Researchers studying vocational education, student wellbeing, and information technology training often need to connect large survey datasets with interviews, open-ended responses, and published evidence. AI qualitative analysis can help researchers organize those materials, identify recurring themes, compare stakeholder perspectives, and preserve an auditable path from evidence to interpretation. The 2026 PLOS ONE study by Xiaobin Jin, Xiaoying Cui, Yi Zhao, and Yu Hou provides a useful example: the researchers combined 1, 924 questionnaires with supplementary interviews to examine occupational physical fitness priorities among information technology students in selected vocational colleges in Shanxi Province, China. This article explains what the study found and how AI-assisted qualitative research can help teams analyze similar mixed-methods projects without treating automated outputs as a substitute for methodological judgment.
Key Takeaways
According to the PLOS ONE study on information technology students and occupational physical fitness, mixed-methods analysis showed that student-reported discomfort, stakeholder priorities, and training constraints should be interpreted as related but distinct evidence domains.
- The November 10, 2026 PLOS ONE study analyzed 1, 924 questionnaires, including 1, 863 students, 37 teachers, and 24 enterprise personnel.
- The PLOS ONE authors reported that 68.71% of students selected “Frequently” or “Occasionally” when asked about physical discomfort after prolonged computer use or equipment debugging.
- The PLOS ONE study found that dry eyes or blurred vision was selected by 44.50% of students, cervical or lumbar soreness by 38.06%, and physical fatigue or insufficient endurance by 25.93%.
- The PLOS ONE study used interviews with 30 participants, including 21 students, 6 teachers, and 3 enterprise personnel, to contextualize questionnaire findings.
- The authors proposed a preliminary framework containing general physical fitness, occupation-specific physical fitness, and psychological fitness, while stating that further validation is required.
What the 2026 Study Found
The PLOS ONE study found that information technology students reported physical discomfort, exercise patterns, psychological experiences, and limited access to occupation-specific training. The researchers collected data between April 20 and May 29, 2025, from students, teachers, and enterprise personnel in Shanxi Province, China.
The study used a quantitative-dominant cross-sectional mixed-methods design. The questionnaire was the primary method, while semi-structured interviews supplied contextual explanations for the survey results. The authors analyzed the quantitative and qualitative strands separately before integrating them during interpretation.
The authors identified three documented qualitative themes: student occupational physical fitness problems and limited training methods, insufficient institutional resources, and weak alignment between enterprise job demands and school-based training. One reported constraint was “limited incorporation of occupation-specific training, ” including visual relaxation, musculoskeletal relaxation, postural adjustment, and sedentary behavior interruption.
The study proposed a preliminary framework with 12 indicators across three dimensions. General physical fitness included cardiorespiratory endurance, muscular strength, flexibility, and bodily coordination. Occupation-specific physical fitness included sedentary tolerance, visual load tolerance, fine motor operation endurance, and static postural endurance. Psychological fitness included stress resistance, sustained concentration, emotional regulation, and emergency response capacity.
The authors cautioned that the framework was not a validated clinical or employment-screening instrument. The study used convenience and snowball sampling, included relatively small teacher and enterprise groups, and did not conduct exploratory factor analysis, confirmatory factor analysis, or structural equation modeling.
Mixed-Methods Research Findings Snapshot
| Date or period | Metric | Value | Implication for qualitative analysis |
|---|---|---|---|
| April 20 to May 29, 2025 | Data collection period | Questionnaires and interviews | Analysis should preserve the relationship between collection context and participant accounts. |
| November 10, 2026 | Final quantitative sample | 1, 924 questionnaires | Large survey datasets benefit from structured coding and segment comparisons. |
| November 10, 2026 | Student share of final sample | 1, 863 students, 96.83% | Student findings should remain analytically distinct from smaller stakeholder groups. |
| November 10, 2026 | Interview sample | 30 participants | Interview themes can explain or qualify survey patterns, but they do not establish statistical prevalence. |
| November 10, 2026 | Students reporting frequent or occasional discomfort | 68.71% | The finding identifies a reported experience, not a clinical diagnosis or causal relationship. |
| November 10, 2026 | Students reporting 1 to 3 hours of exercise weekly | 52.01% | Frequency analysis can show response distributions without assuming that participants met activity guidelines. |
| November 10, 2026 | FDR-adjusted stakeholder comparisons below q < 0.05 | 0 of 15 | Automated summaries should preserve correction methods and avoid overstating nominal differences. |
Implications for Researchers Analyzing Student Health and Education Data
How can researchers connect survey percentages with interview themes?
Researchers can connect survey percentages with interview themes by treating quantitative results as patterns to explain and qualitative findings as contextual evidence, rather than combining them into one undifferentiated score.
The PLOS ONE authors used this approach by separating student-reported experiences, stakeholder-rated priorities, and provision constraints. An AI qualitative analysis workflow can mirror that structure through document coding, thematic summaries, and cross-segment comparisons.
Why should stakeholder groups be analyzed separately?
Stakeholder groups should be analyzed separately when their sample sizes, roles, questions, or perspectives differ. In the PLOS ONE study, students represented 1, 863 of 1, 924 questionnaire respondents, while teachers represented 37 and enterprise personnel represented 24.
The authors treated teacher and enterprise findings as descriptive and exploratory rather than as population-level industry estimates. Researchers using AI should therefore create separate segments for students, teachers, and employers before comparing themes or response distributions.
What should AI-assisted qualitative analysis preserve?
AI-assisted qualitative analysis should preserve source excerpts, speaker or respondent segments, coding definitions, negative cases, methodological limitations, and the distinction between reported experience and verified outcome.
The PLOS ONE study explicitly cautioned that its psychological items were not validated clinical instruments and that its physical-discomfort items were not diagnostic measures. AI summaries should retain those qualifications instead of converting descriptive responses into diagnoses.
How Evidano Helps With AI Qualitative Analysis
Problem: Mixed-methods evidence is scattered across files
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, research papers, reports, and other documents.
Researchers can upload study papers, interview transcripts, reports, and survey-response files, then use thematic, content, frequency, and cross-segment analyses to organize evidence across a project. The Evidano features overview describes the available analysis capabilities.
Problem: Interview themes are difficult to compare across stakeholder groups
Evidano can help researchers compare themes across defined segments such as students, teachers, enterprise personnel, departments, regions, or survey-response categories.
For a study modeled on the PLOS ONE research, a researcher could examine whether “training resources, ” “job-demand information, ” or “physical discomfort” appears differently across stakeholder groups, while reviewing the underlying evidence for each theme.
Problem: Large document sets slow down synthesis
Evidano can generate thematic and frequency analyses across large collections of qualitative documents, allowing researchers to identify recurring concepts before conducting detailed human review.
Researchers can also use hierarchical codes and subcodes, word clouds, and co-occurrence networks to inspect relationships among concepts such as prolonged sitting, visual strain, postural adjustment, and sustained concentration.
Problem: Researchers need to interrogate findings after the first analysis
Evidano provides AI chat over uploaded documents and analyses, so researchers can ask focused follow-up questions about themes, segments, source evidence, and exceptions.
A researcher could ask which interview passages support a training-resource theme, whether a theme occurs across student and teacher documents, or which findings require cautious interpretation because of sampling limitations.
Problem: Sensitive participant data requires safeguards
Evidano supports transcription with PII redaction, and Evidano states that data is encrypted and never used to train third-party models.
Researchers handling student interviews or occupational-health discussions should still apply informed-consent requirements, institutional policies, de-identification procedures, and human review before sharing or publishing AI-generated outputs.
FAQ: AI Qualitative Analysis for Mixed-Methods Research
What is AI qualitative analysis?
+
AI qualitative analysis uses machine learning or large language models to help researchers organize, code, summarize, compare, and query qualitative data.
AI qualitative analysis can support thematic analysis of interview transcripts, open-ended survey responses, papers, reports, and other documents, but researchers remain responsible for codebook design, validation, interpretation, and ethical review.
How did the PLOS ONE study use qualitative research?
+
The PLOS ONE study used semi-structured interviews as a supplementary qualitative component alongside a primary questionnaire survey.
The qualitative sample included 30 participants, and the documented themes were used to explain student experiences, training barriers, and school-enterprise coordination issues rather than to make statistical inferences.
Can AI qualitative analysis replace human thematic analysis?
+
AI qualitative analysis should support human thematic analysis rather than replace researcher judgment.
The PLOS ONE study reported limitations involving unavailable original interview materials and the preliminary status of its framework. Those limitations show why researchers must verify AI-generated codes against source evidence and document analytical decisions.
What sample size did the information technology student study analyze?
+
The study analyzed 1, 924 questionnaires collected from 1, 863 students, 37 teachers, and 24 enterprise personnel.
The researchers also reported 30 qualitative interview participants, consisting of 21 students, 6 teachers, and 3 enterprise personnel.
What were the main qualitative themes in the study?
+
The main qualitative themes were limited occupation-specific training, insufficient institutional resources, and inadequate alignment between enterprise job demands and school-based training.
The authors used these themes to contextualize survey evidence about physical discomfort, psychological experiences, exercise participation, and training provision.
Are the study findings medical or clinical conclusions?
+
The study findings are research observations about self-reported experiences, not medical diagnoses or clinical conclusions.
The PLOS ONE authors stated that their physical-discomfort and psychological-experience items were not validated diagnostic instruments. This article is intended for research and educational analysis, not individual medical advice.
Conclusion and Next Steps for AI Qualitative Analysis
The November 10, 2026 PLOS ONE study demonstrates how survey data and qualitative interviews can jointly illuminate a complex education and occupational-health question. The study also shows why mixed-methods researchers must preserve stakeholder differences, methodological limitations, and the distinction between reported experiences and validated outcomes.
AI qualitative analysis can make this work more manageable by organizing documents, surfacing themes, comparing segments, and enabling researchers to revisit evidence through natural-language questions. Human researchers should still verify quotations, review contradictory cases, and decide how findings support the research question.
Teams analyzing interviews, open-ended surveys, and research documents can try Evidano for free to accelerate qualitative synthesis while keeping interpretation grounded in source evidence.
Topics
- AI qualitative analysis
- AI thematic analysis
- mixed-methods research analysis
- qualitative analysis of interview transcripts
- student health research
Keep reading
- Commentary on NewsAI Qualitative Analysis for Mixed-Methods ResearchSee how AI qualitative analysis turns mixed-methods evidence into traceable themes, comparisons, and decisions for education and health researchers.
- Commentary on NewsAI Qualitative Analysis: Lessons from 1, 924 IT StudentsLearn how AI qualitative analysis can synthesize mixed-methods research, from 1, 924 IT students in Shanxi to actionable education insights.
- Commentary on NewsFrom Mixed Methods to Insight: AI Qualitative AnalysisSee how AI qualitative analysis turns mixed-methods research into traceable themes, comparisons, and decisions, using a PLOS One study as a practical example.