Tags

, , , , , , ,

US Representative Lamar Smith (R-TX) has proposed a new bill to set funding criteria for the National Science Foundation (NSF). According to the bill any award of funding by the NSF must meet the following three criteria.

1) “… in the interests of the United States to advance the national health, prosperity, or welfare, and to secure the national defense by promoting the progress of science;

2) “… the finest quality, is groundbreaking, and answers questions or solves problems that are of utmost importance to society at large; and

3) “… not duplicative of other research projects being funded by the Foundation or other Federal science agencies.”

ScienceInsider

Even the suggestion of these criteria demonstrates a completely lack of understanding about science methodology. Let’s start with the first criteria and work our way through all of them.

1) “… in the interests of the United States to advance the national health, prosperity, or welfare, and to secure the national defense by promoting the progress of science;

This criteria is nothing but hollow rhetoric. What constitutes advancing health, prosperity, welfare or defense? Those areas are so broad that they encompass everything. There is little out there which could not be argued to advance those interests in some manner or another. The problem is that frequently legislators and the public don’t understand what the research is doing enough to judge whether it would meet this first criteria.

2) “… the finest quality, is groundbreaking, and answers questions or solves problems that are of utmost importance to society at large; and

It would be nice to fund nothing but groundbreaking novel research but the fact is that very little of science is groundbreaking. It takes hundreds of thousand of man hours to research questions which many people would consider trivial in order to build a foundation for groundbreaking research. The truly novel discoveries depend on having a base level of knowledge about the area of interest.

Few would consider research in to number theory to be groundbreaking or of great importance. Yet John Tate’s work on number theory was used as the basis for creating internet encryption which made internet business possible.

In 1952 few would have labeled the x-ray diffraction image taken by Rosalind Franklin to be groundbreaking. But that image became the basis of our modern understanding of DNA which transformed our understanding of biology and medicine.

These are but two examples of seemingly unimportant research at the time which later completely changed our lives. There is no way to tell at the outset what research will lead to. In fact it may take decades in order to understand the importance of research. It is rather common for individuals to receive Nobel prizes decades later because their research became the foundation for new innovations. It is completely unreasonable to expect ALL research to APPEAR to be groundbreaking before even starting the research.

3) “… not duplicative of other research projects being funded by the Foundation or other Federal science agencies.”

Finally the third criteria shows that Representative Smith has a complete lack of understanding about science methodology. This criteria requires that research not duplicate any other research. Yet replication of research is absolutely necessary for scientific advancement. Each and every experiment must be designed such that any other researcher can replicate the experiment with the same results as the original. This is a requirement on ANY experimental design. Why? Because every human endeavor is error prone. There is always a possibility of error; that might be human error in execution of the study; it might be random sampling error; it might be an uncontrollable factor that can’t be accounted for. The source of error doesn’t matter, that fact is that humans are error prone. To compensate for that researchers replicate studies that have already been done in order to verify the results. A single study does not make something true in science. That study must be replicated in order to demonstrate that the original study was accurate.

If replication is not funded then how do we know that the original work is correct? The original could have found an effect which does not exist (type 1 error) or it could fail to find an effect that does exist (type 2 error). There would be no way of knowing without replication. So if no effect was found then it would be erroneously assumed that no effect exists and then no further research in that area would be funded. Whereas if an effect was found that does not exist then research could continue in that area but it would only waste money since it is based on erroneous information. Instead replication serves as the means by which science can correct itself. Replication helps to root out errors made in science so they can be corrected and so that researchers don’t continue to propagate those errors.

In conclusion Representative Lamar Smith (R-TX) has no clue about science. He has put forth nice sounding rhetoric about making science funding about advancing the health, prosperity, welfare and defense of the country. But he has no clue about the importance of replication. He also fails to understand that not all science can be groundbreaking. Groundbreaking discoveries are built on years of mundane research. Those that create policy about science should have a solid understanding and background in science. Otherwise they are creating laws about something they do not know and thus they should not be involved in.

Advertisements