Effectiveness of Web-based versus live plant identification tests

Setup and administration of comprehensive live plant identification (ID) tests in horticulture classes is time-consuming and costly. The curricular goal of this study was to integrate Web-based plant ID self-tests and computer-graded tests into floriculture potted plant production classes to potentially replace live plant ID tests. This research was conducted during 2000 and 2001 with students enrolled in Hort 4051 at the University of Minnesota, St. Paul. All plant ID tests were mandatory, constituting 12% of the grade. In 2000, only Web-based ID tests were used, while both Web-based and live plant ID tests were used in 2001. Two separate self-tests were designed as study aids with 34 randomized photographs/test. Correct spelling was mandatory to receive full credit for genus, species, and family. Self-tests could be taken ten times each per student. Students then completed two for-credit (graded), unmonitored Web-based tests. Students completed a Website evaluation form at the end of the semester. The two live plant ID tests were conducted with the same materials and were monitored. Mean student scores for the Web-based ID tests in 2000 ranged from 73.5 to 99.5% with a class average of 91.9%; there were no significant differences among students' scores. Student Web-based ID test scores for 2001 had a similar range with a high class average of 93.8%. In contrast, the 2001 live plant ID tests had a wider score range of 21.7% to 100.0% and lower class average (72.2%). Web-based and live plant ID tests, students, and their interaction were all highly significant. Web site course evaluations demonstrated interesting trends in student perception of Web-based and live plant testing. The implications for future class use and potential modifications for continued Web-based instruction are presented.