5 Ethics in research
You have learnt how to ask a RQ, and identify different types of studies to obtain data. In this chapter, you will learn to:
- list common ethical issues to be considered in research design.
- understand the purpose of reproducible research.
5.1 Obtaining ethical clearance
All research must be ethical, and must meet ethical guidelines, to minimise risk of harm to the environment, property and to participants, and to preserve the well-being, dignity, rights and safety of participants (including animals). Practically every university and research organisation in the world promotes and enforces responsible research practices.
Most research studies require a comprehensive ethics approval process before research begins, which must be approved by an ethics committee. We only give brief comments about research ethics.
Example 5.1 (Ethics) Ethics are important for all studies, not just those involving people or animals. For example:
- Rubbo et al. (2019) found \(238\) engineering articles published between 1945 and 2015 were retracted, mostly for unethical research practice.
- In the chemical sciences, \(331\) retractions were reported in 2017 and 2018 due to ethical issues, such as falsification of data and plagiarism (Coudert 2019).
5.2 Ethical issues to consider in research design
Ethical issues to consider when designing research studies include:
- Acknowledgement: All those who contributed to the research should be acknowledged, including those who prepare figures, take photographs, or have helped collect data.
- Analysis: The analysis must use appropriate methods.
- Confidentiality: Data should be kept confidential and secure.
- Consent: When appropriate, people should consent to being in the study, and hence should be told what the study involves. People should also be able to withdraw from the study without penalty.
- Economic risks: Financial loss to participants should be avoided. Reimbursements of reasonable costs for participating may be appropriate.
- Environmental risks: Environmental impacts and damage should be avoided or minimised.
- Funding: Sources of funding should be disclosed. Any studies funded by, or sanctioned by, companies or organisations with vested interests need to be carefully scrutinised. These may lead to, or give the impression, of conflicts of interest.
- Incentives to participate: If participants are offered incentives to participate (above reimbursement of costs), these should be acknowledged as it may (perhaps unconsciously) cause participants to influence the results.
- Legal risks: Participants should not be put in the position of breaking laws.
- Plagiarism: The work of others should be appropriately acknowledged and not claimed to be original.
- Physical risks: Physical harm or discomfort (to researchers, participants or bystanders) should be avoided or minimised.
- Psychological risks: Psychological harm or discomfort (to researchers, participants or bystanders) should be avoided or minimised.
- Resourcing: The study should not waste resources, time or money (e.g., if the answer to the RQ is already known, the study is not necessary).
- Sample size: The study should not use more individuals than necessary.
- Social risks: Social harm or discomfort (to researchers, participants or bystanders) should be avoided or minimised.
- Storage of data: Data should be stored securely, kept for the required amount of time, then securely disposed.
Example 5.2 (Poor ethics) In the Tuskegee syphilis experiment, conducted between 1932 and 1972, treatments were actively withheld from men with syphilis (Corbie-Smith 1999). The men's wives and children often were affected. The men were lied to about their treatments, and prevented from seeking other treatments. This study was highly unethical, and could not be conducted now.
Example 5.3 (Poor ethics in analysis) In 1986, the American space shuttle Challenger exploded just after launch, killing all seven astronauts on board. A review (Dala, Fowlkes, and Hoadley 1989) found that the cause was partly that the engineers failed to use some data that should have been used. This was unethical.
5.3 Reproducible research
One way to ensure that research results are reliable and trustworthy is to encourage reproducible research: enabling someone else to repeat the study and analysis. For research to be reproducible, the methods, data, analysis methods and any relevant computer code must be made available (Laine et al. 2007).
Methods for ensuring reproducible research are often discipline dependent, and beyond the scope of this book. Different journals also have different expectations regarding reproducibility. Nonetheless, realising that reproducibility is advantageous is important.
Using point-and-click interfaces for statistical analysis (e.g., spreadsheets) is not recommended, since results are not reproducible. The importance of reproducibility in the analysis phase is crucial:
There are serious medical consequences to errors attributable to the effects of spreadsheet programs and software operated through a graphical user interface [...] that could have been avoided through a reproducible research paradigm...
--- Simons and Holmes (2019), p. 471
Rather than using spreadsheets (also see Sect. 2.12), using tools which encourage reproducible research are recommended. Statistical software packages, such as python, R, SAS, SPSS and Stata, are recommended.
5.4 Chapter summary
Studies must be ethical, and any formal study must obtain ethical approval before beginning. Ethics covers issues including, but not restricted to:
- acknowledgements.
- analysis methods.
- confidentiality.
- consent.
- economic risks.
- environmental risks.
- funding
- incentives for participants.
- legal risks.
- plagiarism.
- physical risks.
- psychological risks.
- resourcing.
- sample size.
- social risks.
- storage of data.
5.5 Quick review questions
Are the following statements true or false?
- Ethics apply for any type of study.
- Ethics only refer to the interactions of the researchers with participants in the study.
- Ethics only apply when people are the individuals.
- Ethics only apply when people or animals are the individuals.
- Ethics can extend to storage of data.
- Ethics only apply to the design of the study.
- Ethics apply even to the analysis of the data.
- Ethical clearance is provided by (for example) the University Ethics Office.
5.6 Exercises
Answers to odd-numbered exercises are available in App. E.
Exercise 5.1 Consider this conundrum (Crozier and Schulte-Hostedde 2015):
A research team has an extraordinarily successful long-term study of a population of bighorn sheep (Ovis canadensis) on Ram Mountain [...]
The population contains marked individuals for which the research team has incredibly detailed data [...] this research has lead to numerous important publications.
Recently, however, a cougar (Puma concolor) that has learned to specialize on these sheep is slowly but surely eating all of them. This is a study of a natural population, which includes predation, but this cougar is drastically reducing the sample size of the study.
Since it is legal to hunt cougars in the region where this study is taking place, one option is to try to kill the predator; however, even if a cougar were successfully hunted, this would not ensure that it was the correct one.
What action would you recommend, from an ethical point-of-view?
Exercise 5.2 Suppose a research group is testing a new drug, with the potential to cure a debilitating illness. The researchers could (a) use a control group that does not receive the new drug, and so obtain stronger evidence for using the drug if it works; or (b) not use a control group, so that everyone in the study potentially benefits from the using the new drug.
What would you decide? Explain.
Exercise 5.3 Suppose a very deadly and highly contagious disease breaks out. Is it ethical to use of a new drug to treat those affected, even though the drug is experimental and the potentially harmful side effects are unknown? Discuss your point-of-view.
Exercise 5.4 Is it ethical to lie to the subjects in a study? Deception is used in some disciplines, and may be approved by ethics committees under certain circumstances (such as the potential benefits of the study, and whether the deception is likely to cause physical or psychological discomfort to the participants).
Is it ethical to tell participants that they are taking an active medication, when it is actually ineffective (a 'placebo')? Discuss the advantages and disadvantages.