Search Results

Author: Shenhav, Na'ama
Resulting in 3 citations.
1. Miller, Douglas L.
Shenhav, Na'ama
Grosz, Michel Z.
Selection into Identification in Fixed Effects Models, with Application to Head Start
NBER Working Paper No. 26174, National Bureau of Economic Research, August 2019.
Also: https://www.nber.org/papers/w26174
Cohort(s): Children of the NLSY79, NLSY79 Young Adult
Publisher: National Bureau of Economic Research (NBER)
Keyword(s): Educational Attainment; Head Start; High School Completion/Graduates; Modeling, Fixed Effects; Panel Study of Income Dynamics (PSID)

Many papers use fixed effects (FE) to identify causal impacts of an intervention. In this paper we show that when the treatment status only varies within some groups, this design can induce non-random selection of groups into the identifying sample, which we term selection into identification (SI). We begin by illustrating SI in the context of several family fixed effects (FFE) applications with a binary treatment variable. We document that the FFE identifying sample differs from the overall sample along many dimensions, including having larger families. Further, when treatment effects are heterogeneous, the FFE estimate is biased relative to the average treatment effect (ATE). For the general FE model, we then develop a reweighting-on-observables estimator to recover the unbiased ATE from the FE estimate for policy-relevant populations. We apply these insights to examine the long-term effects of Head Start in the PSID and the CNLSY. Using our reweighting methods, we estimate that Head Start leads to a 2.6 percentage point (p.p.) increase (s.e. = 6.2 p.p.) in the likelihood of attending some college for white Head Start participants in the PSID. This ATE is 78% smaller than the traditional FFE estimate (12 p.p). Reweighting the CNLSY FE estimates to obtain the ATE produces similar attenuation in the estimated impacts of Head Start.
Bibliography Citation
Miller, Douglas L., Na'ama Shenhav and Michel Z. Grosz. "Selection into Identification in Fixed Effects Models, with Application to Head Start." NBER Working Paper No. 26174, National Bureau of Economic Research, August 2019.
2. Miller, Douglas L.
Shenhav, Na'ama
Grosz, Michel Z.
Selection into Identification in Fixed Effects Models, with Application to Head Start
Journal of Human Resources (15 November 2021): DOI: 10.3368/jhr.58.5.0520-10930R1.
Also: https://jhr.uwpress.org/content/early/2021/11/03/jhr.58.5.0520-10930R1.abstract
Cohort(s): Children of the NLSY79, NLSY79 Young Adult
Publisher: University of Wisconsin Press
Keyword(s): College Enrollment; Head Start; Modeling, Fixed Effects; Panel Study of Income Dynamics (PSID)

Many papers use fixed effects (FE) to identify causal impacts of an intervention. When treatment status only varies within some FE groups (e.g., families, for family fixed effects), FE can induce non-random selection of groups into the identifying sample, which we term selection into identification (SI). This paper empirically documents SI in the context of several family fixed effects (FFE) applications with a binary treatment. We show that the characteristics of the FFE identifying sample are different than the overall sample (and the policy-relevant population), including having larger families. The main implication of this is that when treatment effects are heterogeneous, the FE estimate may not be representative of the average treatment effect (ATE). We show that a reweighting-on-observables FE estimator can help recover the ATE for policy-relevant populations, and recommend its use either as a primary estimator or as a diagnostic tool to assess the importance of SI. We apply these insights to re-examine the long-term effects of Head Start in the PSID and the CNLSY using FFE. When we reweight the FFE estimates, we find that Head Start leads to a 2.1 percentage point (p.p.) increase (s.e. = 5.9 p.p.) in the likelihood of attending some college for white Head Start participants in the PSID. This participants' ATE is 83% smaller than the traditional FFE estimate (12 p.p). We also find that the CNLSY Head Start participants' ATE is smaller than the FE estimates. This raises new concerns with the external validity of FE estimates.
Bibliography Citation
Miller, Douglas L., Na'ama Shenhav and Michel Z. Grosz. "Selection into Identification in Fixed Effects Models, with Application to Head Start." Journal of Human Resources (15 November 2021): DOI: 10.3368/jhr.58.5.0520-10930R1.
3. Shenhav, Na'ama
Essays on Gender Gaps and Investments in Children
Ph.D. Dissertation, Department of Economics, University of California, Davis, 2016
Cohort(s): Children of the NLSY79, NLSY79
Publisher: ProQuest Dissertations & Theses (PQDT)
Keyword(s): Behavior Problems Index (BPI); Child Health; Earnings, Wives; Family Income; Geocoded Data; Household Income; Obesity; Parental Influences; Peabody Individual Achievement Test (PIAT- Math); Peabody Individual Achievement Test (PIAT- Reading); Wage Gap

Permission to reprint the abstract has not been received from the publisher.

The third chapter builds on the findings made in the first two chapters, and explores the implications of changes in male and female wage opportunities for child achievement. It contributes to a large literature that has shown that a child's academic success and physical development are strongly influenced by family income, but which has less evidence on whether the source of income also matters. The empirical strategy takes advantage of national shifts in the return to occupations over this time period as a source of exogenous convergence of wages across sexes in a marriage market. In contrast to previous findings, the results do not show that a higher female to male wage ratio significantly improves children's outcomes, although the confidence intervals allow for an important positive or negative effect. Auxiliary analyses which use observed relative household income produce a qualitatively different, negative and statistically significant effect of relative wages on children's development, which is likely a reflection of an omitted variable bias. Sources of the imprecision in the estimation are discussed.
Bibliography Citation
Shenhav, Na'ama. Essays on Gender Gaps and Investments in Children. Ph.D. Dissertation, Department of Economics, University of California, Davis, 2016.