Computer Based Testing: The way of the future?

October 8, 2009 at 3:51 pm | Posted in uncategorized | 25 Comments

(authored by Brian Slocum)

When is the last time you took a standardized test that was not computer based? For most of us, it was most likely when we took the SAT during our junior or senior year in high school. Tests required for post-graduate studies, such as the GRE (Graduate Requirements Examination), MCAT (Medical College Admission Test), GMAT (Graduate Management Admission Test) are all taken on a computer. Are standardized tests the only ones that that should be considered for computer based assessment? Can we find ways to integrate computer based testing into courses that are just normal parts of a high school or college curriculum? In some cases, we already have. We have seen a variety of tools such as Course Compass and MyMathLab which enable students to take online quizzes. Having increased access to computers nowadays allows us to use them in almost any environment. When considering computer based testing, the questions we must investigate are:
1) Why the shift from a paper and pencil based test to a computer based test?
2) Is there a significant difference between the results when someone takes a paper and pencil based test or a computer based test?
3) Are current paper and pencil based tests (such as the SAT) eventually going to be given on the computer?

There are numerous reasons for moving from a paper and pencil based test to a computer based test. One of the major benefits is that the software can automatically score the exam as it is being taken and then provide the students with their immediate results as opposed to a turn-around of several weeks. On standardized tests, this also alleviates the need to wait for the allotted time to pass before everyone can continue on to the next section. Once a student is completed with a section and feels that they have answered every question to the best of their ability, they are simply a click of the mouse away from continuing on to their next section.

Several studies have shown that there is little to no difference when comparing exam scores that test the same material, some of which are taken with a paper and pencil and others at a computer. In analyzing test results from an accounting course, Anakwe (2008) states

The study reveals that in three different accounting courses, there were no differences in student test scores between the online tests and the in-class tests. The study also revealed no correlation between a student’s gender or class and the student’s test performance.

Another study was conducted in which students actually took the SAT at a computer. They were given scrap paper and pencils but all answers were recorded one question at a time on the computer. Tools such as an on-screen calculator were provided. Interviewing the director of guidance for grades 6-12 in the district where this SAT was taken, an article from eSchool News (2001) states

Of the students who’ve taken this so far, I think 75 [percent] or 80 percent have said they prefer this to a paper and-pencil test.

This study was conducted in 2001, and 8 years later students are still taking the SAT at a desk with paper and pencil. Perhaps the College Board feels that the format of the test is fine in its current form and is in no rush to change, but it is something that must be considered as new generations grow up very familiar and comfortable with using a computer. Of course there are still disadvantages in this case. The SAT requires enough computers for the hundreds of students that take the test at one time at each location. This would require some heavy thought as to possibly restructuring the entire exam. Perhaps each test could be unique and offered more frequently so fewer students would take the exam at once, similar to the GRE. However, these should be viewed as modifications to be made that would improve the quality of the test, not disadvantages preventing us from adopting a new format. After all, the aforementioned study stated that the majority of students involved preferred the computer based test. If students are the ones taking the exam, shouldn’t we be listening to their thoughts on the matter?

References

Anakwe, B. (2008). Comparison of Student Performance in Paper-Based Versus Computer-Based Testing. Journal of Education for Business, 84(1), 13-17. Retrieved October 8, 2009, from ABI/INFORM Global. (Document ID: 1580346501).

Schools try giving SAT via computer. (2001). eSchool News. Retrieved October 7, 2009 from http://www.softwaresecure.com/pdf/SATsOnComputer_020501_.pdf