Computer-based testing could close the gender gap in test scores, according to a UK study in the Journal of Computer Assisted Learning.
While girls tend to score higher on many paper tests than boys, boys and girls appear to be even in computer-based testing. The study compares girls’ and boys’ performances on conventional testing and on paper and computer versions of the LASS Secondary and LASS Junior modules.
“Gender differences favoring females that are evident in conventional literacy tests are not evident in the LASS Secondary and LASS Junior reading and spelling tests,” concludes British researcher J. Horne.
The report was based on three small-scale studies involving 242 junior high and high school students. Girls outperformed boys in reading and spelling in conventional and paper testing in these three studies, but the girls’ advantage disappeared with computer-based testing.
Horne, from the Department of Psychology at the University of Hull, Hull, UK, attributes the computer-based results to several factors. First, boys tend to have more computer experience than girls. “It has been suggested that females may perform more poorly than males on computerized tests due to computer anxiety, although it has also been argued that males under-perform on such tests as they practice less,” Horne reports. The researcher notes that if differences in computer interest exist, then this could bias the results of computerized assessments in favor of boys.
Second, Horne says, in group testing with pen-and-paper tests, boys may be trying to rush through the tests so that they finish before others. This competitiveness may not come into play with more individualized computer testing where boys do not feel they are in a race with others. Under less pressure to finish the tests quickly, boys may make fewer mistakes, the researcher reasons.
Advantages of adaptive testing
One of the principal advantages of computer-based testing, Horne writes, is that it is adaptive testing. “In an adaptive test, individuals are moved quickly to the test items that will most efficiently discriminate his or her capabilities making assessment shorter, more reliable, more efficient and often more acceptable to the person being tested,” the researcher writes.
More protracted paper testing may have a more negative impact on boys, who could be more motivated to do well on computer-based tests, Horne says. Previous research has found both that girls obtain lower scores on computerized assessments and that there are no gender differences on computerized tests.
The three small-scale studies were:
Study 1: LASS Secondary vs. conventional tests (71 students)
LASS Secondary is a computerized test with seven modules: Sentence reading, spelling, reasoning, visual spatial short-term memory, auditory sequential short-term memory, non-word reading (phonic decoding) and syllable segmentation. Girls outperformed boys in spelling the conventional paper tests, but there was no gender difference in the LASS Secondary computerized tests.
Study 2: LASS Secondary computerized vs. pen-and-paper LASS (126 students)
Students took both versions eight weeks apart with each school completing the tests in a different order. The modules were sentence reading, spelling, and reasoning. Girls performed better than boys on reading and spelling paper tests, but there were no gender differences in the computerized versions.
Study 3: LASS Junior computerized vs. pen-and-paper (45 students).
Students took both tests with a four-week gap in between with half the pupils taking the computerized test first and half the paper test first. There was no gender difference on the reading and spelling tests for the computerized version, but girls outperformed boys on the paper spelling test.
“Gender differences in computerized and conventional educational tests” by J. Horne. Journal of Computer Assisted Learning 23, 2007, 47-55.