Talk:Anchorage, Alaska/Census debate
From Wikipedia, the free encyclopedia
"As of the census of 2000, there are 260,283 people, 94,822 households, and 64,099 families residing in the municipality."
So many problems... OK, for one, it's not even proper English. As of 2000, there were whatever. Secondly, it's an estimate, it shouldn't be presented as an exact number. Third, it's redundant. It's already mentioned at the top of the article. Anthony DiPierro 19:00, 20 Mar 2004 (UTC)
- I agree with you that the numbers are too precise. I think rounding to about the nearest 100 is good, and adding approximately. For percentages, round to the nearest whole percent. moink 19:08, 20 Mar 2004 (UTC)
- As this applies equally to tens of thousands of articles, it should be discussed elsewhere, and if there is consensus about any change of the format, a bot would have to be approved to change all articles. I refuse to even discuss it here on the talk page of a specific city. --Wik 19:13, Mar 20, 2004 (UTC)
-
- I think we should try this on a limited basis before extending it to 10,000 articles. Anthony DiPierro 19:20, 20 Mar 2004 (UTC)
The numbers are not too precise - those are the exact numbers reported by the census. If you have issues with the accuracy of the census, that I can understand. But the way to resolve that is to make the Demographics section say "according to the 2000 census", as the top of the page already does. Then we're just reporting what the census said, and that's indisputable fact. We have no basis for adjusting the census figures unless you've got some other sources that aren't being cited. --Michael Snow 00:24, 21 Mar 2004 (UTC)
I agree with Michael. These were the reported numbers from the census. They're the best numbers we have, there's no need to change them, especially not in every article the bot created. RickK | Talk 00:29, 21 Mar 2004 (UTC)
The numbers reported by the census are estimates, not exact numbers. If you're going to say that the census report is claiming these are exact numbers, then you need to show me where this claim is made. If you can show where the census report claims that the numbers are exact, then and only then is changing "As of the census of 2000" to "According to the census of 2000" proper. And you still have to change "are" to "were". And listing the number of people a second time is still redundant. Anthony DiPierro 20:39, 21 Mar 2004 (UTC)
- Where did I say they were exact numbers? Please re-read what I wrote: "They're the best numbers we have". And stop being a baby and trying to cause trouble because people don't let you get your way. RickK | Talk 01:47, 22 Mar 2004 (UTC)
-
- I wasn't responding to you on that point. Michael said "those are the exact numbers reported by the census." They're not exact numbers. They're approximate, and should be reported as such. And no more significant digits than are known should be presented. I'm sorry you consider trying to fix this article to be causing trouble. Anthony DiPierro 01:59, 22 Mar 2004 (UTC)
-
-
- Try going to the Census Bureau website, for example at [1]. The 2000 census reported a population of 260,283 for Anchorage. The census DID NOT report a population of 260,000. When I said "exact", that was exact as in identical (an exact replica), not exact as in correct. As I've said already, I know there are arguments about the reliability of census data. But even if you consider the census numbers an estimate or an approximation, the census number is STILL 260,283 (there is no rule that estimates or approximations have to be in round numbers). As for significant digits, what basis do you have for deciding how many significant digits are, or are not, known? Why not four (260,300), or one (300,000)? Personally, I think applying the concept of significant digits to the census misunderstands what significant digits are for. There are much better statistical methods available for evaluating census data. --Michael Snow 03:56, 22 Mar 2004 (UTC)
-
-
-
-
- That may be the number they present, but it is not claimed to be the exact population. Presenting it as such is incorrect. If you want to rephrase it in some way to avoid presenting the approximate data as exact, then feel free to do so. My suggestion is to round the figures and use the word "approximate". Actually, a better solution for this particular situation is to move the census data to a different article entirely, maybe Anchorage, Alaska census data. Because the data are completely superfluous here. Anthony DiPierro 10:52, 22 Mar 2004 (UTC)
-
-
-
-
-
-
- Superfluous? Do you think nobody who reads this article would be interested in knowing the population of Anchorage?
-
-
-
-
-
-
-
- I didn't say we should present 260,283 as the exact population, however you may be using that word; I've only ever used "exact" on the talk page. I said we should report that this data is "according to the 2000 census". Do you object to that language, too, or are you just objecting to the word "exact"? --Michael Snow 17:34, 22 Mar 2004 (UTC)
-
-
-
-
-
-
-
-
- As for the data being superfluous, perhaps a very small minority of people might be interested in this information. For those people, we could link to the government page which presents it. Wikipedia is not a source repository, after all.
-
-
-
-
-
-
-
-
-
- "As of the census of 2000, there are 260,283 people, 94,822 households, and 64,099 families residing in the municipality." Sounds exact to me. How do you propose we fix it? Anthony DiPierro 21:30, 22 Mar 2004 (UTC)
-
-
-
-
-
-
-
- This is a complicated question. On the one hand, we can be said to be quoting the census. And they reported a number which is overly precise for its accuracy. I've been taught in engineering school that you never ever give an extremely precise number when you don't know that the data is that accurate. And I highly doubt that the census is accurate to the person or even to the tens or hundreds. Thus, if I were the census, I would round it. But I'm not the census and they idiotically give too much precision to their results. But if we're actually quoting them, we should use their numbers. I'm undecided, I guess. moink 17:58, 22 Mar 2004 (UTC)
- I would say that we should use the census's numbers exactly as they appear in the census. There is no reason to round them off because we are reporting the official figures. Find the census figures questionable? Good for you. Then question them for yourself. What Wikipedia shouldn't get in the habit of doing is thinking for its readers -- by rounding off to a particular significant digit, we are suggesting "we know more about how inexact census figures are than you do: trust us, it's somewhere around here". Let's give people the numbers, make it clear they're based on the census, and give them the courtesy of assuming they can think about it as critically as they want. No one will sue us because they printed up 260,283 flyers to advertise a concert to the people of Anchorage and discovered subsequently that there were, in fact, 260,417 people. Someone might, though, get legitimately upset if their geography project was docked 1/2 a point because they didn't give the 2000 census's population number, but rather an intelligent estimate of it. Jwrosenzweig 18:07, 22 Mar 2004 (UTC)
- Agree with this completely. It is well known that census figures have a built-in inaccuracy, but it is highly controversial, and attempting to judge this accuracy for ourselves by assuming a stastical rounding error would do a disservice to readers.-- Decumanus | Talk 18:09, 22 Mar 2004 (UTC)
- You make a very good point. Perhaps the solution is to link the word census and then put an intelligent discussion of census accuracy (which I admit I'm not too capable of) in census. moink
- The word census is already linked on the page. So is United States Census Bureau. And yes, those would be the right places to discuss census accuracy. There's some discussion of the issue there, although I'm sure it could be improved. --Michael Snow 19:07, 22 Mar 2004 (UTC)
- I would say that we should use the census's numbers exactly as they appear in the census. There is no reason to round them off because we are reporting the official figures. Find the census figures questionable? Good for you. Then question them for yourself. What Wikipedia shouldn't get in the habit of doing is thinking for its readers -- by rounding off to a particular significant digit, we are suggesting "we know more about how inexact census figures are than you do: trust us, it's somewhere around here". Let's give people the numbers, make it clear they're based on the census, and give them the courtesy of assuming they can think about it as critically as they want. No one will sue us because they printed up 260,283 flyers to advertise a concert to the people of Anchorage and discovered subsequently that there were, in fact, 260,417 people. Someone might, though, get legitimately upset if their geography project was docked 1/2 a point because they didn't give the 2000 census's population number, but rather an intelligent estimate of it. Jwrosenzweig 18:07, 22 Mar 2004 (UTC)
- This is a complicated question. On the one hand, we can be said to be quoting the census. And they reported a number which is overly precise for its accuracy. I've been taught in engineering school that you never ever give an extremely precise number when you don't know that the data is that accurate. And I highly doubt that the census is accurate to the person or even to the tens or hundreds. Thus, if I were the census, I would round it. But I'm not the census and they idiotically give too much precision to their results. But if we're actually quoting them, we should use their numbers. I'm undecided, I guess. moink 17:58, 22 Mar 2004 (UTC)
-
-
-
-
-
-
-
- Again, I don't see why we need to include this data in the first place. At the very least, it should be moved to a separate page. As long as this page still has this crap in it, it will never be brilliant prose in my mind. Anthony DiPierro 21:33, 22 Mar 2004 (UTC)
- While we should of course aim for brilliant prose where this is practical, an encyclopedia's goal is ultimately not to delight with prose genius but to inform. The information you object to is useful and desired information which we will continue to provide on this article unless the community wishes differently. I have seen no evidence yet that anyone but Anthony wishes differently. Furthermore, characterizing factual information as "crap" indicates to me a disdain for the very nature of an encyclopedia. If this is indeed your position, Anthony, I suggest that being involved in the creation of a factual encyclopedia may be a project you will ultimately be disappointed by. Jwrosenzweig 22:12, 22 Mar 2004 (UTC)
- Again, I don't see why we need to include this data in the first place. At the very least, it should be moved to a separate page. As long as this page still has this crap in it, it will never be brilliant prose in my mind. Anthony DiPierro 21:33, 22 Mar 2004 (UTC)
-
-
-
-
Okay, it seems pretty clear to me that Anthony simply refuses to accept the numbers as they stand ("too exact"). Reviewing the discussion above, I think most people participating in the discussion want to keep the census data without changing the numbers, but perhaps modifying the language slightly. Sounds like an impasse - should we move on and start a poll on the subject? --Michael Snow 21:41, 22 Mar 2004 (UTC)
- This issue has only been discussed for 2 days. That's far too soon for a poll. See Wikipedia:Polling guidelines. As for your accusation of what I will and won't accept, no one has even proposed an alternate phrasing. Anthony DiPierro 22:07, 22 Mar 2004 (UTC)
-
- I propose "According to the 2000 United States census records, there are 260,283 people, 94,822 households, and 64,099 families residing in the municipality." I am open to suggestions for modification and improvement. Jwrosenzweig 22:15, 22 Mar 2004 (UTC)
-
- That sounds like pretty fair wording to me. DJ Clayworth 22:19, 22 Mar 2004 (UTC)
-
- I agree. I already suggested the "according to" phrasing on a couple of occasions, but perhaps that wasn't explicit enough for everyone to recognize it as a proposal. --Michael Snow 22:23, 22 Mar 2004 (UTC)
-
- First of all, "are" should be "were". Secondly, that's not what the US census records say. The US census records say that 260,283 people (households, etc) were counted, they do not say that that is how many people reside in the municipality. If you're going to say that figure represents the actual number of households (the US census records do not make that claim), you have to say "approximately". Anthony DiPierro 00:10, 23 Mar 2004 (UTC)
-
- As far as wording, that sentence would be fine with me (for the first sentence, the rest would have to be changed similarly). Another issue is that the "people" figure is redundant (already in the first paragraph). And the other issue is that I feel this whole section should be moved to a separate article. Anthony DiPierro 00:15, 23 Mar 2004 (UTC)
-
-
- The rest of the data are mostly percentages and averages, not counts. How would you word those? As for moving the census data to a separate article, this article is not substantial enough to warrant splitting it. --Michael Snow 00:27, 23 Mar 2004 (UTC)
-
-
-
-
- I would word them as "approximate". Anthony DiPierro 04:28, 23 Mar 2004 (UTC)
-
-
-
-
-
-
- How would you do that? Without telling us the context for your proposed use of the word "approximate", I don't think we can evaluate this as a proposal. I can't tell what the effect would be without seeing how the word is used in a sentence. How would you write the disputed section so that it's NPOV, factually accurate, and takes into account both our concerns and yours? --Michael Snow 07:08, 23 Mar 2004 (UTC)
-
-
-
-
-
-
-
-
- I've already done it. It's the reason this page was protected in the first place. Substitute the "exact" numbers in for the rounded ones. anthony 07:15, 23 Mar 2004 (UTC)
-
-
-
-
-
-
-
-
-
-
- Okay, I'll operate on the assumption that your last edit of the page is your proposal, but with the different numbers. Given that the word "approximately" appears numerous times in the first two paragraphs of the demographics section, and not at all in the last two, I have to believe the proposal is incomplete. And if the last two paragraphs are edited in a similar fashion, the section would be positively littered with the word "approximately". I don't think we need to beat people over the head with our doubts about census data. As a result, I still prefer Jwrosenzweig's proposal.
-
-
-
-
-
-
-
-
-
-
-
- As Jwrosenzweig also pointed out above, we can simply state the fact that the census reported certain statistical information, and let readers make their own judgments about the value of it. I would add that restating the census data, with the source clearly identified, is purely factual and NPOV. On the other hand, rounding the numbers or fudging them by calling them "approximate" is POV (specifically, it introduces POV about the census). In my opinion, the place to discuss the multiple POVs about the census is on the census pages, which are linked from this one. --Michael Snow 17:35, 23 Mar 2004 (UTC)
-
-
-
-
-
I propose leaving the damn thing exactly the way it is, unless somebody wants to create a bot to change EVERY SINGLE ONE of the city articles to match whatever other wording somehow gets voted on. And this poll should DEFINITELY not be here, since it affects every location in the United States that we are reporting on. RickK | Talk 04:15, 23 Mar 2004 (UTC)
- I'll create the bot. But I think this should be done on one page first, before doing it on all pages. Anthony DiPierro 04:27, 23 Mar 2004 (UTC)
-
- I don't think the wording of every article created by the bot needs to stay exactly the same forever (again, exact as in identical, not as in correct). A little stylistic variety can be accepted. So we need not convert this into a policy discussion that affects the whole encyclopedia. We're just talking about Anchorage here, nothing more. --Michael Snow 07:08, 23 Mar 2004 (UTC)
- We're talking about Anchorage here and now, but if Anthony gets his way here, you know he's going to want to expand it. RickK | Talk 02:31, 24 Mar 2004 (UTC)
- Well, wik was the one bringing up the other pages. If it all works out, then I'd like to expand it, but as I've said above, I think we should try it on a limited basis first. anthony 21:14, 25 Mar 2004 (UTC)
- We're talking about Anchorage here and now, but if Anthony gets his way here, you know he's going to want to expand it. RickK | Talk 02:31, 24 Mar 2004 (UTC)
- I don't think the wording of every article created by the bot needs to stay exactly the same forever (again, exact as in identical, not as in correct). A little stylistic variety can be accepted. So we need not convert this into a policy discussion that affects the whole encyclopedia. We're just talking about Anchorage here, nothing more. --Michael Snow 07:08, 23 Mar 2004 (UTC)
I don't see any need to change anything on this page or any other page on any other city anywhere in the world. It's fine just like it is, and this whole dispute is a tempest in an igloo. Doovinator 14:24, 25 Mar 2004 (UTC)
I moved the useless crap into a table. That way it isn't phrased as a sentence and the fact that it is merely census data is evident from the table caption. anthony (this comment is a work in progress and may change without prior notice) 14:29, 3 Apr 2004 (UTC)
- While I think calling census data "useless crap" is needlessly provocative, I'm willing to accept this as a resolution of the dispute as it applies to this page. --Michael Snow 22:34, 9 Apr 2004 (UTC)
-
- Good thing there isn't a rule against being needlessly provocative. I don't consider this a resolution. I still feel the information should be removed. But at least the POV and inaccurate aspects of the information have been relieved. It's also slightly less obnoxious as a table. anthony (see warning) 22:41, 9 Apr 2004 (UTC)
- First of all, it is against the spirit of Wikilove to needlessly provoke people, Anthony -- not a rule, but take that as a very good guideline. Secondly, the issue was resolved for everyone but you -- we do things by consensus here, and consensus was (and is) against removing the census data. I think the table is acceptable (though not really preferable to the previous solution), but I would like you to stop agitating to remove the data from this page. You make good contributions to corners of this project, and I know you can apply that attitude here and recognize that the census data is seen as useful and important by this community, even if it isn't by you. Jwrosenzweig 22:47, 9 Apr 2004 (UTC)
- The consensus among an extremely small portion of the community is to keep the census data. In fact, I haven't heard anyone's opinion on whether or not the data could be moved to a separate page, so as not to clutter this one. But yes, I don't see anyone else pushing to remove this data (any more, rambot has been accused of adding "gazetteer entries and not encyclopedia articles", so much in fact that it's a freqently asked question), so that's why I've come up with a compromise. But a better compromise would be to move the data to a different page, rather than pollute this one. anthony (see warning) 23:16, 9 Apr 2004 (UTC)
- First of all, it is against the spirit of Wikilove to needlessly provoke people, Anthony -- not a rule, but take that as a very good guideline. Secondly, the issue was resolved for everyone but you -- we do things by consensus here, and consensus was (and is) against removing the census data. I think the table is acceptable (though not really preferable to the previous solution), but I would like you to stop agitating to remove the data from this page. You make good contributions to corners of this project, and I know you can apply that attitude here and recognize that the census data is seen as useful and important by this community, even if it isn't by you. Jwrosenzweig 22:47, 9 Apr 2004 (UTC)
- Good thing there isn't a rule against being needlessly provocative. I don't consider this a resolution. I still feel the information should be removed. But at least the POV and inaccurate aspects of the information have been relieved. It's also slightly less obnoxious as a table. anthony (see warning) 22:41, 9 Apr 2004 (UTC)
-
-
- Well, it's a resolution in the sense that it's Anthony's edit, and I'm leaving it where it is, and I hope that other people will too. It should be clear, as Jwrosenzweig says, that the consensus is running against removal of that information, so I'm counting on Anthony to live with that whether or not he considers the dispute "resolved". As for rules about being needlessly provocative, it's far too subjective to create a specific rule against it (I believe Anthony was making that comment to be funny anyway, though I'm not sure everybody got the joke). --Michael Snow 23:01, 9 Apr 2004 (UTC)
- There are two issues: whether (and to what extent) the data should be included, and how best to phrase it. Assuming the table stays the way it is, I consider the phrasing dispute resolved. As for whether to include the data in the first place, I only removed it once, and I seem to be so overwhelmingly outnumbered that I immediately gave up on that. In hindsight, I actually think it would probably be better to move the data elsewhere rather than delete it. And presenting it in a table to some extent lessens that need. Maybe the solution is to some of the table (geography, population, maybe a few others), and then link to the rest. But then the rest is really going to be source text, and maybe it should be moved to wikisource and linked there. I don't know. anthony (see warning) 23:31, 9 Apr 2004 (UTC)
- So that you know, I objected above to moving the data to a separate page, and I maintain that objection. The article is not substantial enough at this time to warrant splitting it into multiple pages. --Michael Snow 23:38, 9 Apr 2004 (UTC)
- Second that. Leave it like it is, and leave all other pages like it like they are. Census data may be boring or irrelevant to some, but it's fascinating to others. When I was a kid I'd make up lists of cities, states etc. that I'd been to, and sort them by all sorts of data--by population, by altitude, by acreage of parkland, percentages of ethnic types, number of cars, age, etc. etc. etc. No harm in keeping it easily available.Doovinator 02:02, 10 Apr 2004 (UTC)
- The harm, in my opinion, is that it distracts from the more interesting information. As for census data being fascinating to some, that may be true, but I think it's clear we should draw the line somewhere. There is a lot of the census data that we're already not including. Institutionalized persons, people in College dormitories, Military quarters, Emergency shelters for homeless , Visible in street locations, Other noninstitutional group quarters, race of householder, contract rent, Persons per Occupied Housing Unit by Tenure, the list goes on and on and on. I think we've drawn the line too far on the inclusive end, for the main page. anthony (see warning) 02:35, 10 Apr 2004 (UTC)
- Second that. Leave it like it is, and leave all other pages like it like they are. Census data may be boring or irrelevant to some, but it's fascinating to others. When I was a kid I'd make up lists of cities, states etc. that I'd been to, and sort them by all sorts of data--by population, by altitude, by acreage of parkland, percentages of ethnic types, number of cars, age, etc. etc. etc. No harm in keeping it easily available.Doovinator 02:02, 10 Apr 2004 (UTC)
- So that you know, I objected above to moving the data to a separate page, and I maintain that objection. The article is not substantial enough at this time to warrant splitting it into multiple pages. --Michael Snow 23:38, 9 Apr 2004 (UTC)
- There are two issues: whether (and to what extent) the data should be included, and how best to phrase it. Assuming the table stays the way it is, I consider the phrasing dispute resolved. As for whether to include the data in the first place, I only removed it once, and I seem to be so overwhelmingly outnumbered that I immediately gave up on that. In hindsight, I actually think it would probably be better to move the data elsewhere rather than delete it. And presenting it in a table to some extent lessens that need. Maybe the solution is to some of the table (geography, population, maybe a few others), and then link to the rest. But then the rest is really going to be source text, and maybe it should be moved to wikisource and linked there. I don't know. anthony (see warning) 23:31, 9 Apr 2004 (UTC)
- Well, it's a resolution in the sense that it's Anthony's edit, and I'm leaving it where it is, and I hope that other people will too. It should be clear, as Jwrosenzweig says, that the consensus is running against removal of that information, so I'm counting on Anthony to live with that whether or not he considers the dispute "resolved". As for rules about being needlessly provocative, it's far too subjective to create a specific rule against it (I believe Anthony was making that comment to be funny anyway, though I'm not sure everybody got the joke). --Michael Snow 23:01, 9 Apr 2004 (UTC)
-
