IN ITS SIMPLEST SINGLE-ANALOGY FORM, DEDUCTIVE POLICY EVALUATION INVOLVES SAYING XI POLICY OR SITUATION (WHICH HAS BEEN ADOPTED) HAS BEEN FOUND TO CAUSE Y EFFECT; X2 POLICY (WHICH HAS NOT BEEN ADOPTED) IS LIKE X1 POLICY WITH REGARD TO ALL THAT WE KNOW TO BE RELEVANT TO THE OCCURRENCE OF Y EFFECT; AND THEREFORE WE PREDICT X2 POLICY WILL ALSO CAUSE Y EFFECT IF IT IS ADOPTED. IN A SLIGHTLY MORE SOPHISTICATED DOUBLE-ANAOLOGY FORM, THE DEDUCTIVE APPROACH INVOLVES SAYING X1 POLICY HAS BEEN FOUND TO CAUSE Y1 EFFECT; X2 POLICY HAS BEEN FOUND TO CAUSE Y2 EFFECT; X3 POLICY IS IN THE MIDDLE BETWEEN X1 AND X2 WITH REGARD TO ITS RELEVANT CHARACTERISTICS; AND THEREFORE WE PREDICT THAT X3 POLICY WILL CAUSE Y3 EFFECT WHICH IS IN THE MIDDLE BETWEEN Y1 AND Y2. (AUTHOR ABSTRACT)