Papers - Method Flashcards
(39 cards)
Hanna & Linden (2012) Discrimination in Grading
Randomly assign caste-gender labels to identical scripts; each script graded by multiple teachers. OLS with grader fixed effects, fractional logit, and quantile regressions compare marks across labels; interactions with grader caste, grading order, and question difficulty test heterogeneity; placebo gender test validates design.
Lavy (2008) Do Gender Stereotypes Reduce Girls or Boys Human Capital Outcomes?
Compute within-student gap between teacher and blind external scores across subjects; regress bias on gender with student FE, subject dummies, classroom covariates. Teacher-switching, sibling-FE and IV for expected effort test robustness; quantile regressions explore distributional effects.
Alesina et al. (2024) Revealing Stereotypes: Evidence from Immigrants in Schools
Randomly varies timing and personalisation of IAT feedback before grading; compares immigrant-native marks.
Bertrand & Mullainathan (2004) Are Emily and Greg More Employable than Lakisha and Jamal?
Logit and linear-probability models with job-ad FE measure callback differences by name race and resume quality; interactions test credential returns. Randomisation inference and cluster-bootstrapped errors provide rigorous inference.
Altonji, Bharadwaj & Lange (2012) Changes in the Characteristics of American Youth
Applies DiNardo-Fortin-Lemieux reweighting to map 1979 wage structure onto 1997 covariates; Shapley decomposition allocates changes to demographics, parental education, AFQT and school quality; bootstrapped SEs confirm precision.
Abramitzky & Lavy (2014) How Responsive is Investment in Schooling to Changes in Returns?
Staggered-adoption triple-difference interacts cohort, exposure grade and reform intensity; synthetic controls and placebo pre-trend checks bolster causal inference; IV for predicted wage gain addresses measurement error.
Altonji, Bharadwaj & Lange (2008) The Anemic Response of Skill Investment to Skill Premium Growth
Dynamic-programming model with tuition, borrowing cap and wage uncertainty predicts enrolment; state-cohort panel regressions instrument college premium using demand shocks; robustness to aid, unemployment and tuition changes.
Jensen (2010) The Perceived Returns to Education and the Demand for Schooling
Cluster-randomised info sessions in 200 schools; difference-in-means and IV using treatment as instrument link belief updates to dropout; attrition reweighting ensures balance.
Acemoglu, Johnson & Robinson (2002) Reversal of Fortune
Quadratic OLS and IV regressions of 1995 income on 1500 density; settler mortality mediates channel; spatial autocorrelation corrected; Blinder decomposition assesses share explained by institutions.
Acemoglu, Johnson & Robinson (2001) The Colonial Origins of Comparative Development
2SLS with settler mortality as instrument for institutions; jackknife and over-ID tests validate exclusion; robustness to region, legal origin and outliers.
Nunn & Qian (2011) The Potatos Contribution to Population and Urbanization
Country FE DD using post-1700*potato suitability; IV with altitude; event-study plots verify parallel trends; placebo on sweet potatoes checks specificity.
Imberman, Kugler & Sacerdote (2012) Katrinas Children: Peer Effects from Hurricane Evacuees
School-grade FE regressions; IV for evacuee share using parish destruction; dynamic panel and quantile analysis test heterogeneity; placebo pre-Katrina sample.
Lavy & Schlosser (2011) Mechanisms and Impacts of Gender Peer Effects at School
Instrument female share with adjacent cohort sex ratio; school*grade FE and cohort trends; mediation via classroom climate indices; placebo on independent schools.
Lavy & Sand (2018) Social Networks and Student Outcomes
Random class assignment instruments friend exposure; multi-level models with student FE; network centrality metrics; Manski bounds for spillovers.
Gould, Lavy & Paserman (2004) Immigrating to Opportunity: School Quality for Ethiopian Israelis
Regression of outcomes on pre-migration school quality with community FE; IV uses native-teacher share; survival analysis for dropout; migration robustness checks.
Duflo & Hanna (2005) Monitoring Works: Getting Teachers to Come to School
Cluster RCT; DID with school FE; compliance via camera logs; cost-effectiveness compared to extra teachers; two-year follow-up.
Lavy (2009) Performance Pay and Teachers Effort, Productivity and Ethics
Regression discontinuity around 45 % pass cutoff; diff-in-diff across subjects; teacher surveys on pedagogy; manipulation and placebo tests ensure validity.
Katz & Murphy (1992) Changes in Relative Wages, 19631987: Supply and Demand Factors
CES demand for college vs non-college labour; decompose wage changes into supply and demand within/Between industries; dynamic adjustment model.
Autor, Levy & Murnane (2003) Skill Content of Recent Technological Change
Link DOT task data to CPS and equipment prices; regress computer adoption on routine task share; task-demand matrix converts shifts to labour demand.
Goos, Manning & Salomons (2014) Explaining Job Polarization
Three-task CES demand estimated via GMM; decompose within/Between-industry contributions; offshoring intensity via input-output tables.
Autor & Dorn (2013) Growth of LowSkill Service Jobs and US Polarization
Spatial-equilibrium model; regress 1980 routine share on 2000-05 service growth; IV uses 1950 industry mix; bootstrapped SEs.
Deming (2017) Growing Importance of Social Skills in the Labor Market
Wage regressions with occupation and cohort FE; interact social intensity with decade dummies; Bartik IV for IT adoption; quantile analysis.
Cavounidis et al. (2024) Nature of Technological Change 19602016
Markov transitions estimate skill-productivity growth; cross-validate with O*NET; simulate skill complementarity; bootstrapped SEs.
Cavounidis et al. (2024) Obsolescence Rents: Truckers & Impending Computerization
Dynamic OLG model calibrated to trucking; event-study diff-in-diff comparing high- vs low-risk jobs around tech news.
