Finding a Research Question

Last updated on January 13, 2023

Deciding what to research can often be the hardest challenge for students. Often they know generally what they want to do but struggle to figure out exactly what their question is. For example, they might know that they want to do something on gender inequality in family but find that every question they can think of has already been answered. This is especially hard for new scholars because they have yet to really learn the literature and know what has been studied and what is already pretty well established.

Here are a few suggestions for how to start this:

Find the Question in the Literature

1. Go to the Annual Review of [Your Discipline] (e.g. Annual Review of Sociology) and find as many recent articles as you can on your topic.
2. If you don’t find much recent there (recent meaning maybe the past 5-7 years), do a similar search using your library’s data base, or even Google Scholar, and pull out maybe 10 or so articles that sound interesting to you, the more recent the better.
3. Starting with the most recent review article, go to the end of each article and look for a section, a paragraph, or even a sentence that says something along the lines of “Future research needs to blah, blah, blah…” That is exactly how I found the topic for my dissertation. When I read a George Farkas article that ended by saying something to the effect of “no one knows when this thing starts,” I decided to find out when that thing started.

Find the Question in the Data

This technique is helpful if you want to do a quantitative analyses with data that has already been collected by a national statistical agency like Statistics Canada or the U.S. Census Bureau.

1. Download the questionnaire or code book and look at all the variables in the list. Do any look interesting to you? Find a few variables that you’re curious about and check to make sure that they have a lot of missing data or are weird.
2. Do a lit review like above, but this time use the name of the data set as your key word. See what others have used the data for and if you can ask a new question that no one else has looked at.

Talk to People

Often times the first person we might ask would be the professors in your classes or your research supervisor. But there are lots of other people you could ask for suggestions:

1. Email the author of an article you find interesting. I did this once for a different kind of question, to the hugely famous Anthony Giddens in my first year of graduate school, not realizing that he was a big deal… like a consultant to presidents and prime ministers big deal. My office mate made fun of me, but Giddens actually sent me back a really nice but short email saying he didn’t have an answer to my question. I once emailed a far less famous academic after a conference I attended where I was interested in her stuff and asked for advice, and she never replied. I too have received some emails I’ve replied to and sometimes when I’m really swamped and just honestly don’t have the time. So, try not to be hurt or offended if they don’t reply.
2. Talk to other more advanced graduate students. Other graduate students can be a huge resource but I think are probably pretty underutilized. I know when I was in grad school, I often thought I had to prove I knew as much as them, so I didn’t always seek their help. But advanced graduate students are in the process of completing or just completed a number of major literature reviews for their comprehensive exams and/or dissertation proposals, so the literature is likely fresh in their minds and they may have a lot of wisdom as to where the gaps are. They also have probably been experiencing a lot of grilling about their weaknesses through these various defenses, so they might welcome someone seeing them as a person who knows something rather than as a person who needs to know more.
3. Talk to your friends, family, wait staff in restaurants, people on the street, your kids’ daycare providers, your Uber driver or anyone else out there. Pay attention to debates you get into or discussions where people are making claims that may or may not be valid. Is it true that breastfeeding is the most important thing ever, as your kids’ babysitter says? Is Uber really helping to combat traffic congestion as your Uber driver tells you? Did you overhear your waitress complain to a coworker about their sexist manager? Maybe these are questions you want to explore more. Come up with an answer to your know-it-all uncle, so next Thanksgiving you can begin the annual family feud with the phrase “Well, actually…”

Borrow a Research Question

If all else fails, or even if not, you can also borrow a research question from your supervisor. This is actually, extremely common in the hard sciences when someone is part of a lab, I believe. In those fields, students tend to work on research that is very closely aligned with their professors, often leading to co-authored publications. In the social sciences, however, students often will carry out projects further afield from their professors’ work. For instance, I have supervised (or am currently supervising) students for the following topics, among others: An ethnography of the Nova Scotia employment assistance office by a student who worked there in a co-op program; a quantitative analysis comparing the gender wage gap among Canadian military members and the general Canadian population; a qualitative study on women professors’ fashion choices; a content analysis of body image portrayals on Instagram; and a theoretical analysis of how BDSM play can be feminist. Only two students of mine have ever specifically looked at breastfeeding (and one left our program, so that was just for a year), although the vast majority of my published research in the last 10 years has been on breastfeeding. I have at least 5 questions popping around in my head at any moment on breastfeeding or gender inequality in general that I’d happily hand over to a student to use for their own study. If I did a lot of work on it for turning it into a publication after a defense, I’d probably expect to be a second author on the paper. But if you do most of the work both during the program and for turning it into a publication, just a thanks in the acknowledgements would be appropriate.

I actually did this for my MA thesis when I was delightfully surprised to find out that I was pregnant in my first year. I decided I wanted to finish quickly before the baby came because I had no maternity leave and was worried I might not come back if I didn’t finish my MA first. So, my then supervisor said, how about you use these data I collected with a student a few years ago and look at gender differences in organizational commitment. “Done!” and I ended up defending a month before the baby was born, a semester early. Thankfully, I kept on going though, although the findings of the study were kind of weak so I never went on to publish it.

Remember, you don’t have to do this all alone. Yes, it’s good to find a question that you are interested in enough to sustain you and drive you to keep going when you’re tired of getting feedback and just want to finish. But, often times you just need to get going; eventually, you’ll have more research questions than you’ll know what to do with.