379

National economic statistics are well established with the government, the media, and the public. The government also produces many data on social issues such as education, health, housing, and crime. Social statistics are less complete than economic statistics. We have good data about how much money is spent on food but less information about how many people are poorly nourished. Social data are also less carefully produced than economic data. Economic statistics are generally based on larger samples, are compiled more often, and are published with a shorter time lag. The reason is clear: economic data are used by the government to guide economic policy month by month. Social data help us understand our society and address its problems but are not needed for short-term management.

There are other reasons the government is reluctant to produce social data. Many people don’t want the government to ask about their sexual behavior or religion. Many people feel that the government should avoid asking about our opinions—apparently it’s OK to ask, “When did you last visit a doctor?” but not “How satisfied are you with the quality of your health care?” These hesitations reflect the American suspicion of government intrusion. Yet issues such as sexual behaviors that contribute to the spread of HIV and satisfaction with health care are clearly important to citizens. Both facts and opinions on these issues can sway elections and influence policy. How can we get accurate information about social issues, collected consistently over time, and yet not entangle the government with sex, religion, and other touchy subjects?

The solution in the United States has been government funding of university sample surveys. After first deciding to undertake a sample survey asking people about their sexual behavior, in part to guide AIDS policy, the government backed away. Instead, it funded a much smaller survey of 3452 adults by the University of Chicago’s National Opinion Research Center (NORC). NORC’s General Social Survey (GSS), funded by the government’s National Science Foundation, belongs with the Current Population Survey and the samples that undergird the CPI on any list of the most important sample surveys in the United States. The GSS includes both “fact” and “opinion” items. Respondents answer questions about their job security, their job satisfaction, and their satisfaction with their city, their friends, and their families. They talk about race, religion, and sex. Many Americans would object if the government asked whether they had seen an X-rated movie in the past year, but they reply when the GSS asks this question. The website for the GSS is gss.norc.org.

380

This indirect system of government funding of a university-based sample survey fits the American feeling that the government itself should not be unduly invasive. It also insulates the survey from most political pressure. Alas, the government’s budget cutting extends to the GSS, which now describes itself as an “almost annual” survey because lack of funds has prevented taking samples in some years. The GSS is, we think, a bargain.