Difficulty reading research papers?

[Loric]

I think a large part of the problem is that you have people who equate the extensive use of jargon with somehow conveying intelligence.

 

It actually does quite the opposite.

[ProfLorax]

All these tips are really helpful. Thanks, everyone! On a different note, do you wear glasses? I was having a hard time in the beginning of the semester staying focused while reading; I got headaches and was confused and constantly distracted. Turns out, I needed glasses, which I had never needed before. I got a pair of prescription glasses, and my reading is much smoother and faster than before! 

[olsenolsen]

I think these are all excellent tips! I would suggest a couple of additional things (forgive me if these are repeated):

1) I generally pair the experiments with its corresponding results. This allows me to do two things: first, it allows me to have a much more "linear" understanding of the experiments. Once paired you can place the methods and results within the context of the rest of the paper which makes the whole thing much more enjoyable to read. 

2) I very briefly skim over a review article that is cited in the paper. This allows me to not only contextualize the research, but also see how the author has chosen to expand on the field of interest.

3) Recently, I have started to read with much more purpose. For example, if I want ton try a new protocol, I only read the methods and skim the discussion. This is particular helpful if I'm trying to bridge together several protocols/methods: instead of spending hours reading 5-6 relevant papers and then working to build a new protocol, I can excise the protocols from those papers, build a new one, and tweak as I read the rest of the papers more comprehensively.

 

Also +1000x on Mendeley. 

[TakeruK]

I think a large part of the problem is that you have people who equate the extensive use of jargon with somehow conveying intelligence.

 

It actually does quite the opposite.

 

Jargon is almost always used pejoratively and while I agree that just spouting out all the jargon (e.g. some TV science fiction scripts!) does not make you sound intelligent, jargon is absolutely necessary for quick and effective communication of information. So it is incorrect to make a blanket statement claiming that uses of jargon (even extensive jargon) is undesirable. Scientists should definitely avoid jargon when speaking to people outside of their field, but an academic journal is exactly the right place to be as jargony as necessary.

 

A scientific journal article is almost never designed to be read on its own. Like Istoleart said, it is but one sentence or one thought in a long continuing story. There are exceptions of course. For example, the journal "Annual Reviews of Astronomy and Astrophysics" is a publication that is issued once a year with about a dozen review articles. These massive articles basically attempts to summarize all the current knowledge and experiments on one particular topic. These articles are meant to be read stand-alone, they do not assume any knowledge of jargon but will usually define important terms and use them so that the reader becomes familiar with them, and they are perfect for a newbie or a non-expert to get acquainted with a topic. 

 

But the "regular" journal article that just tells one sentence? They should not need to explain everything. Otherwise, we will have hundreds of articles with identical paragraphs defining a whole bunch of terms. If I just plucked out an article in physics but not in my field (e.g. semiconductors), I should not expect to understand much more than a very basic level. I would have to reread it and read related articles to really understand what the problem they were tackling and what solution they are proposing. The idea of grad school though (and Journals Club) is to get e.g. astronomy grad students, to the point where if they did pick up a random article in Astronomy, they should be able to be get a little more out of it than a complete newbie to the field. Journal Clubs helps us build a broad foundation of knowledge.

[Cookie]

I do agree that not all authors are good writers and many articles are esoteric, but most journals are to present the cutting edge research in a particular field, so it is absolutely important to build up basic knowledge required in that field, as well as the skills to read and understand them.

 

One problem I have noticed is that most students don't start reading articles and follow journals until AFTER they get in grad school. That's why it's stressful and difficult to read 3 papers in one sitting!

[Arcadian]

That's probably because most research papers are overly-verbose and filled with unimportant details (at least, unimportant for the reader's purposes). Unless you really need to know something about the particular methods they used, why bother reading the methods? Abstract > Introduction > Discussion is the way to go most of the time.

[Guest |||]

That's probably because most research papers are overly-verbose and filled with unimportant details (at least, unimportant for the reader's purposes). Unless you really need to know something about the particular methods they used, why bother reading the methods? Abstract > Introduction > Discussion is the way to go most of the time.

 

Isn't neuroscience notorious for misuse of statistics?

[Arcadian]

Isn't neuroscience notorious for misuse of statistics?

 

lol That's a pretty broad statement. Experimental psychology is having a bit of a crisis with data replicability and availability, I can tell you that. Some major journals such as Psychological Science are changing their requirements for statistical reporting. And there is also a push to abandon null-hypothesis significance testing in favor of estimation techniques.

Edited February 12, 2014 by Arcadian

[Guest |||]

lol That's a pretty broad statement. Experimental psychology is having a bit of a crisis with data replicability and availability, I can tell you that. Some major journals such as Psychological Science are changing their requirements for statistical reporting. And there is also a push to abandon null-hypothesis significance testing in favor of estimation techniques.

 

There isnt a push to abandon null-hypothesis testing, but to supllment it with range estimation statistics.

 

Which for the most part amounts to confidence intervals for effect sizes.

[RedPill]

If there isn't a reading club at your school, I highly suggest starting one. There isn't one at my school, I tried very hard to start one, and thus I'm looking for another place to do my PhD. I think it speaks highly to the type of student that is in your department and the type of professor that is in your department. 

 

Agreed

[danieleWrites]

Something to understand is that every paper you read is part of a conversation. So, opening that article up? That's you walking into the middle of a conversation. I can't remember who wrote this, so, can't cite, but imagine you've arrived late to a dinner party where small groups of people are standing around in deep conversations that may or may not be related to the conversations other groups are having. This party has been going on for a long time, and will continue long after you leave. You find a group speaking about something that interests you, listen in until you feel you have the gist, and begin to offer your own opinions. Some people leave, others arrive, some change groups. Eventually, you leave for a while, until you have time to come back later and join in the conversation.

A scholarly article is a fragment of one of those conversations. Some of it you simply will not get because it's replying to others that you have not read. You can't read everything in your field. You will be a bit lost.

The key is to figure out the argument the writer(s) are adding to the conversation, and how that argument fits into the parts of the conversation you already know.

[zipykido]

A lot of papers are poorly written and have horrible formatting and worse figures, but you do get used to reading them. Jargon in your field eventually gets easier to understand and once you know enough it's simple to read a paper all the way through without needing to stop and look for specific techniques and acronyms. Of course you eventually learn to read papers for the information you want to extract from them, just practice and like people have suggested join a journal club. They'll teach you how to read papers and the more experienced members will point out nuances that you may have missed. They are also excellent opportunities to learn to properly write a research paper that is clear and readable.

[danieleWrites]

http://proswrite.com/2013/04/09/the-genre-of-research-articles-introduction-sections/

It's a way to understand an introduction for a paper, and that will often make the rest of the paper make some sense. While this is written more as a model for writers, it can be applied to reading as well.

[pears]

Something to understand is that every paper you read is part of a conversation. So, opening that article up? That's you walking into the middle of a conversation. I can't remember who wrote this, so, can't cite, but imagine you've arrived late to a dinner party where small groups of people are standing around in deep conversations that may or may not be related to the conversations other groups are having. This party has been going on for a long time, and will continue long after you leave. You find a group speaking about something that interests you, listen in until you feel you have the gist, and begin to offer your own opinions. Some people leave, others arrive, some change groups. Eventually, you leave for a while, until you have time to come back later and join in the conversation.

A scholarly article is a fragment of one of those conversations. Some of it you simply will not get because it's replying to others that you have not read. You can't read everything in your field. You will be a bit lost.

The key is to figure out the argument the writer(s) are adding to the conversation, and how that argument fits into the parts of the conversation you already know.

 

Very well said! "Orienting myself" to readings is probably my most commonly encountered issue, in terms of both orienting myself with regards to the "conversation"(who said what, and when? Has anyone responded?) and orienting myself to the "jargon" and relevant terms (what does a [insert mouthful of a theory name here] perspective entail? How is/isn't it compatible with [other theoretical perspective]?). Google is my best friend when it comes to dense readings.

 

Also, if I'm seriously struggling, I'll shut myself in a quiet room and read the piece out loud as slowly as possible, sometimes scribbling notes along the way in the margins. When the most opaque, dense, borderline illegible papers are getting the best of me, I "translate" them from oldoverlyscholarlybritishguyinthe1970s-ese to the way I would try to paraphrase it if someone wanted a play-by-play of each sentence.

[annoyed]

Thanks for posting this RedPill. I'm glad I'm not the only one. I know I have a problem with reading comprehension. I think it's mainly because I haven't read enough. I never liked reading for pleasure, the only books and articles I ever read are for school. The internet has also destroyed my attention span, I find it difficult to read anything of great length.

 

The other part of the problem are the authors. I notice PhD's like to make simple concepts seem complicated.

[victorydance]

I don't see any point in focusing on how people write or jargon or whatever. 

 

The first step in reading any scholarly article is to identify the argument. More specifically, the relationship. What two (or more) variables are related to each other and how are they related? Does changes in X mean changes in Y? This is usually located right after the literature review. 

 

You should read the intro first just to get a general feel for the paper. Then find the argument. Then read the paper and understand their reasoning behind why they are arguing something. How is the person measuring the variables? Do you agree with the measurement? If it's qualitative, do you agree with their reasoning? If it's quantitative, do you agree with the method they are using?

 

You should be able to identify what they are saying without the conclusion. In many cases, the conclusion is useless except to see how they summarize their paper in their own words. 

 

So basically:

 

-find relationship

-how did they arrive at this relationship?

-what are the variables?

-how did they measure the variables?

-is their method sound?

 

Keep in mind this is from a social science perspective, I don't have much experience with pure science articles but the previous should apply broadly.

Edited March 13, 2014 by HopefulComparativist

[victorydance]

Also, I don't take notes on my first reading, and I almost always read the whole thing from top to bottom through. I make mental notes in my head.

 

Then, if I really care about what the article is about, I go back and take notes on the argument, and all the other stuff.

 

A task that can help with this kind of stuff is making annotated bibliographies.

 

Also, if your program allows you to enrol in independent reading classes this is where you will really hone your skills. Basically it's one on one with a prof and he gives you stuff to read every 1 or 2 weeks and you go in and discuss it. It will force you to prepare and understand the text because, trust me, there is nowhere to hide in these classes.