Bias in Science, Part 3

From: Randy Isaac <>
Date: Sun Jun 12 2005 - 08:54:07 EDT

In this third posting on bias in science, I'd like to ponder the influence of religion on the practice of science. In Part 1 we talked about the danger of prejudicial bias and the value of scientific bias. In Part 2 we took a closer look at Baumgardner's paper and argued that the paper ought to be rejected solely on the basis of poor scientific methodology.

When anyone believes that religious ideas are in any way connected with the results of a scientific analysis, the potential for prejudicial bias, in either direction, is greatly increased. It is of utmost importance that the most stringently controlled methodologies be used. How can we tell whether the scientific result or the rejection of a paper is due to prejudicial bias based on religion and not to scientific bias against poor methodology?

A charge that scientists have a bias due to religious implications may run along the following line:

1) Some scientists want to deny the theistic claim that God exists and that he created the physical universe
2) Some people think that a young earth is a key indicator that the Bible is true and that God created the heavens and the earth
3) Therefore all results indicating a young earth/universe or that contradict an old earth/universe must be rejected

Meanwhile, the scientist reviewer without an atheistic bias might reason as follows:

1) Many independent foundational experiments have documented the old age of the earth
2) A paper purporting to find a young age of the earth and refuting all previous measurements is most likely incorrect.
3) Therefore such a paper can be rightly rejected, even if the error in methodology is not clearly discernible

The dilemma is, if one only knows that the paper was rejected, how can one tell if the scientist followed the first or the second mode of reasoning? Unfortunately, I don't think we can, at least not in the case of an anonymous reviewer. But we can say that if the second path of reasoning is valid in a particular case, then the existence of some scientists who follow the first path doesn't mean the decision is wrong.

In the example of Baumgardner's paper, it is possible that some reviewer might reject the paper on the basis of religious prejudice. That doesn't mean the rejection was wrong--provided it can be shown, as we did in Part 2, that the methodology is flawed. A similar example is the recent removal of Robert Gentry's paper from the arXiv web site. (see and Based on his paper in the December 2004 PSCF, I would assert that his flaws in methodology are sufficient to warrant such removal.

In practice, many scientists have pursued the following logic:

1) YEC sponsored papers claim either a young age of the earth/universe or an error in a measurement that indicates an old earth/universe.
2) All YEC papers that have been analyzed so far have been found to be fatally flawed with regard to scientific methodology
3) Therefore, any new YEC paper with the same claim can be rejected as being flawed, even without identifying the error in methodology.

This logic seems to be widely held in this discussion list which is why most of the Acts and Facts comments are summarily dismissed, often with disdain, without further analysis. A detailed analysis is certainly preferable but may not always be possible.

Another kind of logical reasoning can lead to rejection of papers due to anti-religious bias:

1) Many theists believe that identifying flaws in evolution is tantamount to finding evidence for the existence of a Creator.
2) Some scientists do not wish to support evidence for a Creator nor to fuel the debate nor to be branded a creationist.
3) These scientists therefore reject any paper criticizing evolution and suppress any of their own criticism of evolution

Never mind that there are many theists who do not agree with premise #1 and many scientists who do not agree with #2, the action in #3 occurs and may even set back legitimate scientific progress in studying evolution.

The possibility that a scientific result may have an implication for a religious belief makes the prospect of a prejudicial bias very strong. It means that extraordinary measures must be put in place to ensure that only the proper methodology is followed. The suspicion that an atheistic bent will cause a bias in one direction and a theistic bent will cause a bias in another direction means that such an experiment is likely to generate some tough discussions. Both sides need to be doubly cautious and highly disciplined in carrying out and explaining the analysis. In particular, charges that the entire established scientific knowledge base in a specific area is faulty due to a religious bias are highly suspect since such charges themselves are characterized by a strong religious preference. It must also be remembered that established scientific knowledge is almost always a trustworthy basis for future scientific endeavor.

Prejudicial bias is prevalent due to many sources. Religious bias is only one source of such bias, but its implications are more far-reaching in that they affect not only the scientific community but the religious community. Most devout adherents are not scientifically literate and will choose to listen to their spiritual leader rather than a scientific or logical argument. When such spiritual conviction is used to affect the outcome of scientific work or to opine on the validity of a scientific publication, the result can cause damage to both science and religion. It is of utmost importance that we maintain a healthy scientific bias and avoid any hint of prejudicial bias by insisting on the highest standards of integrity in scientific methodology.

Received on Sun Jun 12 09:01:33 2005

This archive was generated by hypermail 2.1.8 : Sun Jun 12 2005 - 09:01:34 EDT