Fairly new to research

[MKarmoose]

Hello all,

I'm sorry if there is some sort of redundancy in this topic!

I'm fairly new to research. I recently joined a research group in Cognitive Radio Networks (I'm an Electrical Engineering grad student). I was asked to get familiar with the topics at first, and I started reading a lot. I have a very good background about the topic now from a very broad perspective. But now I need to find myself a good research point, and this I don't know how to do. I got me some more specific subtopic under the large umbrella of Cognitive Radio Networks which I found to be appealing to me, and I started reading papers that tackles this point in specific, and now I'm stuck at this phase. So my questions now are:

1- When do I stop reading? I've reading for two weeks now with a total of 15 papers or so, all regarding this subtopic I've chosen. Is this enough?

2- How do I get research ideas? I always try to question the work in each of the papers I read, but then I don't actually come up with something very valuable.

3- One thing I'm trying to do is to make a tree of the current work in the topic I'm investigating. This can help me find some sort of gaps in the field which can eventually highlight some opportunities for research. However, this tree requires much much more reading and investigating, which again brings us to the first question.

Thanks

[fuzzylogician]

Maybe this is a good time to talk to your advisor, or to older students or postdocs in your lab. Explain to them why you are interested in the topic you chose and ask for advice how to proceed. Maybe they can point you to other papers, or point out open issues that they find interesting or have thought about in the past. One way to find a more concrete topic without talking to anyone is to look for recent conference/workshop presentation on your topic. Look at abstracts that were accepted to get an idea of what people are currently working on. It's good to try and question what you read - be critical of the actual question being asked, the methodology pursued in trying to answer the questions, the interpretation of the data. Try and sketch that out in details for the papers you think are good and convincing, and ask yourself why that is. That may get you somewhere too.

Re: number of papers, I don't know. For my field it's not a lot - but do you mean actually read in detail (understand everything from design to interpretation of results and everything in between) or skim for the general ideas and arguments? For your stage in your grad program, I imagine you skimmed a lot of paper and read in detail only the few you found exciting -- which is what I think you should be doing. But then, you could spend some more time digging up both seminal work and the most current work to see if you can find a more detailed question that you are interested in.

[Dal PhDer]

fuzzy has given you some really great advice!

For me, the first few steps in generating a research project are as followed

1) I read a few papers in the area that I am interested in...if you do a good broad search, 15 may be enough- this is really discipline dependent.

2) Once I know the 'area' that I am interested in, and have my papers, I look at the discussion section and see what kinds of limitations these studies faced or listing as gaps in the current knowledge. This helps me see what the current issues/gaps are in this area.

3) once I know the gaps and have more detail in the area that I am interested in, I can start forming a question.

For me, it was helpful to have a question to take to my supervisor to begin with. It kind of gives you a working point to start off of.

Key things I think about when I read the literature and think about my question are: what will be my dvs and ivs, what is my population of interest, is there a specific methodology/method I want to use, what kind of data might I need to answer my question, who might I need to work with..etc. Figuring out these kinds of things will help you narrow your question.

IMO, I think it's better to to a broad lit search to help you figure out your question. Once you have your question down, then it's time to really dive deep into the literature. I have recently changed my dissertation topic, and came up with a new question and objectives. I have written a 7 page proposal with easily 35+ papers referenced in it. Some of these are skimmed, or for specific items- on average, I try to read 4-5 articles a day, and synthesize/write along with it.

Hopefully this helps!

[MKarmoose]

Thank you for the follow-ups. The main issue here is that this project is interdisciplinary, in which I and two of my colleagues are solely responsible for the Electrical Engineering part of the program. So we're on our own here! However, I believe that at some point I'm gonna end up this way, so I don't mind if I do a little earlier

Anyways, thank you!

[xjenx]

@MKarmoose: Good thing that you've finally decided on what to do with your project.

@fuzzylogician: I've learned something from you. thanks for the post.

@Dal PhDer: Your advice is great and I've learned from it too. Thanks.

[Tall Chai Latte]

1- When do I stop reading? I've reading for two weeks now with a total of 15 papers or so, all regarding this subtopic I've chosen. Is this enough?

2- How do I get research ideas? I always try to question the work in each of the papers I read, but then I don't actually come up with something very valuable.

3- One thing I'm trying to do is to make a tree of the current work in the topic I'm investigating. This can help me find some sort of gaps in the field which can eventually highlight some opportunities for research. However, this tree requires much much more reading and investigating, which again brings us to the first question.

Thanks

1. It depends, but I don't think there's an end to reading. You need to stay current on the literature, so that means reading often.

2. This is an important skill to learn. I'm still learning how to do this, but according to my PI, you can start from reading review articles published in your field (and not just some random review articles -- the ones published in prestigious journals are the most ideal). In some of these review articles, authors will explicitly state what are the unanswered questions in the current field, and that can give you some leads. Or you can simply read a lot and make connections between findings, sometimes an interesting question would pop up. The latter method is harder in my opinion.

3. Having a mind map definitely helps. At your current stage, you can use it to track general informations rather than jumping straight to opportunities for research. At one point I had one of those mind maps, but it was too labor-intensive for me to go back and forth between reading and making the map. Now I just take simple notes to help me remember things I read.

Edited May 18, 2012 by Tall Chai Latte

[Cici Beanz]

fuzzy has given you some really great advice!

For me, the first few steps in generating a research project are as followed

1) I read a few papers in the area that I am interested in...if you do a good broad search, 15 may be enough- this is really discipline dependent.

2) Once I know the 'area' that I am interested in, and have my papers, I look at the discussion section and see what kinds of limitations these studies faced or listing as gaps in the current knowledge. This helps me see what the current issues/gaps are in this area.

3) once I know the gaps and have more detail in the area that I am interested in, I can start forming a question.

For me, it was helpful to have a question to take to my supervisor to begin with. It kind of gives you a working point to start off of.

Key things I think about when I read the literature and think about my question are: what will be my dvs and ivs, what is my population of interest, is there a specific methodology/method I want to use, what kind of data might I need to answer my question, who might I need to work with..etc. Figuring out these kinds of things will help you narrow your question.

IMO, I think it's better to to a broad lit search to help you figure out your question. Once you have your question down, then it's time to really dive deep into the literature. I have recently changed my dissertation topic, and came up with a new question and objectives. I have written a 7 page proposal with easily 35+ papers referenced in it. Some of these are skimmed, or for specific items- on average, I try to read 4-5 articles a day, and synthesize/write along with it.

Hopefully this helps!

Thanks for posting this! I am starting my MS this fall and have set a goal of 2 published (or publishable, at least) papers- one per year. This was exactly how I'd planned to prepare, and your post has somehow upped my confidence/enthusiasm.

[Elimba]

Hey MKarmooze,

You have already a great bunch of advice from the Fuzzy and Dal PhDer. Reading, reading ... you never finish doing that.

PhDer told you to pay attention to the discussion section of papers you read. I will add to that, pay even closer attention to the perspective (and conclusion) section of the paper. You can easily hand pick like flowers in a field, nice ideas and suggestions of unfinished business. There are hints in this particular section to help you start off where the other guy left over. But keep in mind that you really need to be grounded in a solid context by following like a curious dog the most important references made at the end of papers.

To buid confidence in yourself, find a workshop or an easy conference (generous acceptance rate) in the field, submit a short paper (3 or 6 pages) or at least a poster with one or two ideas. Anything is OK. Think like Tracy Chapman who said in her song Fast Cars "starting from zero and having nothing to loose" . You only need to brush out your ideas.

Wait and see what happens.

Your paper is accepted, then go and defend your ideas and collect direct feedbacks from peers in the field. Your paper is not accepted, then it's OK too. At least you made an effort to run the Boston marathon (I am in Boston . Then you can read the feedbacks (praises and nasty punches) from the referees in the program committees and reflect on that. My idea here is that you really need to interact with others in your field as soon as possible.

Later, you can build on this short experience to climb higher. You do not need to shoot for the stars or aim for a grand prize. You really need to narrow down quickly your research subject around two or three questions stated strongly in the introduction of your papers. Writing a paper in science is no different from story telling in that you are leading your readers on a journey, a wonderful one of discovery and enchantment. The guys in program committees are sensitive to that. I know that for a fact because when I was doing my PhD, I always got a praise about my writing style. Do not forget that side, or your paper may be boring. In spite of its great content, it may not make the cut for publication. Take a look at the Academic Joy site, PhD Candidate section for more detailed resources and tools.

I hope this helps you in some way.

Stay strong and good luck.