This study compared the test scores of AP Calculus students. Two methods were used to work the calculus problems: the traditional pencil and paper method and the graphing calculator method. Four researcher-constructed assessments on various calculus topics were administered over a six-week period to two sections of high school AP Calculus students. Each assessment consisted of two sections: a non-calculator section and a calculator-active section. A t-test for dependent samples when the means are related was performed on each of the four assessments at the 0.05 level of significance. The t-test showed that there was a statistically significant difference between the two sections of students' test scores on the first two assessments. Two sections of the students' test scores showed no statistically significant difference on the last two assessments. Contains 43 references. (Author/NB)