Hello Li,

If your principal aim is that of determining whether a relationship exists between your chosen set of independent variables/predictors and your dependent variable measure (some sort of indicator of burnout, I presume), and you are convinced that OLS regression is appropriate, then:

1. Determine what is the smallest, non-ignorable, degree of relationship (e.g., R-squared) you'd like for your study to be capable of detecting, should it exist. This is one way to characterize your target effect size (ES).

2. Identify the target number of IVs you'll use in your model.

3. Identify the intended type I error risk you'll use for hypothesis tests ("alpha level").

4. Identify how much statistical power (between 0 and 1.0; higher is better) you'd like your study to have.

5. Enter the identified bits of information (#1-#4, above) into the g*power program (freely available: https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower), using "F-tests" family / Linear multiple regression: Fixed model, R^2 deviation from zero" options.

That will give you a plausible sample size (lower bound) to guide your data collection efforts.

Good luck with your work.

See it:

https://www.statisticshowto.com/probability-and-statistics/find-sample-size/

Dear colleague.

It is primarily a type of sample.

Secondary is the size. Due to the error rate of the sample, it is necessary to perform a calculation. You can use e.g. : https: //www.surveysystem.com/sscalc.htm.

Good luck