I am designing a study that has 2 independent variables (cultural intelligence, social judgment) and a dependent variable (hiring decision). The 2 independent variables are continuous, rated on a scale, and the dependent variable is categorical(yes/no).
I plan on using logistic regression to analyze the data, but how do I determine what sample size to use?
Johnny Lu An a priori power/sample size planning analysis for logistic regression can be conducted in the free G*Power software, see
Article DETERMINING SAMPLE SIZE IN LOGISTIC REGRESSION WITH G-POWER
Please check the following:
Article Sample Size Guidelines for Logistic Regression from Observat...
https://www.real-statistics.com/logistic-regression/logistic-regression-sample-size/
@Johnny you can refer to below references for your cause.
Hsieh, F. Y., Bloch, D. A., Larsen, M. D. (1988) A simple method of sample size calculation for linear and logistic regression. Statistics in Medicine
McConnell, B. and Vera-Hernandez, M. (2015) Going beyond simple sample size calculations: a practitioner’s guide. Institute for Financial Studies.
Johnny Lu You can calculate the odds ratio as exp(B_j), where "exp" indicates exponentiation and B_j is the logistic regression slope coefficient for predictor j.
Your field should have an established sample size for what you are studying. If it doesn't, then a sample of 500 is generally understood to be the minimum.
Sample size required depends on several factors, one of the most important being the minimum expected difference size you want to detect. It is usually referred to as a power test.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size