Discussion Topic:
Case Study: Accuracy of information
Abi is a researcher at an institute and also a statistical programmer. Abi has received a project from a manufacturer to review the nutritional value of a new cereal, Whizzz. Having collected the necessary data, he now needs to perform the appropriate analyses and print the reports for him to send to the manufacturer. Unfortunately, the data Abi has collected seems to refute the claim that Whizzz is nutritious, and, in fact, they may indicate that Whizzz is harmful.
Abi also realises that some other correlations could be performed that would cast Whizzz in a more favourable light. “After all,” he thinks, “I can use statistics to support either side of any issue.”
Abi must not alter any data values as he needs to respect the authenticity and accuracy of the data as well as the output coming out of it. He also needs to keep a certain level of integrity. According to the Ethical Guidelines for Statistical Practice from the American Statistical Association (ASA), regardless of personal or institutional interests or external pressures, the statistical practitioner does not use statistical practices to mislead any stakeholder. (“Ethical Guidelines for Statistical Practice,” n.d.)
I do not believe that exposing different conclusions to the customer is unethical. However, Abi needs to be transparent and demonstrate the actual output to the manufacturer. I think he should present the result as it is without considering an alternate version of the result that might better suit the stakeholder.
Also, I do not think that Abi will be held responsible for what the manufacturer decides to do with the data and the results. Nevertheless, if it goes against his beliefs, he can report the case to his manager and discuss a possible solution that will respect both his ethical opinion and his work duties.
References:
Ethical Guidelines for Statistical Practice, n.d. 8. Default. (n.d.). Ethical Guidelines for Statistical Practice. [online] Available at: https://www.amstat.org/your-career/ethical-guidelines-for-statistical-practice.
Hi David,
Thank you for your insight regarding the case study at hand. You have brought up very valid points regarding research integrity and ethics.
I agree that the statistical practitioner should not mislead any stakeholder regardless of personal or external pressures.
Like the American Statistical Association guidelines, the Code of Practice for Statistics outlines the cornerstones of a common quality framework within the UK that can be referenced when evaluating the professional conduct of researchers. The code is based on three pillars, which are Trustworthiness, Quality and Value (UK Statistics Authority, 2018).
Within the Trustworthiness pillar, it is important that there is confidence in the people that produce data and research. An important facet of building confidence is to ensure those handling the data are acting with honesty and integrity and guided by ethics. The results should be presented impartially and objectively regardless of external factors (UK Statistics Authority, 2018).
The Quality pillar assures that produced statistics fit their intended uses and are not materially misleading. This includes principles that ensure suitable data sources as well as a methodology that is sound and based on good practice (UK Statistics Authority, 2018).
References:
UK Statistics Authority (2018) Code of Practice for Statistics. Available from: https://code.statisticsauthority.gov.uk/wp-content/uploads/2018/02/Code-of-Practice-for-Statistics.pdf [Accessed 29 April 2022].
Hi David,
I completely agree with your standpoint of Abi maintaining professional integrity by demonstrating the actual research results, regardless of the positive/negative outcomes that they portray. As an independent researcher, I agree that Abi could not be held liable for the decisions and way in which the company chooses to use the published data. Although, if key information is omitted, the company could blame the lack of ‘thorough research’ on Abi’s part as a part cause of the mis-advertising.
Although focused on news articles, a 2021 article produced by JV Consulting on behalf of the Reuters Institute for the Study of Journalism shows that impartiality is a key factor that can affect a user’s trust and opinion of a service/provider. This is a statement that appears to be echoed in the UKRI Framework for Research Ethics, a principle of which is “research should be conducted with integrity and transparency” (UKRI, n.d.). Both of which demonstrate that Abi should disregard any professional bias for the company in question and publish the full results of the research, regardless of whether the complete outcome is favourable or not.
References:
JV Consulting. (2021) The relevance of impartial news in a polarised world. Available from: https://reutersinstitute.politics.ox.ac.uk/sites/default/files/2021-10/Vir_the_relevance_of_impartial_news_in_a_polarised_world_FINAL_0.pdf [Accessed 1st May 2022].
UKRI. (n.d.) Framework for research ethics. Available from: https://www.ukri.org/councils/esrc/guidance-for-applicants/research-ethics-guidance/framework-for-research-ethics/our-core-principles/ [Accessed 1st May 2022].
As part of the module, we all discussed and debated Abi's ethical dilemma. Abi is a researcher and a statistical programmer. The company required him to perform multiple analyses and submit the resulting report to them.
It has become apparent that my initial judgment was correct. Abi should not perform any additional analyses to paint a more positive picture of the product. Marzio and Kieran brought up interesting points. We have all agreed that Abi should not alter any data and keep a certain level of integrity. We all conclude that Abi needs to be transparent and provide the actual output without any alteration.
In an ideal world, research data should be reported objectively (Marco and Larkin, 2000). The final result of a statistical analysis must never be modified to better suit the stakeholder situation.
References:
Marco, C.A., Larkin, G.L., 2000. Research Ethics: Ethical Issues of Data Reporting and the Quest for Authenticity. Acad Emergency Med 7, 691–694. https://doi.org/10.1111/j.1553-2712.2000.tb02049.x