영문목차

Contributors=xvi

1. Introduction and Summary/Gary Burtless=1

School Differences and Student Achievement=4

Educational Resources and Earnings=12

Reconciling the Results=16

The Evidence in This Book=20

School Resources and Student Performance=21

The Relation between School Resources and Student Performance=24

Evidence from Fifteen Schools in Austin=26

Labor Market Effects of School Quality=27

Is There a Link between School Inputs and Earnings?=32

Does Measured School Quality Really Matter?=36

Conclusion=40

References=41

2. School Resources and Student Performance/Eric A. Hanushek=43

The Aggregate Story=44

Approaches to Analyzing the Use of School Resources=52

Educational Production Function Estimates=54

Wage Determination Models with School Resources=62

Conclusions=68

References=70

3. Have Times Changed? The Relation between School Resources and Student Performance/Larry V. Hedges;Rob Greenwald=74

Aggregate Data=74

Educational Production Function Estimates=80

Improving Research on Resource Utilization=88

References=90

4. Evidence from Fifteen Schools in Austin, Texas/Richard J. Murnane;Frank Levy=93

5. Labor Market Effects of School Quality: Theory and Evidence/David Card;Alan B. Krueger=97

A Theoretical Model of School Quality, Education, and Earnings=100

Summary of the Literature=104

Lessons from the Experiences of Southern-Born Blacks=128

Conclusions=133

References=136

6. Is There a Link between School Inputs and Earnings? Fresh Scrutiny of an Old Literature/Julian R. Betts=141

The Direct Impact of School Inputs on Earnings=143

Five Hypotheses to Explain the Observed Patterns=163

The Impact of School Inputs on Educational Attainment=178

Suggestions for Future Research=182

Conclusion=183

Appendix A. A Response to Card and Krueger=185

References=188

7. Does Measured School Quality Really Matter? An Examination of the Earnings-Quality Relationship/James Heckman;Anne Layne-Farrar;Petra Todd=192

The Empirical Plan and the Framework for Organizing the Evidence=198

A Comparison of the Aggregate-and Individual-Level Data Approaches=207

Migration: The Source of Identifying Information in Aggregate-Data Models=209

Testing Monotonicity of the Quality-Earnings Relationship=213

An Empirical Exploration of Two Representative Models=220

Conclusion=252

Appendix A. Description of the Analysis Samples=254

Appendix B. Schooling Quality Data Sources=257

Appendix C. Additional Tables=260

Appendix D. Rejoinder to David Card and Alan Krueger=281

References=288

Index=290

1-1. Spending per Pupil and Average Teacher Salaries in Selected States, 1993-94=2

2-1. Real Current Expenditure on Public Schools, Selected Years, 1890-1990=45

2-2. Sources of Growth in Instructional Expenditure, by Period, 1890-1990=47

2-3. Percentage Distribution of Estimated Effects of Key Resources on Student Performance, Based on 377 Studies=54

2-4. Summary of the Estimated Relationship between Student Performance and Various Components of School Expenditure, Based on Thirty-Nine Value-Added Studies of Individual Student Performance within Individual States=59

2-5. Generalized Least Squares Estimates of School-Specific Educational Attainment=65

2-6. Generalized Least Squares Estimates of State-Specific Educational Attainment=66

2-7. Distribution of Estimated Effect of Teacher-Pupil Ratio and Expenditure per Pupil, by State Sampling Scheme and Aggregation=67

3-1. Trends in National Assessment of Educational Progress Achievement Scores, Selected Years, 1970-92=78

3-2. Indicators of Social Capital, Selected Years, 1940-70=79

3-3. Summary of the Production-Function Coefficients Used in the Combined Significance Analyses of Longitudinal Studies=82

3-4. Results of Combined Significance Tests of Longitudinal Studies=85

3-5. Summary of Results of Combined Significance Tests of Longitudinal Studies=86

5-1. Summary of Studies that Estimate the Effect of Expenditures per Student on Earnings=110

5-2. Summary of Studies that Estimate the Effect of Pupil-Teacher Ratio on Earnings=114

5-3. Effect of Pupil-Teacher Ratio on Return to One Year of Schooling=119

5-4. Summary of Heckman, Layne-Farrar, and Todd's Study of Effect of School Quality on Return to Education=125

5-5. Effect of Pupil-Teacher Ratio on Return to Education, Based on Heckman, Layne-Farrar, and Todd=126

6-1. Percentage of Wage Regressions with Significant Results at 5 Percent, by Author and Data Source=144

6-2. Estimates of Elasticity and Standard Error of Earnings with Respect to Spending per Pupil, by Author and Data Source=151

6-3. Estimates of Elasticity and Standard Error of Earnings with Respect to the Teacher-Pupil Ratio, by Author and Data Source=153

6-4. Estimates of Elasticity and Standard Error of Earnings with Respect to Teachers' Education, by Author and Data Source=154

6-5. Estimates of Elasticity and Standard Error of Earnings with Respect to Teachers' Salary or Relative Salary, by Author and Data Source=156

6-6. Estimates of Elasticity and Standard Error of Earnings with Respect to Length of the School Year, by Author and Data Source=157

6-7. Internal Rate of Return and Ratio of Net Benefits to Costs from Increasing Given Type of School Expenditure, Calculated from Average Elasticities of Earnings with Respect to School Inputs and Average Male Wage Profiles, 1992=158

6-8. Internal Rate of Return and Ratio of Net Benefits to Costs from Increasing Given Type of School Expenditure, Calculated from Average Elasticities of Earnings with Respect to School Inputs and Average Wage Profiles of All Workers, 1992=161

6-9. Internal Rate of Return and Ratio of Net Benefit to Cost for Spending per Pupil versus Spending on One Additional Year of Schooling=162

6-10. Percentage of Educational Attainment Regressions with Results Significant at 5 Percent, by Author and Data Source=179

6-11. Estimates of Increase in Educational Attainment with Respect to Spending per Pupil and Elasticity=180

6A-1. The Impact of a 1 Percent Increase in the Pupil-Teacher Ratio on the Percentage Returns to Education in the Log Earnings Equation=186

7-1. Proportion of States for Which the Average Educational Quality or Average Weekly Wage of Migrants Exceeds That of Nonmigrants, by Birth Cohort, 1910-59, Census Years 1970, 1980, 1990=211

7-2. Distribution of Regional Migration for Birth Cohort, 1930-39, with Varying Education Levels, 1980 Census=212

7-3. Kendall-Tau Correlations between Region-of-Birth Quality Rankings and Log-Wage Rankings, by Birth Cohort, 1920-49, with Twelve Years of Education, 1980 Census=216

7-4. Kendall-Tau Correlations between Region-of-Birth Quality Rankings and Log-Wage Rankings, by Birth Cohort, 1920-49, with Four or More Years of College, 1980 Census=217

7-5. Nonparametric Tests for Agreement in Region-of-Birth Average Wage Rankings across Regions of Residence, by Birth Cohort,: 1910-59, with Varying Education Levels, Census Years 1970, 1980, 1990=219

7-6. Kendall-Tau Correlations between Different Quality Measures, by Birth Cohort, 1920-49=222

7-7. Effect of Schooling Quality on the Rate of Return to Education, Linear Model, 1980 Census=227

7-8. Effect of Hypothetical Improvements in Quality on Earnings for High School Graduates (HSG) and College Graduates (CG), Extension of Linear Model, Census Years 1970, 1980, 1990=229

7-9. Effect of Schooling Quality on Fixed Effects, Linear Model, 1980 Census=233

7-10. F-Tests for Linear Functional Form by Birth Cohort, 1910-59, and by Education Level, Census Years 1970, 1980, 1990=235

7-11. Effect of Schooling Quality on the Linear Base Return to Education, Nonlinear Model, 1980 Census=240

7-12. Effect of Schooling Quality on the Marginal Rate of Return to One to Three Years of College, Nonlinear Model, 1980 Census=241

7-13. Effect of Schooling Quality on the Marginal Rate of Return to Four or More Years of College, Nonlinear Model, 1980 Census=242

7-14. Effect of Hypothetical Improvements in Quality on Earnings for High School Graduates (HSG) and College Graduates (CG), Nonlinear Model including Region-Education Interactions, Census Years 1970, 1980, 1990=243

7-15. F-Tests for Significance of Region of Birth-Region of Residence Interactions, Both in Intercepts and in Slopes for the Linear Model, by Birth Cohort 1910-59, Census Years 1970, 1980, 1990=244

7-16. Effect of Schooling Quality on the Return to Education, Linear Model with Region of Birth-Region of Residence Interactions, 1980 Census=247

7-17. F- Tests for the Restriction That Quality Effects Are Equal across Regions of Residence in the Linear Second-Stage Model with Interactions, Census Years 1970, 1980, 1990=249

7A-1. Summary of Sample Restrictions, Census Years 1970, 1980, 1990=255

7A-2. Sample Sizes for Birth Cohorts, Census Years 1970, 1980, 1990=256

7C-1. Effect of Schooling Quality on the Rate of Return to Education, Linear Model, 1970 Census=261

7C-2. Effect of Schooling Quality on the Rate of Return to Education, Linear Model, 1990 Census=262

7C-3. Effect of Schooling Quality on Fixed Effects, Linear Model, 1970 Census=263

7C-4. Effect of Schooling Quality on Fixed Effects, Linear Model, 1990 Census=264

7C-5. Effect of Schooling Quality on the Linear Base Return to Education, Nonlinear Model, 1970 Census=265

7C-6. Effect of Schooling Quality on the Marginal Rate of Return to One to Three Years of College, Nonlinear Model, 1970 Census=266

7C-7. Effect of Schooling Quality on the Marginal Rate of Return to Four or More Years of College, Nonlinear Model, 1970 Census=267

7C-8. Effect of Schooling Quality on the Linear Base Return to Education, Nonlinear Model, 1990 Census=268

7C-9. Effect of Schooling Quality on the Marginal Rate of Return to One to Three Years of College, Nonlinear Model, 1990 Census=269

7C-10. Effect of Schooling Quality on the Marginal Rate of Return to Four or More Years of College, Nonlinear Model, 1990 Census=270

7C-11. The Effect of Schooling Quality on the Return to Education, Linear Model with Region of Birth-Region of Residence Interactions, 1970 Census=271

7C-12. The Effect of Schooling Quality on the Return to Education, Linear Model with Region of Birth-Region of Residence Interactions, 1990 Census=273

7C-13. Effect of Quality on Educational Attainment, by Birth Cohort, 1920-49, and Education Level, 1980 Census=275

7C-14. Effect of Quality on Educational Attainment, by Birth Cohort, 1910-49, and Education Level, 1970 Census=277

7C-15. Effect of Quality on Educational Attainment, by Birth Cohort, 1930-59, and Education Level, 1990 Census=279

7D-1. Effect of Pupil-Teacher Ratio on Return to One Year of Schooling, Census Years 1970, 1980, 1990=284

1-1. Spending per Pupil, U.S. Average, 1940-90=6

1-2. Pupil/Teacher Ratio, U.S. Average, 1940-90=7

1-3. Relationship of School Attainment and Earnings in Two School Systems=28

1-4. Relationship of School Attainment, School Quality, and Earnings=36

2-1. Instructional Staff and Other Current Expenditure per Student, Selected Years, 1890-1990=46

2-2. Scholastic Aptitude Test Scores, Total and by Race, Selected Years, 1966-93=48

2-3. Science Achievement of Seventeen- Year-Olds by Race and Ethnicity, Selected Years, 1970-90=49

2-4. Reading Achievement of Seventeen- Year-Olds by Race and Ethnicity, 1971-90=50

2-5. Mathematics Achievement of Seventeen- Year-Olds by Race and Ethnicity, 1973-90=51

5-1. Earnings-Schooling Relationship in High- and Low-Quality Schools=103

5-2. Return to Single Years of Education, White Males, Nationwide=105

5-3. Pupil-Teacher Ratios by Race, North Carolina and South Carolina=129

5-4. Difference in Returns to Schooling versus Difference in Pupil-Teacher Ratios for Blacks and Whites, by State=131

6-1. Classification of Studies by Significance of School Inputs and Years in Which Students Received Grade 1-12 Education=149

6-2. Number of School Districts in the United States, 1940-90=167

7-1. Trend in Average Schooling Quality over Time, 1895-1985=194

7-2. Convergence in Schooling Quality over Time, 1895-1985=195

7-3. Comparison of Wages of Persons with Different School-Quality Levels=214

7-4. Evidence of Sheepskin Effects in the Return to Education, 1930-39 Cohort, 1980 Census=236

7-5. Plots of Estimated Education Returns for Four States, 1930-39 Cohort, 1980 Census=239