Guidelines for Critiquing Research
Research is a process of seeking information or the solution
to problems based on evidence of a scientific nature. In emergency medicine, as in any professional field, the
practitioner must be able to judge the importance, quality, and applicability
of research reported in professional journals.
The fact that a research study has been published is no
guarantee that the study was well done or that the results are accurate, or
even useful. Some reviewers and editors
fail to be discerning regarding certain aspects of the articles which are
submitted to them, or an article may be important in one context (i.e., the
study conditions), but not in the way that the reader wants to use it
(extrapolation of findings to their own setting). And no author is going to tell the reader that the study was
unimportant, of questionable quality, or not applicable to settings outside
their own study site, although any of these may be true. The reader must learn to judge for
himself. The following questions are
designed to guide you in critiquing research articles.
Variables
- Identify
the key variables in this study.
If there are independent and dependent variables, label them as
such.
- How
does the author operationally define each of the key variables?
- Is
each operational definition specific enough that you could duplicate
it? If not, what more information
do you need?
- Give
examples of potential extraneous variables that may potentially lead to a
confounding relationship between the dependent and independent
variables? Were data collected on
these variables? Were they
analyzed statistically to confirm/exclude their effect on the relationship
between the dependent and independent variables?
Literature Review
- What
is the focus of the literature review?
- Does
the literature review convince you that there is a need for the study?
- Considering
the problem under study, does the literature review appear to be fairly
comprehensive (i.e., does it include references from an appropriate
selection of topics, sources, time periods, and of differing conclusions)?
- Does
the literature review appear to be up-to-date? Can older references, if any, be justified?
Hypotheses
- Do the
hypotheses flow logically from the problem statement and literature
review?
- Are
the hypothesized relationships among the variables clearly specified?
- For
each hypothesis, indicate its direction and magnitude.
Ethics
- Are
there ethical issues that you think may have been of concern to the review
committee for the protection of human rights?
- Did
the authors state how they obtained informed consent?
- Did
the authors receive IRB approval from their institution(s)?
Design
- What is
the basic research design employed in this study?
- What
are the strengths and weaknesses of this design?
- What
is the treatment variable, if there is one?
Sampling Procedures
- What
type of sampling is used in this study?
- What
is the author’s target population?
- Did
the author truly sample from his intended population, to the exclusion of
others?
- How
well are the study subjects described?
- Is it
possible that the sampling procedures employed may have excluded subjects
that should have been included, and included others who should not have
been?
Validity and Reliability
- What
evidence does the author give that the data collection instruments (those
used to measure the dependent variable) are reliable? Is this convincing evidence?
- What
evidence does the author give that the data collections instruments are
valid? Is this convincing
evidence? Were the instruments
piloted and validated in a previous study?
Data Analysis
- List
the statistical tests that are used in the data analysis.
- Does
the data analysis address all the hypotheses or research questions?
- Does
the data analysis address questions that were not hypothesized?
- If the
data analysis includes significance tests, what was the null hypothesis
that was tested?
- If the
data analysis includes significance tests, describe their major findings.
- How do
the results compare with what the author hypothesized? How do they compare with the findings
of previous studies?
Conclusions and Recommendations
- Are
the conclusions presented by the author supported by the findings of the
study?
- Are
the author’s recommendations for practice justified by the findings?
- Does
the study make an important contribution to our body of knowledge, either
in establishing theory or in guiding practice?