800 RSVP to Census plan to cut marriage questions

The Census Bureau has gotten at least 800 written comments about its proposal to drop five marriage questions from its annual survey.

A bureau official declined Tuesday to say whether most of the comments were in opposition to the proposal, but leading family research groups have been unanimously against cutting the questions from the American Community Survey (ACS).

Tuesday was the last day to send in written comments on the proposal; the bureau will be issuing a detailed report in coming weeks, said Gary Chappell, manager for the ACS content-review project.

The five marriage-related questions ask people whether, in the last 12 months, they have married, divorced or been widowed; how many times they have married; and what year did they last marry.

Some 3.5 million people fill out the ACS each year.

Census officials say there must be “compelling justification” from the federal government to ask each of the 72 numbered questions in the ACS — and there isn’t such support for the five marriage questions.

Researchers who study the American family say the detail about people’s marital history captured in the ACS is crucial to understanding trends and formulating policies, both national and local.

The ACS “is the only way” to estimate annual, national divorce rates, as well as marriage and divorce rates for counties and cities, said researchers with the National Center for Family and Marriage Research at Bowling Green State University (BGSU) in Ohio.

Already, data from ACS has helped debunk the idea that there’s a “seven-year itch” in marriage — it’s actually more of a “12-year itch,” since roughly half of first marriages end 12 years after the wedding, the BGSU center said.

Moreover, it added, the ACS will be “the only way to estimate levels and trends in same-sex marriage and divorce.”

Six states no longer report data on marriages and divorces — making national estimates very difficult — and other major federal surveys don’t include older populations, aren’t taken every year or have other serious limitations for these kind of vital statistics, wrote Brookings Institution scholars Isabel Sawhill, Ron Haskins and Richard Reeves, who lead the think tank’s Center on Children and Families.

“The landscape of family formation is changing rapidly, with important implications for child and family well-being, economic inequality and mobility, and policy decisions at both the state and federal level,” they said in a Dec. 23 comment to the Department of Commerce, which oversees the Census Bureau.

“We urge you to reassess the decision to remove these questions,” the Brookings scholars wrote.

The five marriage questions — plus ones about the type of bachelor’s degree a respondent has and whether a respondent’s property houses a business — were all deemed to be unjustified after a meticulous review of the ACS questions, Census Bureau officials explained.

“Every ACS question must have compelling justification,” defined as a legal or regulatory basis, programmatic need or local impact on a community, the bureau told an April gathering. Collecting data that are “interesting to know” or “needed for research” is not reason enough to keep a question on the ACS, it said.

Over the next few months, the bureau will review the written comments and prepare a report for the White House Office of Management and Budget (OMB), Mr. Chappell said.

“We do intend to report on every point that was raised for or against the proposal, and the volume of support for each point,” he said.

The OMB will review the bureau’s recommendations — and will likely call for a second round of public comments — before arriving at a final decision by June.
Any changes in the ACS would go into effect in 2016.