Picking a Research Question

Starting a personal research program can be a daunting task. For many (most?) scientists, this is the most difficult part. Academia is littered with the corpses of brilliant PhDs who graduated, got a job and then couldn't establish their own personal research agenda. Of course, with Citizen Sky we aren't looking for a career-long research program. However, even coming up with somewhat modest research ideas can be challenging.

Dr. Doug Welch gave a very good talk at the 2nd Citizen Sky workshop titled "Picking Topics for Scientific Investigation". Unfortunately, the California Academy of Sciences lost the video they recorded from that day of the workshop. Doug has provided the powerpoint slides of his talk which can be downloaded here. (We hope that he can record an audio narration of the slides in the future.) The slides are descriptive even without the accompanying video so I recommend viewing them, especially the last four slides. 

I'd like to focus a bit on the issue of determining a research question, which is your guiding light throughout the process. The question sets your starting point, direction and boundaries along the path. There is no one best way to determine your research question. But here are some common ways people come up with their own Eureka moments:

• Serendipity: Sometimes an idea pops up related to another idea proposed by someone else. It can be an extension of the original idea, a new take on the idea or even a refutation of it. But the genesis is related to an outside idea. If you work like this, then the best thing to do is seed yourself with as many outside ideas as possible. Just keep reading and following others until something catches your eye.
• Personal Interest: It is very important that you are passionate about your research question. This is true in all of science but is especially true in citizen science. So think about subjects that you like to read about or observing programs you look forward to participating in. Then think about questions, challenges, issues in those area that have always annoyed you or caused you to wonder about. Try to come up with a question that you personally would love to see answered.
• Targets of Opportunity: Sometimes things just drop in your lap. Maybe you have access to a unique piece of equipment or observing location that few others can replicate. Or maybe you read about a newly discovered object/idea that needs to be tested/observed and no one else has started it yet. These are opportunities to do some type of unique experiment, which often leads to cool results. Ask yourself "What can I do that no one else can?".
• Brute Force: Those of us with short attention spans prefer to just dive in. You can begin with just a general idea and start collecting tons of data and look for some signal in the noise. For example, you could say "I like flare stars" and decide to take a ton of high-cadence, multi-color photometry on a few flare stars every night, all night long for a month. Then look at the data to see if anything is there. This is a very inefficient method, but it can be the most fun if you are the type who prefer to get their hands dirty. If you want to sound fancy, you can refer to this method as a type of exploratory data analysis.

These methods to develop a research question are not mutually exclusive. In fact, they almost always overlap in some way.

The research question itself is usually written to be as narrow as possible. They also should not be  yes/no, either/or questions. A poor question is "What is the precision of visual variable star observations?". An entire book can be written on that topic. A better question is "How is the precision of visual variable star observations by amateur astronomers related to properties of the stars being observed?" or ""How has the precision precision of visual variable star observations by amateur astronomers changed over time?".

Research questions will change over time. As you learn more about the subject you may find the original path has already been taken, no longer makes sense, or just doesn't interest you anymore. No harm, no foul! Changing and even throwing out research questions is part of the game. The professional goal is quality research that drives the field. And hopefully your goal is personal growth and enjoyment in the process. So adjust the goalposts as needed.

For me, writing a good research question is the hardest part of the scientific method. It is frustrating because it forces me to place constraints on vague ideas that I am excited about. However, it is a very valuable exercise in discipline and will save you a ton of pain later on.

If you have any ideas that you'd like feedback on, feel free to post them here. Remember, DocArne provided a long list of interesting variable stars here. And we have some other ideas here. Remember a Citizen Sky team does not need to focus on a purely scientific research project. It can address any part of the scientific method. And you don't have to work on a  team with others. Solo projects are also possible.

A formal call for submissions for a dedicated issue of the Journal of the AAVSO will be made in the next couple of months with a likely deadline of spring 2012. So now is a time to join a team or start a team!