Omitted Variable Sensitivity Analysis with the Annotated Love Plot

The goal of this research is to make sensitivity analysis accessible not only to empirical researchers but also to the various stakeholders for whom educational evaluations are conducted. To do this it derives anchors for the omitted variable (OV)-program participation association intrinsically, using the Love plot to present a wide range of possible values in a compact yet intuitive manner. At the same time it calibrates anchor points for the OV-outcome association extrinsically, drawing on published descriptions of nationally representative samples. The overall approach requires a quasi-experimental comparison using matching or weighting for covariate adjustment. Study results report that what is gained in return is a vast widening of the audience for sensitivity analysis. Broadening participation increases the pool of information brought to bear on the inherently subjective assessments required to assess omitted variable sensitivity. One figure is appended.