Bridgeman, B., Harvey, A., & Braswell, J. (1995). Effects of calculator use on scores on a test of mathematical reasoning . Journal of Educational Measurement , 32 (4), 323–340. https://doi.org/10.1111/j.1745-3984.1995.tb00470.x

Half of the sample took the test with the use of a calculator while the other half took the test without the use of a calculator.

The sample consisted of 11,457 college-bound high school juniors from a total of 257 high schools. In 19 of these schools, 40% or more of the student body consisted of African Americans.

Four measures were used:
-A 70-item test that included all item types proposed for the mathematics portion (SAT-M) of the new Scholastic Aptitude Test (SAT) (i.e., regular mathematics, quantitative mathematics, and student-produced response)
-A questionnaire to determine the extent of calculator use among the students, what types of calculator the students normally used, and whether they used calculators on school math tests
-Students also went back through the test to mark questions on which they had used a calculator and noted whether they thought the calculator had been very helpful, somewhat helpful, or not helpful for each of those questions. In the no-calculator, students indicated items that they thought would have been easier with a calculator, using the same 'helpfulness' categories as above.
-A background information questionnaire

The use of calculators resulted in a modest score increase, although effects on individual items ranged from positive through neutral to negative-Calculator effects were found on items at all difficulty levels, and calculators were beneficial for students at all ability levels. Prior experience in using calculators in testing situations appeared to be very beneficial.