Research Fellow: AI-moderated Interviews at Scale: The Trust in Healthcare Study
Required skills
- AI-moderated Interviews
- Trust in Healthcare
- dataset
- qualitative dialogues
- conversational interviews
- peer-reviewed publication
- analytical plan
- journal manuscript
- analysis code
Nice to have
- computational methods
- transcript data
- interviewer design
- conversation dynamics
- external data integration
Role summary
Surgo Health is seeking proposals for research projects utilizing their dataset on healthcare trust, which includes qualitative dialogues from AI-moderated interviews and standard survey data. The role of Research Fellow focuses on advancing the understanding of trust in healthcare, particularly exploring the value of conversational interview data. Projects should leverage the existing dataset to answer research questions, aim for peer-reviewed publication, and demonstrate practical application. Key research areas include trust drivers, survey methodology, interviewer design, and computational methods for analyzing transcript data. Selected fellows will receive a $6,000 stipend, dataset access, and methodological support.
- Advance the narrative on trust & healthcare —contribute to contemporary thinking on this subject in the U.S.
- Leverage the conversations — projects should examine the reasoning, experiences, or context revealed when respondents elaborate, going beyond what closed items already capture.
- Use the dataset — research questions should be answerable with the current dataset and not require funding for new data collection
- Publishable — proposals should aim for peer-reviewed publication, with rigorous methodology.
Research Focus Areas
We welcome proposals in areas below, but also encourage you to submit proposals that are adjacent to these areas.- Trust in healthcare — drivers of trust erosion and recovery; public perspectives on AI in healthcare; decisions made against medical advice; institutional versus individual trust; and perceptions of the pharmaceutical industry.
- Survey methodology — variations in response quality by education, age, income, geography, and political views; order and carry-over effects across modes; effort signatures; construct validity of trust items; and turn-level drop-off in conversational surveys.
- Interviewer design and conversation dynamics — probing strategies and sequencing; respondent responses (elaboration vs. new disclosures); interviewer neutrality; and managing sensitive disclosures.
- Computational methods — analysis pipelines for transcript data, including predicting closed-ended responses from conversational text (and vice versa).
- Core contribution — how the research builds beyond existing literature
- Use of qualitative data — what insights the conversational responses provide beyond standard closed-ended items.
- Practical application — key decisions or practice changes informed by the findings.
- Analytical plan — key variables, panel-specific estimates, and analytical approaches for randomized and observational factors.
- Publication goals — target journals and expected outputs.
Program Structure
-
Surgo Health will select two to three proposals based on:
- Practical relevance and research depth
- Feasibility with available data
- Alignment with Surgo Health research priorities
- Potential for publication in high-impact journals
- Stipend amount: $6,000 in compensation upon successful submission of draft to Surgo
- Weekly touch-point meetings on the project
- Full access to the study dataset
- Methodological and analytical support from Surgo's research team
- Onboarding resources, including the data dictionary and reporting standards
- Lead authorship and publication support
Timeline, Milestones, & Expected Output
Timeline & Milestones
- Earliest project start date: 26th October 2026
- Paper completion: 19th February2027
- Final manuscript submission: 26th February 2027
- Approved analytical plan (Weeks 1–2)
- Preliminary results review
- Draft manuscript internal review
- Final manuscript submission
Expected Outputs
- One journal manuscript prepared for peer review
- Documented analysis code reproducing all reported figures from the dataset
Originally posted on Himalayas
