Tips For
Reading Scientific Papers
Remember when reading a paper be
critical but also pay attention to exciting findings, novel insights, and
creative ideas. It’s easy to criticize, but hard to praise!
General: Keep in mind the whole time
you read the paper.
-
What is the central question/hypothesis the author is proposing?
-
What critical data/experiment would you
do to evaluate this hypothesis?
- You should form an opinion before reading
what they in fact did!
- What assumptions are made both when proposing the hypotheses and when evaluating them in light of the data collected.
-
What data do they collect to assess their hypothesis?
-
What is their conclusion given the data?
-
Do you agree with their interpretation or are there other compelling alternatives
given the data?
Specific: As you read each section,
focus on some of these specifics.
Introduction:
-
What is the main question they are interested in pursuing?
-
What background research/pattern/theoretical prediction motivates this
question?
-
Why is this question interesting in light of the background they discuss?
-
Do they offer one hypothesis or more than one?
If more than one hypothesis is offered, is each
exclusive, meaning that it proposes a distinctly alternative explanation that
is incompatible with the others, or could some of these hypotheses operate
simultaneously?
-
What assumptions are made when proposing the hypotheses?
Methods:
-
What assumptions are made about the effectiveness of their experiments or the
accuracy of their data?
- Do
their proposed methods critically test their hypotheses?
-
Are any of their methods confounded?
-
Did the authors use a creative method to evaluate their hypothesis?
-
Are their methods simple and elegant or complicated and convoluted?
-
Did they come up with a new technique to better evaluate a problem others have
struggled with?
Results:
-
What does the data say about the hypotheses?
-
Is there only one interpretation of the data?
-
Are there any big surprises / unexpected results?
Discussion:
-
Does the author say that they support or reject the hypothesis?
-
Do you agree with the author's interpretation of the data?
-
What novel insights are gained from the results?
-
What do the results imply more generally for the field of interest? For other
fields?