Library / Research Method Practice Principles Framework
Jump to passage
In this reading

Link to current text

Published source confirmed at last check

Source changed 2026-10-03 02:22:15 UTC · snapshot created 2026-10-03 03:38:22 UTC · last check 2026-10-03 03:55:20 UTC

RMP.Preface:12 - Conditions for a usable combined result

Use one receiving question, intended episteme change, and useful stop across the combination. The source frame, rival accounts, applicability envelope, and supplied owner results must be compatible with that same use. Two locally plausible results cannot be combined when, for example, the brief concerns field seal degradation but the design comparison quietly assumes a different material or temperature mechanism. Requalify the changed question or obtain the missing relation before treating the design as its answer.

Feasibility concerns the whole selected strategy. For a mixed design, RMP.2’s family comparison requires a reason for each component and a warranted relation among its results. Check shared participant burden, access windows, specialist capacity, resources, and incompatibilities over the same intended period. Define what a component’s failure or disagreement would do to the claim. An extra component earns its place by the uncertainty it can change; agreement produced by shared data or assumptions is not independent corroboration merely because two Methods were named.

Consider a constructed mixed inquiry into the stream anomaly. The same qualified investigator pool has twelve available investigator-days in the relevant window. Field comparison requires six, archival/source comparison five, and a computational sensitivity component four; assume these demands cannot overlap or share effort. Every pair fits, but all three need fifteen days. If all three are necessary for the proposed criticism, the whole option is infeasible with the supplied twelve-day capacity. Compare a feasible worthwhile arrangement or finish the current allocation with no new study and the evidence limitation retained. Additional capacity is an option to appraise, not an automatic acquisition. Removing one component changes the design whenever it removes a necessary challenge. Lack of capacity does not establish that the current evidence is sufficient.

If a capacity result remains necessary to select a useful strategy and its plausible values would change that choice, RMP.2 returns ResearchDesignHold. When the current allocation instead finishes with no new study, retain the unresolved capacity as a limitation rather than an awaited acquisition. When several owner results remain unresolved, apply RMP.2’s joint-input test in RMP.2:4.3 before treating the remaining needs as PostSelectionNeed. The RMP.2 result rule supplies the required record and the action when a value arrives outside the admitted range.

These conditions first qualify the selected design and then constrain its operationalization. RMP.3 applies the same joint-input condition to the proposed conduct; RMP.4 preserves the actual choices and changes. Neither a completed protocol nor an extensive trace rescues a lost comparison or unsupported measurement relation. Preserve the completed question/design and unaffected material, identify the failed dependency, and return it to its owner. Competent execution and actual analysis require their own Methods and results. RMP.5 and RMP.6 can use the supported material to return a bounded claim and credibility decision; RMP.7/RMP.8 supply synthesis and broader revision when needed; RMP.9 addresses a distinct Method difficulty and returns only its actual consequences.

Constituent actions in ongoing work. During a study, classifying one observation can constitute part of constructing the analytical comparison, within an ongoing inquiry into a stated explanation. If the inquiry changes from describing recorded cases to explaining their occurrence, a classification that ignores how cases were selected may no longer support it. Recording skill and statistical calculation can both be available while the intermediate operationalization is inadequate. RMP.3 addresses that operationalization; the design and analysis Methods retain their own questions. Recover the vertical with B.1.5.EW and account for the access, specialist capability and shared observation burden needed by the combined work. Earlier collection is not assumed to continue during every later calculation.