Ethical considerations in research: types and examples

Ethical considerations in research are the rules and choices that protect the people in your study and keep your work honest.

research ethicsnoun

The principles that guide how research is planned, carried out and reported, so that participants are respected and protected and the results can be trusted.

In India, the main guide for research with people is the ICMR's National Ethical Guidelines for Biomedical and Health Research Involving Human Participants, published in October 2017. They are not only for doctors: they also cover social and behavioural science research for health, which is where Ritika's study sat.

They rest on four basic principles: respect for persons, beneficence, non-maleficence and justice. In plain words: respect people's choices, do good, do no harm, and share the burdens and benefits of research fairly. These grow into 12 general principles, among them voluntariness, privacy and confidentiality, and risk minimisation.

Informed consent is the heart of research ethics. ICMR describes it as a continuous process: give people the information they need, make sure they understand it, and make sure their decision is voluntary, all before any study activity begins.

Before anyone agrees, the information sheet must tell them, in a language they understand:

• that this is research, and its purpose and methods in simple words
• the possible benefits, and any risks, discomfort or inconvenience
• how far their records will be kept confidential
• any payment or reimbursement for their time
• that they are free to refuse or withdraw at any time, without penalty or loss of anything they are entitled to
• who to contact with questions, including someone at the ethics committee for complaints

Written consent is the norm. If a participant cannot read, ICMR asks for an impartial literate witness, a person not connected to the research, to be present through the whole process. Oral consent, with no signature or thumb impression, is allowed only in exceptional cases with the ethics committee's approval. Ritika's English form went first. She rewrote the information sheet in Hindi and Chhattisgarhi and read it aloud.

Voluntary was the harder word. "If the man who pays them asks them to come," Dr. Dewangan said, "how many will feel free to say no?" The guidelines note that power differences in India make open refusal hard, and ask researchers to read silence, one word answers or restlessness as discomfort, and not to press on.

People tell researchers things they would not tell their neighbours. Three words about protecting that trust are often mixed up:

In an interview study the researcher knows who said what, so promise confidentiality, with codes or pseudonyms. Promise anonymity only when you collect nothing that identifies anyone.

• Photographs need care. ICMR says publications must not show photographs or details that reveal who someone is. Ritika kept her photographs of the sorting piles and left out every face.
• Collect only what you need. She removed the question asking for each woman's exact address.
• Keep the key apart. The list linking codes to names stays separate from the data.
• Be honest about the limits. ICMR asks researchers to tell participants that confidentiality has limits; a threat to someone's life, for example, may have to be reported.

Harm in research is not only physical. For social and behavioural studies, the ICMR guidelines name several other kinds:

• Harm to dignity: people feel humiliated, dismissed, or treated as a means to someone else's end.
• Psychological harm: a question makes someone relive a painful or frightening experience.
• Social harm: taking part leads to stigma, or trouble with family, an employer or the community.
• Informational harm: identifiable information leaks out. For much social research, this is one of the main risks.

For Ritika, social harm was the real danger. If the thekedar learned which women had complained about him, some could lose their work. So she met the women through a self help group in their basti instead, at times they chose, and never told him who took part.

Then there was money. Every hour a woman spent answering questions was an hour not sorting, and sorting is paid by the kilo. ICMR allows researchers to pay participants for travel, time and inconvenience, including loss of wages. But it must not become an undue inducement, a sum so tempting that people agree against their judgement, and the ethics committee must approve it. Ritika proposed about an hour's earnings.

ICMR calls people vulnerable if they are disadvantaged and open to exploitation, cannot decide freely, or could be pushed into saying yes by hope of a benefit or fear of refusing. Ritika's participants fitted more than one of these.

Vulnerable does not mean excluded. Such groups have an equal right to be included, so research benefits reach them too, and a study that recruits only a vulnerable group should answer that group's own needs. Ritika's did. This is justice at work: no one should carry the risks of research only for the benefit of people better off.

Children need two permissions: consent from a parent or guardian, and the child's own agreement, called assent.

A few teenage girls sorted waste beside their mothers. Ritika's question was about adult women, so she limited the study to women aged 18 and above, and wrote that decision and its reason into her proposal.

Who decides whether a study is ethical enough? Not the researcher. ICMR says all biomedical, social and behavioural research for health involving people must be approved by a properly constituted ethics committee, and that a researcher cannot decide that her or his own proposal is exempt. The committee chooses one of three kinds of review:

Ritika's study involved a vulnerable group, so it went to full review with her information sheet, consent form, interview guide and planned payment. It was approved after one change: the committee's phone number had to go on the information sheet, so the women could complain to someone other than Ritika.

Under the New Drugs and Clinical Trials Rules, 2019, institutions doing biomedical and health research with people must have an ethics committee, registered with the Department of Health Research. Outside health research, practice varies between universities.

Ethics does not end when the interviews do. The ICMR guidelines name three kinds of research misconduct:

• Fabrication: making up data or results and recording or reporting them.
• Falsification: manipulating materials, equipment or processes, or changing, leaving out or suppressing data or results without a scientific or statistical reason, so the research record is no longer accurate.
• Plagiarism: presenting another person's words, ideas or work as your own, including reusing your own published work without saying so.

Ritika's test came in the last week. Eleven women had skipped the question on hours worked, and a classmate suggested filling the blanks with the average. Done quietly, that is falsification; honest ways of handling missing data are declared in the methods. Ritika reported the eleven as missing.

The same honesty applies to writing. Cite every idea you borrow; see the types of plagiarism that catch students unawares. If you use AI tools, say so and check every claim, as our guide on using AI in research honestly explains.

An ethics section should describe what you did or will do, not promise good behaviour. Include:

1. The approval.Name the ethics committee, and give the approval number and date if your university or journal asks for them.
2. Consent.Who consented, in which language, and whether a witness or parent was involved.
3. Risks.The real risks to your participants, and what you did about each.
4. Confidentiality.How identities were protected and data stored.
5. Payment and conflicts.Any payment or reimbursement, and any interest that could affect the study.

The study was approved by the university's Institutional Ethics Committee. Participants were women aged 18 and above who sort waste for a living. The information sheet was read aloud in Chhattisgarhi or Hindi, and written consent or a thumb impression was taken in the presence of an impartial literate witness. Participants could withdraw at any time. Names were replaced with codes, and the contractor who buys their waste was not told who took part. Each participant received a payment equal to about an hour's earnings, approved by the committee.

In a paper it usually closes the methods, as our guide on how to write a research methodology shows; in a research proposal it gets its own heading.

Dr. Dewangan read it twice and signed the letter to the committee. "Now it says something," she said. Ritika's interviews took five weeks instead of one, and every woman she spoke to had chosen to be there.