Draft version 1.7 beta — Revised October 26, 2026

← Return to the table in Chapter 4, § 5

Bias in scientific investigation

A topic page linked from Chapter 4, § 5, Examples of Reported Violations of Enforceable Ethics Codes.

Definition

Some of the different types of bias are discussed in Chapter 3, section 3.2. There, I asserted that bias distorts the scientific process by negatively impacting a scientist's objectivity. You should be particularly vigilant about avoiding confirmation bias¹. According to Claude Bernard,²

C’est ce qui nous a fait dire ailleurs qu’il ne fallait jamais faire des expériences pour confirmer ses idées, mais simplement pour les contrôler.

We must never make experiments to confirm our ideas, but simply to test them.

There are two specific mentions of bias in the list of ethical issues in Chapter 4, § 5: bias in scientific investigation and bias in interpretations or conclusions. It is quite possible to engage in a scientific investigation that is not affected by bias to any significant extent, but that is diminished by bias in the interpretation of results and the conclusions drawn from the study. Hence, we will consider bias in interpretation and conclusions a separate issue. Indeed, an author's bias is often most clearly evident in their interpretations and conclusions.

A well-designed and competently executed scientific investigation and its raw (uninterpreted) data results are valuable on their own, independent of interpretation and concluding statements. Bias in the design and execution of an investigation diminishes or invalidates its scientific utility, and will certainly infect the interpretation and conclusions with that bias. 

One common example of investigatory bias occurs when the funder of an investigation explicitly or implicitly indicates the results the investigation should generate, before the investigation occurs. In that case, the investigation is likely to compile evidence in favor of the pre-determined outcome and ignore evidence to the contrary. This is not a scientific investigation; rather, it is some form of advocacy or marketing. In contrast, science is a human effort to discover reliable information about the world through collection of reproducible observations and their explanation through testable hypotheses.

In severe cases, bias in scientific investigation and consequent interpetation and conclusions might sink to the level of fraud. As defined in Black's Law Dictionary ²,

fraud, n. (14c) 1. A knowing misrepresentation or knowing concealment of a material fact made to induce another to act to his or her detriment... 2. A reckless misrepresentation made without justified belief in its truth to induce another person to act.

Relevant provisions in geoethics codes

AGI – § 1.3; GSA – § 3.1a.

Code Abbreviations

• AGI — American Geosciences Institute, AGI Guidelines for Ethical Professional Conduct (2015), Link to current document
• GSA — Geological Society of America, GSA Code of Ethics & Professional Conduct (2023), Link to current document.

Notes and References

← Return to the table in Chapter 4, § 5