All Things Census
Facing similar obstacles of cost and people’s reluctance to participate in national enumerations, some European countries are trying innovative ways to count their populations, according to an article on census-taking in Europe published (in English) by the Institut national d’études démographiques (INED). Among the new methodologies are expanded use of administrative records from municipal population registers and information collection from ad hoc or existing sample surveys.
The article cites the U.S. Census Bureau’s American Community Survey as one example of a new way of gathering demographic, social and economic data about households. The Census Bureau confronts similar problems to those of European countries, in the form of growing costs and increased public reluctance to participate. The Census Bureau makes limited use of administrative records to collect population data, and has done research into expanded uses, including an experiment during the 2000 Census. Although the results were not definitive due to the limited nature of the experiment, researchers said use of administrative records has great potential in a number of realms.
Census Bureau data can help answer the question of how likely it is that a long-lasting marriage, such as the 40-year-union of Al and Tipper Gore, will come to an end.
Only a small share of long-duration marriages end in divorce each year, but the likelihood that a first marriage remains intact dwindles with each anniversary. According to a new Pew Research Center report on long-duration marriages and divorce, only about half the first marriages begun in the early 1970s lasted until their 25th anniversaries.
Rates of intermarriage have risen in the United States, but trends differ markedly for different race and ethnic groups, according to a new Pew Research Center report that makes extensive use of U.S. Census Bureau data. Among blacks and Asians, there also are notable variations by gender.
The new report also includes a discussion of geographic variations in intermarriage patterns, and an interactive map displays data for states and regions. The report also includes a section on public opinion about intermarriages.
The Delaware House of Representatives passed a bill this week that would count prisoners at their home addresses, not the places where they are incarcerated, for purposes of redistricting after the 2010 Census. Inmates who lived out of state before they were incarcerated would not be counted for redistricting purposes. The bill now goes to the state Senate.
In the past, when legislative districts were redrawn using census data, inmates have been included at their prison addresses. Some state officials and legislators say that gives disproportionate power to rural areas with large prisons. The Census Bureau announced in February that it would release detailed numbers earlier than usual to give more power to states to use 2010 Census data to exclude prisoners from being counted at their prison locations for redistricting purposes.
Legislation has been enacted in Maryland to count prisoners at their last known home addresses if they are in state or federal prisons and were state residents before they were incarcerated. According to the Prison Policy Institute, which advocates for counting prisoners at their home addresses, similar legislation has been proposed in New York and Rhode Island.
How do respondents’ answers to a Census Bureau question about their race vary depending on the type of question asked? How do response rates and costs vary depending on the data collection mode for the American Community Survey? And how might better question wording improve inaccurate responses to Census survey questions about same-sex marriage?
These and other census-related topics were among those explored in presentations at the recent annual conference of the American Association of Public Opinion Research. Abstracts of the papers are available at the AAPOR website, along with email addresses of the authors.
For general readers who want to dig further into how the decennial Census has changed over the years, here is a short list of selected books that explore its past. Included are general histories of census-taking and demographic findings, memoirs of Census Bureau directors, detailed analyses of changes in race and ethnicity categories and several books about the politically charged debate over census undercounts.
The reading list was provided by Margo J. Anderson, professor of history and urban studies at the University of Wisconsin-Milwaukee. Her own book, “The American Census: A Social History” (Yale University Press, 1988) is a thorough and well-written account of the census from colonial days through the 20th century.
In suggesting the list, Anderson wrote, “The materials listed below provide a window into a much larger literature on the history and politics of census taking, American demographic history and the political uses of the census in the United States. All of them have extensive footnotes and bibliography for the reader searching for more information. Follow their leads! The census is one of those invisible American institutions (like the Post Office) that spans our nation’s history and thus rewards the intrepid explorer with great insights about American life.”
In addition to Anderson’s reading list, the Census Bureau website includes a history section that includes overviews of questions asked through the decades, links to legislation pertaining to the Census Bureau and other material. Read more
Among Americans who have not obtained a regular high school diploma, Hispanics are less likely than members of other major U.S. race and ethnic groups to acquire a General Educational Development (GED) credential, according to a Pew Hispanic Center analysis that is based on newly available educational attainment data from the Census Bureau’s 2008 American Community Survey.
The relatively low level of GED attainment among Hispanics, the nation’s largest minority group, is notable because Hispanics have a much higher high school dropout rate than do blacks or whites, the analysis points out.
The Census Bureau today released a report summarizing levels of voting and voter registration in the November 2008 presidential election. Based on the November 2008 Current Population Survey Voting and Registration supplement, the Census Bureau reports that over 131 million people cast a vote in 2008, up from 126 million in 2004.
The 2008 electorate was the most diverse in U.S. history, according to a Pew Hispanic Center analysis of the same data. Nearly one-in-four voters was non-white in 2008, up from about one-in-five in 2004.
The Census Bureau does not ask U.S. residents for their immigration status when they are counted in the 2010 Census or other population surveys. A major reason is that bureau officials fear it would make people reluctant to respond, despite the promise that their information is confidential.
So when numbers are cited for the size or characteristics of the undocumented immigrant population, they are estimates that are arrived at indirectly. The Wall Street Journal recently published an interesting article and a blog posting explaining the ins and outs of these estimates. Among the estimates discussed are those produced by the Pew Hispanic Center, including a 2009 report that featured estimates for states.
A snapshot of the lowest-responding neighborhoods in the 2010 Census shows that more than two-thirds are in cities, and they tend to be more racially or ethnically diverse than higher-responding areas. The analysis, by the Center for Urban Research at the Graduate Center of the City University of New York, is based on mail participation rate data for census tracts released by the Census Bureau last week. (A summary of the analysis also is available.)
The analysis includes a look at the 9,546 census tracts (14% of the total) that had a mail participation rate of less than 60%. Census-takers are in the field now to seek responses from households from which forms were not received.
The analysis also looks at the impact of replacement questionnaire mailing to low-responding areas, and concludes that the second questionnaires had a “strong impact” in boosting the mail participation rate top 72%, which equals the response in the 2000 Census.