I have a dataset with 5000 observations with class 0 and 300 observations with class 1. I would like to run some statistical analysis, for example on the average length of strings, the number of words, ... I was calculating the average. Do you think it makes sense to calculate and compare these values with a similar difference In classes?
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
The answer to this question really depends on the insights you are after with these statistics.
Generally speaking however, it is wise to investigate per class in such a situation. Taking your examples this would entail comparing average string lengths for both classes. This will help you understand how both classes are (dis)similar given the dataset.
