Have you ever seen a report about a new scientific study that really scared you, only to find out later that it wasn’t what you thought? Often this happens as a result of misinterpreting words.
Researchers and statisticians in scientific fields such as medicine and psychology use words and phrases that are common to other fields but may be used differently, which can cause some confusion if you are not familiar with how they use those terms.
This is a brief guide to how some of those terms may mean something different than what you think they do:
- Phrases like “relationship between,” “link between,” “connection between,” “correlated,” “joined”—many people assume that this means there is a cause-and-effect relationship between the factors being discussed. This is not necessarily the case. What this often means is that these factors seem to happen together, but the type of relationship may not be known. One may cause the other, there may be one or more other things that is causing both to happen, or it could be coincidence that they tend to happen together.
- Stating that something is “significant”— statistical significance is not the same as how we usually use the word “significant,” and often when these studies are being discussed, the type of significance they are referring to is statistical significance. In medicine and psychology, the most common thresholds for statistical significance are 1% and 5% (described as p values; p>.01 or p>.05). This means is that there is still a 1-5% chance that the results are due to coincidence and that there is no real relationship between the factors. As you can see, this is different than our usual idea of what significant means (i.e. important, noteworthy, and indicative of something in particular).
- The term “population”—the non-researcher usually thinks of the world “population” as being all-inclusive. In research, however, a population can actually be very limited by age, ethnicity, location, condition, or other factors. When you see the word “population,” do not assume that they are necessarily talking about everyone or almost everyone.
Other things to look for:
- Sample size— if the sample size is really small compared to the size of the population, this may indicate that it is a preliminary study into an issue, so one may not want to jump to conclusions until more extensive studies are performed.
- Level of statistical significance— check what level of statistical significance was used to perform the study; if the number is higher than 5% (p>.05), then there is a higher likelihood that the results are due to chance.
If you have any other questions or would like to schedule an appointment, contact us today.