Mixed-Methods Insight: AI Qualitative Analysis

The PLOS One study, published on November 10, 2026, shows how mixed-methods research can combine questionnaire data with interviews to examine occupational physical fitness among information technology students in selected vocational colleges in Shanxi Province, China.

AI qualitative analysis can support this workflow by identifying themes across interviews and documents, comparing stakeholder segments, and linking qualitative interpretations to the quantitative context without claiming that interview themes provide statistical inference.

• In data collected from April 20 to May 29, 2025, the study retained 1, 924 questionnaires: 1, 863 from students, 37 from teachers, and 24 from enterprise personnel.
• In the 2026 PLOS One study, 68.71% of students selected “Frequently” or “Occasionally” for physical discomfort after professional practical operations such as prolonged computer use or equipment debugging.
• In the same study, 30 people participated in supplementary interviews: 21 students, 6 teachers, and 3 enterprise personnel.
• The researchers identified three qualitative themes: student occupational physical-fitness problems and limited training methods, insufficient institutional resources, and weak alignment between enterprise demands and school-based provision.
• The study concluded that its three-part framework, general physical fitness, occupation-specific physical fitness, and psychological fitness, remains preliminary and requires further validation.

The 2026 PLOS One study examined perceived occupational physical-fitness priorities among information technology students, teachers, and enterprise personnel in selected vocational colleges in Shanxi Province, China.

According to the study authors, the research used a quantitatively dominant cross-sectional mixed-methods design. The questionnaire described physical discomfort, exercise, psychological experiences, training provision, enterprise involvement, and perceived importance ratings, while semi-structured interviews supplied contextual explanations.

The study authors reported that 68.71% of students selected “Frequently” or “Occasionally” when asked about physical discomfort after professional practical operations. The most frequently selected categories were dry eyes or blurred vision at 44.50%, cervical or lumbar soreness at 38.06%, and physical fatigue or insufficient endurance at 25.93%.

The study authors wrote that occupational physical fitness “should not be simply understood as general sports ability.” The authors instead organized the proposed framework around general physical fitness, occupation-specific physical fitness, and psychological fitness.

The source study also reported important constraints. The teacher and enterprise samples were relatively small, recruitment used convenience and snowball sampling, and the qualitative audio files, verbatim transcripts, field notes, and original coding records were no longer available for independent retrospective verification.

Mixed-methods researchers should analyze qualitative themes as contextual evidence and keep them distinct from survey estimates, because the 2026 PLOS One study explicitly treated its interview findings as supplementary explanatory evidence rather than as a basis for statistical inference.

The first implication is separation of evidence types. The source study distinguished student-reported experiences, stakeholder-rated priorities, and provision constraints instead of combining them into one undifferentiated measure of need.

The second implication is segment-aware coding. The source study included students, teachers, and enterprise personnel, so an AI qualitative analysis workflow should preserve stakeholder labels and compare how each group describes discomfort, training, resources, and job demands.

The third implication is auditability. The source study stated that original interview recordings, transcripts, field notes, and coding records were unavailable for retrospective verification. Researchers using AI should therefore preserve source excerpts, coding decisions, definitions, and review notes while analysis is in progress.

The fourth implication is cautious framework building. The source study used literature analysis, expert content evaluation, questionnaire results, and interview themes to propose a framework, but it did not present the framework as a fully validated latent structure. AI-generated themes should similarly be treated as research inputs that require human review and methodological validation.

The 2026 PLOS One study shows how mixed-methods research can combine 1, 924 questionnaires with 30 interviews to examine a complex educational and occupational-health topic. AI qualitative analysis can make this type of work more manageable by accelerating thematic coding, preserving stakeholder distinctions, comparing segments, and connecting qualitative evidence with quantitative context. Responsible use requires human review, transparent evidence trails, careful treatment of limitations, and a clear separation between reported experiences and causal or clinical claims. Researchers can try Evidano for free to explore AI-supported analysis for interviews, open-ended surveys, and research documents.