Learning about research, or how to research, is a lot like learning anything else.
There are a few basics, some “jargon” people use, some methods and procedures, and some big controversies they fight over. One in particular lies with “academicians” who look down their noses at historians, and whether historical research is valid due to the interpretative nature of research projects and whether “history” can actually be tested by the Scientific Method.

They are two realms for research. The first is the realm of theory. That is what goes on inside of our heads as researchers. It is were we keep our theories about what the Civil War era was actually like in reality.
The second is the realm observations. It is the real world into which we translate our ideas- our discoveries, findings, measurements, and observations. It is also in how that translates into our impressions.

When we “do research,” we are continually jumping back and forth between the two realms; between what we think was true about the Civil War era, and what had actually happened or existed in it.
Research tends to be either “theoretical” or “empirical.” Much research is theoretical, and deals with developing, exploring or testing the theories or ideas that clinical or social researchers have about how the world operates. The type of research we, as reenactors, living historians, and historians tend to do is empirical, and deals with observations and measurements of the reality of life in the Past. Yet, in some ways, our research can be a blending of these two- as a comparison of our theories about how the life during the Civil War “was,“ and our observations or “discoveries” of how it actually was.

Although there are three basic types of questions that research projects can address, Descriptive, Relational, and Casual- most of our research tends to be Descriptive and tries to describe what was going on or what existed during the Civil War. This can apply to the basic areas that go into an “impression,” or makes up a “persona:” the material culture, the mental and physical man, and the world and activities around him.

Research usually, but not always, beings with an “hypothesis” which is a specific statement of prediction. For our Civil War purposes, that is instead usually a “question.”
Research, as in logic, usually works from two broad methods of reasoning as either deductive and inductive approaches. Deductive reasoning is done from the more general to the more specific. Inductive reasoning works from the other way, moving from specific observations to broader generalizations and theories.
Inductive reasoning, by its nature, is more exploratory, especially at the beginning. Deductive reasoning is more narrow and is concerned with testing or confirming hypotheses.

Most of our Civil War research is inductive, meaning we have questions about the physical nature of an article or clothing or item or gear; or the way or manner in which something was done, etc. Often times, in our research we use both methods. For example, we explore original blouses for raw materials; methods of production; patterns, styles, and sizes; and markings. Then we take that and look to apply to how and when and under what circumstances were these blouse issued so we can transfer that to our individual impressions/personae for specific units at specific times and at specific places.
Used together, we use both in a methodology that continually cycles and recycles from theories, down to observations, and back up again to theories.

To borrow from the world of statistics, what we do when we “research” is that we examine artifacts and primary sources to “measure” our theories, to confirm or deny our observations, or to test whether the inferences and applications made from our discoveries or observations are “reliable and valid.”
Elaborate formulae exist in clinical and social research realms for computing how large a sampling has to be to “reliable and valid.” In “Civil War” related research these formulae rarely apply, but the principles still do. Meaning, for example, the rate (fraction, proportion, or percent) at which something is found to occur is a more valid measurement than a simple numerical count of occurrences. An observation or finding is a “valid” measure of something if it is relative or appropriate as representative of that property. “Reliability” only says that the result or observation is repeatable.

In plainer terms, take jaguar skin trousers. Based upon a surviving period photograph, we know of them being worn (under one set of circumstances, at one time, and at one place). A “sample pool” of one IS valid if our theory is that jaguar skins were worn during the Civil War. It IS NOT valid if the theory is that all Confederate soldiers wore jaguar skin trousers. Is not reliable, in that the incidences of documented jaguar skin trousers are not repeatable. (This translates out, in applications for impressions, that if we were “doing” that man, at the that time and place, in that studio, before that photographer- that jaguar skin trousers may be appropriate as researched and documented.)
However, taking that question or theory further, may lead to further research and investigation, etc., etc.

And, once having “documented” (proving or disproving the question or theory), that becomes a “hinge point” around which our collective Civil War knowledge rests- until additional research and documentation comes to light that forces us to reevaluate and reexamine our beliefs about life in the Civil War past.