We are inundated with research studies that are intended to make sense of the complex world we live in. Marketers use it to understand and influence our purchasing decisions, while social scientists use it to understand human behaviors. But how do you know if the research you read about is valid? How do you know if it can be trusted?
Know who paid for the research study: There is an inherent level of bias when conducting research. The ways you ask questions, the order questions are asked, how findings are interpreted, and more, rely on a human element. You want to know who paid for the study to understand what pressure they might have put on the researcher. Does the financier have anything to gain or lose by the study’s results? Although many researchers do their best to eliminate human bias, if a tobacco company financed a study that found cigarettes are good for your health, be cautious.
Know who is talking about the research: When you learn about research results, chances are you are hearing from someone other than the researcher. Most people do not actually read the researchers published paper, and therefore rely on a journalist, blogger, or other writer to make the research findings understandable. The thing is, many journalists have never taken a basic statistics class, much less have formal training in how to interpret research findings or critically evaluate research methodology. Be wary of anyone that interprets research findings without having learned how to properly do so.
Know the difference between causation and correlation: One of the hardest concepts students must grasp when learning how to conduct research is the difference between causation and correlation. Causation refers to one thing causing another, also called cause and effect. For example people, who smoke develop lung cancer – ie. smoking caused cancer. Correlation refers to two things happening at the same time, however if one caused the other is uncertain. For example, girls who watch soap operas are more likely to develop an eating disorder. Did watching soap operas cause the eating disorder, or did the eating disorder cause them to watch soap operas? Or was it that the girls had poor body images, and therefore watched soap operas and had an eating disorder? It is very difficult for a research study to show causation. Most research results show correlations. Anytime someone tells you that a recent research study showed one thing caused another, be critical.
Know that statements of fact are not research: Another difficult lesson students of research much learn is that the research process does not produce statements of fact. You may hear “The research study proved that…” when in reality, the study simply found support for the researcher’s theory. Research typically relies on averages, so research shouldn’t use words like: never, always, none, all. Look for “softer” language in research such as: most, many, some, few.
Let’s see what you have learned. Click this link to see a commercial by 5-hour ENERGY® where the actress discusses a research study they conducted. Pay attention to the four lessons you just learned. What do you see?
The main takeaway for me is that the commercial would lead you to believe that 73 percent of doctors would recommend 5-hour ENERGY®. If you listen closely (and read the fine print) you realize that only 56 percent would specifically recommend 5-hour ENERGY® to their healthy patients that are already taking energy supplements. How many doctors have healthy patients? And how many of those healthy patients are already taking an energy supplement? Most importantly, how did the researcher frame the question? The doctors might have simply responded that if their health patients were taking an energy supplement, they would recommend it be low calorie. And since 5-hour ENERGY® is low calorie, it would meet these criteria.
So what do you think? Do you have other examples of bad research? What about research that you felt was poorly interpreted by a writer? Share your thoughts in the comments!