Qualitative analysis of mentorship matters to researchers and program leaders who want to know who mentors undergraduates, where mentors come from, and which mentoring interactions actually provide support. The PLOS One study collected survey data in Spring 2024 (March–May 2024) and provides concrete numbers, making it possible to apply AI-enabled qualitative methods to synthesize mentoring networks and recommendations. This post translates the PLOS One findings into actionable qualitative research steps and shows how AI tools can speed synthesis for mentoring program design.
Key Takeaways
According to PLOS One, published August 19, 2026, a Spring 2024 survey of 2, 501 biology and biomedical undergraduates found that "most BBS students at both institutions reported having a mentor." The PLOS One study sampled 1, 327 students at Institution A and 1, 174 students at Institution B and reports institution-level differences in mentor sources and interaction patterns.
- 55% of respondents at Institution A and 65% at Institution B reported having a mentor in Spring 2024, per the PLOS One study (p = 6.92 × 10−8).
- 38.1% of Institution A students and 43.1% of Institution B students reported mentors outside the institution, according to PLOS One.
- Peer mentors were reported by 25.2% at Institution A and 34.4% at Institution B in the PLOS One survey.
- The PLOS One analysis found that perceived mentoring support clustered into two validated scales, Degree and Career Support and Existence of a Role Model, with mean DCS = 5.11 (Institution A) and 4.90 (Institution B) in Spring 2024.
- The PLOS One team reported configural and metric measurement invariance but not scalar invariance, so cross-institution mean comparisons should be interpreted cautiously.
What Happened and How the PLOS One Study Worked
The PLOS One study surveyed 2, 501 undergraduates in Spring 2024 (March–May 2024) at two public R1 institutions to measure mentoring sources, mentor-mentee correlates, and perceived mentoring support.
The PLOS One questionnaire used a broad working definition of mentorship, explicitly telling respondents that "A mentor is a person who supports, advises, and guides you, " which the authors used to capture diverse mentoring relationships.
The PLOS One team combined the Process-Oriented Model of Mentoring and the College Student Mentoring Scale, measured experiential similarity, interaction frequency, relationship length, and adapted CSMS items for Degree and Career Support and Existence of a Role Model.
The PLOS One authors ran confirmatory factor analyses and multi-group tests: they achieved configural and metric invariance across institutions but failed scalar invariance, per the PLOS One results published August 19, 2026, which means correlations are interpretable across groups but raw mean differences may reflect response patterns.
Findings Snapshot
| Date | Metric | Value | Implication |
|---|---|---|---|
| March–May 2024 | Total respondents | 1, 327 (Institution A); 1, 174 (Institution B) | Large cross-sectional sample of 2, 501 undergraduates enables descriptive and comparative analyses, per PLOS One. |
| Spring 2024 | Percent reporting any mentor | 55% Institution A; 65% Institution B (p = 6.92 × 10−8) | Mentorship is common but uneven across institutional contexts, per PLOS One. |
| Spring 2024 | Mentor located outside institution | 38.1% Institution A; 43.1% Institution B | Students frequently rely on non-academic mentors, suggesting multi-source networks, per PLOS One. |
| Spring 2024 | Peer mentor prevalence | 25.2% Institution A; 34.4% Institution B | Peer mentoring is a substantial component of undergraduate support systems, per PLOS One. |
| Spring 2024 | Interaction frequency modal response | Most common: "Less than once a month" (higher in Institution B) | Low-frequency, targeted meetings can still coincide with high perceived support, per PLOS One. |
| August 19, 2026 | Measurement validity | Two-factor CSMS model (DCS and ERM) fits each institution; scalar invariance not established | Comparable factor structure exists, but mean-level comparisons require caution, per PLOS One. |
Implications for mentoring researchers and program administrators
Researchers and program administrators should measure mentoring networks across institutional contexts and capture multiple mentor roles rather than assuming faculty-only mentorship, per PLOS One.
- Design surveys to let respondents report multiple mentors, because the PLOS One study shows many students name mentors outside the institution.
- Include mixed methods: pair the PLOS One style CSMS quantitative scales with open-ended interview prompts to capture content of interactions, since PLOS One found weak correlations between frequency and perceived support, implying content matters.
- Collect institutional context data (student-to-faculty ratio, program funding), because the PLOS One authors attribute cross-institution differences partly to institutional structures and resources.
- Use validity testing: follow PLOS One by running CFAs and measurement invariance tests before comparing groups across institutions.
How Evidano Helps: AI-enabled qualitative research for mentoring studies
Problem: large mixed-format mentoring data slow synthesis
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents.
Evidano automates ingestion of survey spreadsheets and interview transcripts, which speeds coding and thematic synthesis for a dataset like the PLOS One Spring 2024 sample.
Evidano supports mixed-methods workflows with thematic, content, frequency, and cross-segment analyses, so teams can link CSMS scale scores to qualitative quotes.
Problem: divergent instruments and cross-institution comparisons
PLOS One highlights the need for mentoring-centric instruments that validate across contexts; Evidano helps by enabling rapid cross-site codebook harmonization.
Evidano lets researchers upload instruments and transcripts, run automated coding suggestions, and compare theme frequencies across segments, reducing the manual work of measurement invariance follow-ups.
See how Evidano features support this workflow in our features page.
Problem: extracting actionable quotes and examples for program design
PLOS One found that content, not just frequency, likely explains perceived support; Evidano extracts representative quotes, links them to codes, and quantifies prevalence across subgroups.
Evidano transcription and survey ingestion reduce preparation time for qualitative analysis, and Evidano’s AI chat over documents helps teams surface illustrative quotations and reconcile coder differences quickly.
Security and ethics for mentoring research
When working with student data, PLOS One authors removed identifying demographics before sharing; Evidano supports secure projects and provides resources about data handling and anonymization on our data security page.
Evidano’s platform supports PII redaction during transcription and encrypted data storage to align with institutional IRB expectations.
FAQ: qualitative analysis of mentorship
What proportion of undergraduates reported having a mentor in the PLOS One study?
Answer: Most undergraduates reported having a mentor, with 55% at Institution A and 65% at Institution B in Spring 2024, per PLOS One.
Supporting detail: The PLOS One survey collected responses from 1, 327 students at Institution A and 1, 174 at Institution B during March–May 2024, which yields a total sample of 2, 501 students used for the descriptive claims.
Where do undergraduates find mentors according to PLOS One?
Answer: Many undergraduates identified mentors outside their academic institution, with 38.1% at Institution A and 43.1% at Institution B reporting outside-institution mentors, per PLOS One.
Supporting detail: PLOS One also reports substantial peer mentoring (25.2% at Institution A, 34.4% at Institution B), indicating that mentoring networks are multi-source.
Does meeting frequency predict perceived mentoring support?
Answer: Meeting frequency had only weak correlations with perceived mentoring support in the PLOS One dataset.
Supporting detail: The PLOS One analysis reported a weak negative correlation between interaction frequency and both Degree and Career Support and Existence of a Role Model, suggesting targeted lower-frequency meetings may still deliver high perceived support.
How should researchers instrument mentoring across institutions?
Answer: Researchers should validate mentoring scales and test measurement invariance before comparing means across institutions, per PLOS One.
Supporting detail: PLOS One achieved configural and metric invariance for the two-factor CSMS subscales but did not achieve scalar invariance, so they recommend caution when interpreting cross-institution mean differences.
How can AI accelerate qualitative follow-ups to the PLOS One findings?
Answer: AI-enabled qualitative platforms can rapidly code open-ended responses, extract representative quotes, and compare themes across subgroups to support the PLOS One call for richer mentoring instruments.
Supporting detail: Using a platform that ingests transcripts and survey spreadsheets lets teams move from descriptive statistics to nuanced thematic analysis faster, enabling the qualitative studies PLOS One recommends for understanding mentor role differences.
Conclusion & Next Steps
The PLOS One study published August 19, 2026, demonstrates that most biology and biomedical undergraduates report mentors, but mentorship sources and interaction patterns vary by institution and role.
For researchers and program leaders, the PLOS One recommendations point to measuring multiple mentors, validating instruments across sites, and collecting open-ended data on interaction content.
If you want to accelerate the qualitative synthesis PLOS One recommends, Evidano helps teams ingest surveys and transcripts, run thematic and cross-segment analyses, and extract representative quotations; learn more on our features page.
Next step: Try Evidano for free to prototype a mixed-methods workflow for mentoring research and get reproducible, auditable qualitative outputs.
Topics
- qualitative analysis of mentorship
- AI qualitative research mentorship
- undergraduate mentoring analysis
Keep reading
- Commentary on NewsFaster Photovoice Qualitative Analysis with AIHow AI accelerates photovoice qualitative analysis: extract themes, frequencies, and policy-ready insights from community photos and narratives. Learn methods and try Evidano.
- Commentary on NewsAI-assisted qualitative analysis for museum observationsHow AI-enabled qualitative analysis accelerates coding and insight from museum visitor observations. Learn methods and stats from PLOS ONE, and try Evidano.
- Commentary on NewsAI qualitative insights: epilepsy medication adherenceAI-enabled qualitative analysis of epilepsy medication adherence in Uganda: 277 patients, 66.5% adherence, and actionable insights for researchers and care teams.