A cognitive skills course for secondary school students called Problem Solving was implemented as an elective in the science curriculum for eleventh and twelfth grades at a rural, public, secondary school in Maryland during 1986-88. Problem Solving used the Whimbey and Lochhead (1982) think-aloud pair problem solving (TAPS) strategy to teach precise processing of information in verbal and mathematical problems. This investigation determined the effects the Problem Solving course had on college bound students' analytical problem solving skills, logical reasoning skills and SAT scores. Over a 3-year period the study compared: (a) the mean change scores on the Whimbey Analytical Skills Inventory (WASI) for 148 subjects in the treatment and control groups, (b) the mean change scores on the New Jersey Test of Reasoning Skills (NJTRS) for 80 subjects, and (c) the mean PSAT to SAT change scores for 234 subjects. A before-after, nonequivalent control group design was used to compare pre- to posttest change scores for students who had the Problem Solving class with those who did not have the course. Treatment and control group change scores were analyzed using ANOVA and ANCOVA statistical techniques. The Problem Solving course had a statistically significant impact on the analytical-problem-solving test change scores and logical-reasoning-test change scores (p < .01). An ANOVA of the treatment group's PSAT to SAT change scores showed a statistically significant mean SAT gain of 119 points; the comparison group had a mean SAT gain of 85 points. An ANCOVA which controlled for differences in race and sex, reading ability, and grade-point-average revealed that the Problem Solving course showed a marginally positive effect on verbal SAT scores and little effect on math SAT scores. Participants' affective reaction to the Problem Solving course was highly positive.