Cohen, A. S., & Kim, S. (1992). Detecting calculator effects on item performance . Applied Measurement in Education , 5 (4), 303–320. https://doi.org/10.1207/s15324818ame0504_2

On 2 forms, items were classified into 4 groups: (a) computation only, (b) computation possible but answer is misleading, (c) algorithm needed to solve without computation, and (d) algorithm and computation needed. Students used calculators only on the second half of the form.

Participants included 1,490 students (765 on form 1 and 725 on form 2) enrolled in calculus and precalculus math courses.

The dependent measure was a 28-item test using operational items from the precalculus sections of a standardized university mathematics placement test.

Calculator effects were detected in 12 items: (a) On 2 items, computation problems were easier with calculators. (b) On 2 items, the function key on the calculator made them easier. (c) On 8 items, use of a calculator impeded performance due to inappropriate calculator use rather than lack of mathematical skill.