Determine the number of participants that the study includes. If the study is based on a relatively small number of participants, then evidence in the study may only be inferential and not considered to be conclusive. The larger the number of participants or population considered in the study, the more likely the study will reflect the actual outcomes it is designed to evaluate.
Determine the type of study that you are evaluating. Questions that can be considered include what populaton group was used for the study and what methods were used in its analysis. Determine whether the study is a meta-analysis that combines the research information of many studies. If so, the researcher would want to consider how many studies were combined in the meta-analysis and what was the source and quality of those studies. When considering clinical studies, determine if the study is a randomized, double-blind study or another type. A randomized double-blind study is considered to be the gold standard for medical trials, especially those involving drugs and other supplements. The reason for this is that neither the participants nor the administrators of the study know which individuals are receiving the actual drug or supplement, and which are receiving a placebo. The results, then, are more likely to be unbiased.
Double-check the mathematics of the study for accuracy. Try to perceive if the conclusions reached based on the math are objective or subjective. If the title of the study makes a broad claim, consider whether the numbers really support that claim. Determine what variables were and were not accounted for in the study.
Consider the length of time that the study documents. If the study is short-term, then in some cases, the true facts, effects, side effects or conclusions reached may be obscured. Certain trends which may be true over the short term might not reflect the real long-term situation. Some studies that make sweeping claims and project dramatic conclusions might be true only for a limited time period in the absence of studies that consider the same matter over the long term.
Research who sponsored the study. If the study was sponsored by a for-profit company that has potential economic interest in the results of the study, then there is a stronger likelihood that the design of the study or the way that the study is interpreted by those conducting the study will result in a biased outcome. Howard Brody, MD, in the book, "Ethics, the Medical Profession and the Pharmaceutical Industry," documents that the purpose of many clinical studies can be "driven by marketing, rather than by scientific aim." This can be true for studies sponsored by any special-interest groups.