ABSTRACT
Duncan and Raudenbush (chap. 8, this volume) provide an excellent summary of the most pressing methodological difficulties in assessing neighborhood effects. The authors rightly note that although there is a growing interest by social scientists in understanding the roles that neighborhood and community influences play in the lives of children, adolescents, and families, the research thus far has not been very successful at capturing these neighborhood effects or providing much insight into the processes that might account for them. Duncan and Raudenbush attribute much of this problem to deficiencies related to method, measures, and analytic techniques. Although they do an excellent job overviewing a number of these difficulties and suggesting some possible solutions, we believe the problem is more fundamental and resides earlier in the research process.
