[R] Significance of Svyrepdesign Object Warning

[Anthony Damico]

hi, great example.  i am ccing survey package author/maintainer dr.
lumley.  why do you have `na.action=na.exclude`?  if you remove it, things
work as expected--


    library(RCurl)
    library(survey)
    data <- getURL("
https://raw.githubusercontent.com/cbenjamin1821/careertech-ed/master/elsq1adj.csv
")
    elsq1ch <- read.csv(text = data)
    #Specifying the svyrepdesign object which applies the BRR weights
    elsq1ch_brr<-svrepdesign(variables = elsq1ch[,1:16], repweights =
elsq1ch[,18:217], weights = elsq1ch[,17], combined.weights = TRUE, type =
"BRR")
    elsq1ch_brr
    #Logistic regression call which yields a warning regarding svyrepdesign
object

    # your warning
    a <-
svyglm(formula=F3ATTAINB~F1PARED+BYINCOME+F1RACE+F1SEX+F1RGPP2+F1HIMATH+F1RTRCC,family="binomial",design=elsq1ch_brr,subset=BYSCTRL==1&G10COHRT==1,na.action=na.exclude)
    summary(a)

    # works fine
    a <-
svyglm(formula=F3ATTAINB~F1PARED+BYINCOME+F1RACE+F1SEX+F1RGPP2+F1HIMATH+F1RTRCC,family="binomial",design=elsq1ch_brr,subset=BYSCTRL==1&G10COHRT==1)
    summary(a)



    the mismatch of vectors generating that warning happens inside

    debug(survey:::summary.svrepglm)

    [..snip..]

    Browse[2]> length(presid)
    [1] 12614
    Browse[2]> length(object$survey.design$pweights)
    [1] 8397


    and including vs excluding the na.action=na.exclude gives you a
slightly different dispersion parameter calculation

        (Dispersion parameter for binomial family taken to be 0.7756235)

        (Dispersion parameter for binomial family taken to be 0.7849244)


not sure if the two survey:::residuals.sv* methods should deal with the
na.action= parameter?


thanks

On Sun, Oct 23, 2016 at 11:56 AM, Courtney Benjamin <cbenjami at btboces.org>
wrote:

> Hello R Users,
>
> I am using Lumley's Survey Package in R to analyze complex survey data
> that involves 200 balanced repeated replicate (BRR) weight variables.  I
> have ensured that my svyrepdesign object that specifies the application of
> the BRR weights to the data set is accurate and I have matched the
> published standard errors of the data set.
>
> When doing a logistic regression through the svyglm call, I receive the
> following warning:
>
> In object$survey.design$pweights * presid^2 :
>   longer object length is not a multiple of shorter object length?
> I have search around quite a bit online and have not been able to find any
> good interpretation of its meaning.  I want to be sure that I am not making
> some type of mistake that is causing this warning to be produced.  Any
> advisement is greatly appreciated.
> The following is an MRE that can be pasted into the R console:
> library(RCurl)
> library(survey)
> data <- getURL("https://raw.githubusercontent.com/
> cbenjamin1821/careertech-ed/master/elsq1adj.csv")
> elsq1ch <- read.csv(text = data)
> #Specifying the svyrepdesign object which applies the BRR weights
> elsq1ch_brr<-svrepdesign(variables = elsq1ch[,1:16], repweights =
> elsq1ch[,18:217], weights = elsq1ch[,17], combined.weights = TRUE, type =
> "BRR")
> elsq1ch_brr
> #Logistic regression call which yields a warning regarding svyrepdesign
> object
> svyglm(formula=F3ATTAINB~F1PARED+BYINCOME+F1RACE+F1SEX+
> F1RGPP2+F1HIMATH+F1RTRCC,family="binomial",design=
> elsq1ch_brr,subset=BYSCTRL==1&G10COHRT==1,na.action=na.exclude)
> allCC <- summary(svyglm(formula=F3ATTAINB~F1PARED+BYINCOME+
> F1RACE+F1SEX+F1RGPP2+F1HIMATH+F1RTRCC,family="binomial",
> design=elsq1ch_brr,subset=BYSCTRL==1&G10COHRT==1,na.action=na.exclude))
> allCC
>
> #Session Info
> #R version 3.3.1 (2016-06-21)
> #Platform: x86_64-w64-mingw32/x64 (64-bit)
> #Running under: Windows >= 8 x64 (build 9200)
>
> #locale:
> #  [1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United
> States.1252
> #[3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C
> #[5] LC_TIME=English_United States.1252
> #attached base packages:
> #  [1] grid      stats     graphics  grDevices utils     datasets
> methods   base
> #other attached packages:
> #[1] survey_3.31-2   survival_2.39-4 Matrix_1.2-6    RCurl_1.95-4.8
> bitops_1.0-6
> #loaded via a namespace (and not attached):
> #[1] tools_3.3.1     splines_3.3.1   knitr_1.14      lattice_0.20-33
>
>
> Courtney Benjamin
>
> Broome-Tioga BOCES
>
> Automotive Technology II Teacher
>
> Located at Gault Toyota
>
> Doctoral Candidate-Educational Theory & Practice
>
> State University of New York at Binghamton
>
> cbenjami at btboces.org<mailto:cbenjami at btboces.org>
>
> 607-763-8633
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
>

	[[alternative HTML version deleted]]