Credibility of research methods


A recent article in The Conversation highlights the issue of unreliable methods in scientific studies as well as the resulting inaccurate reports that can occur in the media. The article talks about yet another prank study set up by science journalist John Bohannon, submitted and initially published on a website that looks like a legitimate open access medical journal but on closer inspection is not. The article describes a published study that had a small sample size of 15 participants which can lead to some spurious findings.

The participants were divided into three groups, a control group who made no changes to their diet, a second group that ate a low carbohydrate diet and a third group that ate a low carbohydrate diet, plus a 42g chocolate bar daily. The researchers measured 18 factors including sleep quality, blood pressure, weight and cholesterol. The large number of factors and the small sample size selected meant the researchers were likely to find something worth reporting despite the questionable significance of the results. The “chocolate diet” study highlights the need to examine a research article for valid methods and analysis of results.

If you are developing a scientific study or experiment as part of your studies, then the methods in related scientific studies that you read may provide some insight into what is appropriate. In addition, the library contains resources that can help guide you on research methods and/or experimental design and analysis of the results. Some notable examples are:

Understanding the reliability and validity of various research methods and the subsequent analysis of results is important when critiquing a research article. Being able to critique a research article will add credibility to your research assignments. See the following information on critiquing journal articles, including their research methods: Critiquing a Journal Article