Complex sampling survey _ Use of survey package

Hi Sylvie,

You still use too much tilde signs (~) as Thomas Lumley already pointed 
out.

Replace your line

clustot <- svydesign(id=~num+ ~ Id_An, fpc=~fpc1+~Totanim, data=tab1)

with

clustot <- svydesign(id=~num+Id_An, fpc=~fpc1+Totanim, data=tab1)

Hope this helps,
Tobias

P.S.

"~ num + Id_An" is called a formula in R and is omnipresent as
an interface in R functions


Ahoussou Sylvie wrote:

--------------------------------------------------
From: "Ahoussou Sylvie" <sylvie.ahoussou at antilles.inra.fr>
Sent: Friday, September 12, 2008 9:48 AM
To: "Thomas Lumley" <tlumley at u.washington.edu>
Subject: Re: [R] Complex sampling survey _ Use of survey package

Thanks for your answer

I think I made a mistake when I recopied the 5 first rows of my database

here is the table with the comlums of interest

num esp fpc1 Totanim Id_An
2045 G 551 12 10
2046 C 551 68 11
2070 G 551 9 50
2070 S 551 9 51
2070 S 551 9 52



yes Totanim is the total number of animals in the farm and num is the 
total number of herds

I keep on obtaining this error message

clustot<-svydesign(id=~num+ ~ Id_An, fpc=~fpc1+~Totanim, data=tab1)

Erreur dans as.fpc(fpc, strata, ids) :
 FPC implies >100% sampling in some strata.


--------------------------------------------------
From: "Thomas Lumley" <tlumley at u.washington.edu>
Sent: Friday, September 12, 2008 12:02 AM
To: "Ahoussou Sylvie" <sylvie.ahoussou at antilles.inra.fr>
Cc: <r-help at r-project.org>
Subject: Re: [R] Complex sampling survey _ Use of survey package

On Thu, 11 Sep 2008, Ahoussou Sylvie wrote:

Data base = tab1 here are the five first rows of the database (nrow = 
11792)




    num
    esp
    Quarters
    Totcat
    Totshp
    Totgt
    Tbtpos
    fpc1
    Totanim
    Id_An

     10
    2045
    G
    01-Q1
    0
    0
    12
    1
    551
    10
    10

     11
    2046
    G
       01-Q1
    8
    0
    60
    1
    551
    11
    11

     50
    2070
    G
       01-Q1
    0
    3
    6
    1
    551
    50
    50

     51
    2070
    S
       01-Q1
    0
    3
    6
    1
    551
    51
    51

     52
    2070
    S
    01-Q1
    0
    3
    6
    1
    551
    52
    52

This has 11 data values in each row but only 10 variable names.  If we 
assume the first number doesn't belong it, the fpc1 column lines up 
correctly, so I will do that.

It is surprising that Id_An is the same as Totanim for each of these 
five records

- First strata = herds

herd's id column = num

sampled in 551 herds (column fpc1)



- Second strata = animals

animal id column= Id_An

sampled in Totanim



I tried :

clustot <- svydesign (id=~num+Id_An, fpc=~fpc1+~Totanim, data=tab1)

This should be
  clustot <- svydesign (id=~num+Id_An, fpc=~fpc1+Totanim, data=tab1)
but that doesn't explain the problem

Your svydesign() call looks correct if fpc1 is the total number of 
herds in the population and Totanim is the total number of animals in 
the particular herd. I am not sure from your explanation whether this 
is how the variables are defined.

      -thomas

and I get this error message

clustot <- svydesign(id=~num+Id_An, fpc=~fpc1+~Totanim, data=tab1)

Erreur dans as.fpc(fpc, strata, ids) :

 FPC implies >100% sampling in some strata.



I don't have any strata with 100% sampling so do you know which 
arguments of the fonction svydesign I'm supposed to use ?



Thanks in advance



AHOUSSOU Sylvie
V?t?rinaire Epid?miologiste
CIRAD Domaine Duclos
97 170 Petit-Bourg
tel : 05 90 25 59 47
[[alternative HTML version deleted]]

Thomas Lumley Assoc. Professor, Biostatistics
tlumley at u.washington.edu University of Washington, Seattle