For some time now, I’ve been trying to drive the point home that Darwinists employ methods that are insufficient to support their conclusions. Their theories have many causal predictions (e.g., random mutation + natural selection causes new species or rm+ns results in new complex functions/information). In science, when your theory makes causal statements, you are limited to the experimental method in order to test the prediction. Some supportive information can be gleaned from correlational studies (i.e., how much one variable is associated with another), but the theory is not truly tested unless an experimental design is utilized. The experimental method is used extremely rarely in evolutionary research.
This brings us to the reason for my post. A recent study found that a statistical method used in many studies of natural selection is invalid.
Scientists at Penn State and the National Institute of Genetics in Japan have demonstrated that several statistical methods commonly used by biologists to detect natural selection at the molecular level tend to produce incorrect results. “Our finding means that hundreds of published studies on natural selection may have drawn incorrect conclusions,” said Masatoshi Nei, Penn State Evan Pugh Professor of Biology and the team’s leader.
Of course, nearly all of these hundreds of potentially invalid studies do not employ the experimental method, but simply use statistical analyses of DNA/gene sequences and so forth. And from this, they often draw causal conclusions (RM+NS resulted in this new function). However, it was found that the statistical methods were incorrect in their predictions.
In real fields of science, the experimental method is performed routinely. In psychology, the experimental method is practiced very often. Not so with evolutionary biology. The reason given in this press release was:
Nei said that to obtain a more realistic picture of natural selection, biologists should pair experimental data with their statistical data whenever possible. Scientists usually do not use experimental data because such experiments can be difficult to conduct and because they are very time-consuming.
Yes, real science is difficult to conduct and is time consuming, but I submit to you that it would be worth it to have properly designed, and properly conducted research. I think an interesting statistic would be to determine how much money has been wasted on improperly conducted research by evolutionary biologists. I’m not saying that correlational research is not science, but it’s not science to make causal inferences from methods that do not support causal inferences. Even by using the experimental method, the best you can hope for is a reduction in bias. Bias is nearly impossible to eliminate from studies, because studies are conducted by scientists (i.e., biased human beings).
Here is the take home message for Darwinists:
1). When you make causal predictions and wish to draw causal inferences, you must use the experimental method.
2). If you are not going to use the experimental method, but a correlational method, your conclusion must not go beyond the method you are using (e.g., this caused that). The best you can say is: this is related to that or this is associated with that.
3). Imagination is not an experimental or even scientific method.
4). If you use statistical methods to make predictions, the methods must first be validated. Otherwise your predictions are of unknown validity.