Gauging the effect size helps audiences evaluate whether a program produced meaningful results that were worth the time and energy.
An effect size measures the magnitude of a research finding. At its most basic level, the effect size is the difference between those who received a particular intervention (the “treatment group”) and those who did not (the “control group”). For example, let’s say a researcher is studying the effects of a new online course that prepares students for the ACT college-admissions exam. Half of the 11th graders in a school district take the course. Their average score is 22. Half the 11th graders in the district do not take the course. Their average score is 21. So the effect size of the course is 22 minus 21, which equals 1 point on the ACT scale.
When reporting on studies of interventions like this hypothetical ACT course, you should always try to include the effect size. That helps audiences evaluate whether the program produced meaningful results that were worth the time, effort and resources required to implement the program.
If researchers report effect sizes in ways that are difficult to understand (e.g. “5 percent of a standard deviation”), ask them to translate the effect size into units (like points on the ACT scale) that you think your audience will recognize. Also, ask them to explain how the effect size of their study compares to the effect sizes identified by other studies of similar programs or programs that might be used in similar ways. To return to the hypothetical ACT study, you might ask the researchers about the average effect size for online ACT courses studied by other researchers.
– Holly Yettick