Showing papers by "Joshua D. Angrist published in 2006"

Long-Term Educational Consequences of Secondary School Vouchers: Evidence from Administrative Records in Colombia

Abstract: Colombia's PACES program provided over 125,000 poor children with vouchers that covered the cost of private secondary school. The vouchers were renewable annually conditional on adequate academic progress. Since many vouchers were assigned by lottery, program effects can reliably be assessed by comparing lottery winners and losers. Estimates using administrative records suggest the PACES program increases secondary school completion rates by 15 to 20 percent. Correcting for the greater percentage of lottery winners taking college admissions tests, the program increased test scores by two-tenths of a standard deviation in the distribution of potential test scores. (JEL: I21, J12, I28)

Quantile Regression under Misspecification, with an Application to the U.S. Wage Structure

Abstract: Quantile regression (QR) fits a linear model for conditional quantiles just as ordinary least squares (OLS) fits a linear model for conditional means. An attractive feature of OLS is that it gives the minimum mean-squared error linear approximation to the conditional expectation function even when the linear model is misspecified. Empirical research using quantile regression with discrete covariates suggests that QR may have a similar property, but the exact nature of the linear approximation has remained elusive. In this paper, we show that QR minimizes a weighted mean-squared error loss function for specification error. The weighting function is an average density of the dependent variable near the true conditional quantile. The weighted least squares interpretation of QR is used to derive an omitted variables bias formula and a partial quantile regression concept, similar to the relationship between partial regression and OLS. We also present asymptotic theory for the QR process under misspecification of the conditional quantile function. The approximation properties of QR are illustrated using wage data from the U.S. census. These results point to major changes in inequality from 1990 to 2000.

Instrumental variables methods in experimental criminological research: what, why and how

Abstract: Quantitative criminology focuses on straightforward causal questions that are ideally addressed with randomized experiments. In practice, however, traditional randomized trials are difficult to implement in the untidy world of criminal justice. Even when randomized trials are implemented, not everyone is treated as intended and some control subjects may obtain experimental services. Treatments may also be more complicated than a simple yes/no coding can capture. This paper argues that the instrumental variables methods (IV) used by economists to solve omitted variables bias problems in observational studies also solve the major statistical problems that arise in imperfect criminological experiments. In general, IV methods estimate causal effects on subjects who comply with a randomly assigned treatment. The use of IV in criminology is illustrated through a re-analysis of the Minneapolis domestic violence experiment. The results point to substantial selection bias in estimates using treatment delivered as the causal variable, and IV estimation generates deterrent effects of arrest that are about one-third larger than the corresponding intention-to-treat effects.

New Evidence on the Causal Link between the Quantity and Quality of Children

Abstract: A longstanding question in the economics of the family is the relationship between sibship size and subsequent human capital formation and economic welfare If there is a causal "quantity-quality tradeoff," then policies that discourage large families should lead to increased human capital, higher earnings, and, at the macro level, promote economic development Ordinary least squares regression estimates and a large theoretical literature suggest that this is indeed the case This paper presents new evidence on the child-quantity/child-quality trade-off Our empirical strategy exploits exogenous variation in family size due to twin births and preferences for a mixed sibling-sex composition, as well as ethnic differences in the effects of these variables and preferences for male births in some ethnic groups We use these sources of variation to look at the causal effect of family size on completed educational attainment, fertility, and earnings For the purposes of this analysis, we constructed a unique matched data set linking Israeli Census data with information on the demographic structure of families drawn from a population registry Our results show no evidence of a quantity-quality trade-off, though some estimates from one subsample suggest that first-born girls from large families marry sooner

Is Spanish-Only Schooling Responsible for the Puerto Rican Language Gap?

Abstract: Between 1898 and 1948, English was the language of instruction for most post-primary grades in Puerto Rican public schools. Since 1949, the language of instruction in all grades has been Spanish. We use this policy change to estimate the effect of English-intensive instruction on the English-language skills of Puerto Ricans. Although naive estimates suggest that English instruction increased English-speaking ability among Puerto Rican natives, estimates that allow for education-specific cohort trends show no effect. This result is surprising in light of the strong presumption by American policymakers at the time that instruction in English was the best way to raise English proficiency. This has implications for medium of instruction policy in former colonies as well as U.S. education policy toward immigrant children.

Multiple Experiments for the Causal Link between the Quantity and Quality of Children

Abstract: A longstanding question in the economics of the family is the relationship between sibship size and subsequent human capital formation and welfare. If there is a causal "quantity-quality trade-off," then policies that discourage large families should lead to increased human capital, higher earnings, and, at the macro level, promote economic development. Ordinary least squares regression estimates and a large theoretical literature suggests that this is indeed the case. This paper presents new evidence on the child-quantity/child-quality trade-off using quasi-experimental variation due to twin births and preferences for a mixed sibling-sex composition, as well as ethnic differences in the effects of these variables and preferences for male births in some ethnic groups. For the purposes of this analysis, we constructed a unique matched data set linking Israeli Census data on human capital, earnings, and other outcomes with information on the structure of families drawn from a population registry. Our sample includes groups with very high fertility. An innovation in our econometric approach is the juxtaposition of results from multiple instrumental variables (IV) strategies, capturing the effects of fertility over different ranges for different sorts of people. To increase precision, we also develop an estimator that combines different instrument sets across partially-overlapping parity-specific sub-samples. The resulting variety of evidence addresses the question of the external validity of a given set of IV estimates. Our results are remarkably consistent in showing no evidence of a quantity-quality trade-off across samples and experiments. We do find, however, that girls from larger families marry sooner.

Lead Them to Water and Pay Them to Drink: An Experiment with Services and Incentives for College Achievement

Abstract: High rates of attrition, delayed completion, and poor achievement are growing concerns at colleges and universities in North America. This paper reports on a randomized field experiment involving two strategies designed to improve these outcomes among first-year undergraduates at a large Canadian university. One treatment group was offered peer advising and organized study group services. Another was offered substantial merit-scholarships for solid, but not necessarily top, first year grades. A third treatment group combined both interventions. Service take-up rates were much higher for students offered both services and scholarships than for those offered services alone. Females also used services more than males. No program had an effect on grades for males. However, first-term grades were significantly higher for females in the two scholarship treatment groups. These effects faded somewhat by year's end, but remain significant for females who planned to take enough courses to qualify for a scholarship. There also appears to have been an effect on retention for females offered both scholarships and services. This effect is large enough to generate an overall increase in retention. On balance, the results suggest that a combination of services and incentives is more promising than either alone.

Lead Them to Water and Pay Them to Drink: An Experiment with Services and Incentives for College Achievement

Abstract: High rates of attrition, delayed completion, and poor achievement are growing concerns at colleges and universities in North America. This paper reports on a randomized field experiment involving two strategies designed to improve these outcomes among first-year undergraduates at a large Canadian university. One treatment group was offered peer advising and organized study group services. Another was offered substantial merit-scholarships for solid, but not necessarily top, first year grades. A third treatment group combined both interventions. Service take-up rates were much higher for students offered both services and scholarships than for those offered services alone. Females also used services more than males. No program had an effect on grades for males. However, first-term grades were significantly higher for females in the two scholarship treatment groups. These effects faded somewhat by year's end, but remain significant for females who planned to take enough courses to qualify for a scholarship. There also appears to have been an effect on retention for females offered both scholarships and services. This effect is large enough to generate an overall increase in retention. On balance, the results suggest that a combination of services and incentives is more promising than either alone.

Is Spanish-Only Schooling Responsible for the Puerto Rican Language Gap?

Abstract: Between 1898 and 1948, English was the language of instruction for most post-primary grades in Puerto Rican public schools. Since 1949, the language of instruction in all grades has been Spanish. We use this policy change to estimate the effect of English-intensive instruction on the English-language skills of Puerto Ricans. Although naive estimates suggest that English instruction increased English-speaking ability among Puerto Rican natives, estimates that allow for education-specific cohort trends show no effect. This result is surprising in light of the strong presumption by American policymakers at the time that instruction in English was the best way to raise English proficiency. This has implications for medium of instruction policy in former colonies as well as U.S. education policy toward immigrant children.