All Things Census
The United Kingdom will conduct a traditional census next year, but it may be the last of its kind, according to the British Cabinet minister in charge of the count. Francis Maude told the Daily Telegraph newspaper that there are ways to “provide better, quicker information, more frequently and cheaper.”
Britain has conducted a census each decade since 1801, with the exception of 1941 during World War II. But Maude said the count is out of date by the time its results are published the following year. He noted that 1.5 million households did not fill out census forms during the 2001 census, making the count incomplete. Maude suggested that one alternative would be to use administrative data in government records or from private databases to assemble a population count, and that it might be done more frequently than once each decade.
The government’s Office of National Statistics has issued a statement that changes to the traditional census are being considered as part of a project called Beyond 2011, whose missions is to “develop a range of options for the production of population statistics beyond the 2011 Census. It is considering how a number of different data sources could be used to produce the key information needed to support effective decision-making.” Read more
The head of Statistics Canada has resigned over the government’s decision to drop the mandatory long form in the 2011 Census, stating that the voluntary survey that will be instituted in its place is not an adequate substitute. The resignation statement by Munir A. Sheikh declined to explain what advice he and his agency gave the government about its plan.
The Conservative government’s announcement last month that the long form would be dropped from the census after 35 years of use provoked criticism from Canada’s three opposition parties, and legislators plan to hold hearings next week on the issue. The Cabinet minister who oversees the census, Tony Clement, issued a statement after Sheikh’s resignation reaffirming the decision to drop the long form “because we do not believe Canadians should be forced, under threat of fines, jail, or both, to divulge extensive private and personal information.”
Many Americans were puzzled or irritated by the questions about race and Hispanic ethnicity on the 2010 Census form. Some did not want to declare a race, others did not think they fit the available categories and still others wondered why the government should ask the question. Kenneth Prewitt, who directed the Census Bureau during the 2000 Census, also has concerns about the way the government asks about race and ethnicity. The race question on the census form “does not work well and we can do better,” he writes in a commentary in USA Today.
Instead of the two questions about race and Hispanic ethnicity, Prewitt suggests that the next census ask the following: “What national origin, ethnicity, tribe, language group or ancestry do you consider yourself to be? (List all those important to you.)” This question, he says, could be paired with others on immigration status: “Where were you born, and where were your parents born?”
Prewitt, professor of public affairs at Columbia University, says that a race question must be asked in order to shed light on disparities and discrimination based on color, ancestry or immigrant status, “but only if we draw the portrait more carefully than that produced by the 2010 Census.” He argues that his suggested questions allow people to define their own identities (which can include multiple races) and at the same time allow statisticians to group people into categories for comparison purposes. The immigration question, when paired with the race question, “tells us how immigrant status interacts with national origin, ethnicity or language group” so it can be seen how well newcomers are faring, he writes.
The race and ethnicity questions on the census form were assembled in response to specific demands over time, he argues, rather than being carefully designed to “provide information relevant to the public purposes that justify asking the questions in the first place.” (For additional background, see this history of racial categorization in the census and this history of counting Hispanics.)
Prewitt, who has a long-standing research interest in racial classification, also has argued his case in academic journals (such as this article from Daedalus in 2005 that suggests a less dramatic rewrite of the race question). He is writing a book about race and racial classification.
As the 2010 Census information-gathering phase winds down and the Census Bureau turns to quality-checking and data-processing, Director Robert Groves offered some statistics at a recent operational briefing to assess how the national count has gone thus far. One indicator, the quality of the address list, appears to have improved since the 2000 Census. Another, the share of proxy interviews–information obtained from neighbors and building managers instead of the householders themselves–has worsened.
The foundation of a good census is having a complete list of addresses because Americans are counted at their homes or the other places they are living. The quality of the address list is important in aiding census-takers who head out on follow-up visits to people who did not return their mailed-out questionnaires.
During the recent non-response follow-up operation, Groves said, census-takers found fewer non-existent addresses on their rounds in 2010 than their counterparts had in 2000. In 2000, 6 million non-existent addresses were deleted from the list because census-takers could not find them. In 2010, 4.1 million were deleted. During follow-up visits, census-takers also are supposed to look for addresses that are not on the official list, so they can be added. In 2010, Groves said, “we had fewer adds proportionately” compared with 2000, although he said this is not as much of a “hard quality indicator” because it could mean that census-takers did not follow procedures for including new addresses.
On another quality measure, Groves said census-takers who were trying to collect information at addresses from which census forms were not received had to rely more heavily on neighbors and building managers than was the case during the 2000 Census. In 2010, about 22% of interviews were from proxies, not from the householders themselves, compared with 17% in 2000. This is of concern because proxy data traditionally have been less accurate and complete than information that people provide about themselves. Groves said “this fits the expectation we had with regard to the cooperation of the American public.” Some people were never home during repeated visits by census-takers; others refused to provide information about themselves.
In addition, Groves said census operations so far have been on schedule and “significantly under budget.” Previously, census officials had reported a better-than-expected mail participation rate–72% of occupied households returned their forms, the same as in the 2000 Census.
Getting under way now are a series of quality-check operations “to make sure we’ve gotten it right,” in Groves’ words. This page will have more information on these operations in a later posting.
What would happen if Americans were not required by law to respond to census surveys? This is of interest because Canadian government officials have decided to drop the mandatory long form in next year’s census and replace it with a voluntary household survey. All Canadian households will still be required to answer basic questions on the census short form.
The Census Bureau tested this idea a few years ago when it researched how response rates to the American Community Survey would change if survey response was not required. The American Community Survey, a monthly survey that has replaced the census long form, includes questions about topics that include educational attainment, commuting, citizenship, income and housing costs. The bottom line: There is a “significant drop in mail cooperation” when respondents are not required to respond, Census Bureau researchers reported. Read more
Statistics Canada has announced that the nation’s 2011 Census will include the same eight basic questions that were asked of everyone in the 2006 count, and that the mandatory long form will be replaced with a voluntary survey.
The census long form went to one-in-five households and had been part of the national census for 35 years. The voluntary National Household Survey will be sent to one-in-three households within a month after the May 2011 census is conducted. Government officials say they hope the larger mail-out will result in an adequate number of responses. The household survey will ask the same questions about citizenship, ethnicity, religion, income, work, housing and other topics that had been on the long form. However, it will not include a request for consent to release individual-level data after 92 years, which had been included on the long form.
The short form includes questions asking for the name, age and gender of each person living in each household, as well as each person’s relationship to “Person 1.” Among the relationship options are “same-sex married spouse” and “same-sex common-law partner.” For each person in the household, there also is this language question: “What is the language that this person first learned at home in childhood and still understands?” (Here is a link to the list of short-form questions.)
According to some news accounts, the long form is being replaced because a growing number of Canadians do not want to answer personal questions, but some statisticians are concerned that a voluntary survey will not achieve the same data quality as a mandatory one. Some genealogists have criticized the decision to stop releasing individual-level data after 92 years, saying that it will impose severe limits on historians and people seeking to do family research.
In a posting on his blog, Census Bureau Director Robert Groves says that census-takers are nearly done with knocking on doors of households from which 2010 Census forms were not received. As of June 28, 99.6% of the interviews were completed.
The bureau is conducting quality-assurance audits to make sure the interviews were conducted correctly (see this report about 10,000 households in Brooklyn whose information will be redone because two managers allegedly used online databases to fill in census forms rather than contacting the householders directly). Additional quality-assurance operations will examine the status of addresses deemed to be vacant on April 1, Census Day; addresses that could not be found and were recommended for deletion from the address list; and addresses added too late to be visited by census-takers during the non-response follow-up phase.
A new report on childless women from the Pew Research Center uses data from the Current Population Survey to track recent trends and describe this group’s demographic characteristics. The report finds that childlessness among 40-44-year-old women has risen in all racial and ethnic groups, and most education levels, but has declined over the past decade among women with advanced degrees. Highly educated women still are the most likely to be childless.
The Census Bureau today released 2009 population estimates for cities, villages, boroughs and minor civil divisions that will be the last such numbers published for these incorporated places before the 2010 Census results are available.
Here is the Associated Press take on these estimates, and here is a first look from USA Today. The themes of their coverage include growth in Texas and decline in the Rust Belt.
The Census Bureau is clearing its data cupboard to make room for results of the 2010 Census. Today, the bureau released 2009 state and county housing unit estimates, the last ones before decennial results are compiled.
Last week, the bureau released population estimates by age, sex, race and Hispanic origin for the nation, states and counties. The baseline for these estimates is the 2000 Census, updated with data from the American Community Survey and recent government records such as birth and death certificates, immigration statistics and building permits. The reference date for these estimates is July 1, 2009.
All sets of estimates can be found on the Census Bureau’s population estimates page.