Overview

The lab approaches the delivery of social and health services as a question of equity and implementation: how services and interventions reach, adapt to, and are sustained in communities that have been marginalized and underserved. Its work spans three levels of the service system, from the delivery system, through the populations served, to the practitioners who provide care. Findings are published in health, social work, migration, and computational social science journals, and in peer-reviewed natural language processing and social computing conferences.

The lab runs on volunteer and paid research assistants and meets every other Friday at 1pm. Members take on tasks that fit their interest, time, and experience, and credit follows the contribution they make. The full authorship and contribution policy is on a separate page.

Background: language access in health and social services

The U.S. Census Bureau classifies a person as having limited English proficiency (LEP) if they speak English less than "very well." More than 25 million people, roughly 9 percent of the population over age five, are LEP, and that population has grown substantially since 1990 (Zong & Batalova, 2015). Language barriers are a documented obstacle to health and social services, including mental health care, for immigrant and refugee populations, alongside cost, insurance, stigma, and the limited availability of language-concordant services (World Health Organization, 2023; Mohammadifirouzeh et al., 2023).

Federal law treats language access as a civil rights matter. Title VI of the Civil Rights Act of 1964 and Section 1557 of the Affordable Care Act require health programs that receive federal funding to take reasonable steps to provide meaningful access to individuals with LEP (U.S. Department of Health and Human Services, n.d.). The HHS Office of Minority Health maintains the National Standards for Culturally and Linguistically Appropriate Services (CLAS), which set expectations for language assistance in health and health care (HHS Office of Minority Health, 2013). The lab treats the availability of language-concordant services as a structural determinant of health and well-being and builds data and methods to measure where it is present and where it is missing (Yoo, Hong, & Choi, 2025; Mannan et al., 2025).

Research areas

The lab's work spans three levels of the service system.

• The delivery system. Workforce composition, the availability of culturally and linguistically concordant services, and the state-level policy environments that shape language access. The lab builds the underlying indices, a national census of immigrant-serving organizations, an access score and a language access score, and a state immigrant-policy inclusion index, and maps them in interactive tools such as the US Immigrant & Refugee Support Dashboard in the United States and the Immigrants in Korea Dashboard (Yoo, Park, & Chang, 2025; Yoo, under review).
• The populations. The well-being, coping, help-seeking, and support networks of immigrant and minority communities, and their use of digital and AI-mediated services and support, from online communities and social media to AI chatbots (Yoo, Rodwin, et al., 2026; Yoo & Jang, 2026).
• The practitioners. Capacity-building and the development of interventions for culturally and linguistically appropriate services, in collaboration with the community organizations and local providers that serve immigrant communities (Yoo, Kang, et al., 2026).

Across these levels, the lab treats the delivery of social and health services as a matter of equity and implementation, asking how services reach, adapt to, and are sustained in communities that have been marginalized and underserved.

Methods and approaches

Projects combine quantitative and computational methods with community-engaged research. The lab partners with communities and community organizations at specific stages, such as framing questions, building measures, and interpreting findings. This community-engaged mode is distinct from community-based participatory research, and it does not claim equal community partnership at every stage of a study (Ward et al., 2018). The lab is particularly invested in natural language processing and large language models in social work and migration studies, across three uses: in research, as methods for work with diverse populations; in practice, as technologies that shape assessment, documentation, and decision-making; and in education, as a growing area of training in the field. Most studies combine more than one of the methods below.

• Natural language processing (NLP) and large language models (LLMs). Computational analysis of text to classify content, detect themes, and measure constructs at scale, an approach increasingly used in mental health research (Malgaroli et al., 2023). The lab applies NLP and LLMs to clinical notes (Stanhope et al., 2024; Yoo et al., 2024) and to social media and online communities (Youm et al., 2026; Yoo, Rodwin, et al., 2026).
• Geospatial analysis. Mapping the spatial distribution of need and services, including spatial accessibility measures such as the two-step floating catchment area (2SFCA) method, a gravity-based model that combines supply, demand, and travel distance (Ngui & Apparicio, 2011), applied in the lab to map the clinical social work workforce (Yoo, Park, & Chang, 2025).
• Decomposition. Partitioning a group difference in an outcome into a part explained by measured characteristics and an unexplained part, using the Blinder-Oaxaca decomposition for continuous outcomes (Rahimi & Hashemi Nazari, 2021) and the Fairlie extension for binary outcomes (Fairlie, 2005), used in the lab to explain language-related disparities in care (Yoo, Hong, & Choi, 2025).
• Network analysis. Representing actors as nodes and relationships as edges to study referral and support structures (Northwestern University, n.d.).
• Survey methods. Analysis of complex survey data using sampling weights so that estimates represent the target population rather than the sample alone (National Center for Health Statistics, n.d.).

How the lab works

Work on a project is organized as a ladder of contribution, from foundational support to lead intellectual work. A project is usually carried by one or two people who share its tasks, and members can take on more over time. No prior research experience is needed to start. Where a member's contribution sits on the ladder determines whether it is recognized through acknowledgment or through authorship. Foundational support is recognized through acknowledgment; intellectual contribution to design, analysis, or writing earns authorship.

Research ethics and authorship

The lab follows the authorship standards of the International Committee of Medical Journal Editors (ICMJE) and the American Psychological Association (APA). The ICMJE defines an author as someone who meets four conditions: a substantial contribution to the design of the work or to the acquisition, analysis, or interpretation of data; drafting or critically revising the work; approving the final version; and agreeing to be accountable for it. Contributors who do not meet all four are acknowledged rather than listed as authors (International Committee of Medical Journal Editors, n.d.). Under the APA Ethics Code, authorship credit reflects relative scientific contribution regardless of status, and holding an institutional position does not by itself justify authorship (American Psychological Association, 2017). Each project keeps a contribution log using the Contributor Roles Taxonomy (CRediT), an ANSI/NISO standard of fourteen roles such as conceptualization, methodology, data curation, formal analysis, and writing (National Information Standards Organization, 2022).

The lab does not practice gift or honorary authorship, which lists someone who did not meet the criteria, or ghost authorship, which omits someone who did. Both are documented as common in academic publishing, including among social scientists (Pruschak & Hopp, 2022). Following guidance from the Committee on Publication Ethics (COPE), authorship is discussed early in a project (Committee on Publication Ethics, n.d.). The full policy, including author order, paid versus volunteer work, special situations, and the journals and conferences the lab targets, is on the authorship and contribution policy page.

Meetings and joining

The lab meets every other Friday at 1pm. To get involved, mark interest in a current project on the projects page, or bring a project you are interested in to a lab meeting. Anyone taking on intellectual-tier work has an explicit conversation about authorship early in the project. Volunteers are free to step away at any time without penalty. If you want to lead your own study, see the guide to independent projects.

References

1. American Psychological Association. (2017). Ethical principles of psychologists and code of conduct (2002, amended effective June 1, 2017), Standard 8.12. apa.org/ethics/code
2. Committee on Publication Ethics. (n.d.). Authorship. publicationethics.org
3. Fairlie, R. W. (2005). An extension of the Blinder-Oaxaca decomposition technique to logit and probit models. Journal of Economic and Social Measurement, 30(4), 305–316. rfairlie.sites.ucsc.edu
4. International Committee of Medical Journal Editors. (n.d.). Defining the role of authors and contributors. icmje.org
5. Malgaroli, M., Hull, T. D., Zech, J. M., & Althoff, T. (2023). Natural language processing for mental health interventions: A systematic review and research framework. Translational Psychiatry, 13, 309. nature.com
6. Mohammadifirouzeh, M., Oh, K. M., Basnyat, I., & Gimm, G. (2023). Factors associated with professional mental help-seeking among U.S. immigrants: A systematic review. Journal of Immigrant and Minority Health, 25(2), 1–19. PMC10063938
7. National Center for Health Statistics. (n.d.). NHANES tutorials: Weighting module. Centers for Disease Control and Prevention. cdc.gov
8. National Information Standards Organization. (2022). ANSI/NISO Z39.104-2022, CRediT, Contributor Roles Taxonomy. credit.niso.org
9. Ngui, A. N., & Apparicio, P. (2011). Optimizing the two-step floating catchment area method for measuring spatial accessibility to medical clinics in Montreal. BMC Health Services Research, 11, 166. PMC3142205
10. Northwestern University Institute for Public Health and Medicine. (n.d.). Social network influence in public health and how to map it. feinberg.northwestern.edu
11. Pruschak, G., & Hopp, C. (2022). And the credit goes to … Ghost and honorary authorship among social scientists. PLOS ONE, 17(5), e0267312. journals.plos.org
12. Rahimi, E., & Hashemi Nazari, S. S. (2021). A detailed explanation and graphical representation of the Blinder-Oaxaca decomposition method with its application in health inequalities. Emerging Themes in Epidemiology, 18, 12. PMC8343972
13. U.S. Department of Health and Human Services. (n.d.). Section 1557: Ensuring meaningful access for individuals with limited English proficiency. hhs.gov
14. U.S. Department of Health and Human Services, Office of Minority Health. (2013). National standards for culturally and linguistically appropriate services (CLAS) in health and health care. Federal Register. federalregister.gov
15. Ward, M., Schulz, A. J., Israel, B. A., Rice, K., Martenies, S. E., & Markarian, E. (2018). A conceptual framework for evaluating health equity promotion within community-based participatory research partnerships. Evaluation and Program Planning, 70, 25–34. PMC6077092
16. World Health Organization. (2023). Refugee and migrant mental health [Fact sheet]. who.int
17. Zong, J., & Batalova, J. (2015). The limited English proficient population in the United States. Migration Policy Institute. migrationpolicy.org