jump to navigation

Paper consumption November 11, 2010

Posted by mareserinitatis in Uncategorized.
trackback

How long should it take to read a journal paper or technical article?

I recently read a blog post by another grad student lamenting how long it takes her to read papers, on the order of several hours per paper. I completely understand where she’s coming from. It’s hard when you’re a new grad student, and you are getting into the literature for the first time. I also have the unfortunate experience of changing projects and getting into a new area, and suddenly the literature became unreadable again. After all, most people in grad school are rather intelligent and feel like they shouldn’t have to struggle with something as simple as reading comprehension. What I’ve come to understand, however, is that reading papers is a progression that gets easier with experience.

I’m currently in a very interesting situation. While the papers in my new area can require a long time to read and digest, I am also working in an area related to what I did for my MS. In fact, right now, I am working on a project where they want me to write a document compiling what research has been done on our project thus far. I also have to keep track of my hours for payroll, so I discovered something quite interesting: I can often read through and comprehend the gist of these papers in about 30 minutes for a 4-6 page article with a few diagrams thrown in.

So why do those differences exist? I think there are several reasons, beginning with the papers themselves.

As I’ve become familiar with the research in the area, I can often skim the background section because that will often contain only one or two pieces of new information. When I first started out, often the related research/background areas took a long time to sort through although they were often the most accessible portion of a paper. At some point, I could look at what other papers were being referenced and know exactly what they were talking about because I’d read the reference or seen it referenced in other papers, significantly reducing the time to cover that section.

Reading the research methods section can vary wildly. If I am reading a paper that is discussing something algorithmic, the math can take a while to wade through. If it is a paper reporting an experimental set-up, it will move considerably faster, especially if there are a number of pictures.

And finally, we have the results. It really depends on how those are written. Some of them are succinct with lots of plots. Those are fast to read, but I find I really like the ones that give a good amount of analysis as well as discuss the implications, i.e. the longer ones.

Aside from the structure of the paper, there are the grammar and readability of the paper. You would hope that some of this would be ironed out during the review process, but that’s not always true. In engineering, I often pull a lot of conference papers. My experience is that conference papers, while containing some really excellent technical information, will often be less rigorously reviewed and are more prone to errors. On one particular occasion, I had a paper which contained numerous devolutions into Engrish (some of which ended up as facebook status updates). While the paper was still understandable, it takes a bit more time to parse the meaning. This, by the way, is not meant as a slight to non-native English speakers. I find just as many errors in papers from people whom I assume use English as their mother tongue but can’t seem to avoid mixed metaphors or mismatches between subjects and verbs.

The other big issue is familiarity. As I mentioned above, familiarity will change the length of time required to reach various sections of a paper. However, another issue that resolves itself with familiarity is how much jargon one understands. When first reading papers in a particular field, I will spend a lot of time looking up terms I don’t understand and trying to get the necessary background. While one would hope that much of this would be figured out during coursework in grad school, my experience is that this is often not the case. A lot of the stuff I need to know from papers of often found toward the back of the textbooks. In other words, most of the things in papers are the topics that are often not covered in courses due to time constraints. As I gain familiarity with the issues, I don’t have to spend so much time looking at all the background information necessary to understand the issues.

Finally, with enough familiarity, you start to know what to expect. When first reading through papers, you don’t understand how the experiment was set up, whether the method is valid, how the results should be plotted, etc. After reading enough papers, you start to see how these issues are addressed. The plots become second nature to read and you have a clear idea how and why they proceeded with a particular method. When reading papers from the same author, they will often take material from older papers, especially from the background section. Further, they are often inclined to present results in a similar manner from one paper to another, making it easier to comprehend their point.

So how long does it take to read a paper? My answer: it depends. Regardless, it does get easier and becomes less time consuming with experience.

Advertisements

Comments»

1. Strigiformes - November 12, 2010

Very true and something I’ve only realized recently, when reading papers my current PI recommends vs reading paper my undergrad advisor suggests.

When we are trained to read research papers we were always told to look at the figures, think critically, and never believe anything until you think that the authors have convinced you with their data. It all make sense, but in reality really hard to do, if you can barely get out from a paper about an unfamiliar topic what the people had actually done, let along criticize their data and/or conclusion.

But then you know you’ve grown if you can criticize the them in a legitimate way (instead of just saying that the error bars are too big…), and relate it to your research. That is always a nice feeling.

Like

2. Understanding presentations | "I'm in the lab…" - November 19, 2010

[…] I read this post a while ago on how long it takes to read a research paper and what factors determine the time […]

Like


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: