A colleague from Holland has kindly sent me some information that
answers some of the questions I asked about the EU-SILC sampling method,
see below - lucky its in English!
Jay Ginn


Specific information on  the Dutch EU-SILC:
                  a)  Survey  model.  An integrated model for the
cross-sectional and longitudinal survey.
b) Sampling frame. The response part of Labour Force Survey (LFS) will
be used as the sampling frame for EU-SILC. In this way the LFS functions
as a panel survey. As a result of this, a substantial reduction of the
questionnaire can be achieved. In addition, this option will enable
Statistics Netherlands to use CATI as interview mode which is very
efficient with respect to the total cost of interviewing. Potential
selectivity in the response part of the LFS shall be corrected for by
adjusting the drawing probabilities to the requirements of a simple
random sample.
c) Sampling design. A rotational design based on 4 years. The sample
consist of 4 sub-samples, each similar in size and design and
representative for the whole population living in private households.
After the first year one sub-sample shall be dropped, the second
sub-sample shall be retained for two years, the third for three years
and the fourth for four years. In principle the first sub-sample which
shall be dropped after the first year but in case of additional
non-response in the longitudinal wave, some elements from the first
sub-sample shall be used for completing the other three sub-samples to
the required representative size.
d) Sample size. A simple random sample of 9000 persons aged 16 years or
over will be drawn from the respons part of the Labour Force Survey. Our
test showed  that 80 percent is willing to participate in EU-SILC and 75
percent actually participated in our test run of EU-SILC. This means
that the net sample will contain more than the 6,560 households
stipulated in the regulation.
e) Data sources. The required EU-SILC data shall be derived from
register information (especially the income information and most of the
demographic variables of the population register), the administrative
accounting system of the interview processing (including data like
duration of the interview), the response sample of the LFS (especially
the labour force information and the education variable) and the CATI
questionnaire.
In the CATI interview the head of the household or his/her partner shall
be asked the necessary information for households (e.g. housing, housing
environment, child care, possession of durables, subjective income
evaluation variables, housing and non-housing related arrears,
possibilities of affording holiday, a meal with meat/fish, etc). The
selected sample person (ssp) shall be asked the personal questions about
health and the special questions about the period when ssp was teenager.
f)  Income reference period.  The income data of EU-SILC 2005 refer to
the year 2004.
g) Income definition. The register information from the tax
administration shall be transformed in the required net income
definition (disposable household income) and net income components. The
paid/received income transfers between households (alimony for children,
parental grants for outside home students, financial support for
relatives outside the household) are not included in the tax based
income definition. This components will be collected in the CATI
questionnaire and subsequently included in the EU-SILC income definition

Jay Ginn
Visiting Professor
Surrey University
Guildford, GU2 7XH, UK
01737 559341


-----Original Message-----
From: Jan Simons [mailto:[log in to unmask]]
Sent: 19 April 2005 10:53
To: J.Ginn
Subject: Fw: RE:Inkomen_bestedingen_vermogen [#54064]

Good mrning Jay,

Attached please find a short document on EU-SILC (fortunately in English
!)
as obtained from our Dutch Statistics Bureau.
I am certainly not an expert on statistics and sampling methodology, but
to
me it looks like the Dutch (and other EU countries?) are using the same
sampling method at least for this particular survey.

I hope this answers your original question.

Kind regards,   Jan

      ----- Original Message -----
From: <[log in to unmask]>
To: <[log in to unmask]>
Sent: Monday, April 18, 2005 4:20 PM
Subject: RE:Inkomen_bestedingen_vermogen [#54064]


>
> Goede middag Mr. J. Simons,
>
> Zie bijgevoegd documentje.
>
> Met vriendelijke groet,
> Jos Haslinger
> Infoservice
> tel: 0900 0227 (50 cent per minuut)
> fax:070 337 5994
> E-mail: [log in to unmask]
>
> --Original Message--
> From: [log in to unmask]
> Date: 04/14/05
> To: [log in to unmask]
> Subject: Inkomen_bestedingen_vermogen[#54064]
>
>     formID: 1001
>       Naam: Mr. J. Simons
>   Geslacht: man
> Emailadres: [log in to unmask]
> Telefoonnr: 0180 - 516114
> submitTime: 14-4-2005 21:36:00
> Branche:
> Particulier
> Onderwerp: Inkomen_bestedingen_vermogen
> Vraag:
> EU doet onderzoek naar "Income and living conditions". Aannemend dat
ook
Nederland hierbij is betrokken,wil ik graag weten, welke methodologie
(sampling method) hiervoor wordt gebruikt.  In het Verenigd Koninkrijk
wordt
1/4 deel van de "sample" jaarlijks vervangen door nieuwe groep personen.
Na
4 jaar wordt een geheel nieuwe groep gekozen.
> Volgt Nederland dezelfde methode ?
>
>
> ---------------
> Aan de inhoud van dit e-mailbericht kunnen geen rechten worden
ontleend.
> De informatie verzonden in dit e-mailbericht is uitsluitend bestemd
voor
de geadresseerde. Het Centraal Bureau voor de Statistiek staat niet in
voor
de juiste en volledige overbrenging van de inhoud van een verzonden
e-mailbericht noch voor tijdige ontvangst daarvan.
> No rights may be derived from the contents of this e-mail message.
> The information in this e-mail message is intended only for the
addressee.
Statistics Netherlands cannot vouch for the correctness and completeness
of
the contents of e-mail messages, nor for the timely receipt thereof.
>
>

******************************************************
Please note that if you press the 'Reply' button your
message will go only to the sender of this message.
If you want to reply to the whole list, use your mailer's
'Reply-to-All' button to send your message automatically
to [log in to unmask]
*******************************************************