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

2021 ◽  
Author(s):  
Martin E Andresen ◽  
Martin Huber
Keyword(s):  
Instrumental Variables ◽  
Treatment Effects ◽  
College Graduation ◽  
Average Effect ◽  
Exclusion Restriction ◽  
Unit Level ◽  
Policy Relevance ◽  
Induced Changes ◽  
Local Average

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.

scholarly journals Tolerating defiance? Local average treatment effects without monotonicity

10.3982/qe601 ◽  
2017 ◽  
Vol 8 (2) ◽  
pp. 367-396 ◽  
Author(s):  
Clément de Chaisemartin
Keyword(s):  
Treatment Effects ◽  
Average Treatment ◽  
Average Treatment Effects ◽  
Local Average Treatment Effects ◽  
Local Average

scholarly journals Identification and Estimation of Local Average Treatment Effects

Econometrica ◽  
1994 ◽  
Vol 62 (2) ◽  
pp. 467 ◽  
Author(s):  
Guido W. Imbens ◽  
Joshua D. Angrist
Keyword(s):  
Treatment Effects ◽  
Average Treatment ◽  
Average Treatment Effects ◽  
Local Average Treatment Effects ◽  
Local Average

Estimating local average treatment effects in aggregate data

2016 ◽  
Vol 24 (11) ◽  
pp. 762-765 ◽  
Author(s):  
Nick Huntington-Klein
Keyword(s):  
Treatment Effects ◽  
Aggregate Data ◽  
Average Treatment ◽  
Average Treatment Effects ◽  
Local Average Treatment Effects ◽  
Local Average

Use of instrumental variables in electronic health record-driven models

2016 ◽  
Vol 27 (2) ◽  
pp. 608-621 ◽  
Author(s):  
Luca Salmasi ◽  
Enrico Capobianco
Keyword(s):  
Electronic Health Records ◽  
Electronic Health Record ◽  
Instrumental Variables ◽  
Treatment Effects ◽  
Ad Hoc ◽  
Health Record ◽  
Health Records ◽  
Negative Effect ◽  
Electronic Health ◽  
High Degree

Precision medicine presents various methodological challenges whose assessment requires the consideration of multiple factors. In particular, the data multitude in the Electronic Health Records poses interoperability issues and requires novel inference strategies. A problem, though apparently a paradox, is that highly specific treatments and a variety of outcomes may hardly match with consistent observations (i.e., large samples). Why is it the case? Owing to the heterogeneity of Electronic Health Records, models for the evaluation of treatment effects need to be selected, and in some cases, the use of instrumental variables might be necessary. We studied the recently defined person-centered treatment effects in cancer and C-section contexts from Electronic Health Record sources and identified as an instrument the distance of patients from hospitals. We present first the rationale for using such instrument and then its model implementation. While for cancer patients consideration of distance turns out to be a penalty, implying a negative effect on the probability of receiving surgery, a positive effect is instead found in C-section due to higher propensity of scheduling delivery. Overall, the estimated person-centered treatment effects reveal a high degree of heterogeneity, whose interpretation remains context-dependent. With regard to the use of instruments in light of our two case studies, our suggestion is that this process requires ad hoc variable selection for both covariates and instruments and additional testing to ensure validity.

scholarly journals Beyond Just the Average Effect: Variation in Head Start Treatment Effects on Parenting Behavior

AERA Open ◽  
2020 ◽  
Vol 6 (4) ◽  
pp. 233285842096969
Author(s):  
Christina M. Padilla
Keyword(s):  
Head Start ◽  
Treatment Effects ◽  
Parent Engagement ◽  
Parenting Behaviors ◽  
Parenting Behavior ◽  
Local Alternatives ◽  
Impact Study ◽  
Random Assignment ◽  
Average Effect ◽  
Head Start Programs

Parent engagement has been a cornerstone of Head Start since its inception in 1965. Prior studies have found evidence for small to moderate impacts of Head Start on parenting behaviors but have not considered the possibility that individual Head Start programs might vary meaningfully in their effectiveness at improving parenting outcomes. The present study uses the Head Start Impact Study to examine the average effect of random assignment to and participation in Head Start on parenting outcomes as well as variation in that effect across Head Start programs. Findings reveal that Head Start is effective on average at promoting parents’ daily reading and overall literacy and math activities with children but that effects vary significantly for parents’ literacy and math activities, with some programs much more and some much less effective than their local alternatives. Findings also demonstrate that Head Start has consistent near-zero impacts across centers on parents’ disciplinary interactions with children.

scholarly journals Using Instrumental Variables for Inference about Policy Relevant Treatment Effects

2017 ◽  
Author(s):  
Magne Mogstad ◽  
Andres Santos ◽  
Alexander Torgovitsky
Keyword(s):  
Instrumental Variables ◽  
Treatment Effects

Successfully scaled solutions need not be homogenous

2020 ◽  
Vol 5 (1) ◽  
pp. 80-89 ◽  
Author(s):  
DILIP SOMAN ◽  
TANJIM HOSSAIN
Keyword(s):  
Real World ◽  
Research Finding ◽  
Treatment Effects ◽  
Pilot Project ◽  
Policy Relevance ◽  
Research Findings ◽  
Multiple Interventions

AbstractAl-Ubaydli et al. point out that many research findings experience a reduction in magnitude of treatment effects when scaled, and they make a number of proposals to improve the scalability of pilot project findings. While we agree that scalability is important for policy relevance, we argue that non-scalability does not always render a research finding useless in practice. Three practices ensuring (1) that the intervention is appropriate for the context; (2) that heterogeneity in treatment effects are understood; and (3) that the temptation to try multiple interventions simultaneously is avoided can allow us to customize successful policy prescriptions to specific real-world settings.

Sign in / Sign up

Export Citation Format

Share Document