How To Interpret Effect Sizes

How can I interpret the effect size from a study into the impact a practice might have if I implement the practice in my classroom or school?

Why is the question important? The gap between research and practice has been a traditional dilemma holding back the effective implementation of an evidence-based education system. Researchers publish results of whether an intervention has an intended effect, is the result “statistically significant”. On the other hand, Practitioners have been interested in how much of an impact an intervention has when the new practice is used with students in real classrooms. For example, how much will a particular researched reading program increase reading scores beyond the currently adopted reading program? Practitioners want to know if the intervention’s effects will be large or small, meaningful or inconsequential. Effect size was developed as a tool to assess the magnitude of an intervention’s effect to help researchers and practitioners answer the question of impact. Despite the best efforts of researchers and statisticians, effect size remains a challenge for those interpreting research. The above chart is one method to assist practitioners better understand the results from educational studies.

How to Interpret Effect Sizes

 
Effect Size Increase in Percentile Scores
 0.1  50 to 54
0.2 50 to 58
0.3 50 to 62
0.4 50 to 66
0.5 50 to 69
0.6 50 to 73
 0.7 50 to 76
0.8 50 to 79
0.9 50 to 82
1 50 to 84


Source: Johns Hopkins University's Center’s The Best Evidence Encyclopedia

Citation: National Bureau of Economic Research. 1050 Massachusetts Avenue, Cambridge, MA 02138-5398. Tel: 617-588-0343; Web site: http://www.nber.org/cgi-bin/get_bars.pl?bar=pub