Talk:Census

--139.76.128.71 18:25, 6 January 2006 (UTC)--139.76.128.71 18:25, 6 January 2006 (UTC)UK Census:

1981 was too early for Poll Tax (should we use "community charge" here? The name Poll tax is not neutral. But then not using it isn't either.) The poll tax was near the end of Thatcher! It didn't hit the 2001 census too badly from what I've heard, but there is some lingering suspicion. The (famous London) poll tax riot was March 1990.

Early UK census: wouldn't that part of the UK under Roman occupation have been censused? (So that the claim of the AD C7 as the earlies would be false?)

Muppet 10-Jun-2003 16:09 BST

Census quote.

There was some quote used when I was in school pertaining to the census and government. Somthing along the lines of the census is the foundation of a government or the first thing it does. I don't know the exact quote, but I feel it would be a good idea to add it to the article.

Ambiguity in first paragraph

The last two sentences of the first paragraph contain an ambiguity:

[Taking a census] can be contrasted with sampling in which information is only obtained from a subset of a population. As such it is a method used for accumulating statistical data.

The "it" in the last sentence: is this supposed to refer to taking a census or sampling? As it appears in the "Census" article, I have to assume it refers to taking a census. But census data would not, as far as I'm concerned, be "statistical data". You don't (usually) do statistics on data from an entire population, you do it on samples. Right? I'm currently having a very similar discussion in Talk:Statistics. Someone please come agree with me. <g> - dcljr 04:13, 13 Sep 2004 (UTC)

You are thinking of inferential statistics. It is quite possible to have descriptive statistics from an entire population. I agree that sentence is not as clear as it could be. zzuuzz ^(talk) 10:03, 22 August 2005 (UTC)

---

zzuuzz is absolutely correct. Statistical analysis using descriptive statistics (central tendencies, frequency distributions [histograms, paretos et al.]etc..)can be performed on a sample or a population. Statistical analysis using inferential statistics can only be performed on a sample. The behavior of the population can be "inferred" from the statistical analysis of the sample. The confusion stems from calling a "parameter" a "statistic". From the Triola book "An Introduction to Statistics", a Population is the complete collection of all elements to be studied. A Census is the collection of data from every element in a population. A Sample is a subset from the population. A Parameter is a numerical measurement describing some characteristic of a population. A Statistic is a numerical measurement describing some characteristic of a sample. For example, in the 1984 US Presidential election, Reagan received 58.8% of the popular vote. Since this numerical value is from the all the popular votes cast during the election, 58.8% is a parameter. Exit polls during the 1984 election showed Reagan with 61% of the popular vote. Since Exit polls are based on samples, 61% is a statistic. statzman 2006 Jan 06

Ancient censuses

Edited the Ancient censuses section: separated the Biblical references from the historical statement about the Romans and cleaned up the Biblical paragraph.

Both paragraphs could use more work, especially the Roman one. Hopefully, more information, and from other cultures, can be added to the section to make it less dependent on a religious text. SpacemanSpork 21:16, 2005 Feb 23 (UTC)

United States census

The US section contains the following sentence:

But there is not a federal census legislation (nor for federal voting).

I would edit it to clarify but I don't know what it means. Can someone else help? Chick Bowen 22:12, 20 September 2005 (UTC)

Methods of performing the census

I feel this page is lacking. I was looking for information on countries where the census is performed by querying the government databases rather than by filling forms (such as my native Finland). Couldn't find any information on this as based on this page, most countries still live in the dark ages.—This unsigned comment was added by 213.169.28.108 (talk • contribs) .

It's true, most countries that have a census, perhaps by definition, send forms out to the population. There are certainly some recent developments with population databases which will spell the end of the census as we know it, which should be mentioned. -- zzuuzz ^(talk) 21:31, 31 March 2006 (UTC)

A census is the process of obtaining information about every member of a population (not necessarily a human population). By this "definition", a database query, which spans the whole population of a country is a census. Being Finnish and never having gone through the process of filling out the forms, anything but the queries just sounds preposterous to me. If the government knows how much they should tax me (and they do, and not simply because I work at a public institution), it shouldn't be too hard to count me as a member of the population. It's just a matter of efficiency.

Actually, the idea of a census is to get info of the entire population of a country at a given time generally whether or not they are paying tax, students, illegal immigrants, criminals, tourists etc etc. In many countries, there are limitations on the amount and type of data that may be stored on a person and how this can be shared (although exceptions could be made for a census of course). There are going to be quite a few people, legally resident in a country with very little info on government databases (only things like birth certificates etc). With very limited However I would say in many countries people would be concerned if the government is so easily able to colate the information in all their databases. You call it efficiency, others may call it big brother. Note also that the census info is expected to be highly confidential and often info is asked that the government possibly shouldn't otherwise know, e.g. religion, ethnicity, country of origin, parternership status (the government may perhaps know marriage but possibly little else), living situation, household income (this is not necessarily the same thing as how much you are legally taxed), place of current residence (AFAIK in many countries it is quite legal for a person to use a PO box for most things although you will usually have to reveal your place of residence for voting) and other stuff I can't remember. And the census also in theory anyway makes it more likely that the truth will be known since it's supposed to be competely confidential so people are hopefully more likely to tell the truth then whatever facade they may have been keeping up for whatever reason. I'm also somewhat doubtful whether it would be more efficient to collate all the data in the pre-computer age. Some of it like residence type for example may be known from local council records if you know the address. But we're talking about a lot of diverse databases here and I personally suspect in the pre-computer age anyway it would still have been more efficient to just use forms rather then collate the diverse amounts of data in diverse databases. Note even in Malaysia, where identity cards are compulsory after 12 (and you're supposed to keep your address updated) census are still carried out. Sure Malaysia has a lot of inefficiencies but I don't personally believe it is simply because of that. There is a lot more censuses are intended to cover then databases do and should have. However if you're sure Finland does conduct a 'database' census or dump then you're welcome to add what you know. Nil Einne 00:41, 30 October 2006 (UTC)

Looking at this page [1] it appears I'm right and in fact it was only recently (1990) that Finland moved to a 'register' based census... BTW, one thing to consider is that Finland is AFAIK a country with a very high level of social welfare and therefore the government tends to have a lot of into on what their citizens. Many other countries have less and therefore the goverment similarly has less info. Universities for example are not necessarily state (government) owned and therefore they may have their own policies about what data they share with who and people may not like it if the goverment forces them to share data. Perhaps it comes down to what I said however. In many countries, especially the English speaking developed. commonwealth countries (including the US here even though it's not technically commonwealth) the people tend to have a greater concern about what info the government has on them. While it may seem like a good idea in theory to make a greater use of available resources and safe time and money this is more likely to concern these people. Even if the information is already there and perhaps it can already be easily be collated, actually doing so may lead to (perhaps irrationally) a great level of concern so it's simply not done. Coming from Malaysia, I don't have so much concern (for example I don't really care about ICs) but I can safely say people here in NZ often do. All countries are moving away from this to some extent in this modern world but at the same time, there is a greater awareness and concern about privacy issues. Sorry if I'm not making sense, I'm tired. Nil Einne 00:46, 30 October 2006 (UTC)

Ok, maybe in other countries people are more sensitive about their governments keeping data on them. Still, my original point was that this page should cover the more modern ways of doing a census. Clearly the page is lacking. Also, I would like to note that actually my privacy is a lot better protected by my government than in many places where the census is still conducted in the conventional way. For example, the US, the Patriot Act gives the government a lot of power, but according to you, the people would still be offended if their data would be in government database from which their information could easily be queried. Actually, those databases actually exist already, but for some reason, they are not used for this purpose. --62.142.198.22 10:34, 24 March 2007 (UTC) (Ok, there seems to some information on Danish census)

Historical populations

I recently created the {{Histpop}} template to make it easier to read and edit historical population tables in demographics sections of articles.   JEK   Tuesday, November 5 2024 at 08:41

Histpop removed in favor of {{USCensusPop}}, but it can still be used for non-US locations. /Timneu22 00:55, 2 March 2007 (UTC)

The Law

Correct me if I missed it, but I saw no mentioning of the laws of taking a census. I mean, I know for a fact that in Canada, it is required by law that every household must complete a census form, and failure to do so could result in some form of criminal prosecution (the ads for our most recent census drilled this into our heads). I don't know what the law is surrounding other countries, but I think this should be looked into and integrated into the article in its appropriate place. -- Reaper X 20:12, 22 August 2006 (UTC)

I believe it's true in most countries or at least most commonwealth countries. It's true in New Zealand and AFAIK also Australia, Malaysia and the UK. N.B. It's mentioned here that it's compulsory in India and in the Australian subpage it's compulsory in India but you're right the info is lacking. Nil Einne 01:08, 30 October 2006 (UTC)

Anniversary Links

Under the heading "canada", the two links quinquennial and decennial both redirect to anniversary, which has no relevance to Canadian census regulations in any way. Either the links should be removed, or should link to articles that are relevant to Canadian censuses. --Nin 08:41, 27 September 2006 (UTC)

and it is also vital to democracy (voting)

This sentence in the introduction has been bugging me for a while. In many countries (such as the UK) there is a completely separate register for voters, and census records are not used for any other purposes than statistical abstracts and perhaps release after 100 years. Are there countries which rely on census records for voting, or does this refer to counting different groups to ensure representation, or does this relate specifically to the counting done for Gerrymandering in the United States? -- zzuuzz ^(talk) 00:56, 30 October 2006 (UTC)

AFAIK voter records are usually seperate. At least in NZ and Malaysia they are as far as I know (although in Malaysia 'voter records' they are part of a wider database). However I think you might be missing the point. Voter records are AFAIK often (usually?) not used to draw electoral boundries because the precise locations of voters may not be recorded in voter records. Census info does record such info and it also records all eligible voters regardless of whether or not they're registered to vote (if you have to register to vote, I believe you don't have to in the US). So the census is usually seen as the best way to draw electroral boundries and this may be required in law. See [2] for example. Note the actual voter records never intersect with the census info and the census info is confidential and only demographical info is released. Nil Einne 01:07, 30 October 2006 (UTC)

Interestingly the UK does not appear to use census data however "The Commission obtain the enumeration date electorate figures from electoral registration officers, either directly, or indirectly through the Office for National Statistics (ONS)" so you're right there Nil Einne 01:17, 30 October 2006 (UTC)

So census data is used to draw electoral boundaries according to counts of particular groups. Is that right? I'd like to stretch this sentence out into a wider point about representation. -- zzuuzz ^(talk) 01:43, 30 October 2006 (UTC)

UK Census information being sold

I changed the reference to UK Census statistical information being 'sold to interested parties' to 'also made available in published reports and on the ONS website.'

Although it's true that if you wanted one of the printed reports you'd have to pay, you could get the pdf of the report free on the web, and you can get hold of any of the results on CD (including the detailed stuff not in the printed reports) on CD free of charge from ONS. I agree that you could have a bit of a detailed quibble about whether the samples of anonymised microdata are charged for or not, but it seemed to be less misleading, without getting into huge detail, to remove the suggestion that you needed to buy the data. Jonesey67 13:29, 13 January 2007 (UTC)

1931 UK Census

You should be aware that there is no 1931 Census. It was destroyed during thr London Blitz. Regards, Paul Norfolk Dumpling 10:24, 19 February 2007 (UTC)