Sources of selection bias in evaluating social programs: An interpretation of conventional measures and evidence on the effectiveness of matching as a program evaluation method

This paper decomposes the conventional measure of selection bias in observational studies into three components. The first two components are due to differences in the distributions of characteristics between participant and nonparticipant (comparison) group members: the first arises from differences in the supports, and the second from differences in densities over the region of common support. The third component arises from selection bias precisely defined. Using data from a recent social experiment, we find that the component due to selection bias, precisely defined, is smaller than the first two components. However, selection bias still represents a substantial fraction of the experimental impact estimate. The empirical performance of matching methods of program evaluation is also examined. We find that matching based on the propensity score eliminates some but not all of the measured selection bias, with the remaining bias still a substantial fraction of the estimated impact. We find that the support of the distribution of propensity scores for the comparison group is typically only a small portion of the support for the participant group. For values outside the common support, it is impossible to reliably estimate the effect of program participation using matching methods. If the impact of participation depends on the propensity score, as we find in our data, the failure of the common support condition severely limits matching compared with random assignment as an evaluation estimator.

Genetic control of social organization in an ant

A central issue in evolutionary biology is the extent to which complex social organization is under genetic control. We have found that a single genomic element marked by the protein-encoding gene Gp-9 is responsible for the existence of two distinct forms of social organization in the fire ant Solenopsis invicta. This genetic factor influences the reproductive phenotypes and behavioral strategies of queens and determines whether workers tolerate a single fertile queen or multiple queens per colony. Furthermore, this factor affects worker tolerance of queens with alternate genotypes, thus explaining the dramatic differences in Gp-9 allele frequencies observed between the two social forms in the wild. These findings reveal how a single genetic factor can have major effects on complex social behavior and influence the nature of social organization.

Social evolution in a new environment: the case of introduced fire ants.

The inadvertent introduction of the fire ant Solenopsis invicta to the United States from South America provides the opportunity to study recent social evolution by comparing social organization in native and introduced populations. We report that several important elements of social organization in multiple-queen nests differ consistently and dramatically between ants in Argentina and the United States. Colonies in Argentina contain relatively few queens and they are close relatives, whereas colonies in the United States contain high numbers of unrelated queens. A corollary of these differences is that workers in the native populations are significantly related to the new queens that they rear in contrast to the zero relatedness between workers and new queens in the introduced populations. The observed differences in queen number and relatedness signal a shift in the breeding biology of the introduced ants that is predicted on the basis of the high population densities in the new range. An additional difference in social organization that we observed, greater proportions of permanently unmated queens in introduced than in native populations, is predicted from the loss of alleles at the sex-determining locus and consequent skewing of operational sex ratios in the colonizing ants. Thus, significant recent social evolution in fire ants is consistent with theoretical expectations based on the altered ecology and population genetics of the introduced populations.