When evaluating scientific claims, it is important to think about bias. Bias is the tendency to favor a particular point of view and to present that view instead of other equally valid alternatives. When presented with new information that can be interpreted in several ways, individuals with different biases may present wildly different conclusions based on that information. Often, issues with multiple interpretations are presented as one-sided.
Here are a few examples of bias that can be found in debates or discussions of genetic engineering that you should look out for:
- Confirmation Bias: The tendency to readily accept conclusions that agree with one’s beliefs, and discard conclusions that disagree with them. This is also known as “cherry-picking.”
- Media Bias: Selectivity in what stories and perspective are covered in the media. This can happen when only certain stories or interpretations are covered, such as those that are sensational. In science reporting, bias can be introduced by emphasizing views not supported by evidence.
- Funding Bias: The tendency of a study or report’s conclusions to favor its financial supporters.
- Belief Bias: The tendency to reject logical arguments with unbelievable conclusions and accept illogical arguments with believable conclusions.
- Selection Bias: This occurs when study participants are not representative of the general population, which can occur through selective inclusion or exclusion of study participants.
If you find or believe there is a bias in argumentation from an individual or group, it does not necessarily mean that the argument is wrong. For instance, people often claim that there is a funding bias when a scientific study is funded by an industry group. That is not reason enough to dismiss the study, but potential bias is good to keep in mind when interpreting its conclusions and comparing them to the conclusions of other studies. To reject them out of hand would constitute a bias in itself. Ultimately, accepting or rejecting a conclusion should be based on the scientific evidence.
One of the goals of the GENetic Engineering Risk Atlas (GENERA) is to reduce the influence of bias on interpreting the outcomes of scientific research, and to help detect bias where it can be found. By building a comprehensive collection of information about studies related to the relative risks of genetic engineering, readers can understand the scientific literature at-a-glance, and evaluate claims made about the literature by others.
For further reading on bias, we recommend the following sites and books:
Reading to evaluate and identify bias(PDF)
Lesson on Bias in Science, by the US Geological Survey
One-Sided Arguments: A Dialectical Analysis of Bias, by Douglas Walton.