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.

A quantitative method for evaluating the stabilities of nucleic acids

We report a general method for screening, in solution, the impact of deviations from canonical Watson-Crick composition on the thermodynamic stability of nucleic acid duplexes. We demonstrate how fluorescence resonance energy transfer (FRET) can be used to detect directly free energy differences between an initially formed “reference” duplex (usually a Watson-Crick duplex) and a related “test” duplex containing a lesion/alteration of interest (e.g., a mismatch, a modified, a deleted, or a bulged base, etc.). In one application, one titrates into a solution containing a fluorescently labeled, FRET-active, reference duplex, an unlabeled, single-stranded nucleic acid (test strand), which may or may not compete successfully to form a new duplex. When a new duplex forms by strand displacement, it will not exhibit FRET. The resultant titration curve (normalized fluorescence intensity vs. logarithm of test strand concentration) yields a value for the difference in stability (free energy) between the newly formed, test strand-containing duplex and the initial reference duplex. The use of competitive equilibria in this assay allows the measurement of equilibrium association constants that far exceed the magnitudes accessible by conventional titrimetric techniques. Additionally, because of the sensitivity of fluorescence, the method requires several orders of magnitude less material than most other solution methods. We discuss the advantages of this method for detecting and characterizing any modification that alters duplex stability, including, but not limited to, mutagenic lesions. We underscore the wide range of accessible free energy values that can be defined by this method, the applicability of the method in probing for a myriad of nucleic acid variations, such as single nucleotide polymorphisms, and the potential of the method for high throughput screening.