Likert Survey -- Stat for comparing single items and sub categories

#1
Hello,

For my dissertation I am using an existing 50 item Likert survey. I am pretty sure I will use a t-test to determine if there are differences between my two groups for the individual items.

What I am not sure about....is the 50 items are divided into 7 subcategories....to determine if there are differences in the subcategories between my two groups (four day schools and five day schools in Missouri) would I still use a t-test?

Thanks!
 

Karabiner

TS Contributor
#2
50 item Likert survey. I am pretty sure I will use a t-test to determine if there are differences between my two groups for the individual items.
Single Likert-type items (in contrast to a Likert scale) are considered ordinal scaled, so the t-test would inappropriate. You'd need to do U-tests.

By the way, what is the goal of 50 (!?) comparisons between the groups? Single items are rarely of interest, and with 50 comaprisons, the risk of false-positve results must be considered.
to determine if there are differences in the subcategories between my two groups (four day schools and five day schools in Missouri) would I still use a t-test
If several Likert-type items are combined into a scale, the resulting scores
are treated as interval, and group compariosn can be carried out using t-tests.

With kind regards

K.