[R] sem by variable x

[Daniel Caro]

Hi Jarret,

Thank you for your answer. I get the following message:

Error in cov(a.df[, 2:7], nrow(a.df)) : incompatible dimensions

The function seems to run for some countries but then appears to stop
when a country has incomplete data (1 var is missing, for example).
How to force the function to continue or skip those countries?

Thank you,
Daniel



On Tue, Jul 20, 2010 at 7:24 PM, Jarrett Byrnes <byrnes at msi.ucsb.edu> wrote:
> You may want to take a look at the lavaan package and use the multigroup analysis there (and see if you even need to group by country as well).
>
> Otherwise, you could do something like
>
> library(sem)
> library(plyr)
>
> cfa_func<-function(a.df){
>
>        cfa<-sem(ses.model, cov(a.df[,2:7], nrow(a.df)))
>        print(summary(cfa))
> }
> d_ply(data, "idcntry", cfa_func)
>
> -Jarrett
>
> On Jul 20, 2010, at 6:22 AM, Daniel Caro wrote:
>
>> Hi R users,
>>
>> I am new in R. I would like to perform confirmatory factor analysis
>> for a data set of countries. My data are:
>>
>> data <- read.csv("ses.raw", header = TRUE)
>> attach(data)
>> names(data)
>>
>> [1] "idcntry" "momed"   "daded"   "dadocu"  "momocu"  "hompos"  "finan"
>>
>>
>> The country id is "idcntry", my model is "ses.model", and variables to
>> be included in the analysis are "momed"   "daded"   "dadocu"  "momocu"
>> "hompos"  "finan" . How can I run
>>
>> cfa<-sem(ses.model, cov(data[,2:7], nrow(data)))
>> summary(cfa)
>>
>> by country? I am able to perform sem on all data by not by country. I tried
>>
>> by(data[,2:7], idcntry, function(x) sem(ses.model, cov(data[,2:7]), nrow(data)))
>>
>> but the output is the same for all countries.
>>
>> Thank you,
>> Daniel
>>
>> ______________________________________________
>> R-help at r-project.org 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.
>
>