Sample size for point biserial
Effect size is the hypothesized association between a categorical and a continuous variable
In order to run an a priori sample size calculation for a point biserial correlation, researchers will need to seek out evidence that provides the proposed correlation between the categorical and continuous variables. The correlation between these two variables constitutes an evidence-based measure of effect size. Find an article in the literature that is methodologically similar to the study of interest or uses the same type of outcome and use those values in the sample size calculation for a point biserial correlation.
A good rule of thumb is to use the effect size criteria identified by Cohen (1988)*.
Small effect size = .10
Moderate effect size = .30
Large effect size = .50
A good rule of thumb is to use the effect size criteria identified by Cohen (1988)*.
Small effect size = .10
Moderate effect size = .30
Large effect size = .50
The steps for calculating sample size for a point biserial in G*Power
1. Start up G*Power.
2. Under the Test family drop-down menu, select t tests.
3. Under the Statistical test drop-down menu, select Correlation: Point biserial model.
4. Under the Type of power analysis drop-down menu, select A priori: Compute required sample size - given alpha, power, and effect size.
5. In the Tail(s) drop-down menu, select One if researchers have a definitive and literature-based reason for believing that the association/correlation that the correlation travels in a certain direction (either positive or negative). Select Two if researchers are unsure whether the association/correlation will be positive or negative.
6. Click on the Determine => button.
Enter ".01" into the Coefficient of determination p2 box if researchers believe there will be a small treatment effect.
Enter ".09" into the Coefficient of determination p2 box if researchers believe there will be a moderate treatment effect.
Enter ".25" into the Coefficient of determination p2 box if researchers believe there will be a large treatment effect.
7. Leave the alpha value at 0.05, unless researchers want to change the alpha value according to the current empirical or clinical context.
8. Enter .80 into the Power (1-beta err prob) box, unless researchers want to change the power according to the current empirical or clinical context.
9. Click Calculate.
2. Under the Test family drop-down menu, select t tests.
3. Under the Statistical test drop-down menu, select Correlation: Point biserial model.
4. Under the Type of power analysis drop-down menu, select A priori: Compute required sample size - given alpha, power, and effect size.
5. In the Tail(s) drop-down menu, select One if researchers have a definitive and literature-based reason for believing that the association/correlation that the correlation travels in a certain direction (either positive or negative). Select Two if researchers are unsure whether the association/correlation will be positive or negative.
6. Click on the Determine => button.
Enter ".01" into the Coefficient of determination p2 box if researchers believe there will be a small treatment effect.
Enter ".09" into the Coefficient of determination p2 box if researchers believe there will be a moderate treatment effect.
Enter ".25" into the Coefficient of determination p2 box if researchers believe there will be a large treatment effect.
7. Leave the alpha value at 0.05, unless researchers want to change the alpha value according to the current empirical or clinical context.
8. Enter .80 into the Power (1-beta err prob) box, unless researchers want to change the power according to the current empirical or clinical context.
9. Click Calculate.
Click on the Statistics button to continue.
Hire A Statistician
DO YOU NEED TO HIRE A STATISTICIAN?
Eric Heidel, Ph.D., PStat will provide you with statistical consultation services for your research project at $100/hour. Secure checkout is available with Stripe, Venmo, Zelle, or PayPal.
- Statistical Analysis on any kind of project
- Dissertation and Thesis Projects
- DNP Capstone Projects
- Clinical Trials
- Analysis of Survey Data
Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.). New Jersey: Lawrence Erlbaum Associates.