Ask when the decision was made
When reading research, distinguish decisions made before examining relevant outcomes from choices made after seeing them. A preregistration records specified aspects of a planned study before the stated analysis stage. Its value depends on what was actually specified and when. A label alone does not establish quality. Read the registered document, its date and its relationship to the reported work rather than treating the word preregistered as a guarantee.
Look for actionable commitments
A useful plan can specify questions, outcomes, exclusions and analysis choices in enough detail for a reader to understand the intended work. The appropriate detail varies by method. Do not assume that every project uses the same registration template or that all valuable inquiry begins with a fixed hypothesis. The central reading question is whether the reported claims accurately describe which decisions were planned and which emerged later.
Treat deviations transparently
Researchers may need to change a plan because an instrument fails, data are unavailable or an assumption proves unsuitable. The change itself does not settle whether the study is credible. Look for an explanation, timing and a clear separation between the original plan and the revised analysis. Undisclosed flexibility is different from a documented adaptation. Read the consequences for interpretation instead of counting every deviation as an identical fault.
Keep exploration visible
An unexpected pattern can be a valuable starting point for further inquiry. The problem is presenting a pattern noticed after inspection as though it were the original prediction. Identify whether a conclusion is confirmatory, exploratory or otherwise qualified by the authors. Those labels should match the documented process. A transparent exploratory finding can inform the next study without being promoted into stronger evidence than the design supports.
Read the plan beside the paper
For a practical reading exercise, choose one paper with an accessible registration linked by its authors. Compare the stated question, main outcome and one analysis choice. Note a match, a documented change and anything you cannot determine from the available record. Avoid speculating about motives. This exercise evaluates traceability of decisions; it does not replace assessment of the study’s measurement, sampling, analysis or substantive argument.
