College Proximity Research Articles

Distance to Degrees: How College Proximity Shapes Students' Enrollment Choices and Attainment Across Race-Ethnicity and Socioeconomic Status

Distance to Degrees: How College Proximity Shapes Students' Enrollment Choices and Attainment Across Race-Ethnicity and Socioeconomic Status

When to use matching and weighting or regression in instrumental variable estimation? Evidence from college proximity and returns to college

Standard two-stage least squares (2SLS) regression remains dominant in instrumental variables estimation of causal effects even though the literature has shown that 2SLS may be inconsistent when effects are heterogenous and the instrument is only valid when conditioning on covariates. To show that this is not merely a hypothetical threat, this paper re-estimates the returns to college using college proximity as an instrument based on the data from Card (Aspects of labour market behavior: essays in honour of John Vanderkamp, University of Toronto Press, Toronto, 1995). The results show that 2SLS yields systematically larger estimates of the returns to college than more flexible estimators based on the instrument propensity score. In the full sample, differences amount to about 50 to 100%. This is due to the implicit conditional-variance weighting performed by 2SLS. Moreover, in line with the theoretical prediction by Sloczynski (When should we (not) interpret linear IV estimands as LATE? IZA discussion papers 14349, Institute of Labor Economics (IZA), 2021), findings suggest that the impact of the conditional-variance weighting is larger when instrument groups are not roughly the same size. Thus, it is advised to use 2SLS with caution and use estimators based on the instrument propensity score instead when groups are of different size and covariates are predictive of the instrument.

Identifying treatment effects in the presence of confounded types

In this paper, I consider identification of treatment effects when the treatment is endogenous. The use of instrumental variables is a popular solution to deal with endogeneity, but this may give misleading answers when the instrument is invalid. I show that when an (unobserved) instrument is invalid due to correlation with the first stage unobserved heterogeneity, a proxy for the instrument helps partially identify not only the local average treatment effect, but also the entire potential outcomes distributions for compliers. I exploit the fact that the distribution of the observed outcome in each group defined by the treatment and the instrument is a mixture of the distributions of interest. I write the identified set in the form of conditional moment inequalities, and provide an easily implementable inference procedure. Under some tail restrictions, the potential outcomes distributions are point-identified for compliers. Finally, I illustrate my methodology on data from the National Longitudinal Survey of Young Men to estimate returns to college using college proximity as a proxy for the instrument low college cost. I find that a college degree increases the average hourly wage of the compliers by 15%–30%.

Instrument-based estimation with binarised treatments: issues and tests for the exclusion restriction

Summary When estimating local average and marginal treatment effects using instrumental variables (IVs), multivalued endogenous treatments are frequently converted to binary measures, supposedly to improve interpretability or policy relevance. Such binarisation introduces a violation of the IV exclusion if (a) the IV affects the multivalued treatment within support areas below and/or above the threshold and (b) such IV-induced changes in the multivalued treatment affect the outcome. We discuss assumptions that satisfy the IV exclusion restriction with a binarised treatment and permit identifying the average effect of (a) the binarised treatment and (b) unit-level increases in the original multivalued treatment among specific compliers. We derive testable implications of these assumptions and propose tests which we apply to the estimation of the returns to college graduation instrumented by college proximity.

INFERENCE IN INSTRUMENTAL VARIABLE MODELS WITH HETEROSKEDASTICITY AND MANY INSTRUMENTS

This paper proposes novel inference procedures for instrumental variable models in the presence of many, potentially weak instruments that are robust to the presence of heteroskedasticity. First, we provide an Anderson–Rubin-type test for the entire parameter vector that is valid under assumptions weaker than previously proposed Anderson–Rubin-type tests. Second, we consider the case of testing a subset of parameters under the assumption that a consistent estimator for the parameters not under test exists. We show that under the null, the proposed statistics have Gaussian limiting distributions and derive alternative chi-square approximations. An extensive simulation study shows the competitive finite sample properties in terms of size and power of our procedures. Finally, we provide an empirical application using college proximity instruments to estimate the returns to education.

Identifying Treatment Effects in the Presence of Confounded Types

In this paper, I consider identification of treatment effects when the treatment is endogenous. The use of instrumental variables is a popular solution to deal with endogeneity, but this may give misleading answers when the instrument is invalid. I show that when an (unobserved) instrument is invalid due to correlation with the first stage unobserved heterogeneity, a proxy for the instrument helps partially identify not only the local average treatment effect, but also the entire potential outcomes distributions for compliers. I exploit the fact that the distribution of the observed outcome in each group defined by the treatment and the instrument is a mixture of the distributions of interest. I write the identified set in the form of conditional moment inequalities, and provide an easily implementable inference procedure. Under some tail restrictions, the potential outcomes distributions are point-identified for compliers. Finally, I illustrate my methodology on data from the National Longitudinal Survey of Young Men to estimate returns to college using college proximity as a proxy for the instrument low college cost. I find that a college degree increases the average hourly wage of the compliers by 15-30%.

College Match and Undermatch: Assessing Student Preferences, College Proximity, and Inequality in Post-College Outcomes

Recently, multiple studies have focused on the phenomenon of “undermatching”—when students attend a college for which they are overqualified, as measured by test scores and grades. The extant literature suggests that students who undermatch fail to maximize their potential. However, gaps remain in our knowledge about how student preferences—such as a desire to attend college close to home—influence differential rates of undermatching. Moreover, previous research has not directly tested whether and to what extent students who undermatch experience more negative post-college outcomes than otherwise similar students who attend “match” colleges. Using ELS:2002, we find that student preferences for low-cost, nearby colleges, particularly among low-income students, are associated with higher rates of undermatching even among students who are qualified to attend a “very selective” institution. However, this relationship is weakened when students live within 50 miles of a match college, demonstrating that proximity matters. Our results show that attending a selective postsecondary institution does influence post-college employment and earnings, with less positive results for students who undermatch as compared with peers who do not. Our findings demonstrate the importance of non-academic factors in shaping college decisions and post-college outcomes, particularly for low-income students.

The Expansion of Higher Education in Turkey: Access, Equality and Regional Returns to Education

This article discusses the recent expansion and structural changes in the Turkish higher education and analyses a number of economic consequences in terms of equity and returns across regions. First, we outline the institutional background of the expansion in order to identify various re-distributive dimensions of the policy. Using Household Labor Force Surveys between 2004 and 2013, we explore whether college proximity affects local families’ access to college. Our results show that this policy had an equity-enhancing effect for daughters of low-educated households in some regions with large-scale expansion. Second, we investigate whether the compositional change has affected local returns to college degrees and contributed to the relative convergence across regions. The estimation results show that despite the increase in college graduates, returns in terms of wages at the local level has increased increasing and that some regional convergence was attained.

Leaving or Staying Home: Predicting Where Students Attend College

Focusing on a cohort of high school students from a Midwest metropolitan region, this study combines multiple sources of data and uses a multinomial logistic regression to model student postsecondary choices with respect to whether and where to attend college. Specifically, we examined the enrollment patterns by proximity to the home region and factors associated with these college decisions. The results suggest that these students’ college choices were a process influenced by both precollege individual characteristics and social contexts. The findings also supported our hypothesis that acquisition of various types of capital and academic success of the school district (as one of several indicators of a college-going culture) were negatively related to student preferences for college proximity. These findings highlight the interplay between individual, family, community, and school at different levels as it influences college decisions of students from the deindustrialized Midwest region and regions alike.

Testing Local Average Treatment Effect Assumptions

In this paper, we propose an easy-to-implement procedure to test the key conditions for the identification and estimation of the local average treatment effect (LATE; Imbens & Angrist, 1994). We reformulate the testable implications of LATE assumptions as two conditional inequalities, which can be tested in the intersection bounds framework of Chernozhukov, Lee, and Rosen (2013) and easily implemented using the Stata package of Chernozhukov et al. (2015). We apply the proposed tests to the draft eligibility instrument in Angrist (1991), the college proximity instrument in Card (1993), and the same-sex instrument in Angrist and Evans (1998).

On falsification of the binary instrumental variable model.

SUMMARYInstrumental variables are widely used for estimating causal effects in the presence of unmeasured confounding. The discrete instrumental variable model has testable implications for the law of the observed data. However, current assessments of instrumental validity are typically based solely on subject-matter arguments rather than these testable implications, partly due to a lack of formal statistical tests with known properties. In this paper, we develop simple procedures for testing the binary instrumental variable model. Our methods are based on existing techniques for comparing two treatments, such as the n}{}t-test and the Gail–Simon test. We illustrate the importance of testing the instrumental variable model by evaluating the exogeneity of college proximity using the National Longitudinal Survey of Young Men.

Exclusion Restrictions and Identification of Returns to Schooling

The paper presents exclusion restrictions that allow identification of returns to schooling using data from the National Survey of Young Men (1966) (NLSYM66). The approach does not use instrumental variables (IV) or control function approaches. Instead, a non-parametric finite mixture model is used to estimate the distribution of latent types. This approach requires two exclusion restrictions to identify the causal effect of attending college on wages. The model identifies two latent types. For the latent type with a low propensity for attending college, receiving 13 or more years of education leads to higher income than having 12 years or fewer years. However, the estimated returns to education for this group is approximately one half of the observed difference and one third of the IV estimates presented in Card (1995) and Kling (2001) using the same data. The difference is due to the difference in the mix of latent types associated with proximity to college and the impact of the college proximity instrument on respondents with a high propensity to attend college.

Testing LATE Assumptions

In this paper, we discuss the key conditions for the identification and estimation of the local average treatment effect (LATE, Imbens and Angrist, 1994): the valid instrument assumption (LI) and the monotonicity assumption (LM). We show that the joint assumptions of LI and LM have a testable implication that can be summarized by a sign restriction defined by a set of intersection bounds. We propose an easy-to-implement testing procedure that can be analyzed in the framework of Chernozhukov, Lee, and Rosen (2013) and implemented using the Stata package of Chernozhukov,Kim, Lee, and Rosen (2013). We apply the proposed tests to the “draft eligibility” instrument in Angrist (1991), the “college proximity” instrument in Card (1993) and the “same sex” instrument in Angrist and Evans (1998).

College Proximity: Mapping Access to Opportunity

The process by which students decide whether and where to attend college has been based most commonly on a college-choice model that is independent of the students' geographic context. However, the ability to attend college close to home is often among the most important factors that U.S. high school students, especially minorities and the socioeconomically disadvantaged, consider. The study presented here mapped the place of residence of a national sample of high school seniors, as well as the location of all colleges within commuting distance of each student, and showed that seniors have a wide range of colleges in proximity. Furthermore, after taking account of important student and zip code-level factors, the study found that each additional college in proximity is associated with a small but significant increase in the odds of applying to college, especially a four-year college. These findings suggest that researchers should stop treating the college-choice process as though it were independent of location and start situating this process within the geographic context in which it occurs.

At the End of the Day, I Want to Be Close to Home: Adult Students' Preferences for College Proximity to Work and Home

ABSTRACT Using 10 distinct samples, analyses were conducted to determine whether empirical support could be garnered for the proposition that working adults selecting a college for part-time studies tend to consider the proximity of the school to their home to be more important than the school's proximity to where they work. On a hypothetical task (first 5 samples), a clear preference for home proximity was evident. However, when real-life choices were studied (last 5 samples), the evidence for a preference for proximity to home was much weaker, perhaps because the research was conducted in a metropolitan region with numerous college choices, so that the perceived proximity to home and work were identical in most instances.

Skill, parental income, and IV estimation of the returns to schooling

Recently, attention has moved away from using parental background variables, such as parental education, in favour of using institutional features of the education system as instruments when estimating the return to schooling. In this paper, these possible instruments are revisited. Using the National Longitudinal Survey of Youth, several specifications of the wage equation are estimated and three types of instruments used - parental education, quarter of birth, and college proximity. It is shown that under some specifications - in particular, by including parental income and individual skill in the wage equation - parental education appears to be a valid and useful instrument. On the other hand, when using the institutional instruments, the weak correlation between the instruments and years of schooling produces imprecise and likely biased estimates of the return to schooling.

Interpreting Instrumental Variables Estimates of the Returns to Schooling

This article synthesizes economic insights from theoretical models of schooling choice based on individual benefits and econometric work interpreting instrumental variables estimates as weighted averages of individual-specific causal effects. Linkages are illustrated using college proximity to instrument for schooling. After characterizing groups differentially affected by the instrument according to family background, I directly compute weights underlying estimation of the overall return. In analyzing the level of schooling at which individuals change their behavior in response to the instrument, I demonstrate that this instrument has its greatest impact on the transition from high school to college. Specification robustness is also examined.

College Proximity and Rates of College Attendance

This paper presents the results of a study on the effect of college proximity upon rates of college attendance among over 20,000 high school graduates of 1966 in the states of Illinois and North Carolina. Results of multivariate dummy-variable regression analyses question both the assumption that college proximity per se is an important factor in college-going, and the often stated belief that the establishment of public junior colleges will help to equalize educational opportunity by providing higher educational access to able children of low status families. In both Illinois and North Carolina, only persons of lower ability appeared to gain in attendance when living in a community with a public junior college.

College Proximity and Rates of College Attendance

College Proximity and Rates of College Attendance