Teaching notes and reuse #
These notes support the workshop and solo route. Read the student packet and worked analysis for your chosen case before leading a discussion. The goal is a defensible decision and a useful test, with uncertainty visible.
The package is a proposed teaching resource. No independent teaching trial or prospective validation study of this package is claimed. The optional record below is a template for future use, not a report of completed delivery.
Source boundaries #
| Material | What it supplies | What it does not supply |
|---|---|---|
| M-PESA packet | Public participant accounts of an early service pilot and its delivery constraints | Complete transaction records, an unbiased comparison population, or a controlled test of Behavioral Strategy |
| Instagram packet | Founders’ retrospective explanation of an early product focus decision | Raw feature-level data, a causal estimate for removing features, or verified framework use |
| Verify packet | A public audit of identity verification, service dependencies, costs, and targets | A randomized comparison or proof that one psychological constraint explains all noncompletion |
| Measurement lab | Invented observations for arithmetic practice, plus two separate research examples | A real participant dataset, an effect estimate for the pack, or a universal success threshold |
Evidence IDs BS-0079 through BS-0083 lead to the evidence ledger. The individual pages link their primary sources and identify retrospective material. The printable guide and slide notes retain the relevant references. Learners should distinguish a reported event from the instructor’s interpretation and proposed action.
Discussion rubric #
This is an informal review aid, not a validated assessment scale. Review each row as clear, needs revision, or unresolved because evidence is missing. Do not add the categories into a numerical fit score or use a cutoff to award certification.
| Element | Look for | Ask for a revision when |
|---|---|---|
| Outcome and action | A named population, observable action, context, and connection to the outcome | The answer gives only a feature, aspiration, or business metric |
| Alternatives | Plausible competing behaviors, including less user work where feasible | The comparison restates the favorite option in different words |
| Actor chain | Recipient, provider, operator, and other dependent actions where relevant | The plan counts only the first user’s visible action |
| Evidence | Sources and observations separated from inference and unknowns | Interest, selection, or hindsight is treated as causal proof |
| Test and measurement | Eligible opportunities, repetitions, support costs, observation window, and prespecified rules | Someone else could count a different result from the same records |
| Next decision | A proportionate action with a stated reversal condition | Expansion follows automatically from activity or a high informal rating |
A missing fact is not automatically a learner error. Credit a response that names the gap, limits its claim, and proposes a proportionate way to resolve it. Challenge a response that silently fills the gap with a favorable assumption.
Common discussion errors #
A case outcome cannot establish why it happened. Ask what the source observed and which causal alternatives remain. A familiar behavior may still require new skills, trust, incentives, or operational support in a new setting.
Reducing user effort can be a good candidate, but it does not make work disappear. Ask who now performs the task, at what cost, and under which service conditions. A concierge test may help discover a viable service; its staffing and exception handling belong in the record.
A null result limits a claim about the tested intervention, population, and outcome. It does not prove that the underlying need is absent or that every possible intervention will fail. Conversely, an intermediate gain does not establish the final outcome.
Host observation record #
Use the editable text record after a session if useful. Keep observations at group level and omit identifying or confidential material. Participation in teaching does not by itself authorize research use or public quotation of learners’ work.
Record the date, pack version, format, group size, prior familiarity in broad terms, actual reading and activity times, and any accessibility or language adaptation. Note which instructions required unplanned explanation. For each rubric row, summarize whether the group could complete it without new help and give an anonymous example of the difficulty.
Record whether the participants changed a decision after the comparison and why. Separate an observed change in a written response from a claimed improvement in real-world outcomes. Note facilitator experience and deviations from the agenda. A future evaluation would need a defined outcome, comparison, consent process where applicable, and a sampling plan before drawing broader conclusions.
There is no collection endpoint or requirement to send this record to the site. Use the corrections page for a specific source or content correction.
Reuse and adaptation #
Original teaching content in teaching pack v1.0 is licensed under CC BY 4.0. This includes this page, the new decision packets and worked analyses linked from the Education hub, the decision workbook, measurement lab and answers, workshop, solo route, and the original text and data in /downloads/teaching-pack/v1/.
You may share, teach, translate, and adapt that original material, including commercially, with attribution and an indication of changes. Outside publications, quotations, trademarks, fonts, and any third-party template elements retain their own rights; linking a source does not relicense it. The two existing methodology guides and older case notes retain their existing site license unless they carry another asset-level notice. See the license map.
Suggested attribution: “Adapted from Jason Hreha, BehavioralStrategy.com, Public teaching pack v1.0, 2026, CC BY 4.0, https://behavioralstrategy.com/education/. Changes: [describe your changes].” For an unchanged copy, replace “Adapted from” with “By” and omit the change description.
Keep the source notes, fictional-data labels, evidence limits, and version with an adaptation. Update the decision, actor chain, denominators, and worked analysis when changing the setting. Remove organization-specific facts that no longer apply. Use the names Behavioral Strategy and DRIVE accurately; the license does not imply endorsement, certification, or affiliation.
The manifest lists the pack files and their checksums. Version 1.0 was prepared on 5 September 2026. Later revisions should carry their own version and change note.