Measurement of Peer Effects

Measurement of Peer Effects - Measurement of Peer E/ects...

Info iconThis preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Measurement of Peer E/ects Manski (1993), Mo¢ tt (2001), Wooldridge 11.5 Classic Example students motivated by classmates peer group formed by classroom g k- group of students in classroom k achieve i = & + ¡ E ( achieve j k ) + u i with i 2 g k achieve- performance on achievement test E ( achieve j k )- expected achieve in classroom k i students (30 in each of 3 classrooms) Finding of Peer E/ects? researchers often &nd ^ & > and statistically signi&cant and might argue "evidence of peer e/ects" Di¢ culties simultaneity (re¡ection problem) correlated unobservables endogenous group selection Re&ection Problem Proposed regression: achieve i = & + ¡ E ( achieve j k ) + u i E ( achieve i j k ) = & + ¡ E ( achieve j k ) E ( achieve j k ) same for all students in k ) E ( achieve i j k ) = E ( achieve j k ) ) E ( achieve i j k ) = & 1 & ¡ achieve i = & 1 & ¡ + u i & and ¡ are not separately identi¡ed Simultaneity & Exogenous E/ects Exogenous e/ect Students motivated by classmates¡parents parent i- achievement of parents of i 2 students in class achieve 1 = & + ¡achieve 2 + ¢parent 2 + £parent 1 + u 1 achieve 2 = & + ¡achieve 1 + ¢parent 1 + £parent 2 + u 2 no excluded exogenous variables order condition violated ) ( ¡; ¢; £ ) not identi¢ed Reduced-Form Identi&cation Reduced form achieve 1 = & 1 + & 2 parent 1 + & 3 parent 2 + v 1 achieve 2 = & 1 + & 2 parent 2 + & 3 parent 1 + v 2 If Cov & parent i ; v j ¡ = 0 , then OLS delivers consistent estimates of ( & 1 ; & 2 ; & 3 ) & 3 = ¡ + ¢£ 1 & ¢ 2 & 3 6 = 0 ) social e/ects Social E/ects Multipliers (Endogenous) Peer E/ects - have a social multiplier...
View Full Document

This note was uploaded on 12/26/2011 for the course ECON 245a taught by Professor Staff during the Fall '08 term at UCSB.

Page1 / 22

Measurement of Peer Effects - Measurement of Peer E/ects...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online