Is there a package for EFA with multiple groups?

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Is there a package for EFA with multiple groups?

Elizabeth Barrett-Cheetham
Hello R users,

I’m hoping to run an exploratory and confirmatory factor analysis on a
psychology survey instrument. The data has been collected from
multiple groups, and it’s likely that the data is hierarchical/has 2nd
order factors.

It appears that the lavaan package allows me to run a multiple group
hierarchical confirmatory factor analysis. Yet, I can’t locate a
package that can run the equivalent exploratory analysis.

Could anyone please direct me to an appropriate package?

Many thanks,

Elizabeth

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Is there a package for EFA with multiple groups?

Joshua Wiley-2
Hi Elizabeth,

In confirmatory factor analysis with multiple groups, the reason one needs
to estimate the models simultaneously is that, typically, one is interested
in applying constraints (e.g., forcing all or some of the factor loadings
to be equal across groups).  In exploratory factor analysis, constraints
are uncommon (they are somewhat un-exploratory).

I would suggest simply using the psych package and subsetting your data to
the particular group, as in:

efa( data = subset(data, Group == "Group1") )

efa( data = subset(data, Group == "Group2") )

etc.

As you noted, lavaan will allow you to test multiple group CFAs, so if/when
you are ready to see whether the same configural factor structure or any
other level of invariance holds across your groups, you can use it.

Sincerely,

Josh




On Mon, Jul 28, 2014 at 2:46 PM, Elizabeth Barrett-Cheetham <
[hidden email]> wrote:

> Hello R users,
>
> I’m hoping to run an exploratory and confirmatory factor analysis on a
> psychology survey instrument. The data has been collected from
> multiple groups, and it’s likely that the data is hierarchical/has 2nd
> order factors.
>
> It appears that the lavaan package allows me to run a multiple group
> hierarchical confirmatory factor analysis. Yet, I can’t locate a
> package that can run the equivalent exploratory analysis.
>
> Could anyone please direct me to an appropriate package?
>
> Many thanks,
>
> Elizabeth
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


--
Joshua F. Wiley
Ph.D. Student, UCLA Department of Psychology
http://joshuawiley.com/
Senior Analyst, Elkhart Group Ltd.
http://elkhartgroup.com
Office: 260.673.5518

        [[alternative HTML version deleted]]


______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Is there a package for EFA with multiple groups?

William Revelle
Dear Josh and Elizabeth,

Josh suggested one way of doing it with the psych package.  As of today, the psych package (version psych_1.4.8) , I have included a new function (faBy) that will work with the statsBy function to do EFA for each of multiple groups.

Basically, it just calls the statsBy function to get correlations for each subgroup, and then applies fa to that output.

Thus,

sb <- statsBy(data, group=“grouping variable”,cors=TRUE)
fb <- faBy(sb,nfactors= how ever many you want)

This version is working its way through the CRAN distribution channels, but can be obtained from the personality-project repository at
http://personality-project.org/r/   (if using a Mac) or http://personality-project.org/r/src/contrib/   if using a PC.
Look for version 1.4.8

Bill



On Jul 28, 2014, at 5:22 AM, Joshua Wiley <[hidden email]> wrote:

> Hi Elizabeth,
>
> In confirmatory factor analysis with multiple groups, the reason one needs
> to estimate the models simultaneously is that, typically, one is interested
> in applying constraints (e.g., forcing all or some of the factor loadings
> to be equal across groups).  In exploratory factor analysis, constraints
> are uncommon (they are somewhat un-exploratory).
>
> I would suggest simply using the psych package and subsetting your data to
> the particular group, as in:
>
> efa( data = subset(data, Group == "Group1") )
>
> efa( data = subset(data, Group == "Group2") )
>
> etc.
>
> As you noted, lavaan will allow you to test multiple group CFAs, so if/when
> you are ready to see whether the same configural factor structure or any
> other level of invariance holds across your groups, you can use it.
>
> Sincerely,
>
> Josh
>
>
>
>
> On Mon, Jul 28, 2014 at 2:46 PM, Elizabeth Barrett-Cheetham <
> [hidden email]> wrote:
>
>> Hello R users,
>>
>> I’m hoping to run an exploratory and confirmatory factor analysis on a
>> psychology survey instrument. The data has been collected from
>> multiple groups, and it’s likely that the data is hierarchical/has 2nd
>> order factors.
>>
>> It appears that the lavaan package allows me to run a multiple group
>> hierarchical confirmatory factor analysis. Yet, I can’t locate a
>> package that can run the equivalent exploratory analysis.
>>
>> Could anyone please direct me to an appropriate package?
>>
>> Many thanks,
>>
>> Elizabeth
>>
>> ______________________________________________
>> [hidden email] mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>
>
>
> --
> Joshua F. Wiley
> Ph.D. Student, UCLA Department of Psychology
> http://joshuawiley.com/
> Senior Analyst, Elkhart Group Ltd.
> http://elkhartgroup.com
> Office: 260.673.5518
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

William Revelle           http://personality-project.org/revelle.html
Professor           http://personality-project.org
Department of Psychology   http://www.wcas.northwestern.edu/psych/
Northwestern University   http://www.northwestern.edu/
Use R for psychology             http://personality-project.org/r
It is 5 minutes to midnight   http://www.thebulletin.org

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.