Research ethics and honest reporting of findings

Notions clés

Approval and consent

Standards 8.01 and 8.02: obtain institutional approval where required, and tell participants the purpose, duration and procedures, their right to decline or withdraw and its consequences, foreseeable risks and benefits, limits of confidentiality, incentives and whom to contact.

Deception and debriefing

Standard 8.07 allows deception only when justified by the study's significant value and when nondeceptive alternatives are not feasible, never about research expected to cause physical pain or severe emotional distress. It is explained as early as feasible, with the option to withdraw data.

Reporting and credit

Standard 8.10 bars fabricating data and requires correcting significant errors; 8.11 bars plagiarism; 8.12 bases authorship on relative contribution, not position; 8.13 bars republishing data as if original; 8.14 requires sharing data with competent professionals for verification.

Questionable research practices

Short of fabrication, practices such as leaving unreported the measures that did not work distort the record. In an anonymous survey of over 2,000 psychologists with incentives for honesty, the share admitting such practices was surprisingly high (John and colleagues, 2012).

Public and media statements

Standard 5.04 requires public advice or comment to rest on professional knowledge and the psychological literature and not to imply a professional relationship. Standard 9.01(b) limits opinions about a person's psychological characteristics to those based on an adequate examination.

Community feedback

Research with communities, such as First Nations, Inuit and Métis communities in Canada, calls for engagement with the community on the research and on interpreting and sharing its results, as Chapter 9 of the Tri-Council Policy Statement (TCPS 2, 2022) sets out.

Exemple commenté

A psychologist presents a pilot study of a brief anxiety program, with 18 participants and no control group, at a conference. Five outcomes were measured; the slides show only the one that improved significantly. Afterward a reporter asks whether the program cures anxiety.

1. Fix the record first: report all five outcomes, including those that did not change, so the slides do not mislead, as Principle C, Integrity, asks.
2. Answer the reporter within the evidence: a small pilot without a control group cannot show a cure, and Standard 5.04 requires public comment to rest on the literature and professional knowledge.
3. State the limits plainly: the sample size, the missing control group and the mixed results, so the story does not overstate the finding.
4. If the published story still misstates the result, take reasonable steps to correct it, as Standard 1.01 requires when one's work is misrepresented.

Report everything you measured and say no more than the data can carry, in every venue.