Designing a Study You Can Defend
Inclusive research design can strengthen scientific validity when it addresses exclusions that would otherwise distort estimates, omit important experiences, or limit the applicability of findings. Learn when inclusion improves evidence, when it introduces methodological trade-offs, and how to evaluate its contribution to validity.
Read Guide →
Designing a Study You Can Defend
Inclusive recruitment can improve the relevance of research evidence, but broader participation may require additional safeguards, accommodations, and resources. Learn how to distinguish necessary restrictions from avoidable barriers while protecting participants and maintaining a feasible, scientifically defensible study.
Read Guide →
Designing a Study You Can Defend
Hybrid effectiveness-implementation designs combine questions about intervention effectiveness and implementation within one coordinated study. Understand the three hybrid types, how they differ from pragmatic trials and mixed-methods research, and how to specify appropriate outcomes, comparisons, and analytical approaches.
Read Guide →
Designing a Study You Can Defend
Using interviews and observations, or several other data collection techniques, does not automatically make a study mixed methods. What matters is the methodological nature of the evidence and how the different components are designed, analyzed, and related.
Read Guide →
Designing a Study You Can Defend
Data triangulation changes where or from whom the evidence comes; methodological triangulation changes how the phenomenon is investigated. A study can use either one, both, or neither.
Read Guide →
Designing a Study You Can Defend
More sources, methods, or measures can strengthen a study when they address different limitations or provide genuinely useful evidence. Simply adding more, however, does not make weak evidence strong.
Read Guide →
Designing a Study You Can Defend
Unplanned research changes should leave a reconstructable trail rather than disappearing into the final methods section. Record what was planned, what changed, why and when it changed, who made the decision, what information was available, and what consequences followed.
Read Guide →
Designing a Study You Can Defend
You can sometimes change a research design after data collection has started, but the timing, reason, consequences, and documentation of the change matter. The key is to adapt transparently without quietly redefining the study around what the emerging data appear to show.
Read Guide →
Designing a Study You Can Defend
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 →
Designing a Study You Can Defend
The strongest research design is not necessarily the most complex one. Choose the simplest design that can answer the research question credibly, and add complexity only when it produces information or protection against bias that matters to the intended inference.
Read Guide →
Doing Research Without Compromising Ethics
Research ethics and research integrity overlap, but they are not interchangeable. Ethics asks whether research is ethically defensible, while integrity concerns whether research is conducted and represented in ways that uphold the standards and values of trustworthy research.
Read Guide →
Doing Research Without Compromising Ethics
A study can be scientifically rigorous yet ethically unacceptable. Strong methodology establishes that research may answer its question, not that researchers are justified in using any means necessary to obtain the answer.
Read Guide →
Doing Research Without Compromising Ethics
Research accountability is usually distributed rather than concentrated in one person. Investigators, institutions, sponsors, sites, funders, and other parties can each remain accountable for different aspects of the same project.
Read Guide →
Doing Research Without Compromising Ethics
Research institutions have responsibilities of their own, including creating appropriate governance systems, supporting research integrity, providing suitable infrastructure, clarifying accountability, and responding when problems arise. These duties do not disappear simply because a principal investigator leads the project.
Read Guide →
Doing Research Without Compromising Ethics
Informed consent can sometimes be waived or altered, but only under specific ethical and regulatory conditions. Researchers generally cannot skip consent simply because a study is minimal risk, retrospective, difficult to recruit for, or scientifically valuable.
Read Guide →
Doing Research Without Compromising Ethics
Protection becomes unfair exclusion when researchers remove a population more broadly than the actual risk requires, particularly when reasonable safeguards could permit participation and the resulting evidence is intended to apply to the people being left out.
Read Guide →
Doing Research Without Compromising Ethics
Convenience sampling is usually discussed as a methodological limitation, but it can also raise ethical questions. The problem becomes ethical when convenience shapes who repeatedly bears research burdens, who is excluded, how power is used, or whether participants' contributions can produce trustworthy knowledge.
Read Guide →
Doing Research Without Compromising Ethics
Community research becomes extractive when researchers primarily take data, knowledge, labor, samples, or access while communities have little meaningful influence over the research or its resulting value. Extraction is a pattern of relationship, not simply the act of collecting data.
Read Guide →
Doing Research Without Compromising Ethics
Exclusion criteria are ethically justified when they have a defensible scientific, safety, methodological, or ethical purpose. Researchers should be able to explain why each important restriction is necessary rather than treating familiar or convenient exclusions as self-justifying.
Read Guide →
Doing Research Without Compromising Ethics
Knowing participants personally does not make confidentiality impossible, but it changes the problem. Researchers must manage information they already know, new information learned through research, role boundaries, contextual recognition, and the possibility that others in the shared setting can identify participants.
Read Guide →
Doing Research Without Compromising Ethics
A participant may be identifiable even when no single detail reveals who they are. Several ordinary characteristics can combine into a distinctive profile that points to one person.
Read Guide →
Doing Research Without Compromising Ethics
Discovering possible illegal activity does not automatically require researchers to report it. Researchers should protect confidentiality, determine whether a specific legal or safety obligation applies, and follow the approved protocol rather than becoming investigators or informants by default.
Read Guide →
Doing Research Without Compromising Ethics
Communities are rarely single voices. Ethical research should identify relevant subgroups, disagreements, power differences, and overlapping interests rather than treating a leader’s view or majority position as what “the community” thinks.
Read Guide →
Doing Research Without Compromising Ethics
Research misconduct commonly centers on fabrication, falsification, and plagiarism, but identifying misconduct requires more than spotting a problematic result. Definitions, intent or recklessness, evidence, and the applicable policy all matter.
Read Guide →
Doing Research Without Compromising Ethics
Research misconduct and questionable research practices are not interchangeable. Formal misconduct is usually more narrowly defined, while other practices can still damage research integrity without meeting that threshold.
Read Guide →