How is the Scientific Method Used in Natural History Study?

Scientists use inference to make general statements about classes of things or the relationships of complex phenomena. Example, by sampling, we can infer from examining 10 marbles from a bag of 1000 marbles that all of them are blue, or half are blue and half are green. By modifying conditions, we can infer forces at work by their occasional absence under certain conditions, then project that all such instances of that complex phenomenon work the same way (weather, chemical reactions, social interactions).

Inferences are OK even if they are wrong if they work for a time until a problem appears. Example, if one of the 1000 marbles is red, we may never know if it does not appear in a sample, but a prediction, that half are blue and half are green, based on scientific sampling, will work for us most of the time. One kind of "inquiry science," one based on sampling, provides predictions. The "why?" question is replaced by "what will happen next?"

Scientists invent a theory, an explanation, about observations of complex phenomena. Scientists accept those explanations that connect all the facts in the simplest way, and use these explanations to guide further study. This does not mean that the simplest explanation is always correct, it means that it has the greatest chance of being correct when further studied. If scientists cannot form an explanation that deals with all the facts, then the facts are re-examined, but otherwise scientists reserve judgment. Theories (or explanations) should have predictive power.

Except for disproving the null hypothesis, scientists are never sure of anything, since absolute surety is not possible in scientific inference. With more knowledge (for example, discovery of the red marble), past inferences must be qualified, expanded or corrected.

Scientists in natural history museums use natural history collections as part of major long-term projects involving sampling to test hypotheses about our environment and the organisms in it. These big studies share many of the same methods as experimental studies in which the sampling is done over a short time period. "Controls," "statistics," "modeling," and "description" are methods that are often part of a scientific study, but the main feature of science is the development of predictive or explanatory theories about classes or complex phenomena.

In Botany, for instance, the hypothesis of a certain multi-institution, multi-decade botanical research project called "floristics" is that plants have particular geographic distributions that are the result of past migrational and dispersal patterns; we have shown that the null hypothesis (that they don't have such patterns) is untrue (because we have specimens whose distributions show such patterns), so we know that such patterns do exist. Inferences based on voucher specimens in the collections help devise hypotheses about patterns that lead to further collecting (sampling) to help develop more detailed theories (predictions or explanations) about why these patterns exist.

One scientific exercise can involve prediction. Predict the contents of a black box and use sampling to develop predictive theories.

Another scientific exercise can involve explanation. Who did something? Detective work develops the simplest explanation that fits all the facts.

More complex studies involve analysis (taking apart) and synthesis (putting together). We take apart a complex phenomenon and devise the simplest explanation that explains the relationship of all the parts. Here is a typical natural history museum method of study:

Example:

a. Analytic method.

Describe fully, using a large data table, the physical attributes of each specimen (organisms, paper fasteners, nuts and bolts, whatever). Point out in the table how the specimens differ from each other.

b. Synthetic method.

Note how each specimen is similar to the others. By noting which specimens are most similar to each other, organize the specimens into groups, and groups of groups. Name the groups. Can the groups be organized by how the specimens were made (evolution, manufacture, provenance)?

METHODS OF SCIENCE are various but include the following:

1. Hypotheses should be falsifiable, that is, they should be presented in such a way that they can be tested. They are more acceptable as explanations to the extent that they survive repeated attempts to prove them false. Scientific hypotheses are not expected to be proven true; they are inferences that simply have not yet been proved false though not through want of trying.
   
2. Observations, including results of experiments, must be reproducible by other researchers.
   
3. Scientific theories and results of experiments must be openly published in sufficient detail to be testable.
   
4. Scientific theories and results of experiments must be free of obvious internal contradictions.
   
5. The scientist must be willing to follow the data where it leads, rather than bending the evidence to fit some preconceived rationale.
   
6. Scientific hypotheses should not include explanations that are unique to a particular event or factors that are not part of observed nature