Search across guide titles, topics, keywords, and content throughout the Researcher's Field Guide.
Showing 801–825 of 1,568 results
A useful deviation report tells readers what was preregistered, what actually happened, why the change occurred, and when the decision was made. The goal is to make differences easy to find and interpret rather than forcing readers to reconstruct them.
Read Guide →Between-subjects designs compare different participants across conditions, while within-subjects designs compare conditions within the same participants. The better choice depends on what is being studied and whether experiencing one condition can influence another.
Read Guide →Recruitment problems do not automatically mean that a study has failed. Diagnose why recruitment is falling short before changing the strategy, because some solutions can alter the sample, introduce bias, or change the study itself.
Read Guide →Falling short of a planned sample size does not automatically make a study worthless, but it may substantially change what the data can establish. The appropriate response is to assess the consequences of the smaller sample rather than retroactively making the original target disappear.
Read Guide →A control or comparison group can fail to function as intended even when recruitment and data collection continue normally. The key is to determine what actually happened in each group, why it happened, and whether the intended comparison still supports the study's research question.
Read Guide →The people you can recruit are not always the people you originally intended to study. Before broadening eligibility or changing recruitment settings, determine whether the accessible population can still provide evidence relevant to the original research question.
Read Guide →A method can be implemented perfectly and still be wrong for the question it is supposed to answer. When that mismatch becomes apparent, reassess what evidence the question requires before deciding whether to adapt the method, narrow the question, or redesign the study.
Read Guide →Not every methodological change creates a new study, but some changes alter the research question, evidence, population, comparison, or inferential logic so substantially that treating the project as unchanged becomes misleading. The boundary is conceptual, not a fixed percentage of procedures changed.
Read Guide →Not every change to a research sample creates a different study. The critical issue is whether the revised sample still provides evidence about essentially the same population and research question, or whether the change alters what the study is actually about.
Read Guide →Not every tiny departure from a research plan deserves equal space in a publication, but deviations that affect interpretation, participant protections, planned analyses, or confidence in the findings generally should be visible. The goal is to let readers understand the study that was actually conducted.
Read Guide →Every real study operates under constraints, so some methodological compromise is unavoidable. The critical question is whether the remaining design can still support the claims you intend to make after those compromises are considered together.
Read Guide →Case study research investigates a clearly bounded case in depth and in relation to its real-world context. Studying one person, classroom, institution, event, or site does not automatically make a project a case study because the design depends on how the case is defined and investigated.
Read Guide →Mixed methods research requires more than collecting qualitative and quantitative data in the same project. The defining issue is whether the components are deliberately related and integrated to produce an understanding that neither component would provide independently.
Read Guide →A study does not become mixed methods simply because its dataset contains both numbers and words. What matters is whether substantive qualitative and quantitative components are intentionally designed, analyzed, and integrated to address the research problem.
Read Guide →Not every research question needs two groups, two treatments, or a before-and-after comparison. The comparison should follow from what you want to know, not from the assumption that stronger research always compares something.
Read Guide →Correlation, association, prediction, and causation answer different research questions. Learn how to match the claim you want to make with the evidence your study is actually designed to provide.
Read Guide →A causal question asks what would have happened under an alternative condition. That unobserved alternative is the counterfactual, and constructing a credible substitute for it is central to causal research.
Read Guide →Observational data do not automatically restrict researchers to non-causal questions. Causal inference may be possible when the causal contrast, design, assumptions, timing, confounding strategy, and analysis are explicitly aligned.
Read Guide →A defensible sample begins with a clearly defined population. Learn how to move from your research question to the people or units you actually recruit without claiming more than your design can support.
Read Guide →Your data usually come from a sample, but your research question may concern a much larger population. Understanding that distinction is essential for choosing participants and interpreting what your findings actually mean.
Read Guide →A sampling frame is the operational source from which a sample is selected, but it may not perfectly match the population you want to study. Learn how frame errors can quietly change who has a chance to enter your sample.
Read Guide →More participants can improve statistical precision and power, but sample size is only one part of research quality. A large sample cannot automatically repair biased recruitment, poor measurement, confounding, or a weak design.
Read Guide →A representative sample is meaningful only in relation to a specified population. Learn what representativeness actually requires, how researchers assess it, and why not every worthwhile study needs a statistically representative sample.
Read Guide →A large sample can reduce random uncertainty while leaving systematic bias almost untouched. Learn why thousands or even millions of observations can produce extremely precise answers to the wrong population question.
Read Guide →There is no universal number of participants that makes a quantitative study adequately powered. Learn what determines sample size, how statistical power fits into the calculation, and why the planned analysis should come before the final number.
Read Guide →