Effect Size
1. Effect Size
- Represents the change measured in standard deviation in an average outcome that can be expected when an intervention is given to a person.
- A standardized measure that can be used to compare different practices across different studies
- Developed to assess the magnitude of an intervention’s effect to help researchers and practitioners answer the question of how powerful is the impact of the intervention.
- Not a method used to establish cause and effect but used to determine the power of an intervention and differences between groups.
2. The generally accepted benchmark for effect size (Cohen’s d) comes from Jacob Cohen. He was hesitant to develop the simplified benchmarks, small, medium, and large because of the potential for misuse of these terms. He nevertheless accepted the risk with the view that the benefits outweighed the problems by offering a common frame of reference for stakeholders to use to understand the results of research.
|
Cohen’s d
|
Effect Size |
|
Small
|
d = .2
|
|
Medium
|
d = .5
|
|
Large
|
d = .8
|
3. A convenient way for educators to interpret effect size is to consider the improvement in percentile scores that would take place if an intervention with a given effect size is implemented. (See “How to Interpret Effect Sizes”)
4. Small sample size studies tend to produce larger effect sizes than large sample sizes and effect sizes in small studies produce more highly variable results than large sample sizes.