Designing a Study You Can Defend
A calculated sample size can exceed the number of people eligible for a study. Learn when finite population correction may be appropriate, why a census does not automatically solve statistical power problems, and how to revise a study responsibly when recruitment cannot meet the original requirement.
Read Guide →
Designing a Study You Can Defend
Recruiting more participants than the calculated sample may protect a study against incomplete responses, unusable measurements, and anticipated exclusions. Learn how to estimate an appropriate allowance, distinguish missing data from genuinely unusable records, and avoid unnecessary recruitment that adds participant burden without improving the analysis.
Read Guide →
Designing a Study You Can Defend
Recruiting more participants may increase the information available for analysis, but it also creates obligations and potential burdens. Learn when additional enrollment becomes ethically difficult to justify, how to balance scientific value with participant welfare, and why an approved maximum is not necessarily a target that must be reached.
Read Guide →
Designing a Study You Can Defend
Research volunteers may be unusually interested, available, confident, or personally invested in the topic being studied. Learn how to recognize motivation-related selection bias, recruit beyond enthusiastic volunteers, and assess what your sample can reasonably represent.
Read Guide →
Designing a Study You Can Defend
Gatekeepers can help researchers access participants, but their decisions may determine who receives an invitation and whose experiences remain unheard. Learn how to define gatekeeper responsibilities, prevent discretionary selection, protect consent, and document recruitment decisions.
Read Guide →
Designing a Study You Can Defend
Researchers do not necessarily need to reveal their exact hypothesis in recruitment advertisements, but they must provide accurate information about the study. Learn when hypothesis disclosure may influence participation or behavior, what informed consent requires, and when limited disclosure needs ethical review.
Read Guide →
Designing a Study You Can Defend
Study advertisements can shape what potential participants expect before they enroll, potentially influencing who volunteers and how they respond. Learn to distinguish recruitment selection from expectancy effects and design accurate, ethically appropriate advertisements that minimize avoidable bias.
Read Guide →
Designing a Study You Can Defend
Participant dropout can reflect study burden, changing circumstances, accessibility barriers, dissatisfaction, or difficulties maintaining follow-up. Learn how to identify plausible dropout risks, distinguish participant characteristics from study-design problems, and develop appropriate prevention and monitoring strategies.
Read Guide →
Designing a Study You Can Defend
Participant retention should be planned before recruitment rather than improvised after participants begin missing assessments. Learn what a research protocol should specify about follow-up, communication, accessibility, withdrawal, retention monitoring, and the prevention of missing outcome data.
Read Guide →
Designing a Study You Can Defend
There is no universal number of reminders or follow-up contacts that suits every research study. Learn how to establish a proportionate contact schedule based on assessment timing, participant preferences, response windows, privacy, and the need to maintain scientifically useful follow-up.
Read Guide →
Designing a Study You Can Defend
Repeated reminders can help participants complete follow-up assessments, but excessive contact may create inconvenience, privacy concerns, or perceived pressure. Learn how to assess cumulative contact burden, establish reasonable boundaries, and maintain research follow-up without compromising voluntary participation.
Read Guide →
Designing a Study You Can Defend
Expected attrition can make a longitudinal study infeasible when recruitment demands, missing outcome data, costs, or selection bias prevent the research question from being answered credibly. Learn how to evaluate these constraints, calculate realistic recruitment requirements, and decide when a study needs redesign rather than a larger initial sample.
Read Guide →
Designing a Study You Can Defend
The best data collection method is not simply the one you know best or can administer most easily. Learn how to work from your research question to the evidence, source, method, and practical design that can actually answer it.
Read Guide →
Designing a Study You Can Defend
Data collection methods make different trade-offs between depth, breadth, standardization, and flexibility. Understanding those trade-offs can help you choose a method that fits what your research question actually needs.
Read Guide →
Designing a Study You Can Defend
Abstract constructs such as motivation, trust, engagement, and anxiety cannot simply be placed into a dataset. Learn how researchers move from a theoretical construct to observable indicators and defensible measurements.
Read Guide →
Designing a Study You Can Defend
A conceptual definition tells readers what a construct means; an operational definition specifies how it will be represented in your study. Keeping the two aligned is essential because what you measure determines what your eventual findings can legitimately mean.
Read Guide →
Designing a Study You Can Defend
Measure, instrument, and indicator are often used as though they mean the same thing, but they describe different parts of the measurement process. Understanding the distinction can make your methods clearer and expose weak links between a construct and the data used to represent it.
Read Guide →
Designing a Study You Can Defend
What people report, what researchers observe, and what systems record are not interchangeable forms of evidence. Each measurement approach can answer different questions and introduce different sources of error.
Read Guide →
Designing a Study You Can Defend
Nominal, ordinal, interval, and ratio remain useful for understanding what values mean, but they should not be treated as a mechanical statistical decision tree. Learn what each level permits and where the familiar framework becomes less tidy.
Read Guide →
Designing a Study You Can Defend
A conceptual framework may include more constructs than a study needs to measure. Learn how to distinguish theoretical context from empirical commitments, determine which relationships require data, and avoid claiming support for pathways that were never examined.
Read Guide →
Designing a Study You Can Defend
Access to more variables does not automatically produce better research. Learn when additional measurements contribute scientific value, when they create unnecessary burden or analytical risks, and how to establish a defensible variable collection plan.
Read Guide →
Designing a Study You Can Defend
A research study needs the variables required to answer its primary question, but identifying them involves more than selecting an independent and dependent variable. Learn how to determine the minimum defensible measurement set, account for study design and confounding, and verify that the planned analysis is possible.
Read Guide →
Designing a Study You Can Defend
Describing a research sample requires more than reporting participants' age and gender. Learn how to select demographic, contextual, and baseline variables that clarify who participated, support interpretation, and help readers assess the applicability of findings without collecting unnecessary personal information.
Read Guide →
Designing a Study You Can Defend
Confounding control requires more than adding demographic characteristics to a regression model. Learn how to identify relevant confounders using causal assumptions, distinguish confounders from mediators and colliders, and determine which variables must be measured to support a defensible causal estimate.
Read Guide →
Designing a Study You Can Defend
A study can collect data successfully yet remain unable to answer its primary research question if essential variables were never measured. Learn how to recognize consequential omissions, distinguish missing measurements from missing responses, and evaluate whether an incomplete dataset can support a defensible analysis.
Read Guide →