HOW TO READ A RESEARCH PAPER
August 30, 2006
Georgia Institute of Technology
At a university, you are often asked to read an academic paper
or article. Here are some guidelines and questions you might ask
yourself while doing so.
Have a look at the book, How
to Read a Book (Mortimer J. Adler and Charles Van Doren,
Simon and Schuster, 1940). It offers lots of useful advice to
improve your reading skills.
You may not be able to understand the whole paper in just one
reading. On the first pass, get the general idea, look up
any vocabulary words you don't understand, and write down your
outstanding questions. Then make another pass in which you fit
the puzzle pieces together.
Identify the paper's thesis and state it as a simple sentence. The
thesis is the main idea that the author is trying to
convince you of. It is sometime surprisingly hard to determine
After reading the paper, ask yourself whether or not the author
has convinced you of the thesis. If not, why not? Was the logic
flawed? Was there not enough evidence provided?
Explicitly summarize the author's argument. That is, write down
the points in the author's argument, and the reasons given for
What is the other side of the issue? Often authors will state this
explicitly, with greater or lesser degrees of bias. Even is the
author doesn't mention the other side at all, you should attempt to
summarize the other side of the argument. After all, if there isn't
another side, then why was the paper written in the first place?
Are there questions that you are left wondering about? Often an
author will explicitly raise "Future Work" that can
be performed to follow up on the work presented. Conversely, the
author may have overlooked some obvious questions left unanswered
by the paper.
More generally, try to characterize the field in which the author
is working. Many times the author will do this explicitly by
talking about "Related Work". And a thorough author
will discuss how his/her work differs from each of the author's
How does the author validate his work? That is, in making a
case, an author should provide evidence that backs it up. Different
fields use different kinds of evidence. For example, mathematicians
provide proofs; psychologists do experiments; and engineers
often build prototypes. Beware of papers that offer opinions
but don't give you the evidence to support them.
What was the author's style? Many academic papers are
fairly (overly) dry, stating just the facts with little
embellishment. Others are more tutorial in nature, using the
second person pronoun ("you") instead of the impersonal
third person ("he"/"she"/"it").
Some use the active voice, some the passive. Some even use an
occasional figure of speech to liven things up.
Here is one other book you might look at (Writing
Arguments: A Rhetoric with
Readings, John D. Ramage, John C. Bean, and June Johnson,
Longman Publishers, 2004). It describes the techniques authors
use (or misuse) in stating their
positions. Understanding the techniques authors use in their writing
can improve how well you read.
Thanks to Weronika Pawlak.
Thanks to Ana Mirilashvili.
Thanks to Elsa Jansson.
Thanks to Sandi Wolfe.
Thanks to Robert Grünwald.