All history research begins with a question that you want to answer about the past. It can be a big question or a small question. Unlike the hard sciences, replicating the past event is impossible. You can't have a lab setting that controls for variables so predicting the future or creating laws or rules is impossible. The "test" if you will, has already taken place. And it cannot ever happen again.

Most likely, your question has already been studied in some way. Or maybe some aspect or many aspects of the question has been studied before by other historians. The historian should try to read the most important works on the question before or during their research process to see what people have already said about the question and how that their answers have changed over time. It might also be necessary to read some social science theory to inform why historical actors behaved in certain ways. This might be sociological or psychological theories just to name a couple.

Once the person has a pretty good sense of how other people have answered the question and how interpretations of the question have changed over time, in an ideal world (rarely if ever is this possible), a historian attempts to gather ALL of the relevant sources that seem to answer their research question. Primary sources are the data that historians use to learn what happened in the past. For some questions, there may be archives and archives full of primary sources. For other questions, there may be just a handful. It might be a written source or it might be a recording/photograph or it might be some sort of object that was created at the time in question or it might a piece of fine art (just to name a few). The historian reads/watches/views/listens to those sources, attempting to understand what each source tells us about our research question, why the source was created, what biases it may contain, and what context informed its creator’s thinking.

At the same time, the historian should investigate other factors at play in the past that might help them answer the research question: the unique set of circumstances that made the event possible, the long and short term lead-up to an event, the cultural understandings from the time period, the social beliefs that people held, and the personality characteristics of individuals living in the period. In a sense, the historian attempts to leave the present world behind. They try to get into the minds of the people of the time: to know what they knew, to believe what they believed, to think what they thought, to be a member of the groups that they were a part of, and to react the way they reacted based on what they knew at the time.

Finally, a historian should answer their question and tell the world about their findings. They should publicly present the conclusions for inspection by their peers, who check to see if the author is correct. The books and articles that they publish include the evidence that they found that supports their conclusion. They often leave footnotes that point the public to where the historian found a particular primary source or what historian they have read that helped them answer the question. This allows others in a sense, to check the data to see if the answers that the historian came up with are correct and are based on good “data.” It also allows other historians to find those sources again or use them in their own projects. This is as close to “scientific” history as a historian can get.

It all sounds so simple on paper. The trouble is---as many theorists have pointed out---that it is very difficult to do this perfectly...basically impossible for various political, social, cultural, and material reasons. For example, maybe some of the key documents that would answer your question did not survive to the present (although there are some niffty workarounds that historians have come up with that sometimes help historians get around this problem). Or maybe a question can’t be answered because the people you want to study didn’t create written sources at all and the people in power didn’t think they were important enough to create sources about them (e.g. the life stories of most slaves). Or maybe the source is false or twisted in some way. For example, a slavemaster might have been called to court for how he treated his slaves. In order to make himself look good, he might leave out key details about a slave’s inhumane treatment that might be very important/interesting for answering your question about the day-to-day life of slaves. Another problem is that we as historians are biased in innumerable ways. We have ideas and conceptions about the world that frame the way we interpret things. One example on Reddit I see a lot is that we think that Europeans were “advanced” while indigenous peoples of the Americas were “behind” Europeans. So the OP asks, why were the Native Americans so far behind? Well just by asking the question in that way, we trapped ourselves into certain stories and certain biases. Our answer automatically assumes that Native Americans were behind in some way. We have subconsciously reached a conclusion before we have weighed the evidence to see if it is really true. Because of the way we asked the question, we might ignore evidence that might change our opinion or we might exaggerate the importance of evidence that actually didn’t matter that much.

That’s why historians are constantly rewriting the past. We ask new questions. We find new primary sources. We try to include more people into our stories. We think of a new method to get around a lack of sources. We learn new things which makes us ask new questions in different ways. So it’s a job that keeps going and going and going…