본문 바로가기 주메뉴 바로가기
국회도서관 홈으로 정보검색 소장정보 검색

목차보기

Title page 1

Contents 1

Abstract 3

1. Introduction 4

2. Framework 7

2.1. Basic econometric model 7

2.2. Conceptual model of examiner decision-making 9

3. Estimation 11

3.1. Two-stage least squares 11

3.2. The case for JIVE 13

3.3. Covariates 14

4. Inference 15

4.1. Clustering 16

4.2. Inference and weak identification 16

5. Extensions to the basic framework 20

5.1. Heterogeneous treatment effects 20

5.2. Heterogeneous treatment effects and heterogeneous rankings 22

5.3. Marginal treatment effects 24

5.4. Multi-valued treatments 27

6. Specification testing 30

7. Case study: Effects of pre-trial detention 33

8. Concluding remarks 39

References 50

Online Appendix 54

A. Examples with monotonicity violations 59

B. Detailed discussion of multiple treatments frameworks 61

References 67

Tables 41

Table 1. Pre-trial Detention Case Study: Defendant-level Summary Statistics 41

Table 2. Assessing Balance 42

Table 3. First Stage Analysis for Pre-trial Release 43

Table 4. Second Stage Analysis for Pre-trial Release 44

Table 5. Robustness to Varying Sample Restriction 45

Figures 46

Figure 1. Weak Instrument Simulation Exercise: IV Rejection Rate, Nominal 5-percent Test 46

Figure 2. Marginal Treatment Effects Illustration 47

Figure 3. Distribution of Judge Leniency Measure 48

Figure 4. Illustration of Test of Pairwise Monotonicity and Strict Exclusion 49

Appendix Tables 55

Appendix Table A.1. Applications of Examiner Researcher Designs in Economics 55

Appendix Table A.2. Pairwise Monotonicity is Violated and Average Monotonicity Holds 60

Appendix Table A.3. Both Pairwise Monotonicity and Average Monotonicity Violated 60

Appendix Table A.4. Response types allowed under ACM, NC, and UPM Assumptions 65

Appendix Figures 66

Appendix Figure A.1. Six Potential Treatment Response Types and Treatment Assignment 66