Jan 12 2016

Real Scientific Literacy

miracleequationWhat does it mean to be scientifically literate? There is no completely objective answer to this question, it can be defined in multiple ways and the bar can be set anywhere along a spectrum.

Many tests of scientific literacy essentially ask a series of scientific facts – they are tests of factual knowledge, but not scientific thinking. This glaring deficit has been pointed out many times before, and was so again in a recent editorial by Danielle Teller. She writes:

There are a number of problems with teaching science as a collection of facts. First, facts change. Before oxygen was discovered, the theoretical existence of phlogiston made sense. For a brief, heady moment in 1989, it looked like cold fusion (paywall) was going to change the world.

I agree. A true measure of scientific literacy should be a combination of facts, concepts, and process. Facts are still important. Concepts without facts are hollow, and facts without concepts are meaningless. Both need to be understood in the context of the process that led us to our current conclusions.

It is shocking that 25% of Americans do not know the Earth orbits the sun. That means that 75% do know the basic fact that the Earth goes around the sun, rather than the sun around the Earth. How many people, however, understand that the Earth and sun actually orbit around their center of gravity (which is within the sun’s surface), that this causes the sun to wobble and provides a method for searching for exoplanets? How many are familiar with Kepler’s three laws of planetary motion? How many people know how we know that the Earth revolves about the sun, and how certain we are of this conclusion? To me, that is scientific literacy.

After a long discussion of the problem, Teller ends with a quick recommendation for what we can do to fix it – we should teach children (to paraphrase) that reproducibility and consensus are important, that data can be manipulated, and that science is more of a process than a collection of facts.

What I want to do is pick up where she left off and discuss in detail what people should know about the process and principles of science. I don’t think it is possible to come up with a list of scientific facts or even concepts everyone should know. That list would be long, subjective, and changing. (At the very least such a list would be book-length, not blog post length.)

Here is my list of what I think everyone should understand (truly understand, not just “know”) about the process of scientific thinking. This will still be a somewhat long list, and may need to be broken up into multiple parts.

True Scientific Literacy. 

1) Scientific Knowledge

The most basic principle to understand is that science itself is not just a collection of facts. It contains a collection of facts, but those facts are organized into theories, laws, and categories which reflect our understanding of those facts and of the universe. Science is a human endeavor, not a static or objective thing. It is our best attempt to understand the world.

In order to think about scientific knowledge it is important to understand that science builds on bits of data or evidence. The term “fact” can be vague, and can refer to data, collections of data, or even theories (like the fact that evolution occurred).

It is also important to understand that there are depths of scientific knowledge. As science advances, usually only in the beginning of a question or area of investigation do our conclusions wholesale change. As a science matures, complete changes in ideas become less and less likely or common. Our knowledge then tends to advance by becoming deeper; not invalidating existing ideas but showing that they are a special case of a deeper more fundamental rule or a more complex system.

2) Science is an empirical and logical process

Science is not one method but a collection of methods that build our understanding of the universe brick by brick. Ideas are tested in science in multiple ways, and those tests comprise the processes of science.

Scientific ideas can be looked at for internal logical consistency – do they make sense? Are there any apparent contradictions? What are the logical implications of a scientific idea? Do the numbers add up (does the mathematical model work)? Is the idea elegant, or is it a complicated mess?

Scientific ideas can also be tested against reality. Is the idea consistent with established scientific facts and existing data? This is actually a low bar and by itself not that compelling, but it is necessary.

Perhaps most importantly, how successful is the idea at predicting new data? Does the idea predict what will happen in the future, the result of future experiments, or the findings of new observations?

Finally (a step often neglected), how does the idea compare to competing ideas? Is it better or worse at predicting new information? Are there any experiments or observations that can distinguish the idea from competing ideas? How many new assumptions does the idea require to work (Occam’s razor) compared to competing ideas?

3) Science uses multiple logical methods

Science is partly, but not entirely, deductive. Deductive reasoning begins with well-established premises and then draws a conclusion that must be true if those premises are true. Often (but not always) deduction begins with general rules and then concludes a specific fact that must be true if those general rules are true.

Science also uses inductive reasoning (perhaps even more than deductive). Inductive reasoning also starts with established facts, but then draws generalized conclusions from those facts – conclusions that may be true but do not have to be true. The more facts support the general rule, and the more accurate predictions we make based on the rule, the more confident we are that it is probably correct.

A conclusion based on inductive reasoning, however, can always be disproved with new information. All swans are white, until we find a black swan.

Science also uses inference – figuring out what is likely to be true based upon what we know to be true. Inference uses facts and logic to come to the most likely conclusion, and can use a combination of induction and deduction. If evolution is true, then what do we expect to find when we look at patterns of genetic inheritance, for example.

4) Science is tentative, and always has error bars

Perhaps one of the most important and neglected aspects of science in the public consciousness is the tentative nature of scientific conclusions. There is a tendency to treat scientific claims with a false dichotomy – true or false, fact or myth. This is a convenient shorthand for claims that are almost certainly true or almost certainly false, but it can obscure the fact that many scientific claims are somewhere in the middle.

Rather, we should teach people to reflexively think of every scientific fact or claim as attached to a degree of certainty. The fact without the certainty is meaningless. Putting a number on it may be difficult, but may help some people grasp current certainty. As an alternative, here are some basic categories:

Rock solid – We know this as much as we know anything. The probability of being true is >99.9%. It can comfortably be treated as a fact.

Very Confident – There is a strong consensus of scientific opinion with no serious competing opinions. The probability is >95%, and we can treat it as likely to be true but there is a tiny possibility future evidence may change our thinking. At this level or higher it is reasonable to take actions based upon the scientific conclusion, and to teach it as probably true in textbooks.

Somewhat Confident – This idea is more likely to be true than not, or at least is our best current hypothesis. There remains serious opposition or alternative views, however, even if they are in the minority.

Controversial – The idea is new, or evidence is ambiguous. Further, there may be two or more competing theories with equal validity and no one theory has the clear upper hand.

Speculative – This is an entirely new idea with little or no evidence. It is just a hypothesis that has yet to be significantly tested.

Minority or Fringe idea – This is a scientific idea that is most likely not true. It has significant problems with evidence and/or logic, is fighting against disconfirming evidence, or requires massive new assumptions that are just not warranted. Ideas in this category may range from damn unlikely to impossible. They are best treated with extreme skepticism.

Disproved and rejected – These are ideas that have been thrown on the trash heap of scientific history. They have already been rejected by evidence or displaced by newer better theories. We can confidently reject them. They may be of only historical interest, or they may linger as fringe ideas.

This concludes part I of this topic. Feel free to give me suggestions for part II in the comments.

17 responses so far