No one instinctively knows how to do research. Just like learning to read or write, there are a certain set of skills which must be acquired in order to do research. When I was in high school, some of the classes I took had requirements to do research papers. I suppose the purpose of those assignments was to teach us how to do research. Unfortunately, no one bothered to explain why we would want to do research in the first place. In addition, the skills they taught, like making note cards, seemed absolutely worthless. I could never understand what I was supposed to do with the note cards and why I was wasting my time writing out cards when I was going to have re-write everything anyway when I wrote the paper.
One notable research effort involved and assignment to research a battle of the U.S. Civil War. At the time I was in high school, you would be absolutely amazed at the total lack of reference material about the Civil War in the school and city libraries. I finally ended up using an article from the Encyclopedia Americana. Fast forward. If one of my grandchildren wrote a report and used the equivalent, Wikipedia, they would be marked down and probably fail. Part of my problem back then was understanding the concept of research. What is unfortunate today is that many of the people I work with in finding their ancestors have the same lack of understanding of the research process that I did in high school. They are still at the level of copying an encyclopedia article.
As I found out over the intervening years, the term "research" is applied to widely divergent activities. I think it is sometimes easier to define what is not research than what is research. In the genealogical context, copying names out of an online user contributed family tree is definitely not research.
In genealogy, it is common to talk about the "research cycle" involving anywhere from five to sixteen or more steps. But before going into lists of "steps," it might be a good idea to have definition of research and what you are trying to accomplish before setting out on the journey through the steps of the cycle. Implicit in the concept of doing research is selecting a hypothesis or specific question to answer. Some people refer to "doing research" as if it were an abstract concept. But whatever the definition, research is an activity that requires results, positive or negative, but results.
Looking at what others have already done i.e. the online user submitted files, is a part of what is usually called the survey step. This is what you do before you actually do any real research. Basically you need to know what has already been done. There is however a classic quote from one of the most famous American genealogists, Donald Lines Jacobus. He said, "Develop a healthy skepticism. Accept nothing unreservedly until proven." (Sounds like he would have done well in the 1960s). Jacobus, Donald Lines. Genealogy As Pastime and Profession. Baltimore: Genealogical Pub. Co, 1968. Electronic version from Google eBooks.
I spent much of my life doing "legal research." The main purpose of this activity was to provide case law support for legal arguments made in court, mostly in the form of written documents. Although we use the term "research" in referring to what I did, the activity consists almost entirely of quotes to support your previously chosen position (or that of your clients). In legal research, the correct answer supports your clients' position. To this extent, legal research is not really research at all, but merely a selective survey.
I once asked an attorney who was much senior to me in age and experience about a difficult legal question. He asked me in return, "How do you want the court to rule?" I told him what I thought and he said, "Then that is the law." In other words, I understood this to mean that by advocating a position, I could influence how the court ruled and thereby make the "law" under our English law derived legal system. Think of chaos in genealogy if genealogists had the same system of research and proof? You want to be related to Charlemagne? Just wait a minute until I come up with my version of your pedigree!
In the case of scientific research it is imperative that the study comprising the research, be based on answering a specific question or proving or disproving a hypothesis of some kind. Genealogy is usually falls somewhere between the legal and scientific worlds. Although, I have found few genealogical researchers that had focused on their subject matter long enough to produce a coherent hypothesis, in the abstract and unlike attorneys, genealogists are trying to find out information they do not already know. But in some cases genealogists use "research" as they call it, to justify their belief that they should be a member of a lineage society or other such thing, just like attorneys doing legal "research."
So what is research? (You might notice that I am repeating the question for emphasis and not senility). Basically, research is finding out something you don't already know. It is not justifying a position or belief you already have.
Hmm. Looks like I probably have more to say about this subject.