[R-sig-ME] NAEP Data and lmer

[Roberts, J. Kyle]

Harold,

I just want to use R to analyze NAEP data in a multilevel framework.  I don't want to actually calculate plausible values (if I understood your question correctly).  I was wondering if I would need to run 5 separate analyses and then average across those values or if there was some built-in function
in R that would let me run all five analyses simultaneously (like how SPSS handles NAEP data with regression and ANOVA).  For my analyses, I am wanting to look at student-level science scores nested within states.  I am wanting to do this to see the effect of states adding science components to
their state mandated standardized assessments (like Texas did a few years back).  This is more curiosity on my part than any rigorous study.  Regardless, I would be very interested in seeing the vignette for the class.acc() function.

Thanks,
Kyle


***************************************
J. Kyle Roberts, Ph.D.
Baylor College of Medicine
Center for Educational Outreach
One Baylor Plaza, MS:  BCM411
Houston, TX   77030-3411
713-798-6672 - 713-798-8201 Fax
jkrobert at bcm.edu
***************************************

-----Original Message-----
From: Doran, Harold [mailto:HDoran at air.org] 
Sent: Thursday, June 07, 2007 10:13 AM
To: Roberts, J. Kyle; r-sig-mixed-models at r-project.org
Subject: RE: [R-sig-ME] NAEP Data and lmer

Kyle:

What do you mean by "handle the plausible values thing in R". The
class.acc() function in the MiscPsycho package is essentially the same expression used for calculating NAEP plausble values. I can send the vignette that shows the computational details. Since for each student you have multiple plausible values, can you give us a sense of what your data structure
looks like and how you're analyzing it in order to answer your question? 

Harold


> -----Original Message-----
> From: r-sig-mixed-models-bounces at r-project.org
> [mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf Of 
> Roberts, J. Kyle
> Sent: Thursday, June 07, 2007 11:03 AM
> To: r-sig-mixed-models at r-project.org
> Subject: [R-sig-ME] NAEP Data and lmer
> 
>  
> Friends,
> 
> Have any of you used either lme or lmer to analyze NAEP data? 
> I have done some things with HLM (the software), but wondered if there 
> was already a set of routines for handling the whole "plausible 
> values" thing in R. Also, for those of you familiar with the NAEP and 
> HLM, does HLM just average the results of the plausible values, or is 
> the final estimate a maximized estimate based on the five plausible 
> values?
> 
> Thanks,
> Kyle
> 
> P.S. - Be on the watch for the first ever "North American Conference 
> on Multilevel Analysis" coming late Spring 2008!!
> 
> ***************************************
> J. Kyle Roberts, Ph.D.
> Baylor College of Medicine
> Center for Educational Outreach
> One Baylor Plaza, MS:  BCM411
> Houston, TX   77030-3411
> 713-798-6672 - 713-798-8201 Fax
> jkrobert at bcm.edu
> 
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list 
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>