Quick answer: Data collection is pending. We have no published study participants, verified quote results or original failure-rate estimates. This framework defines what must exist before results are released.
What can be used today?
The browser-local three-quote planner helps individual homeowners compare scope without transmitting documents. A blank, non-identifying pilot worksheet is also available for private use. Neither generates a research submission or provides evidence of a national price distribution.
Planned research questions
- Standby Generator Quote Study — protocol planning; no results published.
- Generator Ownership Cost Study — protocol planning; no results published.
- Year-3 Reliability Survey — protocol planning; no results published.
- Generator Warranty Experience Study — protocol planning; no results published.
- Installation Scope Study — protocol planning; no results published.
- Maintenance Spend Study — protocol planning; no results published.
- Outage Experience Study — protocol planning; no results published.
The protocol each study must publish
- Research question, version and dates; eligibility, recruitment channels, inclusion and exclusion rules.
- Questionnaire, definitions, observation period and denominators. A year-three study needs ownership duration and exposure, not just complaints.
- Consent, private storage, redaction, retention, withdrawal and deletion process before collection starts.
- Household/project deduplication, linked follow-ups and quote-versus-invoice separation.
- Missing responses, sponsorship or compensation, sampling biases and limits on generalization.
- Analysis plan, sample-size rationale, publication threshold and independent editorial review.
Publication gates
For quote cohorts, fewer than 20 independent verified projects means no distribution is published. This editorial floor does not make larger samples representative. Reliability claims need a separate precision and sampling plan; the quote threshold cannot be reused as proof of a failure rate. No significance claims or national averages will be inferred from self-selected participants.
Future table and chart requirements
Every result must include the question, unit, sample size, denominator, period, geography, configuration mix, missing-data count, inclusion rules and uncertainty. Quote charts will separate proposals from invoices and show median, quartiles and observed range. No placeholder bars or synthetic participants appear as findings.
Privacy and corrections
No survey or upload is open. Research participation will be separate from browser-local tools. Participants will need a documented withdrawal route and retention terms before submitting. Public exports must exclude original documents and private identifiers; small geographic groups require suppression.
See the privacy policy, corrections policy, verified quote project, weakness evidence method and serviceability study method.