You haven't told us nearly enough about
the data and method.
There really aren't any laws on the
books about sample size. A rule of thumb is between 10 and 20 cases
per indicator variable. The common wisdom is 'the more cases the
better.'
How many indicators (items) do you have?
How many factors?
Are any cases missing items?
What type of data are they? What
is the scoring scale?
We assume that you have checked each
item for trouble spots (outliers, skewness, etc.)
SPSS can do path analysis but not CFA..
SPSS does offer the AMOS add-on which will do the bootstrapping for you
as well. EQS, LISREL, MPlus all will do CFA. There is (was,
last time i checked) a free piece of software called Mx that could do CFA.
Don't know if it even exists anymore. It is free, but has a
pretty steep learning curve and is not nearly as user-friendly as AMOS.
***************************************************************************************************************************************************************
Mark A. Davenport Ph.D.
Senior Research Analyst
Office of Institutional Research
The University of North Carolina at Greensboro
336.256.0395
[hidden email]
'An approximate answer to the right question is worth a good deal more
than an exact answer to an approximate question.' --a paraphrase of J.
W. Tukey (1962)
Good Morning list,
I need some help with validating
a model using CFA. My colleague is working on a problem where he
needs to validate his model using CFA. Basically he has had to generate
data using bootstrapping because his sample was too small for CFA; Sample
size below min (200) for CFA; bootstrapped and generated 1000 N, need to
validate results using CFA.
Can someone provide the syntax to
be able to do this in SPSS version 14.0; the data has 10 variables.
TIA
Mike
Locked
