Identification of Endogenous Social Effects: The Reflection Problem

By Charles F. Manski. 1993

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

The Econometrics of Randomized Experiments (2016, methodology): “An important theoretical paper in an observational setting is Manski (1993) who introduced terminology to distinguish between contextual effects, exogenous effects, and endogenous effects.”
Early adoption of conservation agriculture practices: Understanding partial compliance in programs with multiple adoption decisions (2018, background): “This is yet another example of Manski’s “reflection problem” (Manski, 1993).”
Social Networks and the Diffusion of User-Generated Content: Evidence from YouTube (2012, background): “To identify social effects, Manski (1993) recommends that we understand the factors that dictate the composition of social groups.”
Identification and estimation of the linear-in-means model of social interactions (2005, methodology): “The identifying assumptions underlying much of the research were first formally analyzed by Manski (1993), in his now seminal breflection problemQ paper, which established two nonidentification results.”
The Old Boy (and Girl) Network: Social Network Formation on University Campuses (2008, background): “See the working paper for details.”
Under the hood (2002, methodology): “This is only one example of the kinds of identification problems encountered in spatial econometric models, as shown by Manski (1993) and Kelejian and Prucha (1997), among others.”
USING RANDOMIZATION IN DEVELOPMENT ECONOMICS RESEARCH: A TOOLKIT (2008, background): “However, as Manski (1993) has argued, this could be due to many factors other than peer effects, in particular, to the fact that these individuals share the same environment.”
What makes student entrepreneurs? On the relevance (and irrelevance) of the university and the regional context for student start-ups (2016, background): “However, as Manski (1993) has pointed out, there are three different possible explanations for the relationship between group characteristics and the behaviour of individuals who are part of this group.”
Identification of binary choice models with social interactions (2007, background): “Manski (1993) studies identification in linear and nonparametric contexts and emphasizes the absence of identification when little prior information is available.”
Groundwater pumping and spatial externalities in agriculture (2012, methodology): “Measuring interactions between neighbors is complicated by simultaneity-induced endogeneity (individuals affect their neighbors and their neighbors simultaneously affect them) [24,16,34,23]; we use an instrumental variables approach to purge neighbors’ decisions of this endogenous component.”
The effects of financial reporting and disclosure on corporate investment: A review (2019, background): “As Leuz and Wysocki (2016) discuss, empirically identifying and estimating the economic magnitude of peer effects in a plausibly causal manner is particularly challenging because of the reflection problem, as described by Manski (1993).”
Identifying Neighbourhood Effects on Social Exclusion (2001, background): “Manski (1993), in a framework concerned with identifying the effects of social interactions, separates three types of in uence —Endogenous effects—the propensity for an individual to behave in some way varies with the prevalence of that behaviour in the group—for example, individual school…”
Mostly pointless spatial econometrics (2012, background): “That is, if we allow for what Manski (1993) called ‘exogenous’ or ‘contextual’ neighbourhood effects. As is well known in the neighbourhoods literature and as demonstrated by Manski (1993), once we allow for direct effects from X w ' i on i y we can no longer separately identify β and ρ.”
Identification of Peer Effects through Social Networks (2009, unknown): “In a pioneer study, Manski (1993) distinguishes between endogenous effect, i.e., the influence of peer outcomes, exogenous (or contextual) effects, i.e., the influence of exogenous peer characteristics, and correlated effects, i.e., individuals in the same reference group tend to behave similarly…”