A new RAND study strongly challenges those widely held assumptions. Student test scores are rising--rapidly, in the case of minorities--and part of the explanation is improvements in the family environment, according to the study. The rest of the credit--especially for the large gains made by minority students--may belong to those much-criticized public policies and programs of the 1970s and 1980s.
The average math and reading scores of students 13 to 17 years of age increased the equivalent of 3 percentage points for white students, 11 points for Hispanic students and 19 points for African-American students.
Changes in the characteristics of families--notably, a dramatic increase in the education levels of African-American mothers and smaller family size--account for about one-third of the gains in achievement made by African-American youths. The analysis suggests that the remaining gains may be due to policies and programs in effect during these years that were aimed precisely at helping minority and low-income students and families.
In Student Achievement and the Changing American Family, researchers David W. Grissmer, Sheila Nataraj Kirby, Mark Berends and Stephanie Williamson write: "Our analysis supports a more positive picture than is usually drawn of the achievement of American students, the capacity of American families to support that achievement, and the effectiveness of public policies and public investment in education."
The main reason student achievement is thought to be falling is that Scholastic Aptitude Test (SAT) scores relentlessly tell us so. But, as experts in the assessment field point out, this oft-cited measure is not, and was never designed to be, a reliable indicator of nationwide student performance over time. The test is taken by a different mix of students each year and excludes non-college-bound youths, the population in which Grissmer and his colleagues, using other measures, find the greatest gains.
Shunning SAT scores, the RAND team based its conclusions about student performance on scores from the National Assessment of Educational Progress (NAEP), a set of standardized exams administered by the Department of Education each year since the early 1970s to a sample of students representative of the population as a whole. The test can also be used to compare student achievement at different times and across different racial and ethnic groups.
But simple comparisons of test scores across different groups and different years say nothing about how changes in the family affect student achievement. To better understand that, the researchers developed a quantitative model that allowed them to predict the effects on test scores of changes in single-parent status, family income, family size, parents' education, and mother's birth age and workforce participation.
The study found that the single most important factor influencing student achievement was parents' education: Students with one or two college-educated parents performed significantly better than students whose parents were not high school graduates.
Family size and income also were significant. A student with one sibling could be expected to do better than a student with four siblings, whereas a student whose family earned $40,000 could be expected to outperform one whose family earned only $15,000. Likewise, a child born to an older mother is likely to score higher than one born to a younger mother.
The researchers' analysis of national demographic trends between 1970 and 1990 confirmed that there had been changes for the better in two of the most influential family characteristics. For example, mothers in 1990 were better educated than their 1970 counterparts. The data show that 16 percent of mothers of 15- to 18-year-olds in 1990 were college graduates, compared with 7 percent in 1970. Also, only 17 percent of mothers in 1990 lacked a high school degree, compared with 38 percent who lacked the diploma in 1970.
Changes in family size were also dramatic. In 1990, 73 percent of teenage children lived in families with either no siblings or one; in 1970, only 48 percent did.
To gauge how much of the improvement in test scores could be attributed to factors outside the family, researchers subtracted the effects of family characteristics from NAEP test-score gains. They found that although family effects entirely accounted for the test-score gains of white students, they accounted for only about one-third of the large gains made by minority students.
What explains the remaining gains by minority youth? The cause has to be something that affected minorities differentially, says the report, otherwise white students also would have shown gains that were unrelated to family characteristics, and they did not. Public policies that favored equal educational opportunity and a massive investment in schools and children, particularly minority and low-income youth, fit that bill.
"These findings are like a caution light, warning us to go slow in dismissing the public programs of the past 20 years as a waste of resources and a failure of social policy," Grissmer concludes. "They are clear signs that American schools are doing as good a job of educating nonminority students in 1990 as in 1970, and a significantly better job of educating minority students."
Student Achievement and the Changing American Family
David W. Grissmer, Sheila Nataraj Kirby, Mark Berends, Stephanie Williamson,
RAND/MR-488-LE, 1994, 131 pp., $15.00.