[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: FW: st: new problem with table display

From   "Richard Ohrvall" <>
Subject   Re: FW: st: new problem with table display
Date   Thu, 1 Nov 2007 13:35:53 +0100

I think this discussion is drifting away somewhat. Of course I
recognise that figures are often presented with a level of detail that
is not really motivated given the statistical uncertainty or given
uncertainty of other kinds (non-sampling errors etc.) But the degree
of precision of the presented results depends of the specific
situation, sometimes we actually know the exact figures and other
times there might be some other reason to present exact figures, eg.
to keep consistency with other official figures. And there are
certainly occassions where we can present estimates with higher
precision than what -tab- gives us.

I really liked the story about the 1971 census in UK. Thank you for
that one, Nick.


On 10/31/07, Maarten buis <> wrote:
> --- "Lachenbruch, Peter" <> wrote:
> > What wasn't mentioned, but implied here is that census data is not
> > "known" but merely an estimate.  Your other email that indicated the
> > population of the Netherlands is about 16 million suggests this.
> Actually, there hasn't been a census in the Netherlands since 1975,
> these numbers are based on register data collected by municipalities.
> The problem with register data is often more one of (considerable)
> systematic bias than one of uncertainty.
> When you are interested in the entire Dutch population these estimates
> should be very accurate. The only big problem is that, as with any
> register data, illegal activity is not documented. So illegal aliens
> are not represented in this number.
> When looking at population sizes at the municipality level biger
> problems arise: This depends on people who move from one municipality
> to another to register with their new municipality. They are required
> to do so by law, and it can make certain things like getting a passport
> much easier, but going to city hall is a hassle so some people (me)
> tend to postpone that quite a while. This is generally true for
> register data: It depends on people filling out forms, and people are
> not as conscientious as they are required to be by law.
> Register data becomes even worse when it involves a politically
> contentious issue, like unemployment. In those cases the statistics are
> often more representative of politically convenient definitions than of
> real world events. I recently attended a presentation showing German
> unemployment figures which showed that youth unemployment was actually
> lower than unemployment in the general population...
> -- Maarten
> -----------------------------------------
> Maarten L. Buis
> Department of Social Research Methodology
> Vrije Universiteit Amsterdam
> Boelelaan 1081
> 1081 HV Amsterdam
> The Netherlands
> visiting address:
> Buitenveldertselaan 3 (Metropolitan), room Z434
> +31 20 5986715
> -----------------------------------------
>      ___________________________________________________________
> Want ideas for reducing your carbon footprint? Visit Yahoo! For Good
> *
> *   For searches and help try:
> *
> *
> *
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index