Qualitative analysis of girls' GCSE performance

Primary task: move from national aggregates to lived experience, interview girls, teachers and attendance officers to trace pathways from wellbeing to grades.

Design: purposive sampling (persistently absent vs consistently present), compare urban/rural and socioeconomic segments.

Outcome: hypothesis-tested explanations that quantitative data alone cannot prove.

Focus on user-centred signals: how do girls describe distractions, confidence and online harms? Capture verbatim quotes to inform interventions.

Rapid A/B testable ideas: peer-belonging programs, targeted wellbeing check-ins, sleep education piloted in selected cohorts.

Translate local qualitative themes into policy levers: attendance strategies, pastoral staffing, curriculum review.

Use mixed-methods briefs (thematic evidence + segment comparisons) when advising ministers or funding boards.

Consider assessment design; see Ofqual review for historical context: ofqual.blog.gov.uk/2021/05/17/bias-in-teacher-assessment-results/

Upload interview transcripts, survey spreadsheets and school reports directly; use automated transcription (with custom dictionary for school names/terms) and translation where needed.

Automatically generate thematic coding across gender, attendance status and school type; run cross-segment frequency analysis to test whether a theme (e.g., sleep disruption) is concentrated in girls who are persistently absent.

Data is encrypted and not used to train third-party models, useful when handling pupil data or sensitive wellbeing quotes.

PII redaction and codebook import/export help maintain audit trails for ethics and governance.

One-click visualizations (co-occurrence networks, hierarchical themes) and exportable quote packs make reporting to headteachers, governors and policy teams faster and more persuasive.

Collect teacher interviews, pastoral logs, student focus groups, parent surveys and anonymised attendance registers.

Upload files to Evidano, run transcription/translation, apply PII redaction and add school-specific dictionary entries.

Load an initial codebook (wellbeing, sleep, social media, bullying, assessment anxiety) or let Evidano suggest themes from the corpus.

Produce theme prevalence by segment (girls vs boys; persistently absent vs regular attendees) and surface contrasted narratives.

Use co-occurrence networks to find linked drivers (e.g., 'social media' + 'sleep'), and extract representative quotes for each theme.

Share a short report with school staff for face validity, rerun coding on new interviews if needed, and lock final themes for reporting.

Export visuals and an evidence brief for policy teams or SLTs; pair qualitative findings with attendance and attainment KPIs to prioritize interventions.