Sunday, March 9, 2008

Does the Fact that a Scientist Says it is So, Make it So?

We hear a lot about science today and we hear a lot from scientists. Science has brought us many wonderful things and promises to bring us even more in the future. However, there is plenty of disagreement among scientists and most scientists would agree that scientific research and conclusions are more of a journey than a destination. This essay is intended to help the layman sort through everything we hear today in the media in the name of science.

Scientific Method
Scientific method is the process by which scientists make observations, take measurements, conduct experiments, develop a hypothesis, and repeat the cycle in order to prove, disprove, or modify the hypothesis. Often this cycle can take a life time of work without any proof of the hypothesis.

Applied Science
Much of the scientific research conducted has resulted in fantastic discoveries and inventions such as penicillin, nuclear power, flight, the Moon landing, computers, and the list goes on. These inventions and discoveries are the result of applied science or research and development. Some discoveries are quite accidental and others are very intentional. What makes applied science different is that it is verifiable by replication and we see the proof in our day to day lives.

Theoretical Science
There is another field of science that is less tangible. Theoretical science attempts to explain phenomena that are beyond our ability to grasp with our normal senses or it may attempt to project an outcome that we will only know if it is accurate or not after the passage of time. For theoretical scientists the proof of their hypothesis often remains out of reach, at least often beyond their own lifespan. Einstein did not live long enough to see some parts of his Theory of Relativity proven by future experimentation. Other theories have yet to be proven and some are proven to have been wrong.

Scientific Models
Scientists often develop scientific models to predict outcomes or support theories. Models typically use a series of proven principles and link them in such a way that they will accurately predict the outcome of certain actions or natural events. The linkages of various principles involve a set of mathematical relationships called algorithms. These algorithms can become extremely complicated especially when trying to develop models to predict outcomes in complex systems such as the United States economy or the atmospheric science that goes into weather prediction models. As a general rule, most scientists will agree that all models are wrong; it is just that some models are better than others. Certainly, very few weather personalities will stake their lives or reputations on the outcome of the weather forecasting models used today. Some of these models can only be run because of the advent of super computers. Climate change models produce projections and not predictions and they are based on the serial or parallel linkage of several weather and atmospheric science models. These models can be so complicated that once a set a of parameters are input into the model, it may take days or weeks for the model to run before the output is generated even on some of the most powerful computers in the world.

Statistical Research
A lot of scientific research you hear about today is based on statistical analysis. Often the results are described in average numbers, but we all know that nothing is average. To put it another way, going by the average temperature, one could conclude that by standing with one foot in a fire and another foot on a block of ice, that on average that person would be comfortable. The other often reported result comes in the form of a correlation. Results of independent research projects can be run through a computer model to see if there are any statistically valid correlations. We usually hear the results of this research in short sound bites on the daily news. “Research shows that pregnant women who eat an ounce of chocolate a day during their pregnancy have happier kids,” the announcer happily reports. On some days we hear coffee or alcohol in moderation can be good for us and on other days we hear that any amount coffee or alcohol can be bad for our health. The problem with all of this is that a statistically valid correlation does not necessarily mean that there is a cause and effect relationship.

Advocacy Science
This brings us to the most dangerous and a growing field of science—advocacy science. These scientists are so fervent in their beliefs about their field of science that they speak about theories in absolute terms. They have abandoned scientific method and integrity. Advocacy scientists will ignore any empirical proof that their theory may not be correct and they cherry pick the results that support their theory. Scientists should embrace different opinions; they should be open to the possibility that their research may have problems; they should be like steel sharpening steel; they should view dissention as a positive part of the process; and they should embrace the peer-review process. Science advocates, on the other hand, revile anyone who dares to criticize their work; they believe the end outcome or goal justifies their means; they will resort to slander and innuendo to discredit a naysayer; emotion replaces objectivity; and anything short of full endorsement of their unproven theory makes someone the equivalent of a Holocaust denier.

When Theory becomes “Fact”
Scientists have been searching over a century for the “missing link” to proves that mankind evolved from apes, and even though not yet proven conclusively, we often hear of new links that seem to point toward the proof of this theory. However, society as a whole has come to accept this theory as fact because so many scientists are willing to say it is so, notwithstanding the fact that the theory has not been conclusively proven.

Science and Faith
A theory is an unproven hypothesis and theoretical scientists often deal in such hypothetical realms that their theory may never be proven to be true or false. We may never know the origin of matter or prove the Big Bang Theory. And, if the Big Bang Theory is true, what does that mean to humans or all life forms for that matter? We may never know. This raises an interesting dilemma for many scientists, especially those who claim not to be religious. If a scientist cannot prove a theory, yet believes the theory to be true, is that scientist practicing a form of faith? Is not faith the basis of all religions? Can some beliefs in science become a form of religion?

Science versus Policy
Science may provide us with the facts and possible remedies, but there may be a number of ways to address any issue based on the scientific facts and other considerations. This is the realm of policy making. Good policy is informed by science, but often science cannot provide the complete answer to the challenge. The science may not be conclusive, or it may even be contradictory. One scientist’s remedy may be another scientist’s problem. Science is usually focused on one aspect of a larger issue and pursuing one recommended scientific course of action may have other more catastrophic results somewhere else. Scientists tend to be specialists in certain disciplines and they break things down into analytical pieces. It is the role of the policy maker to integrate all the science, all the possible actions, and all possible outcomes in order to make decisions that consider all the pros and cons, all the scientific implications, whether the solution is feasible and practicable, and the economic, cultural, and ecological consequences of the proposed action.

It is never as simple as a scientist says it is so. We live in a complex world where the outcomes of any action may be difficult, if not impossible, to accurately predict. A good dose of healthy skepticism is a principle that any ethical scientist should welcome. Remember, if a scientific report sounds either too good or too bad, chances are it may not be based on experimentally proven scientific method, or there is a body of science that points to a different conclusion.