Manuel B. Garcia

Manuel B. Garcia serves as the Senior Director for Educational Technology and Digital Learning at FEU Institute of Technology, Manila, Philippines. Read More

Contact Info

1607, FEU Tech Building,
P. Paredes St, Sampaloc,
Manila, Philippines
mbgarcia@feutech.edu.ph

Follow Me

Can a Study Be Ethical but Still Be Poorly Designed?

A study can appear ethically careful while still being methodologically weak, but the separation has limits. When poor design substantially undermines the knowledge that justifies participants' burdens or risks, the methodological problem can become an ethical one.

06
Can Ethical Research Be Poorly Designed? Guide 6 of 398
01 · The Question

Can Research Protect Participants but Still Be Bad Science?

Imagine a study with voluntary informed consent, strong confidentiality protections, fair recruitment, minimal foreseeable risk, and appropriate ethics review. Then look at the methodology and discover an inadequate measure, a poorly matched comparison group, serious confounding, or an analysis incapable of answering the research question.

Is this simply an ethical study with bad methods?

Sometimes that distinction is useful. Ethical safeguards do not guarantee methodological quality. But once methodological weaknesses become serious enough to undermine the value expected from the research, the problem may no longer be purely scientific.

02 · The Short Answer

Ethical Protections Do Not Guarantee Good Methodology

In Brief

Yes, research can satisfy important ethical protections while still having methodological weaknesses, but a sufficiently serious design flaw can itself become ethically consequential when it undermines the study's ability to produce the knowledge used to justify its burdens, risks, or use of participants.

Not every limitation makes a study unethical. The key questions are how serious the weakness is, what the study claims it can establish, what burdens or risks it imposes, and whether the remaining design still has a reasonable prospect of producing valuable knowledge.

03 · What You Need to Know

Where Methodological Weakness Becomes an Ethical Concern

Ethical Safeguards and Methodological Quality Are Not the Same Thing

Research can perform well on familiar participant-protection measures and poorly on methodology. A consent process may be excellent even if the questionnaire is incapable of measuring the construct the researcher claims to study. Confidentiality can be carefully protected in an underpowered experiment. Participants can be recruited fairly into a study whose comparison groups cannot support the proposed inference.

This is one reason scientific quality and ethical acceptability should not be treated as synonyms. One does not automatically establish the other.

The distinction, however, should not be pushed too far. Methodological quality can itself contribute to ethical justification.

Scientific Soundness Matters Because Research Asks Something of Others

Research may ask participants to give time, disclose private information, undergo procedures, tolerate inconvenience, accept uncertainty, or assume physical, psychological, social, legal, or economic risk. Even low-risk research consumes participants' effort and institutional resources.

Those burdens are generally accepted for a reason: the study is expected to generate knowledge of sufficient value to justify conducting it.

If a methodological flaw makes that outcome highly implausible, the ethical justification weakens. CIOMS makes this connection explicit for health-related research involving humans. Its international guidelines state that scientific and social value provide the ethical justification for undertaking such research and that proposed studies should be scientifically sound and likely to generate valuable information.

The 2024 Declaration of Helsinki similarly requires medical research involving human participants to have a scientifically sound and rigorous design and execution likely to produce reliable, valid, and valuable knowledge while avoiding research waste.

Belmont Also Connects Study Design to Risk-Benefit Assessment

The connection is not limited to contemporary international health-research guidance. The Belmont Report explains that assessing risks and benefits gives investigators an opportunity to examine whether proposed research is properly designed. It also states that risks should be reduced to those necessary to achieve the research objective and that alternatives should be considered systematically.

Current U.S. Office for Human Research Protections training likewise identifies study design within institutional review board assessment of risks and benefits.

The logic is straightforward. A risk may be reasonable when it is necessary to answer a worthwhile question through a sound design. The same risk becomes harder to justify if the design cannot answer that question in the first place.

Not Every Methodological Limitation Is an Ethical Violation

All research has limitations. Samples may be smaller than ideal. Measurement contains error. Qualitative interpretation involves judgment. Observational studies can leave residual confounding. Exploratory studies may begin with considerable uncertainty. Feasibility studies deliberately answer narrower questions than definitive trials.

Calling all imperfect research unethical would therefore collapse a useful distinction and set an impossible standard.

The more relevant question is whether the limitation is compatible with the study's actual purpose and claims. A sample of 20 participants may be inappropriate for estimating a small population effect precisely but entirely reasonable for a qualitative study seeking detailed accounts of a particular experience. A pilot study need not have the power of a definitive efficacy trial if it does not pretend to provide definitive efficacy evidence.

Methodological quality is relative to what the study is designed and represented as being able to establish.

A Design Problem Becomes More Ethically Serious as the Burden Increases

The ethical importance of scientific weakness depends partly on what participants are being asked to bear.

A poorly designed anonymous survey taking two minutes may waste participants' time and produce weak evidence. That deserves criticism. A poorly designed invasive clinical study exposing participants to substantial physical risk raises a much more serious ethical concern because the stakes are different.

This does not mean low-risk research is exempt from methodological standards. Rather, the ethical consequences of methodological failure should be evaluated in relation to the magnitude and probability of burdens and harms, the importance of the expected knowledge, and the alternatives available.

Using the Wrong Method Can Make Participant Burdens Pointless

Suppose a researcher wants to determine whether an educational intervention causes an improvement in learning but uses a design incapable of distinguishing the intervention's effect from obvious alternative explanations. Participants may still complete assessments, disclose information, and spend hours participating.

If the resulting evidence cannot support the causal claim the study was created to test, those burdens have not purchased the intended knowledge.

The remedy is not necessarily a more complicated design. It is a design appropriately matched to the question and claims. Sometimes that means changing the method. Sometimes it means narrowing the research question or interpreting the findings more modestly.

Small Samples Are Not Automatically Unethical

Sample size is a good example of why methodological criticism needs context.

A sample can be too small for a particular quantitative objective, producing estimates too imprecise to answer the intended question. If participants are exposed to meaningful burdens while the study has little prospect of producing informative results, that can become ethically concerning.

But “small sample” is not itself an ethical diagnosis. Qualitative studies, rare-disease research, pilot studies, case studies, early feasibility work, and other designs may appropriately involve relatively few participants. What matters is whether the sample is defensible for the study's purpose and whether the claims remain proportionate to the evidence.

Overly Large Studies Can Also Create Ethical Problems

More participants do not automatically mean better or more ethical research.

If a study recruits substantially more people than necessary to address its question, additional participants may be exposed to inconvenience, intervention, data collection, or risk without corresponding scientific need. Efficient design therefore concerns avoiding both insufficient information and unnecessary exposure.

The appropriate sample is not simply the largest one the researcher can obtain. It is the sample justified by the design, question, analytical approach, uncertainty, and ethical context.

Research Waste Has an Ethical Dimension

Poor design can waste more than money. It can consume participant goodwill, clinical capacity, community trust, research funding, staff time, biological samples, and opportunities to answer better questions.

The 2024 Declaration of Helsinki explicitly connects rigorous design with avoiding research waste in medical research involving human participants.

This does not make every inefficient decision an ethical violation. Research involves uncertainty, and some well-designed studies inevitably fail to recruit, encounter unforeseen problems, or produce inconclusive results. Ethical criticism should distinguish avoidable methodological deficiencies from uncertainty inherent in genuine inquiry.

An Unexpected or Null Result Does Not Mean the Study Was Poorly Designed

Researchers sometimes judge methodology retrospectively: if the study did not produce a statistically significant result or support the hypothesis, perhaps the design was inadequate.

That inference is mistaken. A scientifically sound study can produce a null, negative, uncertain, or surprising finding. Indeed, discovering that an expected relationship is unsupported can itself be valuable.

Methodological quality should be assessed from the appropriateness of the question, design, execution, analysis, and interpretation, not from whether the findings were exciting or favorable.

Sometimes the Ethical Solution Is to Redesign Rather Than Abandon the Study

Finding a methodological weakness does not automatically mean the research should stop permanently. Researchers may be able to change the measurement strategy, revise eligibility criteria, strengthen comparison conditions, adjust the sampling plan, modify procedures, narrow the research question, or reconsider the intended inference.

The important point is to address the weakness before continuing to expose people or resources to burdens that the flawed design cannot adequately justify.

When the methodological problem is severe enough, the question shifts from whether the study is merely weak to why poor design can itself become an ethical problem.

04 · A Practical Example

When an Apparently Low-Risk Study Has a Serious Design Problem

Hypothetical Example

A Survey That Cannot Answer Its Own Research Question

A researcher wants to determine whether using generative AI causes university students to develop stronger critical-thinking skills. Participants complete an anonymous online survey asking how frequently they use generative AI and how strongly they agree with statements such as “I am a critical thinker.” The researcher plans to use the correlation between the two scores to conclude whether AI use improves critical thinking.

Participant protection Participation is voluntary, the survey presents little foreseeable risk, no direct identifiers are collected, and participants receive clear information before consenting.
Methodological problem A cross-sectional self-report correlation cannot by itself establish that AI use caused changes in critical thinking. The measures may also require stronger justification for the constructs and claims involved.
Ethical significance The immediate participant risk is low, so the ethical stakes differ from an invasive experiment. Still, participants are contributing time and data toward a study whose intended causal conclusion is not supported by its design.
Possible response The researcher could redesign the study to address causal inference more appropriately or change the research question and claims to examine associations rather than causation.
Interpretation The problem is not that a survey is inherently unethical or scientifically weak. The problem is the mismatch between the research question, design, and conclusion the researcher intends to draw.

This distinction matters. Ethical evaluation should not become a generic demand for the most sophisticated method available. It should ask whether the chosen method can reasonably deliver the knowledge being used to justify the research.

05 · What Researchers Often Get Wrong

Common Mistakes About Poor Design and Research Ethics

Misconception

If Participants Are Protected, Methodological Weakness Is No Longer an Ethical Issue

Participant protections remain essential, but they do not exhaust ethical justification. If serious methodological defects leave little prospect of generating the intended valuable knowledge, exposing participants to burdens or risks becomes harder to defend.

Misconception

Every Methodological Flaw Makes a Study Unethical

Research is rarely methodologically perfect. The ethical significance of a limitation depends on its severity, the purpose of the study, the claims being made, the burdens involved, and whether the research retains a reasonable prospect of generating worthwhile knowledge.

Misconception

A Small Sample Is Automatically Unethical

Sample adequacy depends on the design and objective. A small sample may be inappropriate for one statistical inference and entirely appropriate for a qualitative, feasibility, rare-population, or other focused study. The relevant issue is whether the sample supports the intended purpose and interpretation.

Misconception

A Null Finding Proves Participants Were Exposed for Nothing

Well-designed research can produce valuable null or negative findings. Ethical justification concerns whether the study had a reasonable prospect of generating informative knowledge, not whether the preferred hypothesis ultimately succeeded.

Misconception

More Participants Always Make Research More Ethical

Too few participants may leave a study unable to answer its question, but unnecessarily recruiting more people can expose additional participants to burdens without sufficient scientific justification. Appropriate sample size depends on the research objective and design.

06 · What This Means for You

Judge the Design by What the Study Is Supposed to Accomplish

When evaluating a methodological limitation ethically, avoid asking only whether the method is imperfect. Ask what the weakness does to the study's justification.

A simple decision framework

If the limitation is minor and the study can still answer its stated question appropriately
Treat it as a limitation to manage and report rather than automatically as an ethical failure.
If the method cannot support the claim the study is designed to make
Revise the design, research question, or intended claim before asking others to bear unnecessary research burdens.
If participants face meaningful risk or burden
Apply greater scrutiny to whether the design has a reasonable prospect of producing knowledge sufficient to justify that exposure.
If a less burdensome design can answer the same question adequately
Consider whether the additional burden of the proposed method is necessary.
If no feasible redesign can produce worthwhile knowledge at an ethically acceptable burden
Reconsider whether the study should proceed in its proposed form.

This approach avoids two extremes. Methodological imperfection is not automatically unethical, but “the ethics paperwork is fine” is not a sufficient answer when the design cannot accomplish what participants are being asked to support.

07 · A Quick Checklist

Before Separating a Design Problem From an Ethics Problem, Check This

When evaluating methodological weakness, check:
Is the research question sufficiently clear to determine what the methodology actually needs to accomplish?
Can the proposed design support the type of inference or conclusion the researchers intend to make?
Are the measures, sampling strategy, procedures, and analysis appropriate for the study's actual objective?
What burdens or risks are participants assuming in exchange for the knowledge the study is expected to produce?
Does the methodological weakness materially reduce the likelihood that useful or valid knowledge will result?
Could the question, claims, or methodology be revised so that the study remains informative without unnecessary burden?
Are you judging methodological quality according to the actual research design rather than imposing standards from a different type of study?
Would you still consider the burden reasonable if the study produced an inconclusive or null result despite being conducted properly?
08 · Frequently Asked Questions

Frequently Asked Questions About Ethical but Poorly Designed Research

Can a study receive ethics approval even if its methodology has weaknesses?

Review processes and committee remits vary, so approval should not be interpreted as a universal certification of methodological excellence. However, methodological weaknesses can be ethically relevant when they affect risk, anticipated benefit, scientific or social value, or the justification for involving participants.

Does poor methodology automatically make human research unethical?

No. The ethical significance depends on the nature and severity of the weakness, what the study is trying to establish, the burdens and risks involved, and whether the research can still produce worthwhile knowledge. Ordinary limitations should not be confused with designs incapable of serving their stated purpose.

Is an underpowered study unethical?

Not automatically. Adequacy depends on the study objective and analytical framework. An inadequately sized study becomes more ethically concerning when it exposes participants to burdens while having little prospect of producing information sufficient for its stated purpose.

Can pilot studies be ethical even though they cannot provide definitive answers?

Yes. A pilot or feasibility study can have a legitimate purpose precisely because it addresses feasibility, procedures, recruitment, measurement, or other preliminary questions rather than providing definitive evidence of effectiveness. Ethical evaluation should assess it against that purpose.

Is research waste an ethical problem?

It can be. Avoidable research waste may expose participants to unnecessary burdens and consume resources or opportunities without sufficient prospect of valuable knowledge. The 2024 Declaration of Helsinki explicitly requires scientifically sound and rigorous medical research that is likely to produce valuable knowledge and avoid research waste.

Does a failed study mean the original design was unethical?

No. Research outcomes are uncertain by definition. Recruitment difficulties, null findings, unexpected results, or unforeseen complications do not retrospectively prove that a study was unethical. The relevant question is whether the design and ethical justification were reasonable given the information available when decisions were made.

09 · The Bottom Line

Poor Design and Unethical Research Are Different Until They Are Not

The Bottom Line

A study can contain methodological weaknesses while still satisfying important ethical protections, but serious design defects become ethically consequential when they substantially undermine the knowledge that is supposed to justify participants' risks, burdens, or contributions.

Do not demand methodological perfection. Instead, ask whether the design is appropriate for the question and claims, whether the anticipated knowledge remains worth pursuing, and whether the burdens imposed are reasonable given what the study can realistically accomplish.

10 · Sources and Further Reading

Sources and Further Reading

11 · Cite this Guide

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.

Has the Field Guide helped your research?

If a guide helped clarify a question, inform a research decision, or move your work forward, I would love to hear about your experience. Your story may also help other researchers discover the Field Guide.

Share Your Experience
Takes only a few minutes