Pollster.com

Articles and Analysis

US: Census (CNN 3/19-21)

Topics: poll

CNN / Opinion Research Corporation
3/19-21/10; 1,030 adults, 3% margin of error
Mode: Live telephone interviews
(CNN release

National

Do you think the Census is useful or do you think it's a waste of money?
80% Useful, 20% Waste of money

Do you consider the Census an invasion of your privacy, or not?
16% Yes, 83% No

Just your best guess -- how accurate do you think this year's Census will be --
12% Very accurate
55% Somewhat accurate
25% Somewhat inaccurate
8% Very inaccurate

sjt22:

If the Census is "somewhat accurate" then by definition it is also "somewhat inaccurate".

____________________

wes:

heh, good point sjt22.. probably shouldve been "mostly" instead of somewhat.

____________________

Field Marshal:

I would have liked to see a question on statistical sampling or whether or not people favor a cheaper and probably more accurate alternative than counting.

____________________

Sorry Field Marshal. While your question may have merit, it would be unconstitutional to do statistical sampling of the entire census.

Here's the language in the US Constitution governing the requirement that a Census takes place every 10 years:
"Representatives shall be apportioned among the several states according to their respective numbers, counting the whole number of persons in each state."

Thus, for purposes of reapportionment, at least, "counting the whole number of persons in each state," would prohibit statistical sampling. The idea of using some sampling is for other purposes, such as apportionment of some Federal dollars to states. But the key Constitutional reason for a Census is for the determination of the House of Representatives, and by inference, the number of votes each state would get in the Electoral College, for president.

____________________

Stephen_W:

"...whether or not people favor a cheaper and probably more accurate alternative than counting."

Field Marshal, I'm going to go out on a limb and say that there will never ever ever be a more accurate way to count the population than by counting every man, woman and child in person. For the same reason that the most accurate poll on who Americans want as president is the actual election itself. The reason there is statistical sampling is that there's just no way to count everyone (or in the case of polling, call everyone) in the entire country. So you count as many ppl as humanly (or budgetarily possible), and use targeting sampling when it's no longer possible to do that.

As for Nelcon's argument about the constitution, it's a valid point but the founders probably hadn't anticipated an America with 300 million people living in it...

____________________

Field Marshal:

I know that we would need a constitutional amendment to alter the way we census the people, which is why i was hoping they would ask the question; to gauge whether or not there is support for such an endeavor.

Stephen,

There are many a Harvard business school case devote to such a assertion. Many do speculate that statistical sampling would be a more accurate technique for counting the populace. The question is not whether statistical sampling would be error free but if the error term would be less than simple counting.

____________________

Mike_A:

Stephen_W, I think Al Gore supporters might take issue that an election is the most accurate poll on who Americans want as president. In 2000 an estimate 4 to 6 million votes went uncounted and the difference in counted votes was around 500,000. Were the Florida “results” more accurate than a poll? Only if you accept that the votes that were counted were an accurate representation of the views of Florida’s voters. To me the vote was way too close and too many votes were counted to have complete faith in that statement. Counting by humans isn’t guaranteed to be more accurate, as elections, and census results, have shown. With improvements in technology we may be getting closer to a day when every person and every vote is counted, but not today.

____________________

Xenobion:

Considering statistical sampling would require the rule of large numbers some small rural communities could be largely misrepresented.

____________________

GARY WAGNER:

Maybe a question should have been, "Do you think it is appropriate that the government requires that you identify your race?"

I've also always wondered how the census bureau can tell us how many people it didn't count. They even know what race, religion, and political party those people they didn't count belong to. Isn't that like saying, "Will everyone not here please raise your hand?"

____________________

sjt22:

I would have liked to see a question on statistical sampling or whether or not people favor a cheaper and probably more accurate alternative than counting.

If you framed it like that, I'm sure you would get a pretty positive response.

Unfortunately sampling has been continually ruled out due to political pressure. As a result we have had a census which consistently under counts certain people and over counts others.

____________________

Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR