 |
Conclusion: Broadly speaking, unless a test’s goal is to measure mental arithmetic and related simple computational skills, including graphing calculators on tests does no harm. Further, including graphing calculators may increase the validity of the test and enable more accurate measurement of student skills in realistic situations.
SRI International
Three high-quality correlational studies are of particular interest:
|
| • | A study by Educational Testing Service and the College Board (Scheuneman et al., 2002) examined effects of calculator use on the SAT I math test. It concluded that “use of calculators was associated with higher test performance, but more able students were more likely to have calculators and used them more often. Overall, the effects of calculator use were found to be small, but detectable.” The effects were not influenced by gender or ability level, and were not due to simply working faster (a speeded test effect). |
| • | A study of calculator use on the ACT math test reported that average test scores increased for virtually all groups when calculator use was permitted. This effect was found for virtually all genders, ethnicities, income levels, high school grades and courses completed (Colton, 1997). |
| • | A study of the Tennessee Gateway assessment end-of-course test in Algebra I found that “students who responded that they used a graphing calculator performed higher than other groups” but there was no pervasive pattern across all types of calculators (Schwarz et al., 2002). There was no evidence of students failing to complete the tests (speediness) |
 Reference:“ Should graphing calculators be allowed on important tests?” Research Note 2
|
|
SRI International
A review of Ellington’s meta-analysis of 54 high-quality experimental and quasi-experimental studies concluded:
|
| • | There are positive effects both for calculator use on tests and in instruction. However, one cannot separate the effects of curricular innovations made possible by calculators from a technology effect as such. |
| • | There were no reliable negative effects in any of the groups of studies. |
| Reference: (Roschelle and Gallagher 2005)
|
