Sprint 1 planning: research framing and data foundation (Measure phase)¶
How to use: the Scrum Master fills this in during Sprint 1 planning on Thu 2026-09-17. Candidate items become GitHub issues in milestone Sprint 1; record the owner and issue number in the table. Rubric weight: 20%.
Dates: Thu 2026-09-17 to Wed 2026-10-07. Review and retro: Wed 2026-10-07. Check-ins: Thu 2026-09-24, Thu 2026-10-01. Product Owner: Jerad Dunne. Scrum Master: Leticia Aderhold. Developers/Researchers: all four.
Sprint goal¶
Sprint goal:
By Wed 2026-10-07 the P-01 question is fully operationalized and defensible:
DATA_DICTIONARY.mdanddata/README.mdlet a non-author reproduce the sample, the threat model and FMEA inTHREATS_TO_VALIDITY.mdcarry RPN scores, baseline process metrics are captured, and round 1 signal labelling is complete with a written reason on every judgement call.
Capacity¶
| Member | Available hours this sprint | Planned points | Notes (exams, travel, other courses) |
|---|---|---|---|
| Jerad Dunne | fill in | 9 | Owns all 5 open Sprint 1 items. Also carries ~10-11 h of round 1 labelling (147 signal items, 58 lineage pairs, 18 drift pairs) that is not represented as a backlog item yet |
| Leticia Aderhold | unconfirmed | 0 | No response to PR #49 review request or onboarding issue #50 as of 2026-09-17 |
| Allie Hodges | unconfirmed | 0 | No response to PR #49 review request or onboarding issue #51 as of 2026-09-17 |
| Hina Kramer | unconfirmed | 0 | Repository invitation not yet accepted; onboarding issue #52 unstarted |
| Total | 9 | Velocity baseline for Sprint 2. Treat 9 as a one-person figure, not a team velocity |
Point scale: 1, 2, 3, 5, 8. WIP limit: 5 on the board, at most 2 In progress per person. Size the first week for onboarding: new members pair on setup before taking larger items.
Candidate backlog (rubric deliverables for Sprint 1)¶
| # | Item (rubric deliverable) | Acceptance criteria | Estimate | Owner | Issue # |
|---|---|---|---|---|---|
| S1-1 | Selected MSR-inspired topic and precise research question in RESEARCH_QUESTION.md |
Question is one sentence (P-01 section 2), names the dataset, unit of analysis, and outcome; approved by a member other than the author | 3 | Jerad Dunne | #12 |
| S1-2 | Motivation, expected contribution, and at least one competing or simpler explanation | Section in RESEARCH_QUESTION.md and report/draft.md Introduction; competing explanation is testable |
(in #12) | Jerad Dunne | #12 |
| S1-3 | Defined unit of analysis, population, sample, variables, and outcome measures | Variables table in RESEARCH_QUESTION.md with type, source column, and derivation |
3 | Jerad Dunne | #13 |
| S1-4 | Dataset acquisition or sampling instructions and data dictionary | data/README.md has the exact snapshot and steps; inclusion rules implemented in code; DATA_DICTIONARY.md lists every column used; verified by a second member |
3 | Jerad Dunne | #14 |
| S1-5 | First working data-loading and extraction pipeline | make pipeline runs python -m msr_pipeline analyze on the sample from a fresh clone; loader and scanner have tests in tests/ |
5 | Jerad Dunne | #15 |
| S1-6 | At least one exploratory table or visualization generated by code | File in results/ or figures/ produced by the pipeline, referenced from report/draft.md |
2 | Jerad Dunne | #16 |
| S1-7 | Repository with issues, milestones, branch protection or equivalent review practice, and a first pull request | Milestones exist with due dates; main requires 1 approving review; every member has authored and reviewed at least one merged PR |
1 | Jerad Dunne | #17 |
| S1-8 | Retrospective explaining what the group learned and what changed in the backlog | docs/sprints/sprint-1/retrospective.md completed on 2026-10-07 |
2 | Jerad Dunne | #20 |
| S1-9 | Primary risks and threats to validity identified | THREATS_TO_VALIDITY.md status column updated and owners assigned; threat model written; FMEA has RPN scores |
2 | Jerad Dunne | #18 |
| S1-10 | Measurement system check (Lean Six Sigma) | Two members independently run the pipeline on different machines and compare row counts and one summary statistic; agreement recorded in docs/lean-six-sigma/metrics/ |
1 | Jerad Dunne | #19 |
| S1-11 | Updated report draft: Abstract placeholder, Introduction, Research question, Dataset sections | Sections in report/draft.md are prose, not bullets |
2 | Jerad Dunne | no issue yet |
| S1-12 | Every member onboarded | Onboarding issue closed, profile merged, charter section 14 signed, make test green on each member's machine |
Each member | ||
| S1-13 | Decision: sample only, or one streamed full-dataset pass for RQ3 | ADR merged by Thu 2026-10-01 | |||
| S1-14 | Second raters ready for validation | Each second rater has run annotation_kit.py sheet and ui for their kind and labelled the practice items |
Second raters |
Risks pulled from the FMEA register¶
| Risk | RPN | Mitigation this sprint | Owner |
|---|---|---|---|
| Full dataset too large for a laptop (GitSkills 41 GB, R01) | 126 | Work on the official sample; decide any full-dataset pass by ADR (S1-13) | Product Owner |
| Research question not answerable with sample columns | Column-level feasibility check in week 1 | ||
| Uneven contribution (R06) | 147 | Work sign-up rules; every member owns at least one S1 item and one review; contribution shares read at each check-in | Scrum Master |
| Members onboard late in formation (R16) | 108 | Onboarding path in docs/GETTING_STARTED.md; pairing during setup; S1-12 |
Scrum Master |
Measure phase exit criteria (DMAIC)¶
- Every variable in the research question has a documented source column and derivation (operational definitions).
- Baseline data is captured: row counts, missing-value rates, and date range of the sample are in
results/. - Measurement system is checked: two independent runs agree (S1-10).
- Process KPIs have a first baseline in
docs/lean-six-sigma/metrics/DASHBOARD.md(cycle time, PR first-review turnaround, CI pass rate, rework count, contribution balance). - Data collection plan for Sprint 2 (tables, validation sample, second raters, full-dataset decision) is written.
Planning meeting notes¶
- Date: 2026-09-17. Facilitator: Leticia Aderhold.
- Attendees: [ ] Leticia Aderhold [ ] Jerad Dunne [ ] Allie Hodges [ ] Hina Kramer
- Items pulled into the sprint (issue numbers):
- Items explicitly deferred and why:
- Assumptions:
Items this sprint cannot close alone¶
Recorded at planning on 2026-09-17 so the Sprint 1 review and retrospective can point at it rather than reconstruct it later.
| Item | What blocks it | Earliest unblock |
|---|---|---|
| S1-7 (#17) | Acceptance says every member has authored and reviewed a merged pull request. Three members have authored none, and PR #49 merged under an admin bypass with zero reviews | A teammate opens and reviews a PR |
| S1-10 (#19) | A measurement system check needs two people running the pipeline on different machines | A teammate completes onboarding (#50, #51, #52) |
| S1-12 | Onboarding is per member and cannot be done on anyone's behalf | Each member acts |
| Second-rater labelling | Inter-rater kappa is the primary reliability measure and needs an independent human. Kinds were to be assigned at the 2026-09-16 kickoff (#53); none are assigned | A teammate takes a kind, or the instructor reassigns |
The blindness rule makes the last row the sharpest: the primary rater's filled label files for a kind stay uncommitted until the second rater's file for that kind is merged, so an absent second rater blocks the primary rater's work from entering the repository as well as their own.