An earlier study of coaching in private schools showed that special verbal coaching did not affect the SAT-V score; special mathematical coaching did not affect the SAT-M score for students currently taking mathematics courses, but it did improve the SAT-M score by 29 points, on the average, for students not currently taking mathematics courses. The present study, conducted in public schools, showed that special mathematical coaching had an effect of 6 points in one comparison and 18 points in another. Special verbal coaching improved the SAT-V score by about 18 points on the average. The amount of these effects are significant but small compared to the standard error of measurement of the test. Analogies items proved to be most susceptible to coaching. A second part of the study investigated the effect of coaching on items identical to those found in the test. The students who were given the coaching were also given a practice test containing part new items and part items on which they had been coached. It was found that earlier experience with the items raised the scores on these items by 47 College Board points on the average for SAT-V and 15 points for SAT-M. This means that a cram school that was successful enough to coach its students on one-tenth of the items in the actual test could only give its students a 4.7 point advantage on SAT-V and a 1.5 point advantage on SAT-M over those who had never seen the items before.