Results 1 - 10
of
554
Alternative Approaches to Evaluation in Empirical Microeconomics
, 2002
"... Four alternative but related approaches to empirical evaluation of policy interventions are studied: social experiments, natural experiments, matching methods, and instrumental variables. In each case the necessary assumptions and the data requirements are considered for estimation of a number of ke ..."
Abstract
-
Cited by 158 (3 self)
- Add to MetaCart
Four alternative but related approaches to empirical evaluation of policy interventions are studied: social experiments, natural experiments, matching methods, and instrumental variables. In each case the necessary assumptions and the data requirements are considered for estimation of a number of key parameters of interest. These key parameters include the average treatment effect, the treatment of the treated and the local average treatment effect. Some issues of implementation and interpretation are discussed drawing on the labour market programme evaluation literature.
TEACHER QUALITY IN EDUCATIONAL PRODUCTION: TRACKING, DECAY, AND STUDENT ACHIEVEMENT
, 2010
"... Growing concerns over the inadequate achievement of U.S. students have led to proposals to reward good teachers and penalize (or fire) bad ones. The leading method for assessing teacher quality is “value added” modeling (VAM), which decomposes students ’ test scores into components attributed to stu ..."
Abstract
-
Cited by 87 (6 self)
- Add to MetaCart
Growing concerns over the inadequate achievement of U.S. students have led to proposals to reward good teachers and penalize (or fire) bad ones. The leading method for assessing teacher quality is “value added” modeling (VAM), which decomposes students ’ test scores into components attributed to student heterogeneity and to teacher quality. Implicit in the VAM approach are strong assumptions about the nature of the educational production function and the assignment of students to classrooms. In this paper, I develop falsification tests for three widely used VAM specifications, based on the idea that future teachers cannot influence students ’ past achievement. In data from North Carolina, each of the VAMs’ exclusion restrictions is dramatically violated. In particular, these models indicate large “effects” of fifth grade teachers on fourth grade test score gains. I also find that conventional measures of individual teachers ’ value added fade out very quickly and are at best weakly related to long-run effects. I discuss implications for the use of VAMs as personnel tools.
The impact of postsecondary remediation using a regression discontinuity approach: Addressing endogenous sorting and noncompliance (NBER Working Paper No. 14194
- National Bureau of Economic Research
, 2008
"... have improved the paper as well as participants at the Teachers College Society of Economics and ..."
Abstract
-
Cited by 81 (1 self)
- Add to MetaCart
(Show Context)
have improved the paper as well as participants at the Teachers College Society of Economics and
Peer Effects, Teacher Incentives, and the Impact of Tracking: Evidence from a Randomized Evaluation in
- Kenya.” American Economic Review
, 2011
"... To the extent that students benefit from high-achieving peers, tracking will help strong students and hurt weak ones. However, all students may benefit if tracking allows teachers to better tailor their instruction level. Lower-achieving pupils are particularly likely to benefit from tracking when ..."
Abstract
-
Cited by 76 (5 self)
- Add to MetaCart
(Show Context)
To the extent that students benefit from high-achieving peers, tracking will help strong students and hurt weak ones. However, all students may benefit if tracking allows teachers to better tailor their instruction level. Lower-achieving pupils are particularly likely to benefit from tracking when teachers have incentives to teach to the top of the distribution. We propose a simple model nesting these effects, and test its implications in a randomized tracking experiment conducted with 121 primary schools in Kenya. While the direct effect of high-achieving peers is positive, tracking benefited lower-achieving pupils indirectly by allowing teachers to teach at their level. JEL Codes: O1, I20 Duflo: Abdul Latif Jameel Poverty Action Lab (J-PAL), Massachusetts Institute of Technology, Paris School of Economics, National Bureau of Economic Research (NBER), and Bureau for Research and Economic Analysis of Development (BREAD) (eduflo@mit.edu). Dupas: UCLA, J-PAL, NBER, and BREAD. (pdupas@econ.ucla.edu). Kremer: Harvard University, Brookings, Center for Global Development (CGD), J-PAL, NBER, and BREAD (mkremer@fas.harvard.edu). We thank Josh Angrist, Abhijit Banerjee, Michael Greenstone, Caroline Hoxby, Guido Imbens, Brian Jacob, and many seminar participants for helpful comments and discussions. We thank four anonymous referees for their suggestions. We thank the Kenya Ministry of Education, Science and Technology, International Child Support Africa, and Matthew Jukes for their collaboration. We thank Jessica Morgan, Frank Schilbach, Ian Tomb, Paul Wang, Nicolas Studer, and especially Willa Friedman for excellent research assistance. We are grateful to Grace Makana and her field team for collecting all the data. We thank, without implicating, the World Bank and the Government of the Netherlands for the grant that made this study possible. To the extent that students benefit from having higher-achieving peers, tracking students into separate classes by prior achievement could disadvantage low-achieving students while benefiting high-achieving students, thereby exacerbating inequality (Denis Epple, Elizabeth Newton and Richard Romano, 2002). On the other hand, tracking could potentially allow teachers to more closely match instruction to students' needs, benefiting all students. This suggests that the impact of tracking may depend on teachers' incentives. We build a model nesting these effects. In the model, students can potentially generate direct student-to-student spillovers as well as indirectly affect both the overall level of teacher effort and teachers' choice of the level at which to target instruction. Teacher choices depend on the distribution of students' test scores in the class as well as on whether the teacher's reward is a linear, concave, or convex function of test scores. The further away a student's own level is from what the teacher is teaching, the less the student benefits; if this distance is too great, she does not benefit at all. We derive implications of this model, and test them using experimental data on tracking from Kenya. In 2005, 140 primary schools in western Kenya received funds to hire an extra grade one teacher. Of these schools, 121 had a single first-grade class, which they split into two sections, with one section taught by the new teacher. In 60 randomly selected schools, students were assigned to sections based on initial achievement. In the remaining 61 schools, students were randomly assigned to one of the two sections. We find that tracking students by prior achievement raised scores for all students, even those assigned to lower achieving peers. On average, after 18 months, test scores were 0.14 standard deviations higher in tracking schools than in non-tracking schools (0.18 standard deviations higher after controlling for baseline scores and other control variables). After controlling for the baseline scores, students in the top half of the preassignment distribution gained 0.19 standard deviations, and those in the bottom half gained 0.16 standard deviations. Students in all quantiles benefited from tracking. Furthermore, tracking had a persistent impact: one year after tracking ended, students in tracking schools scored 0.16 standard deviations higher (0.18 standard deviations higher with control variables). This first set of findings allows us to reject a special case of the model, in which all students benefit from higher-achieving peers but teacher behavior does not respond to class composition. Our second finding is that students in the middle of the distribution gained as much from tracking as those at the bottom or the top. Furthermore, when we look within 1 tracking schools using a regression discontinuity analysis, we cannot reject the hypothesis that there is no difference in endline achievement between the lowest scoring student assigned to the high-achievement section and the highest scoring student assigned to the low-achievement section, despite the much higher-achieving peers in the upper section. These results are inconsistent with another special case of the model, in which teachers are equally rewarded for gains at all levels of the distribution, and so would choose to teach to the median of their classes. If this were the case, instruction would be less well-suited to the median student under tracking. Moreover, students just above the median would perform much better under tracking than those just below the median, for while they would be equally far away from the teacher's target teaching level, they would have the advantage of having higher-achieving peers. In contrast, the results are consistent with the assumption that teachers' rewards are a convex function of test scores. With tracking, this leads teachers assigned to the lowerachievement section to teach closer to the median student's level than those assigned to the upper section, although teacher effort is higher in the upper section. In such a model, the median student may be better off under tracking and may potentially be better off in either the lower-achievement or higher-achievement section. The assumption that rewards are a convex function of test scores is a good characterization of the education system in Kenya and in many developing countries. The Kenyan system is centralized, with a single national curriculum and national exams. To the extent that civil-service teachers face incentives, those incentives are based on the scores of their students on the national primary school exit exam given at the end of eighth grade. But since many students drop out before then, the teachers have incentives to focus on the students at the top of the distribution. While these incentives apply more weakly in earlier grades, they likely help maintain a culture in the educational system that is much more focused on the top of the distribution than in the United States. Moreover, teacher training is focused on the curriculum and many students fall behind it. Indeed, Paul Glewwe, Michael Kremer, and Sylvie Moulin (2009) show that textbooks based on the curriculum benefited only the initially higher-achieving students, suggesting that the exams and associated curriculum are not well-suited to the typical student. It may also be the case that teachers find it easier and more personally rewarding to focus their teaching on strong students, in which case, in the absence of specific incentives otherwise, they will tend to do that. 2 The model also has implications for the effects of the test score distribution in nontracking schools. Specifically, it suggests that an upward shift of the distribution of peer achievement will strongly raise test scores for a student with initial achievement at the top of the distribution, have an ambiguous impact on scores for a student closer to the middle, and raise scores at the bottom. This is so because, while all students benefit from the direct effect of an increase in peer quality, the change in peer composition also generates an upward shift in the teacher's instruction level. The higher instruction level will benefit students at the top; hurt those students in the middle who find themselves further away from the instruction level; and leave the bottom students unaffected, since they are in any case too far from the target instruction level to benefit from instruction. Estimates exploiting the random assignment of students to sections in non-tracking schools are consistent with these implications of the model. While we do not have direct observations on the instruction level and how it varied across schools and across sections in our experiment, we present some corroborative evidence that teacher behavior was affected by tracking. First, teachers were more likely to be in class and teaching in tracking schools, particularly in the high-achievement sections, a finding consistent with the model's predictions. Second, students in the lower half of the initial distribution gained comparatively more from tracking in the most basic skills, while students in the top half of the initial distribution gained more from tracking in the somewhat more advanced skills. This finding is consistent with the hypothesis that teachers are tailoring instruction to class composition, although this could also be mechanically true in any successful intervention. Rigorous evidence on the effect of tracking on learning of students at various points of the prior achievement distribution is limited and much of it comes from studies of tracking in the U.S., a context that may have limited applicability for education systems in developing countries. Reviewing the early literature, Julian Betts and Jamie Shkolnik (1999) conclude that while there is an emerging consensus that high-achievement students do better in tracking schools than in non-tracking schools and that lowachievement students do worse, the consensus is based largely on invalid comparisons.
The Credibility Revolution in Empirical Economics: How Better Research Design is Taking the Con out of Econometrics
, 2010
"... This essay reviews progress in empirical economics since Leamer’s (1983) critique. Leamer highlighted the benefits of sensitivity analysis, a procedure in which researchers show how their results change with changes in specification or functional form. Sensitivity analysis has had a salutary but not ..."
Abstract
-
Cited by 76 (0 self)
- Add to MetaCart
This essay reviews progress in empirical economics since Leamer’s (1983) critique. Leamer highlighted the benefits of sensitivity analysis, a procedure in which researchers show how their results change with changes in specification or functional form. Sensitivity analysis has had a salutary but not a revolutionary effect on econometric practice. As we see it, the credibility revolution in empirical work can be traced to the rise of a design-based approach that emphasizes the identification of causal effects. Design-based studies typically feature either real or natural experiments and are distinguished by their prima facie credibility and by the attention investigators devote to making the case for a causal interpretation of the findings their designs generate. Design-based studies are most often found in the microeconomic fields of Development, Education, Environment, Labor, Health, and Public Finance, but are still rare in Industrial Organization and Macroeconomics. We explain why IO and Macro would do well to embrace a design-based approach. Finally, we respond to the charge that the design-based revolution has overreached.
The E¤ect of Female Education on Fertility and Infant Health: Evidence from School Entry Policies Using Exact Date of Birth,” NBER Working Paper No
"... This paper uses age-at-school-entry policies to identify the effect of female education on fertility and infant health. We focus on sharp contrasts in schooling, fertility, and infant health between women born just before and after the school entry date. School entry policies affect female education ..."
Abstract
-
Cited by 66 (2 self)
- Add to MetaCart
(Show Context)
This paper uses age-at-school-entry policies to identify the effect of female education on fertility and infant health. We focus on sharp contrasts in schooling, fertility, and infant health between women born just before and after the school entry date. School entry policies affect female education and the quality of a woman’s mate and have generally small, but possibly heterogeneous, effects on fertility and infant health. We argue that school entry policies manipulate primarily the education of young women at risk of dropping out of school. (JEL I12, I21, J13, J16) Education is widely held to be a key determinant of fertility and infant health. From a theoretical perspective, several causal channels have been emphasized. First, education raises a woman’s permanent income through earnings, tilting her optimal fertility choices toward fewer offspring of higher quality (Gary S. Becker 1960; Jacob Mincer 1963; Becker and H. Gregg Lewis 1973; Robert J. Willis 1973). Second, under positive assortative mating, a woman’s education is causally connected to her mate’s education
The Value of School Facility Investments: Evidence from a Dynamic Regression Discontinuity Design." Quarterly Journal of Economics, forthcoming
, 2009
"... Despite extensive public infrastructure spending, surprisingly little is known about its economic return. In this paper, we estimate the value of school facility investments using housing markets: standard models of local public goods imply that school districts should spend up to the point where ma ..."
Abstract
-
Cited by 52 (3 self)
- Add to MetaCart
Despite extensive public infrastructure spending, surprisingly little is known about its economic return. In this paper, we estimate the value of school facility investments using housing markets: standard models of local public goods imply that school districts should spend up to the point where marginal increases would have zero effect on local housing prices. Our research design isolates exogenous variation in investments by comparing school districts where referenda on bond issues targeted to fund capital expenditures passed and failed by narrow mar-gins. We extend this traditional regression discontinuity approach to identify the dynamic treatment effects of bond authorization on local housing prices, student achievement, and district composition. Our results indicate that California school districts underinvest in school facilities: passing a referendum causes immedi-ate, sizable increases in home prices, implying a willingness to pay on the part of marginal homebuyers of $1.50 or more for each $1 of capital spending. These effects do not appear to be driven by changes in the income or racial composition of homeowners, and the impact on test scores appears to explain only a small portion of the total housing price effect. I.
2008): “Peer Effects and the Impact of Tracking: Evidence from a Randomized Evaluation
- in Kenya,” MIT
"... This paper provides experimental evidence on the impact of tracking primary school students by initial achievement. In the presence of positive spillover effects from academically proficient peers, tracking may be beneficial for strong students but hurt weaker ones. However, tracking may help everyb ..."
Abstract
-
Cited by 51 (2 self)
- Add to MetaCart
This paper provides experimental evidence on the impact of tracking primary school students by initial achievement. In the presence of positive spillover effects from academically proficient peers, tracking may be beneficial for strong students but hurt weaker ones. However, tracking may help everybody if heterogeneous classes make it difficult to teach at a level appropriate to most students. We test these competing claims using a randomized evaluation in Kenya. One hundred and twenty one primary schools which all had a single grade one class received funds to hire an extra teacher to split that class into two sections. In 60 randomly selected schools, students were randomly assigned to sections. In the remaining 61 schools, students were ranked by prior achievement (measured by their first term grades), and the top and bottom halves of the class were assigned to different sections. After 18 months, students in tracking schools scored 0.14 standard deviations higher than students in non-tracking schools, and this effect persisted one year after the program ended. Furthermore, students at all levels of the distribution benefited from tracking. A regression discontinuity analysis shows that in tracking schools scores of students near the median of the pre-test distribution score are independent of whether they were assigned to the top or bottom section. In contrast, in non-tracking schools we find that on average, students