Have you ever gone to a wine tasting as an amateur taster? So you do what everyone else is doing…you pick up the wine. You swirl it around in your glass because that is what everyone else is doing. You sniff it. SNIFF! It smells like wine to me. Then you take a sip. That’s pretty much as far as you can go…In the meantime, the people around you are giving their opinions on the wine. And their opinions are not regular , everyday comments. They use an assortment of adjectives (fruity, dry, full body , etc.) As the night wears on, you realize that certain brands of wine are given a rating. People sip the wine, give their opinion, and move on to the next one….(hiccup)!

Believe or not, critiquing research papers in terms of statistics can be like wine tasting. (Okay you can laugh at that idea. ) There is a way to judge a good research paper, just as you would a glass of wine. The statistics have to be sound from the abstract to the final conclusion. Your abstract has to give the reader an idea of what to expect in the paper. What statistical methods were used to arrive at particular conclusions? What were the tools used to execute the analysis?  The methods section should choose appropriate statistical methods and sample populations. And not all research articles are given the same critique. There are different levels of effectiveness. Here is my personal rating system for research papers in terms of statistics:

Rating system the Amy Way

1= Excellent, the researcher presented the correct statistical methods to answer his or her objectives and the sample population is appropriate, not to mention the results and conclusion sections. I come away with a clear picture of how the research was performed. I feel good about this….

2=Moderate, the researcher presents correct statistical methods but it is not entirely clear how the primary objectives were analyzed. There seems to be a few questions about the methods section. I come away feeling a little uneasy about what I read in the article. I feel a little unsure of how the data was analyzed…I compare it to reading a restaurant menu in a dimly lit room. What does that say????

3=Poor, the researcher presents some findings and statistical methods with no corresponding results. The conclusions are not backed up by statistics. Or the sample population is not well-defined. What did I just read????

So there you have it. My opinion but there is a science to critiques and you have to understand the metrics behind the critique- to recognize a good one.

-Moore to follow – Amy