手法を比較

	教育研究における操作変数	局所的平均処置効果（LATE / CACE）
分野	因果推論	因果推論
系統	Regression model	Regression model
提唱年	1991 (canonical education application)	1994
提唱者	Angrist & Krueger (canonical 1991 education application); grounded in IV theory by Wright (1928)	Imbens & Angrist (1994); Angrist, Imbens & Rubin (1996)
種類	Quasi-experimental causal identification	Instrumental-variable causal estimand
原典	Angrist, J. D., & Krueger, A. B. (1991). Does Compulsory School Attendance Affect Schooling and Earnings? Quarterly Journal of Economics, 106(4), 979-1014.	Imbens, G. W., & Angrist, J. D. (1994). Identification and Estimation of Local Average Treatment Effects. Econometrica, 62(2), 467-475.
別名	IV in education, 2SLS in education, education IV, school IV estimation	LATE, CACE, complier average causal effect, Yerel Ortalama Tedavi Etkisi (LATE / CACE)
関連	5	5
概要	Instrumental variables (IV) estimation is a quasi-experimental strategy for isolating the causal effect of schooling or educational interventions when assignment to treatment is confounded by unobserved factors. Pioneered in education economics by Angrist and Krueger's use of quarter-of-birth as an instrument for compulsory schooling, IV finds a source of exogenous variation in exposure to education and uses only that variation to estimate outcomes such as earnings, test scores, or attainment.	The Local Average Treatment Effect is an instrumental-variable estimand, introduced by Imbens and Angrist (1994) and formalised with Rubin (1996), that recovers the average treatment effect for the subpopulation of compliers — units whose treatment status is actually moved by the instrument. It is closely tied to compliance analysis.