I got my data, now I’m working to prepare it for analysis.
And I have to admit, the step I describe here should be done before getting the data itself. The only reason I reversed it is because I already used this date to publish the scientific work.
Raw data are never straight walk towards the amazing conclusions. They need to be double checked for possible errors before one even starts thinking of analysis.
Because I have limited computing resources, I have to thread one small step at the time.
The first step is establishing a base-level of what I wish to observe. I need to make sure that observed increase in oscillations does not happen when there are no registered flares. Because if it does, my whole hypothesis is wrong.
Yes, the basis of the scientific research is looking for the things that might disprove your hypothesis. Not vice versa.
So, now, I am using helioviewer.com to look at what was going on the Sun during the chosen day.
Luckily, this day has 2 and a half hour period where there were no flares observed. Great, that part is solved.
Then I have to decide which areas I wish to observe. There are in total 23 active regions registered for that day. I will start with quiet Sun region at the center of the disk and then slowly go through all other registered active regions for that day. I must include all of them because I do not have a good logical reason to remove any of them.
And that inclusion has to happen if I wish to avoid effects of non-random picking of data; basically, loggical fallacy ‘cherrypicking.’ I cannot do anything that would influence my result, and picking just a few of active regions might do it. So I have to use all registered regions because only then I will know that result I got is a real one and not some distorted image of the truth.
Yes, this approach increases the amount of work I will have to do. But hey, let be honest. I am in pursuit of the truth, and not some imaginary answer that will fit my hypothesis.
This last part, this need to make sure that I did not do anything to influence result is something graduate students learn during their studies.
I remember experiencing unpleasant surprise when I saw my fellow students get cut into pieces during an initial presentation of their project. We all had to present the project we’re working on, explain what we will do. And the first batch of students that got to do that was simply brutally cut down by questions of the audience, the audience consisting of older students, mentors, collaborators, and established scientists. The same people who were extremely friendly and sociable outside that same conference room.
During that first presentations, I did not understand why other older students and scientist were so nasty. And it took me years to learn that is how science works. One scientist presents work, others do whatever they can to find holes in that work.
And, believe or not, this behavior is a favor to the scientist who presents the work. Such nitpicking is what helps scientist see the problem from all angles and recognize their own mistakes and fix them before going public with the work. If this step is missing, and they publish the work with mistakes, their reputation as good scientist suffers. And with it, opportunities for future work.