New in StataNow 19.5: Power Analysis for Logistic Regression
With the update from May 21, 2025, we are pleased to introduce the first new highlight in StataNow 19.5: power analysis for logistic regression. This powerful feature supports researchers in making informed decisions about study design and sample size—without wasting unnecessary resources.
What’s New?
Whether it involves conversion rates in marketing (purchased vs. not purchased), recovery progress in medicine (recovered vs. not recovered), or user behavior in digital products (subscribed vs. not subscribed)—the new power logistic function helps calculate the optimal sample size for logistic regression models.
With this command, you can:
- Calculate sample size, power, or effect size for testing a coefficient in a logistic regression model.
- Specify multiple parameter values simultaneously and display results in both tabular and graphical formats.
For example: Suppose your outcome variable Y is binary and you have two binary covariates—X as the covariate of interest and Z as a nuisance covariate. To calculate the required sample size for 80% power at an α of 5%, the following command is sufficient:
. power logistic 1.5, px(0.3) pz(0.2) py(0.6) pycondx1z1(0.8)But that’s not all: The function also allows for power analyses for more complex models with up to 20 nuisance covariates from 11 different distributions.
What Makes This Feature Special?
The method used by power logistic is considered one of the most precise on the market—and Stata is one of the few statistical packages that offers it at all. Furthermore, the typical Stata user-friendliness and flexibility are maintained: even complex models can be implemented easily.
Who Is This For?
The new function is primarily aimed at users in medicine, epidemiology, public health, and ecology who frequently work with logistic regression models. They benefit from the precise yet user-friendly way to perform power analyses efficiently—whether planning studies or analyzing complex relationships. The feature also offers clear added value in other fields with binary outcomes.
Discover all StataNow innovations.
If you have any questions, please feel free to contact us: Info@dpc-software.de