03.03.2015 Views

2000115-Strengthening-Communities-with-Neighborhood-Data

2000115-Strengthening-Communities-with-Neighborhood-Data

2000115-Strengthening-Communities-with-Neighborhood-Data

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Advances in Analytic Methods for <strong>Neighborhood</strong> <strong>Data</strong> 303<br />

services would fail to produce the types of engagement and participation<br />

called for by the theory. Organizers of such place-based interventions<br />

would undoubtedly anticipate impact on individuals (e.g., increased<br />

knowledge or skills), but they would also be interested in showing effects<br />

on the community as a whole (e.g., increased collective efficacy).<br />

Given the multilevel structure of cluster randomized trials, various<br />

important statistical considerations must be taken into account when<br />

planning these studies. One of the most crucial is the question of the<br />

statistical power to detect impact at the cluster and individual levels.<br />

The power is dependent on several factors. The number of clusters and<br />

the number of cases per cluster are key elements of statistical power.<br />

Covariates can be added at either level to reduce variance due to preexisting<br />

differences, which also contributes to power calculations. Finally,<br />

a consideration that is particular to cluster randomized designs is the<br />

role played by the degree of <strong>with</strong>in- and between-cluster heterogeneity<br />

(Raudenbush 1997). Although a detailed discussion of statistical and<br />

design principles is beyond the scope of this chapter, the W. T. Grant<br />

Foundation has undertaken to provide practical tools for researchers<br />

interested in implementing cluster randomized trials (see http://www.<br />

wtgrantfoundation.org/resources/research-tools).<br />

The evaluation of the Jobs-Plus employment program for public housing<br />

residents is an example of a randomized trial applied to evaluation<br />

of a neighborhood-level intervention (Bloom and Riccio 2005).The goal<br />

of Jobs-Plus was to demonstrate that a place-based and comprehensive<br />

employment-focused intervention could raise employment rates among<br />

public housing residents. The initiative rested on the premise that focusing<br />

financial incentives, employment programs, and resident engagement<br />

in a place would be an effective way to address the employment problems<br />

of public housing residents. Public housing developments in five cities<br />

were randomly assigned to be in an experimental or control group. Jobs-<br />

Plus used a comparative interrupted time series design to create a strong<br />

counterfactual (Bloom 2005). The researchers were able to construct a<br />

multiyear baseline trend and a postintervention trend on employment<br />

rates by using data on all adults living in both the experimental and<br />

control sites. An additional feature of the Jobs-Plus evaluation was that<br />

residents who were exposed to the intervention were tracked even if they<br />

left the public housing development. This design allowed the estimation<br />

of the causal impact of the place-based Jobs-Plus model on individuals<br />

regardless of whether they stayed in public housing the entire time or

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!