Once in a blue moon, you will come across a homework assignment that involves evaluating a research paper in terms of statistics. Relax, this is not rocket science. What you have to realize is that you are actually the main focus of any research paper. The author is hoping that his findings will spark someone’s intellect in the field. Your job is to determine whether or not the article is transparent and meets it’s objectives.
Any research paper will contain an abstract, introduction, methods section, results, and conclusion/discussion. Your abstract is like the crip sheet for the meat of the paper. After reading the abstract you should know the overall premise of the paper. The introduction is really the rationale for why the research was performed – and us statistician appreciate such background information. Good stuff. Next you have the method section. You should familiarize yourself with the statistical procedures and how the data was obtained and from what sample population. If a t-test was performed then you should assume that your outcome measure was continuous. What are your independent and dependent variables? You want to know how the author conducted the experiment. The results section is where the magic happens in terms of statistics. Any tables or graphs should be mentioned in the body of the paper with a reference to the illustration(s). The Conclusions sections are where the results are interpretted by the author.
Okay. You read through an article. Here is the thing that nobody says about reading an article. Vital information can be missing that helps you understand the article. So don’t think you are a dummy just because you read it and stuff is not adding up. You should come away with a strong sense of what analysis was performed and the results. If a statistical procedure was performed but the results do not reflect the method section, that’s not your problem. So read research papers with an open-mind.
-If you enjoyed this blog there are more to follow-Amy