Let's say I have 2 variables for two time points: A and B
At time point A, we measure the height and number of basketball goals of the population.
Over a period of time, we then measure the same variables again from the same population.
From this experiment, I want to know the number of basketball goals that people can make at height x.
For example:
If your height is 5'9'', can this data set tell me number of goals you can make?
What about height of 5'8''?
What if I have more than 2 variables, what kind of tool can I use?
Thanks for all of the help!
Your best tool for questions like that is always multiple regression. If you have only two data points on the time dimension, use a dummy variable (i.e. code A as 0 and B as 1). Make sure you include all variables that are likely to correlate with height and score, e.g. age, sex, hours of basketball training per week, PE grade, etc.
You have measurements on some predictors and an outcome, and you have several measures taken from the same people. The data are therefore clustered by person. You can tackle this as a regression model with clustering. Because goals is the result of a counting process, and probably not properly Poisson, you need to use robust variance estimation
In Stata, you would write
. poisson goals height age, cluster(person) robust
Saeed : Your answer is a copy and paste of my answer. I have reported this to RG admin.
- Badges
- Science topic