How do you generate original research ideas?

[wb3060]

I have completed my first year in a PhD program (in the social sciences - 2 more years of classes to go), and I have this dilemma. I did really well my first year. I excelled in all my classes, impressed a few professors, and developed one really good relationship in particular as a research assistant. 

 

My problem (which I see as pretty critical as a PhD student) is that I really struggle to come up with original research ideas for papers, my eventual dissertation, or potential journal articles. 

 

People always say read the literature... which I am. I read as much as I can, but I have about a million demands on my time (classes, RA and TA work, etc.) and so I can only do so much. 

 

What can I do to get those creative juices flowing? How do you guys come up with good ideas for your papers? 

I feel like whenever I think of something, it has already been done or I think of all the flaws of the idea. 

Edited August 12, 2015 by wb3060

[PoliticalOrder]

A popular myth many people have is that academia is almost always completely original. 99% of literature is some kind of modification or different angle of some other research. No one is expecting a junior scholar or a Ph.D. student to shake the world with their research. The best thing you can do is just take some theories from your discipline and either slightly modify them or test them in some different sub-field. 

 

And when people say ''read the literature'' that is exactly what they are getting at. By reading current debates and old theories in your discipline you will begin to see gaps where no research exists, those gaps are where your research comes in. You should be actively looking for these gaps, or certain theoretical positions that are in major works that haven't necessarily been testing in depth. 

 

It's perfectly normal to piggyback off of famous scholar's ideas that they never actually extensively pursued. 

 

About the flaws dilemma, this is a major problem with grad students. They spend so much time critiquing everything that they find it hard to create something themselves. Research is always flawed, yes you need to be able to see potential flaws to try to mitigate them by using certain approaches but you also need to be able to just create something and continue to improve it, not always doubt yourself. 

Edited August 12, 2015 by PoliticalOrder

[wb3060]

Thanks so much for your advice.

I think you're right. Part of the problem is my mindset. Having only completed a year, I've spent the last two semesters reading all the really famous works and big theories in the field. And then I think, how could I ever come up with something like that. Maybe part of my problem is not reading enough newer stuff yet. 

 

I have actually identified one area that seems to have a gap and is just under-theorized in general. My problem is that I have yet to come up with a good theory. However, this was a fairly recent discovery and I definitely need more familiarity with that literature, so I'm still holding out hope that I could come up with something good there once I pursue it a little more. 

 

 

Also I think another part of my problem is I find so many things interesting within my (broad) sub-field, but I don't have a single area that I find much more interesting than the rest. I think I could be happy doing work in a variety of areas, but since I haven't narrowed it down (although I have some since I began grad school) to something more specific, I keep bouncing from literature to literature instead of just focusing on one. I think this probably is making it harder for me to come up with ideas as well. 

Edited August 12, 2015 by wb3060

[rising_star]

I would say that, in the social sciences, it's pretty rare for PhD students to develop a whole new theory in their dissertation. That might, through a case study, show how multiple theories could be brought together to better understand a particular area (and their case study within that area). If you're having trouble narrowing things down in your subfield, have you thought about doing a project that integrates aspects from multiple narrow areas? My own dissertation did that, bringing together a few things that aren't typically looked at together or in juxtaposition with one another.

 

To get a sense of the scope of what you're expected to do, take a look at recent dissertations in your department and in your discipline as a whole (the ProQuest database is a good place to start). Even skimming their abstracts and introductions should help you get a sense of what it is you're being asked to do once you're ABD.

[mb712]

I'm not exactly seasoned, I start my Ph.D. program this month, but have an idea of what I want to focus my research on so here's my advice:

1) Listen to the people above.

2) Build off of an existing theory and don't try to pull something out of thin air. The research I've started is "original" (it's not a novel idea [is anything?] but it's something that hasn't been published on) and the way I arrived at the topic was combining a theory from a different subfield with a way to respond to cynicism about my subfield from a graduate student in my field. Bizarre, but true. It snowballed into some hypotheses, a research proposal, and now some interesting results that I think I can build off of for my graduate career. Soak in what people are you are talking about, in and out of the classroom. You never know when somebody will say something that gives you an epiphany moment. 

[TakeruK]

You and the other posters are from a social science field but I think what has been said applies to other fields too! 

I also struggle with some of the same issues and feel that I can't come up with original ideas. In the last few months, I've talked more with the postdocs in my field and observed how they come up with their research topics. And I'm glad to see that very experienced people (e.g. rising_star) have said things that alleviated some of my concerns: you don't have to come up with whole new theories and ideas in grad school! 

I have noticed that junior researchers (e.g. senior grad students & postdocs) generally come up with new project ideas by reading the literature and learning about all the different techniques and what has been done so far. Sometimes you will be able to think of a way to use a technique or a solution from one problem to apply to another. I find that going to conferences and listening to talks, especially those that aren't exactly in your field (but related) is very helpful to get new thoughts/ideas flowing. Sometimes the first idea you think of isn't very good and you'll find lots of flaws but discussing or thinking about it further instead of giving up can lead to something else cool. 

I also find it helps to discuss your ideas with other people, but as we have seen in some other threads on TheGradCafe, in some departments, this leads to your colleagues stealing your ideas. So I guess you have to be careful in picking your collaborators/people to bounce ideas off. 

And finally, reading the literature is time consuming and hard! I know it's hard to make time for it with other obligations that are more immediately demanding, such as TA and RA work, but the way I see it, reading the literature is as much of your "work" as a grad student as TA and RA work is. So, assign it a priority and make time for it, as you would other priorities/commitments. There are also other things you can do to make reading more effective (sometimes I find it really dry and boring and get unmotivated). I prefer creating small informal reading groups and discussing papers with other people. I find that the social pressure of actually having read some material and contributing intelligent thoughts does a good job of making sure I actually read and think about the papers!

[rising_star]

 

And finally, reading the literature is time consuming and hard! I know it's hard to make time for it with other obligations that are more immediately demanding, such as TA and RA work, but the way I see it, reading the literature is as much of your "work" as a grad student as TA and RA work is. So, assign it a priority and make time for it, as you would other priorities/commitments. There are also other things you can do to make reading more effective (sometimes I find it really dry and boring and get unmotivated). I prefer creating small informal reading groups and discussing papers with other people. I find that the social pressure of actually having read some material and contributing intelligent thoughts does a good job of making sure I actually read and think about the papers!

I second this completely. You have to read. Also, when professors tell you that you'll never have as much time to read as you do right now, they're not lying. I used to think they were but, I've since learned otherwise. One thing I did while ABD that worked really well was starting my day by reading one journal article that I hadn't previously read. By the time I was ABD, I was reading things that were directly relevant to the dissertation research I was doing and pretty much all of those articles got cited in my dissertation. I would start reading it while eating breakfast and then make sure to finish it before leaving the house. If you can find 30 minutes, you can find the time to at least skim 1 article or book chapter a day. And really, to be successful, that's the minimum you'll need to do.

[bhr]

Read AND observe. If you are in a social science, I assume there is some sort of practicum or field experience component. Take the theories/ideas you have been learning, and look for a new way to apply them, or a group that hasn't been studied using the same procedure. Your methodology doesn't need to be novel, it just needs to work for your purposes.

I'm a big believer in looking for an exigency and trying to find a way to address it. I have a friend who is working with medical anthropology. He's taking work done by another scholar, and applying it to a specific population that wasn't part of the original study. This gives him unique conclusions, without having to defend the theory or methodology as much. (plus, since it's a respected idea in his field, he found it easier to get his original case study published.)

[Lucas at Thinksocialtheory]

Great post bhr. I did Sociology at university and after being familiarised with social theory/theories I had a resource that I could draw upon when it came to doing research projects. For example after learning about Foucault's theory of panopticon I was able to apply that and do a research project which looked at how the panopticon functioned in the call centre that I worked at. Overall it is definitely worth familiarising yourself with some of the key theoretical ideas in sociology as this gives you a theoretical toolkit which you can draw upon regardless of what area of social life you decide to research. In light of this I'd definitely recommend the book "Key Sociological Thinkers" by Rob Stones as this is a great introduction to the contributions of key sociological thinkers have made to studying society, this book is in most university libraries too.

Edited September 2, 2015 by TakeruK
removed link

[rising_star]

I think you should read journal from your niche, not books. You won't find anything new or original there. All specific literature publish only proven theories, issues and ideas. In my opinion specific journals from your field are less conservative and have more fresh ideas, opinions and projects to discuss.

I actually disagree with this. In the social sciences, many of the foundational theories are found in books, not articles, so if you want to read most social theory, you have to read books. You'll find articles which use the theory but, you really should try to read up on the theory itself at some point. Also, some journals are more conservative than others, which you'll discover by reading articles in them and talking to others about the journal. In my area of the social sciences, there are several journals which explicitly seek to publish "radical" or "nontraditional" scholarship while others hew much more closely to the traditions of the discipline which are over a century old at this point. Those journals don't contain the fresh ideas and opinions that Miraly is trying to get you to find. Knowing the differences between various journals will be an important part of your development as a young academic.

[TakeruK]

I actually disagree with this. In the social sciences, many of the foundational theories are found in books, not articles, so if you want to read most social theory, you have to read books. You'll find articles which use the theory but, you really should try to read up on the theory itself at some point. Also, some journals are more conservative than others, which you'll discover by reading articles in them and talking to others about the journal. In my area of the social sciences, there are several journals which explicitly seek to publish "radical" or "nontraditional" scholarship while others hew much more closely to the traditions of the discipline which are over a century old at this point. Those journals don't contain the fresh ideas and opinions that Miraly is trying to get you to find. Knowing the differences between various journals will be an important part of your development as a young academic.

That's interesting. Maybe Miraly is from a physical science field? Most books in my field are just super extended review articles that summarize past knowledge. The only scientific books I've ever read are from my introductory classes and although one can still learn a lot from them, since the newest books can summarize e.g. new scholarship from 2005-2015 which would be very useful. When scientists come up with new theories and ideas, they are published in articles. When these ideas have enough evidence behind them and the community agrees that they are likely to be correct, then they appear in books.

For example, if you read the latest book in planet formation (published in 2010), you will get the wrong impression of the current community's thinking on planet formation. This 2010 book still writes about ideas from the 1990s and while they do a good job of summarizing the models generated in the 1990s and updated in the 2000s and they also do a very good job at pointing out the current puzzles and mysteries, you will not get a full picture of where the field is at. 

In the 2000s, when the book was being written, new ideas to overcome these problems were theorized. But at that point, these ideas are not yet embraced by the community. So they didn't go into the book. An important paper in 2012 was the one that presented this idea with some evidence in a way that really excited the community. That was the start of this new piece being added to our model of planet formation and since then, that's where the community is at.

That said, I still agree with rising_star that the start of your literature search should be books and review articles of past knowledge. For someone looking to solve problems in planet formation, it is essential to read that 2010 book I mentioned and to know how the field got to where it is today. Also, it will help you understand what the major problems are and why this new 2012 idea is required to solve some of these problems. And it will help you understand what problems are still outstanding. Otherwise, you will waste a lot of time reinventing the wheel.

And I also agree with Miraly that (in the physical sciences), after reading the books and review articles, in order to really know where the community is at and what the latest knowledge is, you must read the niche journal articles. You will rarely see new ideas presented for the first time in books. In my field, I guess journals are not very conservative, because basically all theories can be published as long as they are logically sound. A common expression is that a theorist only needs to be right 1 in 20 times! In fact, most journals in my field will not publish anything that is just rehash of old ideas: you must always include something novel in order to merit publication. 

Overall, definitely agree with rising_star too that we need to learn the norms of our field and the differences between each journal. But also note that sometimes these vary because Editors-in-Chief change over time!

[rising_star]

That's interesting. Maybe Miraly is from a physical science field? Most books in my field are just super extended review articles that summarize past knowledge. The only scientific books I've ever read are from my introductory classes and although one can still learn a lot from them, since the newest books can summarize e.g. new scholarship from 2005-2015 which would be very useful. When scientists come up with new theories and ideas, they are published in articles. When these ideas have enough evidence behind them and the community agrees that they are likely to be correct, then they appear in books.

For example, if you read the latest book in planet formation (published in 2010), you will get the wrong impression of the current community's thinking on planet formation. This 2010 book still writes about ideas from the 1990s and while they do a good job of summarizing the models generated in the 1990s and updated in the 2000s and they also do a very good job at pointing out the current puzzles and mysteries, you will not get a full picture of where the field is at. 

What I meant is that the foundation texts in social theory are books. I mean, the work of Max Weber, Karl Marx, Edward Said, Karl Polanyi, Frantz Fanon, Spivak, Stuart Hall, Immanuel Wallerstein, etc. are found in books. While there have been riffs on and texts about the shortcomings of these texts published in journals, it's important to read these theoretical works for yourself. It may not be the best way to generate ideas for new research but, it is important for understanding how the field got to this place. 

Also, in many social science disciplines (anthropology and sociology come to mind, but this is also true of history and many MLA fields), books are a requirement for tenure. So even though early-career scholars do some publishing beforehand, the book must contain substantial new material that hasn't been published elsewhere or it won't be published by reputable academic presses. Consequently, 1-3 chapters may have appeared elsewhere (as book chapters or in peer-reviewed journals) but a lot of the material will not have been published elsewhere and thus would otherwise be inaccessible. For example, there are a couple of anthropologists and sociologists whose work greatly informed mine in the first two years of my PhD. None of these people had the substantial ideas that I was working from then in their journal articles since they'd saved the juiciest materials for their books.

I recognize this as a key difference in scholarship and publishing across the disciplines. I was tailoring my advice earlier in this thread (including my reply to Miraly) to the OP's statement that they are in a social science PhD program. Had they said they were in the natural/physical/life sciences, I would've said to focus on journal articles, especially things published in the Annual Reviews and not mentioned books at all.

[GeoDUDE!]

It may be for different fields, but I feel like if you are having difficulty coming up with original research, you are not reading, talking and collaborating enough. 

For me it flipped, first I had trouble coming up with ideas except for the projects handed to me by my various advisors, then all of a sudden I had read enough literature to start seeing holes myself. But what has been more helpful is talking to people about their projects, and then when you have an interesting question, it keeps rolling into a project. At this point, I have too many problems I want to work on, and am finding it harder and harder to work on dissertation, despite how interesting I found it when I initially came up with it. I've seen it first hand.

I think a lot of students think that most ideas from professors are thought of alone... but how many single authorship papers do you see? Great studies are difficult to do alone, unless your Albert Einstein. As you become a more senior researcher, collaborations happen naturally. I'm entering my 4th year of graduate school and it just started happening for me without pushing. I give advice and some code to people, and "i've just earned myself a spot on their paper". Because my name is on their paper, they don't mind that I get the inside info of their results and do a further work study. And guess what, their name is on that paper. 

This might be a science/modeling thing, but I've noticed it for other people to. 

I am not worried about people stealing my ideas (except my dissertation) because honestly, if you are smart enough to come up with one, you can come up with another!

Edited September 16, 2015 by GeoDUDE!

[knp]

I do like starting with articles, sometimes, because I can crank through a lot more topics a lot faster that way. (Yes, it's slower to read a journal article than a book chapter, but the slowest of all for me are book introductions, and those are usually unskippable.) Then if I like what I'm seeing, I go find the author's book(s).

Second the idea that reading the newest stuff, rather than classics, is very helpful for generating new ideas.

I always, always take notes, too. They don't have to be formal. (Mine are not. "wood engraving insulted as 'mechanical,' vs. new more naturalistic copper/steel engraving (p. 85). huh! v interesting"). I'll usually think of a couple comments I'd make on a paper, like, "this is a great analysis of group A, I wonder whether anything similar happened with group B". If I write them down, I have them! (I always write them in a digitally searchable form, but YMMV). If I do not write those thoughts down, I forget and they fly away as soon as I'm distracted by something else, and I never actually look into whether that theme appears with group B. Practicing finding this sort of "but what about Z" question, and then preserving that question in my notes, has been key for me in my academic life so far, and I expect it to continue to be so.