I have two sets of numerical data describing the degree of independence of children with impaired cognitive functioning. Every number in the set represents one child and is a result of a summation of scores on standard scales used in the field of research. Each set counts 48 values (sample size).

First set of numbers (values of random variable) are obtained prior to training. Second set is obtained after the training. I have stated a couple of hypothesis I'd like to test, but I'm uncertain which test is appropriate.

For example,

H0: There is no significant improvement in the degree of independence after the training.

H1: There is a significant improvement in the degree of independence after the training.

Also,

H0: There is no significant difference in the degree of independence between males and females AFTER the training.

H1: There is a significant difference in the degree of independence between males and females AFTER the training.

And two more I won't bother you with right now.

So basically the same 48 individuals are 'measured' at two different moments in time (time difference is 1 year), before and after the training.

I would appreciate any help, or word of advice.

Thanks in advance.

PS. Experts and experienced personas only. This is important.