|Evaluations of large Federal initiatives that include many sites often target a subset of sites for more intensive study than is financially feasible in all sites. Site selection should be carefully considered, as it has implications for the ability to generalize from and interpret findings. This round-table discussion will address important aspects of site selection, including:
a) Should sites be randomly selected? If so, should randomization be stratified along key dimensions? b) If sites are to be purposively selected, what criteria should be used to identify the information upon which sites will be selected? When sites are selected purposively, how should findings be interpreted? c) To what extent should non-selected sites be included in the evaluation? How should findings from selected sites be used to enhance understanding of non-selected sites?
Researchers from RTI International and the Urban Institute will lead the discussion, using as expamples the following evaluations of multi-site Federal initiatives:
a) the Safe Schools/Healthy Students Initiative (SS/HS) national evaluation, which includes 97 sites; b) the Serious and Violent Offender Reentry Initiative (SVORI) national evaluation, which includes 68 sites; and c) the drug courts national evaluation.
(Return to Program Resources)