Earning and Learning: How Schools Matter
378
Earning and Learning: How Schools Matter
378Paperback(New Edition)
-
PICK UP IN STORECheck Availability at Nearby Stores
Available within 2 business hours
Related collections and offers
Overview
Product Details
| ISBN-13: | 9780815755296 |
|---|---|
| Publisher: | Rowman & Littlefield Publishers, Inc. |
| Publication date: | 08/01/1999 |
| Edition description: | New Edition |
| Pages: | 378 |
| Product dimensions: | 6.00(w) x 9.00(h) x (d) |
About the Author
Read an Excerpt
Chapter One
SUSAN E. MAYER
From Learning
to Earning
No one doubts that adults with a lot of schooling earn more than adults with less schooling. In 1996 the average working-age male who had completed high school earned $28,878, while the average working-age male college graduate earned about $50,000. Employers appear to want workers with more skills and to be willing to pay for those skills. In addition, the returns to education have been increasing. In 1973 college graduates earned 46 percent more than high school graduates. By 1989, the difference was 53 percent and it has risen since.
Not everyone agrees about why schooling improves economic well-being. Some people believe that schooling is associated with higher wages mainly because students who are smart to begin with get more schooling, and employers are willing to pay for such "smarts." Others believe that employers value what students learn in school. This is an important distinction. If employers are willing to pay for what children learn in school, policies that increase the length or quality of schooling could increase wages, at least as long as demand for highly educated workers remains strong. But if employers are willing to pay for students' ability to learn rather than what they have learned in school, and if schooling has a relatively small effect on students' ability to learn, it is less clear how social policy can help. The essays in the first part of this book address this controversy. Combined with other evidence, they provide strong support for the argumentthathow much students learn affects how much they earn.
Aptitude and Achievement
In this chapter I use the term achievement to refer to what people have learned. I use the term aptitude to denote the ability to learn. Aptitude is the result of both genetic endowment and other factors, such as lead poisoning or illness, that have long-term effects on children's ability to learn. Some psychologists might prefer to call achievement "crystallized intelligence" and to call aptitude "intelligence," "generalized intelligence," or "fluid intelligence." I avoid the terms intelligence and IQ mainly because they have acquired emotional and political overtones that can complicate a discussion.
I use the term cognitive skills to denote scores on tests of cognitive skill. The chapters in this book use a variety of tests of cognitive skill. None of these were designed to test general intelligence. Almost everyone agrees that scores on these tests are the result of some combination of ability to learn and opportunity to learn, although there is less agreement about the relative importance of aptitude and achievement to cognitive test scores, and hence to future economic success.
Cognitive test scores are associated with both schooling and wages. Most people think that schooling affects wages mainly because it affects test scores. Therefore much of the debate over how schooling affects wages is really over how schooling affects cognitive skills. Those who think that schooling is associated with higher wages because smart people get more schooling tend to think that cognitive test scores are primarily the result of children's aptitude. Those who think that schooling mainly affects wages because children with a lot of schooling have learned a lot tend to think that cognitive test scores primarily reflect achievement.
In the late 1950s and early 1960s Theodore Schultz, Jacob Mincer, Gary Becker, and other economists developed the theory of human capital. According to this theory, human capital is a function of endowments and investment and, like physical capital, is amenable to accumulation and growth. Educational institutions are the chief mechanism by which such accumulation takes place. Endowments include both genetic factors, such as sex and eye color, and cultural factors, such as language. Investments include both parental investments of time and money in their children and children's own efforts in acquiring human capital. These investments can include both how much schooling a child gets and how hard the child works in school. By emphasizing the role of investments in the accumulation of human capital, this model emphasizes the importance to employers of what students have learned in school. It parallels psychological theories that predict that schooling increases both achievement and aptitude.
Building on human capital theory, social scientists produced a large body of empirical work showing that the longer people remain in school, the higher their future income, and presumably the greater their contribution to the economic wealth of society. The federal Great Society initiative was largely based on this argument. The intent was to fight poverty and inequality with a wide variety of educational and training programs: Head Start, compensatory education, employment and training programs, and Pell grants to low-income college students, to name only a few.
As America's commitment to education edged to its high-water mark, doubt began to be cast on the usefulness of a wide range of educational and social interventions for increasing cognitive test scores. In the mid-1960s, the Coleman report provided an influential scientific challenge to the notion that school quality affects children's cognitive test scores. James Coleman found that differences in expenditures, classroom size, and teacher training had little impact on the amount pupils learned. He concluded that the experience, habits, and values that children acquire from their families and the environments in which they live have more influence on what they learn in school than do school resources. Work by Christopher Jencks seemed to confirm that school quality had little effect on school achievement or educational success. Evaluations of the compensatory education programs and Head Start produced conflicting evidence about their ability to improve either cognitive test scores or economic outcomes. A number of studies seemed to indicate that early improvements in children's cognitive skills "faded out" within a few years.
In addition, some economists argued that the number of years a person remains in school affects their future earnings in part because employers rely on diplomas and credentials as "signals" that enable them to sort job applicants with high aptitude from those with low aptitude. According to this view, more able children get more schooling, giving the appearance that schooling enhances ability. Employers are willing to pay for more able workers, but ability is largely determined either by heredity or by prenatal or very early environmental influences. Very able children learn a lot in school. But the fact that they know algebra or have a large vocabulary is important to employers mainly as an indicator of their aptitude or their ability to learn something new. Neither schooling nor other social interventions have much effect on aptitude. A less radical version of the signaling model holds that while aptitude is mainly hereditary, employers do value the specific skills that students learn in school, such as math and reading. Thus schooling signals both that students have learned these things and that they have a high aptitude for learning new skills in the work place.
This view was bolstered in the mid-1980s by Eric Hanushek's review of research on the effect of school characteristics on achievement. Hanushek concluded that, like the Coleman report, this research showed that no measured characteristic of schools consistently improved students' cognitive test scores. Although recent meta-analyses of the same research show that some school characteristics do have positive effects, these effects are not large.
The economists' signaling model was not the first to emphasize the importance of initial ability to economic success. Some research by psychologists suggests that employers value aptitude and that aptitude is primarily genetically determined.
In The Bell Curve, Richard Herrnstein and Charles Murray contend that the United States is becoming increasingly meritocratic. They argue, as had others before them, that in a meritocratic society family background plays a minor role in economic success, whereas initial aptitude—or IQ—plays a large role. IQ influences economic success by affecting both the quantity and the quality of schooling that children get. Some children simply will not be able to learn as much as others. This inequality in aptitude will inevitably result in economic inequality. To bolster this argument, Herrnstein and Murray point out that although today Americans get more schooling and at higher cost than ever before, the verbal and quantitative test scores of students have changed little over the past two decades, and economic disparities have increased, not decreased. In this view the role of schools is to sort students by their ability to learn new material, and then to help them efficiently achieve the level of learning that is right for them.
Because so much controversy remains about the relative importance of aptitude and achievement to economic success, the chapters in this book reexamine the question using new methods that try to separate the effects of aptitude from ability. These chapters combined with earlier research provide strong evidence that what children learn in school affects their economic success.
The Evidence on the Effect of Achievement
The central methodological issue that arises in virtually every chapter in the first section of this book is selection bias. No one disputes that there are correlations between test scores and earnings and between test scores and years of schooling. The question is why one observes such correlations. Imagine two men who are the same age. Joe graduated from college and earns $25 per hour. Bob graduated from high school and earns $10 per hour. Some people would argue that Joe earns more than Bob because he got more schooling, and hence learned more. Others would say that Joe earns more because he was smarter to start with, which resulted in his both getting more schooling and earning a higher wage. To see whether schooling affects wages, one must control both men's aptitude before they began school. Otherwise, some of the benefit that is attributed to more schooling might really be caused by initial aptitude—put another way, students who get a lot of schooling will be "selected" on their initial aptitude.
The selection issue arises not only in estimating the effects of additional years of schooling, but also in assessing the impact of specific school policies on children's outcomes. For example, studies find that students who take more challenging academic courses subsequently score higher on tests of cognitive skill. This could be because they learned more in the challenging courses. Or it could be that more able students are more likely to choose challenging courses. Because few data sets include both a measure of children's aptitude before they began school and a measure of their wages as adults, researchers have had to develop ways to overcome the problem of omitting (or not controlling) initial aptitude. The chapters in this volume are valuable for the care that they take to overcome the selection problem.
Several recent studies have tried to determine the relative contribution of aptitude and achievement to economic well-being. Most of these have been done by economists interested in estimating the economic returns to a year of schooling. Researchers have used various techniques to overcome the selection problem. For example, some have compared the wages of identical twins who had different amounts of schooling to see if more schooling is associated with higher wages among people who are genetically identical. Most of these studies find that schooling increases wages, holding constant initial aptitude. This implies that what children learn in school pays off in the labor market. Other studies have tried to determine whether additional schooling or particular aspects of schooling increase children's cognitive test scores. These studies tend to show that additional schooling does increase cognitive skill. But because all of these studies are flawed in one way or another, controversy over the importance of achievement remains.
In chapter 2, Christopher Jencks and Meredith Phillips ask whether what children learn between tenth and twelfth grades affects their future educational attainment and wages—or whether the apparent effects on education and wages of cognitive test scores measured in the twelfth grade are really due to aptitude. Jencks and Phillips use the High School and Beyond (HS&B) survey to estimate how much students gained on tests of math and verbal ability between their tenth and twelfth years in high school. They find, surprisingly, that gains in test scores did not depend on students' tenth grade scores. Therefore Jencks and Phillips argue that these gains do not reflect initial aptitude, but instead reflect learning.
They then use these gains to estimate how much what students learned in school between the tenth and twelfth grade affects their eventual education and wages. Their results show that learning one item on the HS&B math test (the mean gain is 1.52 items) is associated with an additional 0.14 year of school and 2.5 percent greater earnings at ages twenty-seven to twenty-eight. A fifth of the wage gain is attributable to the fact that students who learn more math items get more schooling. Verbal scores have a smaller effect than math scores on education and wages, and gains in verbal scores have almost no effect on additional schooling.
If learning affects educational attainment and wages, it is important to know what affects learning. Jencks and Phillips find, not surprisingly, that students who stay in school learn more than students who drop out. Students who take more math classes learn more math, as do students who do more homework. This suggests that between tenth and twelfth grade, learning is responsive to environmental influences and individual effort, rather than simply depending on aptitude.
In chapter 3, Christopher Winship and Sanders Korenman estimate the effects of schooling and test scores on economic well-being. They find that, holding education constant, a 15 point (one standard deviation) increase in scores on the Armed Forces Qualification Test (AFQT), a test of cognitive skill, increases the family income of twenty-five to thirty-two-year-olds by 18.5 percent and the wages of the same age group by 16.9 percent. Holding AFQT score constant, a one year increase in schooling increases family income by 6.2 percent and wages by 7.9 percent. But, like other researchers, Winship and Korenman find that as people get smarter they get more schooling, and as people get more schooling they get smarter. Each increase in schooling increases cognitive test scores, which in turn increases schooling, until the increments are small. This feedback means that holding test scores constant when estimating the returns to schooling understates the total effect of schooling, and holding schooling constant when estimating the economic payoff to test scores understates the importance of test scores. Taking into account this feedback loop, the composite effect of a year of education on family income is 13.4 percent. The composite effect on earnings is 14.9 percent.
In estimating these feedback effects, Winship and Korenman also find that the effects of schooling and test scores on economic well-being are roughly equal. This implies that the claim that employers are only interested in workers' aptitude is as wrong as the claim that employers care only about workers' achievement. Test scores are a function of more or less equal measures of aptitude and achievement, as is economic success.
If one could design an experiment in which students were randomly assigned to groups that got different amounts of schooling, one could compare the eventual wages of these groups to see if the amount of schooling affects wages. While one cannot conduct such an experiment, random variation in the timing of children's enrollment in school arguably provides a rough approximation. In chapter 4, David Knutson and I take advantage of the fact that most states require children to enter first grade if they reach a compulsory school age, usually six years old, by a particular date, such as September 1. As a result, children born in the months immediately following the cut-off date begin school at a later age than those born in the months just before the cut-off date. The difference in age can be nearly one year. States also require children to stay in school until they reach a particular age, usually sixteen years, rather than until they finish a particular grade. Consequently, children who start school when they are younger, because of their birth date, get more schooling than children who start when they are older. This variation in amount of schooling is more or less random.
Like other researchers, we find that beginning first grade at an early age improves adult economic success. Starting school a year earlier increases future wages by 2.7 percent. Some of this increase in wages is due to the fact that children who start school earlier get more schooling. But the timing of schooling also affects cognitive test scores and wages among children with the same amount of schooling.
We find that among children with the same amount of schooling, beginning school at age five rather than at age seven is associated with a 0.64 standard deviation increment in verbal test scores and a 0.44 standard deviation increment in math scores. These effects decline somewhat as children progress through school, but the effect of early enrollment on verbal ability remains substantial at the end of elementary school. If the timing of schooling affects test scores, and the increase in test scores results in higher wages, the implication is that employers are willing to pay for what students learn in school.
This finding also has important implications for compulsory school laws. Currently, the age at which states require children to enter school varies from five years to eight years. Although increasing the length of schooling is expensive, changing the timing of schooling without changing the length is likely to be relatively inexpensive. In fact, if starting school at younger ages improves children's outcomes, shifting the years of compulsory school attendance down by one year could save money, because it generally costs less to educate young children than to educate older children. In addition, working parents with preschool-age children must find someone else to care for them.
The chapters presented in this volume suggest that an additional year of schooling increases cognitive test scores by about one-fifth of a standard deviation, holding constant initial aptitude. Cognitive test scores appear to increase wages among young adults by 10 to 20 percent. If so, an additional year of schooling would increase wages by 2 to 4 percent, net of the effect of aptitude.
The chapters in this book examine the relationship between cognitive test scores and economic success. But cognitive test scores alone do not determine economic success. As the chapters in this book show, cognitive test scores explain only a modest amount of the overall variation in schooling and wages. Most of the variation in wages can be attributed either to factors other than cognitive skills or to cognitive skills not measured by conventional tests. Employers say they want workers who are not only skilled but are also reliable, creative, confident, and honest. But social scientists have devoted little effort to measuring such characteristics, so one cannot tell how important they are to wage differences. One also cannot tell when a school is doing a better or a worse job of fostering these attributes. In contrast, for over a hundred years social scientists have been developing and refining tests of cognitive skills. These tests still have serious problems, especially for assessing the cognitive skills of individuals as opposed to mean skill levels of groups. Nonetheless it is not surprising that researchers try to evaluate schools on the factors that we can measure best, even while it is surprising how little effort has been spent on trying to understand other attributes that employers, parents, teachers, and policymakers seem to think are important.
The fact that cognitive tests imperfectly measure only one important attribute of children does not imply that one should abandon efforts to improve average test scores or reduce inequalities in test scores. Parents and policymakers agree that increasing how much children know is a central goal of schooling, and when all other factors are equal, employers prefer someone who knows more to someone who knows less. Furthermore, economic success is not the only rationale for improving children's cognitive skills. Improving test scores is likely to improve other outcomes besides children's eventual wages. Children with high cognitive skills are less likely to become teenage parents or unwed parents, or to engage in crime. Thus by focusing only on wages, the chapters in this book no doubt understate the overall social benefit of raising children's cognitive skills.
The studies in part 1 of the volume provide strong evidence that what children learn in school affects their economic success. The studies in part 2 try to determine whether particular aspects of schooling or the ways schools are organized affect how much children learn.
References
Ashenfelter, Orley, and Cecilia Rouse. 1999. "Schooling, Intelligence, and Income in America: Cracks in the Bell Curve." Working Paper 6902. Cambridge, Mass.: National Bureau of Economic Research (January).
Barnett, W. Steven. 1985. "Long-Term Effects of Early Childhood Programs on Cognitive and School Outcomes." Future of Children 5(3): 25-50.
Becker, Gary S. 1964. Human Capital: A Theoretical and Empirical Analysis with Special Reference to Education. New York: National Bureau of Economic Research.
Blackburn, McKinley L., and David Neumark. 1993. "Omitted-Ability Bias and the Increase in the Return to Schooling." Journal of Labor Economics 11(3): 521-44.
Bronfenberger, Uri, and others. 1996. The State of Americans. Free Press.
Burtless, Gary. 1996. Does Money Matter? The Effect of School Resources on Student Achievement and Adult Success. Brookings.
Card, David, and Alan Krueger 1992. "Does School Quality Matter? Returns to Education and the Characteristics of Public Schools in the United States." Journal of Political Economy 100(1): 1-40.
—------. 1994. "The Economic Returns to School Quality: A Partial Survey." Working Paper 334. Princeton University, Industrial Relations Section.
Cattell, Raymond B. 1987. Intelligence: Its Structure, Growth and Action. Amsterdam: Elsevier.
Cawley, John, and James Heckman. 1998. "Three Observations on Wages and Measured Cognitive Ability." Unpublished paper. University of Chicago.
Ceci, Stephen. 1991. "How Much Does Schooling Influence General Intelligence and Its Cognitive Component?" Developmental Psychology 27(5): 703-22.
Cicarelli, V. G., J. W. Evans, and J. S. Schiller. 1969. The Impact of Head Start: An Evaluation of the Effects of Head Start on Children's Cognitive and Affective Development. Athens, Ohio: Westinghouse Learning Corporation and Ohio University.
Coleman, James S., and others. 1966. Equality of Educational Opportunity. Government Printing Office
Duncan, Greg, Rachael Dunifon, and David Knutson. 1997. "Long-Run Effects of Motivation and Social Capital on Success." Unpublished paper. Northwestern University.
Hanushek, Eric A. 1986. "The Economics of Schooling: Production and Efficiency in
Public Schools." Journal of Economic Literature 24(September): 1141-77.
Hartigan, John, and Alexandra Wigidor, eds. 1989. Fairness in Employment Testing: Validity, Generalization, Minority Issues, and the General Aptitude Test Battery. Washington: National Academy Press.
Haskins, Ronald. 1989. "Beyond Metaphor: The Efficacy of Early Childhood Education." American Psychologist 44: 274-82.
Hedges, Larry, Richard Laine, and Rob Greenwald. 1992. "Does Money Matter? Meta-Analyses of the Effect of Differential School Inputs on Student Outcomes." Educational Researcher 23(3): 5-14.
Herrnstein, Richard J., and Charles Murray. 1994. The Bell Curve: Intelligence and Class Structure in American Life. Free Press.
Jencks, Christopher. 1998. "Racial Bias in Testing." In The Black-White Test Score Gap, edited by Christopher Jencks and Meredith Phillips. Brookings.
Jencks, Christopher, and others. 1972. Inequality. Basic Books.
Lazar, Irving, and Richard Darlington. 1982. Lasting Effects of Early Education: A Report from the Consortium for Longitudinal Studies. Monographs for the Society for Research in Child Development Series 195, 47(2-3).
Lee, Valerie E., and Susanna Loeb. 1995. "Where Do Head Start Attendees End Up? One Reason Why Preschool Effects Fade." Educational Evaluation and Policy Analysis 17(1): 62-82.
Loeb, Susanna, and John Bound. 1995. "The Effect of Measured School Inputs on Academic Achievement: Evidence from the 1920s, 1930s, and 1940s Birth Cohorts." Unpublished paper. University of Michigan.
McKey, Ruth H. and others. 1985. The Impact of Head Start on Children, Families and Communities. Washington: CSR, Inc.
Mayer, Susan E. 1996. What Money Can't Buy: Family Income and Children's Life Chances. Harvard University Press.
Mincer, Jacob. 1962. "On-the-Job Training: Costs, Returns and Some Implications." Journal of Political Economy 70 (October).
—------. 1974. Schooling, Experience and Earnings. Columbia University Press.
Murnane, Richard J., John B. Willett, and Frank Levy. 1995. "The Growing Importance of Cognitive Skills in Wage Determination." Review of Economics and Statistics 77(2): 251-66.
Murnane, Richard J., and others. 1997. "Does the Self-Esteem of High School-Aged Males Predict Labor Market Success a Decade Later?" Unpublished paper. Harvard University.
Schultz, Theodore W. 1963. The Economic Value of Education. Columbia University Press.
—------. 1971. Investment in Human Capital: The Role of Education and of Research. Free Press.
Spitz, H. H. 1986. The Raising of Intelligence: A Selected History of Attempts to Raise Retarded Intelligence. Hillsdale, N.J.: Erlbaum.
Winship, Christopher, and Sanders Korenman. 1997. "Does Staying in School Make You Smarter? The Effect of Education on IQ in The Bell Curve." In Intelligence, Genes, and Success: Scientists Respond to The Bell Curve, edited by Bernie Devlin and others. New York: Springer-Verlag.
Yoshikawa, Hirozaku. 1995. "Long-Term Effects of Early Childhood Programs on Social Outcomes and Delinquency." Future of Children 5(3): 51-75.
Table of Contents
| Acknowledgments | vii | |
| Part 1 | Schooling, Cognitive Skills, and Future Earnings | |
| 1 | From Learning to Earning | 3 |
| 2 | Aptitude or Achievement: Why Do Test Scores Predict Educational Attainment and Earnings? | 15 |
| 3 | Economic Success and the Evolution of Schooling and Mental Ability | 49 |
| 4 | Does the Timing of School Affect How Much Children Learn? | 79 |
| Part 2 | Improving Schooling | |
| 5 | School Reforms: How Much Do They Matter? | 105 |
| 6 | How Does Class Size Relate to Achievement in Schools? | 117 |
| 7 | The Evidence on Class Size | 131 |
| 8 | The Effects of Math and Math-Related Courses in High School | 169 |
| 9 | Do Hard Courses and Good Grades Enhance Cognitive Skills? | 205 |
| 10 | Nerd Harassment, Incentives, School Priorities, and Learning | 231 |
| 11 | The Effects of School Choice on Curriculum and Atmosphere | 281 |
| 12 | The Effects of School Choice in New York City | 317 |
| 13 | The Costs and Benefits of School Reform | 341 |
| Contributors | 355 | |
| Index | 357 |