I'm currently evaluating clinical data using a logistic regression in R.
Sample data
set.seed(42)
age <- round(runif(100, min = 20, max = 95))
cancer <- sample(c(0, 1), 100, replace = TRUE)
data <- data.frame(age = age, cancer = cancer)
print(data)
Then I created the logistic regression model by
glm(cancer ~ age, data = data, family = "binomial")
For a first inspection I used summary(). As the estimate of the odds ratio for my age variable I get the value 0.08. I know for a fact that age has a positive influence on the likelihood of my outcome (cancer disease) so i would expect the value to be something higher than 1. Previously I worked with Stata and if for every year the RR of cancer increases by .08 it would show as 1.08. The value being lower than 1 makes me wonder if either R has a different way for the output of OR estimate for metric or did I maybe do something wrong during the creation of my regression model?