Tuesday, June 24, 2008

Data Infrastructure

I study higher education data because to me data tell important stories about what we are doing, what we are not doing, what we should be doing, and because it often reveals blatant lies. In an objective manner it establishes reality—often a scarcity in the public policy worlds I work in where ideology, self-interest, issue spinning, partisan politics and convenient myths too often distort or hide reality. Data is a tool to set benchmarks and to frame important public policy discussions. Its presence or absence is itself a statement about our institutional interests in particular subjects, like class issues in higher education.

So when key data to study class issues is eliminated--as the Census Bureau did recently-- this is a red flag issue with me. This spring the Education and Social Stratification Branch of the Census decided to eliminate Table 14 from its annual report on school enrollments. For 36 years the Census Bureau has published this table with extensive information on school enrollment by family income. This table uniquely provides the annual overview of high school graduation, college continuation, college participation, estimated bachelor’s degree completion, and estimated bachelor’s degree attainment by age 24 for dependent 18 to 24 year olds by family income.

In the extreme political context of the Bush Administration, the Census Bureau’s decision appears to be a political act to eliminate embarrassing data. The Chief of the Education and Social Stratification Branch claims that this was not the case. But the history of the Bush administration’s efforts to eliminate Upward Bound services for high school students from low income families, its unfulfilled campaign promises to increase the Pell Grant maximum award, the wasteful spending on the student loan industry, and proposed elimination of many federal student financial aid programs for needy students tells a different story. Any act by the Bush administration—and the Census Bureau is a part of the Bush administration—is a political act. The Chief of the Education and Social Stratification Branch has suggested that limited resources, shifting priorities, and technical issues with the Current Population Survey were all factors in his decision. We have offered to seek additional funds for the production of this table, and we have encouraged addressing whatever technical issues that led him to think a 37th year for this table was not warranted. But he will not change his decision—so the Council for Opportunity in Education has appealed his decision to the head of the Census Bureau, Dr. Steve Murdock, former Texas state demographer and a very serious student himself of the demography of education.

This painful incident is a reminder of the importance of the data infrastructure required for informative policy studies. In my career I have seen data collection stopped when the results were going to be an embarrassment to the collecting agency (the job placement data at the College of Education at the University of Minnesota in the late 1960s). This may or may not be a political decision to stop publishing embarrassing data. The Chief of the Education and Social Stratification Branch has provided detailed assistance to me to retrieve the 2006 data from the Current Population Survey myself. We have published our analyses from the CPS data in the June issue of OPPORTUNITY. And that is the issue: Should this country’s dialogue on the dominance of class and inherited educational opportunity be dependent on an old man who works out of the basement of his house in a cornfield in southern Iowa? Or should such vital data be published by the Census Bureau? We think the answer is obvious, are we are trying to convince the director of the Census Bureau that it is his agency’s job, not ours.

No one should take the data infrastructure for policy analysis for granted. Data is subject to the most depraved political decision making, as I have too often witnessed in my career. If we are to make public policy based on a clear understanding of reality then decisions such as the elimination of Table 14 from the school enrollment reports must be reversed. If additional resources are needed we will help get them. If there are technical issues in the CPS data we will eagerly work to address them. But unilaterally eliminating Table 14 is unacceptable.