The Raspberry Pi Foundation’s August 21, 2025 report on Astro Pi 2024–25 shows clear signals from mentor feedback: increased confidence, strong motivation from the space hook, and operational frictions for less-experienced teams. If your role is research, evaluation, or UX for education programs, this is a tidy corpus to test scalable qualitative methods. This post explains how to run a focused qualitative analysis of Astro Pi using AI tools, extracting themes, comparing segments (Mission Zero vs Mission Space Lab), and turning mentor quotes into stakeholder-ready findings. See the original report at www.raspberrypi.org/blog/code-confidence-and-community-a-look-back-at-astro-pi-2024-25/ and learn how www.evidano.com can map raw survey responses and focus-group transcripts to themes, frequencies, and visualizations in hours, not weeks.
Fast take: qualitative analysis of Astro Pi (source)
Raspberry Pi’s post (published 21 August 2025) reports that 27, 304 young people from 27 countries engaged with Astro Pi 2024–25 and 26, 294 had their code run on the ISS across Mission Zero and Mission Space Lab. Mentors (n=99 for Mission Zero, n=26 for Mission Space Lab) plus five focus-group participants contributed survey and discussion data that reveal what worked and what blocked participation.
- Source: www.raspberrypi.org/blog/code-confidence-and-community-a-look-back-at-astro-pi-2024-25/
- Why this matters: the dataset mixes short survey responses and rich mentor quotes, ideal for thematic + cross-segment analysis.
Findings snapshot
| Metric | Value | Note / implication |
|---|---|---|
| Total young people involved | 27, 304 | Participants across 27 countries |
| Programs that ran in space | 26, 294 | Count across both missions |
| Mentor survey respondents | 99 (Mission Zero); 26 (Mission Space Lab) | Primary source of qualitative feedback |
| Focus groups | 5 mentors | Deeper contextual interviews |
| Non-submitting registered teams (mentor responses) | 55 | Main barriers: lack of time; perceived difficulty |
| Next Astro Pi opening | 8 Sep 2025 | Registration opens, useful operational deadline |
What happened: data & methods in plain English
The Raspberry Pi Foundation combined post-submission mentor surveys and five focus-group discussions to evaluate Astro Pi 2024–25. The dataset is relatively small on the qualitative side (≈130 survey/FG contributors) but high-signal: mentors are experienced observers and their quotes map directly to program design decisions.
- Mixed responses: Likert-style agreement rates (e.g., ~80% Mission Zero mentors see skill gains; ~90% for Mission Space Lab) plus open-text comments.
- Key themes reported: confidence-building, space as motivational context, usefulness of step-by-step resources, and barriers (time/complexity/portal issues).
- Analytical needs: thematic coding, frequency counts, segment comparison (Mission Zero vs Mission Space Lab vs non-submitters), and searchable quotes for reporting.
Implications for researchers and program teams
For evaluation teams
Treat mentor surveys as a targeted qualitative sample: code for outcomes (confidence, motivation), barriers, and resource utility. Quantify theme prevalence before recommending redesigns (e.g., simplify onboarding materials if 'difficulty' appears in >20% of non-submitter responses).
Use cross-segment comparison: the report notes 12% more Mission Space Lab teams achieved flight status year-on-year, explore which support materials correlated with that gain.
For UX / learning designers
Mentor quotes point to friction in the portal and testing tools. Prioritize quick wins: clearer error messages, simplified step-by-step guides for novices, and in-tool examples.
Validate fixes with small rapid cycles: re-run targeted surveys and micro-interviews, then compare theme frequency pre/post change.
For researchers scaling programmatic insight
Small, high-quality qualitative datasets (mentor interviews + selective surveys) scale well with AI-assisted thematic methods: you can compress coding time, maintain audit trails, and produce deliverables (quotes, co-occurrence maps) for stakeholders faster.
How Evidano accelerates qualitative analysis of Astro Pi
Ingest & clean: unify surveys, focus-group transcripts, and forum posts
Import CSV survey exports and upload focus-group transcripts or audio. Evidano transcribes audio (custom dictionary for technical terms like 'Mission Zero' or 'ISS') and redacts PII on import.
Why it helps: saves hours of manual transcription and ensures consistent raw data for coding.
Automated thematic + frequency analysis
Use Evidano’s thematic extraction to surface dominant themes (confidence, motivation, barriers) and get frequency counts per segment (Mission Zero vs Mission Space Lab vs non-submitters).
Output: theme prevalence tables, representative quotes, and exportable codebooks.
Cross-segment comparison and visualization
Run cross-segment analysis to test hypotheses (e.g., are time constraints mentioned more by non-submitters?). Visual outputs include co-occurrence networks and hierarchical code trees that make patterns obvious to stakeholders.
Rapid synthesis and auditability
Evidano preserves traceability: every theme links back to source quotes and original documents so you can defend findings to funders or partners.
Security note: your data is encrypted and not used to train third-party models.
Follow-up data collection
If you need more depth, Evidano supports AI-avatar interviewers for autonomous qualitative data collection, useful for quick post-launch checks before the 8 Sep 2025 registration opening.
7-step checklist: from raw mentor feedback to Action Memo
Step 1: Gather inputs, export post-submission surveys and focus-group transcripts (CSV, text, audio).
Step 2: Upload to Evidano, run transcription, apply custom dictionary (e.g., 'Astro Pi', 'Mission Space Lab').
Step 3: Auto-extract themes, review and refine the suggested codebook; import any existing codes.
Step 4: Run cross-segment frequency analysis (Mission Zero vs Mission Space Lab vs non-submitters).
Step 5: Generate visualizations, co-occurrence network and hierarchical codes to surface root causes.
Step 6: Pull representative quotes and build a 1‑page Action Memo prioritizing fixes (onboarding, portal UX, extra scaffolding).
Step 7: Iterate, deploy small fixes, re-survey a subset, and compare theme prevalence to measure impact.
Wrapping up: next steps
The Raspberry Pi Foundation’s Astro Pi 2024–25 review is a clear example of how small, well-structured qualitative datasets can drive practical program improvements. If you want to reproduce this analysis, merge mentor surveys, focus-group transcripts, and platform logs, then produce a stakeholder-ready report; Evidano shortens the workflow from weeks to days.
- Ready to try this with your Astro Pi or education program data? Start a pilot at www.evidano.com and import your first survey or transcript.
- Original report: www.raspberrypi.org/blog/code-confidence-and-community-a-look-back-at-astro-pi-2024-25/, use it as a blueprint for coding and segment comparisons.
Keep reading
- Commentary on NewsTwo Definitions: Climate Change Acceptance for UndergradsHow a PLoS One Delphi study (Aug 25, 2026) defined climate change acceptance for undergraduate science students, and how AI-enabled qualitative analysis applies it.
- Commentary on NewsResearcher-in-the-loop: AI-enabled UX researchHow the researcher-in-the-loop model governs AI-enabled UX research. Learn practical governance, stats from the August 2026 piece, and how Evidano supports this workflow.
- Commentary on NewsResearcher-in-the-Loop: Governance for AI UX ResearchGovern AI in qualitative UX research with the researcher-in-the-loop model from Jennifer L. Bowie (Aug 25, 2026): practical rules, risks, and tool mappings.