Designing a Study You Can Defend
Integrating quantitative and qualitative evidence during analysis can reveal relationships that separate analyses may overlook. Learn when analytic integration is appropriate, which techniques to use, and how to interpret convergence, complementarity, and disagreement without overstating the findings.
Read Guide →
Designing a Study You Can Defend
Mixed-methods integration can occur primarily during interpretation, but a discussion of separate findings is not automatically meaningful integration. Learn when this approach is defensible, how to construct integrated conclusions, and what evidence is needed.
Read Guide →
Designing a Study You Can Defend
A joint display brings quantitative and qualitative evidence into a structured relationship to support mixed-methods integration. Learn how joint displays differ from ordinary results tables, what they should contain, and why planning their structure before analysis can prevent important integration problems.
Read Guide →
Designing a Study You Can Defend
A mixed-methods study can begin with some uncertainty about its eventual integration procedures, but proceeding without a clear integration purpose creates methodological risks. Learn what minimum planning is needed, when flexibility is defensible, and how to recognize whether a study is genuinely mixed methods.
Read Guide →
Designing a Study You Can Defend
Pilot and feasibility studies both reduce uncertainty before a larger study, but the terms are not simply interchangeable. Feasibility is the broader question of whether and how a future study can be done, while a pilot study typically tests part or all of the intended study on a smaller scale.
Read Guide →
Designing a Study You Can Defend
A pretest usually examines a specific instrument or procedure, a feasibility study investigates whether and how a future study can be done, and a pilot tests part or all of the intended future study on a smaller scale. The boundaries can vary across disciplines, so purpose matters more than the label alone.
Read Guide →
Designing a Study You Can Defend
You can test much more than a questionnaire before the full study begins. Recruitment, consent, participant procedures, intervention delivery, measurements, follow-up, data management, staff workflows, and planned analysis processes may all warrant testing when they contain meaningful uncertainty.
Read Guide →
Designing a Study You Can Defend
A data collection procedure is practical when researchers and participants can complete it consistently under realistic study conditions without unacceptable burden, missing data, resource demands, or operational problems. Testing the full workflow can reveal problems that an instrument alone cannot.
Read Guide →
Designing a Study You Can Defend
A pilot showing that the original design will not work is not necessarily a failed pilot. The finding may prevent a much larger failure by identifying what needs revision, further testing, or reconsideration before the main study begins.
Read Guide →
Designing a Study You Can Defend
There is no universal number or percentage of changes that turns a revised study into a different study. What matters is whether the modifications preserve the core research question and inferential structure or materially change what is being studied, in whom, how, and for what conclusion.
Read Guide →
Designing a Study You Can Defend
A pilot study may be preferable when researchers need to rehearse a smaller-scale version of a proposed definitive study. Learn how pilot studies relate to feasibility research, when broader feasibility assessment is sufficient, and how to justify the preliminary design without treating pilot findings as evidence of effectiveness.
Read Guide →
Designing a Study You Can Defend
Pilot studies can provide valuable evidence about research feasibility, but their findings are frequently overinterpreted. Learn how to distinguish actionable feasibility results from unsupported effectiveness, safety, generalizability, and measurement claims.
Read Guide →
Designing a Study You Can Defend
Not every research procedure needs equally extensive pilot testing. Learn how to identify the procedures most likely to threaten a definitive study, prioritize them systematically, and design a focused pilot that produces actionable feasibility evidence.
Read Guide →
Designing a Study You Can Defend
Pilot testing can reveal whether qualitative interview questions are understandable, appropriately sequenced, and capable of generating relevant accounts. Learn when pretesting is worthwhile, what to examine, and how to revise an interview guide without turning qualitative inquiry into a standardized questionnaire.
Read Guide →
Designing a Study You Can Defend
Data management failures can compromise a study even when recruitment and data collection succeed. Learn when to pilot test data entry, coding, validation, storage, linkage, and export procedures, and how to verify that the resulting data remain accurate, secure, and suitable for analysis.
Read Guide →
Designing a Study You Can Defend
An intervention may be theoretically promising but difficult to deliver under actual research conditions. Learn when intervention delivery should be piloted, what implementation outcomes to assess, and how to distinguish feasible delivery from evidence of effectiveness.
Read Guide →
Designing a Study You Can Defend
Even a well-designed study can produce weak evidence if its outcomes cannot be measured consistently and appropriately. Learn when outcome assessment procedures should be pilot-tested, which feasibility indicators matter, and why successful administration does not automatically establish measurement validity.
Read Guide →
Designing a Study You Can Defend
An analysis pipeline may execute successfully while producing incorrect results. Learn how to test data cleaning, coding, transformations, statistical models, and outputs using synthetic or pilot data, and why technical verification does not establish methodological validity.
Read Guide →
Designing a Study You Can Defend
Not every research project needs a formal pilot or feasibility study. Preliminary work is most valuable when important uncertainties about whether or how the main study can be conducted cannot be resolved adequately from existing evidence or simpler testing.
Read Guide →
Designing a Study You Can Defend
A separate feasibility study may be justified when important uncertainties cannot be resolved through routine planning or limited pretesting. Learn how to determine when independent feasibility research is necessary, what it should establish, and when proceeding directly to the main study may be defensible.
Read Guide →
Designing a Study You Can Defend
Before committing to a definitive study, researchers should establish whether its essential procedures can produce meaningful evidence. Learn which feasibility questions matter, how to select appropriate outcomes, and how to translate preliminary findings into defensible progression decisions.
Read Guide →
Designing a Study You Can Defend
Recruitment rate can provide essential evidence about whether a definitive study can enroll enough participants within its available time and resources. Learn how to define, calculate, interpret, and use recruitment outcomes without confusing enrollment speed, eligibility yield, and participation proportions.
Read Guide →
Designing a Study You Can Defend
Participant retention can determine whether a definitive study produces sufficiently complete follow-up evidence. Learn when retention should be a feasibility outcome, how to distinguish it from attrition and data completeness, and how to interpret preliminary retention findings without assuming they will generalize to the main study.
Read Guide →
Designing a Study You Can Defend
Intervention adherence can reveal whether participants are able and willing to complete the activities required by a proposed intervention. Learn when adherence should be a feasibility outcome, how to measure it meaningfully, and why high adherence does not establish intervention effectiveness.
Read Guide →
Designing a Study You Can Defend
A study can recruit and retain participants yet still lack the information required for analysis. Learn when data completeness should be a feasibility outcome, how to define and calculate it, and why complete records do not automatically establish measurement validity or freedom from bias.
Read Guide →
