Neighbourhood Measures: Quantifying the Effects of Neighbourhood Externalities
In recent years, analyses of neighbourhood externalities have grown with the perceived importance of their influence upon outcomes. Despite this growth, a clear understanding of the role of neighbourhoods in determining outcomes remains elusive. Various attempts have been made to quantify the role of neighbourhoods and limit problems of misspecification that have plagued this literature. Recent research suggests that neighbourhood proxies that measure characteristics similar to the dependent variable may better capture neighbourhood externalities. We explore variation in estimations including distinct neighbourhood proxies by estimating the influence of neighbourhood externalities upon youths' education expectations. Misspecification tests for normality and heteroscedasticity show particular neighbourhood proxies are more susceptible to misspecification. Monte-Carlo experiments show these neighbourhood proxies are also more likely to produce biased estimates if particular family characteristics are not fully captured. We find estimations including neighbourhood proxies measuring characteristics proximate to youths' education are less likely to suffer misspecifications. We also find that different geographic definitions of neighbourhoods can lead to erroneous findings, particularly considering variation in school quality.
|Date of creation:||Feb 2003|
|Date of revision:|
|Contact details of provider:|| Postal: Melbourne Institute of Applied Economic and Social Research, The University of Melbourne, Victoria 3010 Australia|
Phone: +61 3 8344 2100
Fax: +61 3 8344 2111
Web page: http://www.melbourneinstitute.com/
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:iae:iaewps:wp2003n04. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Abbey Treloar)
If references are entirely missing, you can add them using this form.