Login  Register

Using factor scores

classic Classic list List threaded Threaded
2 messages Options Options
Embed post
Permalink
Reply | Threaded
Open this post in threaded view
| More
Print post
Permalink

Using factor scores

DR VEENA Joshi
6 posts
Hi,

I have 20 questions (variables) where likert scale ( 1 to 5 ) is used.
I have got four factors.
One of the variable should be in factor 1 (loading = 0.527)but the
loading for the same variable is higher in factor 2 (loading = 0.547).
Can I retain this variable in factor 1?

I have saved the factor scores. Do I have to give any weights because
I forced this variable to be in factor1?

Can I use my a variable which is also measured on likert scale as
outcome variable and use linear regression?
OR
Neglecting Neutral, can I use logistic regression for disagree (after
combinig strongly disagree and disagree) and agree?(after combinig
strongly agree and agree).

Regds

Veena
Reply | Threaded
Open this post in threaded view
| More
Print post
Permalink

Re: Using factor scores

Hector Maletta
602 posts
If you think the variable's meaning is related to other variables with high loadings on factor 1, try to rotate the factors to get a more interpretable factor structure. If that fails, perhaps your theoretical approach, or your interpretation of that specific variable, is wrong. Perhaps it is measuring something else.

Hector

----- Mensaje original -----
De: DR VEENA Joshi <[hidden email]>
Fecha: Miércoles, Septiembre 6, 2006 12:47 pm
Asunto: Using factor scores

> Hi,
>
> I have 20 questions (variables) where likert scale ( 1 to 5 ) is used.
> I have got four factors.
> One of the variable should be in factor 1 (loading = 0.527)but the
> loading for the same variable is higher in factor 2 (loading = 0.547).
> Can I retain this variable in factor 1?
>
> I have saved the factor scores. Do I have to give any weights because
> I forced this variable to be in factor1?
>
> Can I use my a variable which is also measured on likert scale as
> outcome variable and use linear regression?
> OR
> Neglecting Neutral, can I use logistic regression for disagree (after
> combinig strongly disagree and disagree) and agree?(after combinig
> strongly agree and agree).
>
> Regds
>
> Veena
>