[R] Normality Test and T-Test

[Ferry]

Hello R Users,

Suppose I have data with the structure below:

Group_Name   Pre_Test   Post_Test
Grp_A             xxx          xxx
Grp_A             xxx          xxx
Grp_A             xxx          xxx
...
Grp_B             xxx          xxx
Grp_B             xxx          xxx
...
Grp_Z             xxx          xxx
Grp_Z             xxx          xxx
Grp_Z             xxx          xxx

Number of observations of each group are varies.

I want to conduct Normality test (ad.test for Anderson Darling or
pearson.test for Pearson) for each group by their pre and post values.
Later, I want to do a t-test.

Is there a better way to do normality test for each group without the
need of loop? At this moment, the only thing I can think of is
separating each group (and their pre / post test values) by creating
bunch of smaller set, and do the test by way of looping.

For example:

group_name <- unique(mydata.frame$group_name) ## or something similar
for (each_group in group_name) {
     smaller_set <- subset(mydata.frame, group_name == each_group)
     each_pretest <- ad.test(smaller_set$pre_test)
     each_posttest <- ad.test(smaller_set$post_test)

     print(paste(each_group, "pre_test p-value:",
each_pretest$p.value, sep = ""))
     print(paste(each_group, "post_test p-value:",
each_pretest$p.value, sep = ""))
}

and the same thing with t-test.

Any idea is appreciated.

Thank you.

Ferry