cronbach's alpha help!

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

cal99
Posts: 1
Joined: Thu Apr 21, 2016 11:00 pm

cronbach's alpha help!

Postby cal99 » Thu Apr 21, 2016 11:07 pm

Hi everyone,

I'm currently running a factor analysis and cronbach's alphas as part of a study for my PhD. However, I'm struggling to understand whether I need to run a cronbach's alpha analysis on all the items of my scale (which has 4 subscales) or is a cronbach's alpha for each subscale sufficient?

When I analyse the subscales individually I get alphas ranging from just below 0.6 to 0.8, but when I do it for all scales together I get 0.45 which is worrying.

Thanks for any advice anyone can give.
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: cronbach's alpha help!

Postby GerineL » Tue May 10, 2016 8:49 am

are you going to use the scale as 1 scale, or as 4 subscales?
If you use them as 1 scale, you want to know how reliable that measure is, hence, you compute alpha for the whole scale.
if you use subscales, you want to know how reliable each of the subscales is and you compute them separately.

Who is online

Users browsing this forum: No registered users and 2 guests

cron