Opinion, Berkeley Blogs

Stumbling in the dark

By Claude Fischer

I recently turned to one of the central sources of information about social trends in America, The Statistical Abstract of the United States, described on its web page as “since 1878, the authoritative and comprehensive summary of statistics on the social, political, and economic organization of the United States.” Also on the web page was this notice, in bold red:

The U.S. Census Bureau is terminating the collection of data for the Statistical Compendia program effective October 1, 2011. The Statistical Compendium program is comprised of the Statistical Abstract of the United States and its supplemental products . . . .

Statistical Abstract of the U.S. (cover) The notice goes on to explain that the elimination was a result of fiscal cutbacks and it advises readers to scour footnotes of old tables for the sources of the data and go there.

This budget-cutting – criticized by some on both the left and right – is annoying, but it is just one several trends in making access to information about our society more difficult, more costly. And thereby making it harder to understand what is happening in America.

No need to know

For years now, social scientists have faced efforts in Congress to curtail funding of social science research. The behavioral science part of the National Science Foundation budget is a favorite target. It seems to survive – barely – each round.

What has not survived is the "long form" part of the decennial census. For several censuses in the late 20th century, one of every six American households received a long version of the form. This version asked questions about important social topics ranging from the respondents’ national origins and the languages they spoke at home to their occupations, commuting, housing, and income. It was attacked as too nosy and unconstitutional and, justly or not, eventually dropped beginning in the 2010 census.

To make up for the loss of so much critical information about our society, the Bureau of the Census has started using the “American Community Survey,” which reaches a couple of million households each year. The ACS is valuable, but in various ways it cannot give us the same coverage and accuracy of the old long-form census. And, it too now faces attacks trying to shut it down.

Keep it private

Social research has been boxed in from another angle by what many scholars consider a too-obsessive concern about privacy. For example, my colleagues and I found that it extremely difficult if not impossible to obtain decades-old information about neighborhoods  – for example, the proportion of the workers living in a particular census tract in 1960 who held professional jobs. The fear, baked into U.S. Census rules, is that with enough such general data we might be able to identify a particular person in a census report 50 years ago and find out, say, how much money he made.

Social scientists on campuses have been struggling against zealous IRB’s — university Institutional Review Boards — that must approve any research conducted by faculty on human subjects. Designed quite properly to prevent harm to subjects of studies, particularly subjects of medical treatments, the IRBs in many places have expanded their mission to closely supervising social science research. Some, for example, treat the posing of survey questions – say, asking respondents’ opinions about social issues – or even the gathering of historical records as if they were in the same category as injecting people with drugs.

This zealotry sets up great hurdles that delay or derail social studies. Doctoral students, in particular, can have careers crippled by these restrictions. (See statements by social science organizations here in 2001 and here in 2011.)

(At one time here at Berkeley, the local IRB, according to some reports, entertained the idea of "social harm": research that might impugn a social group — for instance, showing that women get more emotional in some settings than men — ought to be stopped. That would pretty much stop social research altogether.)


Various motivations seem to lie behind these efforts to curtail social science research – concerns about privacy, suspicion of government, suspicion of scholars, cost-cutting – but they combine to increasingly blind policymakers and the public to what is going on in America.

(By the way, journalists are not restricted nearly as severely in interviewing informants or using uncovered private information. And corporate data-gatherers are hardly restricted at all in gathering social information and mixing it in with other data such as individuals’ credit records.)

These restrictions on social science may also reflect yet another attitude held by some, both left and right: the conviction that there is no need to gather the data, because they already know the answers.

Cross-posted from Claude Fischer’s blog, Made in America: Notes on American life from American history.