SSCI 2900U: Guidelines for Article Reviews
The first thing to keep in mind is that no research is perfect. All research must make some compromises and have limitations. The point then is not to think that all research is futile but to critically think about the ways in which limitations may affect the kinds of conclusions that can be drawn from research.
What the reviews are NOT:
1)These reviews are NOT summaries of the articles. The reviews will be graded on the assumption that you have already read the article and understand the key points. Therefore, it is neither necessary nor helpful to summarize the article.
2) These reviews are NOT like film or book reviews. These reviews are not meant to describe whether you liked/disliked the article and what parts you found interesting/boring.
What the reviews ARE:
1) These reviews are critical evaluations and assessments of the key limitations of the research and the ways in which these flaws affect the study’s findings and conclusions.
The reviews should be 4 pages in length, typed, double-spaced, in Times New Roman font, size 12, with regular unaltered margins. You should have a title page with your full name, student ID, TA name, my name, date, course title, and section number.
Focus of the Reviews:
You should begin your reviews by briefly stating what the article is about, noting what the purpose of the research is and what questions the researchers are trying to examine. This should be no longer than two or three lines at the very beginning.
The reviews should focus on the article’s “fatal flaws”, i.e. major problems with measures, sampling, data, and interpretations/conclusions drawn that raise fundamental concerns about the article. However, it is not enough to just identify a problem. For example, it is not enough to just say “there is an issue with the author’s measures”. You need to be specific in identifying which measures are problematic, in what ways they are problematic, and you need to discuss HOW it might have affected the study’s findings and conclusions.
In other words, after having specifically identified the problem, you need to specifically link that problem to the findings/conclusions. Consider a hypothetical study in which there is a blatant problem with the sample and the kinds of generalizations that can be made. Instead of just saying “there is a problem with the sample”, you may want to consider something along the lines of: “The author wishes to generalize to parolees and makes policy recommendations on how parolees should be managed within the criminal justice system. However, she has studied a sample of arrestees. This is problematic because not all arrests lead to convictions; fewer lead to incarcerations; and even fewer are paroled. Therefore, the types of generalizations she wishes to make go beyond her sample and many of her conclusions and recommendations are not supported by her study, as her findings only specifically apply to arrestees.” This could still be considered general, in the sense that it does not link the problem to specific findings. However, you should be able to understand from this example that it is not enough to just say “there is a problem with the sample” or “there is a problem with generalizability”.
The main part of your reviews should focus on fatal flaws. You can then discuss other more minor points with the study. In other words, the space you devote in the reviews should be proportionate to the severity of the problem. The more fatal the flaw, the more space you should use to discuss it. The more minor the issue, the less you should use. Understanding and figuring out which issues are more major and minor will allow you to not only become more critical and analytical in your reading of the articles but also allow you to effectively use the allotted space and keep within the page limits.
All the problems you discuss in the article should focus on problems that could have been specifically avoided had the researchers been more thoughtful, careful, and meticulous, NOT on problems that all research inevitably faces. For example, statements such as “this study relies on official crime statistics and we all know that official data sources miss the dark figure of crime, so this study is flawed” are neither useful nor critical. Recall that all research is limited in some way. Strong critiques address what the researchers have specifically done and are not vague, overly general claims.
The following are some questions you may want to ask yourself as you identify limitations with an article and think about the ways in which the limitations may affect the findings and conclusions of a study. You will not necessarily find all of these problems in every single article.
1) How do the researchers define and measure their concepts? Focus on key independent and dependent variables. Do they make sense given the concepts they are intended to represent? If not, how does this affect their findings and the conclusions they can draw?
2) How have the researchers selected their sample? Does the sampling procedure limit or bias the conclusions that can be drawn from the study?
3) Some studies compare groups to determine if they are significantly different on some key dependent variable. Are these groups sufficiently similar to begin with to ensure that pre-existing differences are not responsible for the differences they find on the dependent variable?
4) Do the researchers present basic descriptive statistics on their sample and measures? Do they suggest any obvious problems? Are there lots of missing data?
5) Does they way they have presented and analyzed their data, whether in tables, charts, etc., make sense?
6) Have the researchers adequately addressed alternative explanations that might account for the findings in their study?
7) Sometimes researchers will acknowledge a problem with their study but go on to draw conclusions or make claims that ignore the problem. Researchers will also sometimes go on to make claims that are either unsupported by or unrelated to the particular study they undertook. Have the researchers adequately taken into account issues that affect their study? Do they make claims that are supported by their study?