To address the research gap, the current study utilizes a dataset from the Rapid Assessment of Pandemic Impact on Development (RAPID) project. I leverage OpenAI's GPT-3, a state-of-the-art NLP model, to examine open-ended responses to the question “What is challenging during the pandemic for your family”. This approach not only offers a richer and more dynamic understanding of the pandemic's impacts but also uncovers disparities in challenges across different racial and income demographics. By identifying challenge categories and analyzing their prevalence, the study would further provide actionable insights for stakeholders, facilitating corresponding support for diverse caregiver groups and aiding long-term early childhood program and policy development.

Author	Zhao, Priscilla
Advisor	Liu, Sihong
Advisor	Fisher, Philip
Advisor	Pea, Roy
Advisor	Smith, Sanne
Department	Stanford Graduate School of Education
Degree granting institution	Stanford University

Subject	Early childhood development
Subject	COVID-19 (Disease)
Subject	Well-being
Subject	Natural language processing
Genre	Text
Genre	Article

Related item	Preferred citation Liu S, Zalewski M, Lengua L, Gunnar MR, Giuliani N, Fisher PA. Material hardship level and unpredictability in relation to U.S. households' family interactions and emotional well-being: Insights from the COVID-19 pandemic. Soc Sci Med. 2022 Aug;307:115173. doi: 10.1016/j.socscimed.2022.115173. Epub 2022 Jun 30. PMID: 35785642; PMCID: PMC9242702.
DOI	https://doi.org/10.25740/yx926fy9172
Location	https://purl.stanford.edu/yx926fy9172

Use and reproduction: User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
License: This work is licensed under a Creative Commons Attribution Non Commercial 4.0 International license (CC BY-NC).

Preferred citation: Zhao, P. (2023). Using Large Language Models to Examine Family Stresses During Pandemic. Stanford Digital Repository. Available at https://purl.stanford.edu/yx926fy9172. https://doi.org/10.25740/yx926fy9172.

Education Data Science (EDS) Capstone Projects, Graduate School of Education

Loading usage metrics...