I am working on a study that aims to assess predictors of insufficient food intake among elderly patients. In the study I am evaluating (9) predictors. Regarding calculating the sample size, I decided to use the rule of thumb, but I am confused between to references that pointed out how to apply it:

-The first one suggested in their multivariate text that the minimum number of subjects for each predictor or independent variable in a regression analysis should be 5-to-1.

-The other one suggested this formula to calculate the sample size N ≥ L / , where can be determined from R2 ( = R2 / 1 − R2; R2 = effect size), and L (lambda) = 6.4 + 1.65 m −.05 m2 (m = number of predictor variables). For a medium effect size (R2 of .13) with 9 predictors, the sample size was calculated as N ≥ 17.2 / .15 = 114.

Thank you