Studies in statistics

Print Friendly, PDF & Email

Following last year’s publication of Turnitin Effectiveness: Plagiarism Prevention in U.S. High Schools, Turnitin has recently published a follow-up research study, Turnitin Effectiveness in U.S. Colleges and Universities.

Similar to the earlier study, the college and university study purports to demonstrate that, although schools and colleges which qualified for the study often experience an increase in the rate of “unoriginal content” in the first year of Turnitin use, most schools and colleges experience a decrease in the rate of unoriginal content in the second year of use, and, on average, all schools experience decreases in the third and subsequent years.

Unlike the earlier study, Turnitin does not use unfounded assumptions of increase in rates of use of unoriginal content in schools which do not use Turnitin in an attempt to demonstrate how effective use of Turnitin can be (see Imagine… (another flawed study)).

However, just as in the earlier study, Turnitin only considers papers with Overall Similarity Index (OSI) scores of 50% or more, those in which an extremely high proportion of submitted text matches content in the Turnitin databases.  By the fourth year of Turnitin use, claims the study, the number of papers submitted to 2- and 4- year schools and colleges with more than half  the content matching other sources is reduced, in some cases dramatically.

Whether the decrease is due to use of Turnitin is moot.  It might be. It might not be.

It might be that Turnitin is an effective deterrent, and students therefore write with this in mind, accurately and appropriately citing and referencing any use of other people’s words, works, and ideas.

It might be that schools which allow students to view their Turnitin reports and then to correct or edit their work enables students to reduce their OSI scores below 50%.  It would be nice to think that use of the reports helps these students reduce their OSIs to 0% – or rather, to a point at which the students’ own voices are better heard, and all remaining unoriginal content, appropriately cited and referenced, is used to support what the student has to say rather than being what the student says.

It might be that Turnitin has helped make students, and teachers and instructors, more aware of the issues and problems of (especially) copy-and-paste, that the “plagiarism plague” is not a plague after all, and more a discovery of a problem that has long been there, leading to better education at ever-earlier ages, in schools which do not use Turnitin as well as in schools which do.

It might be that instructors, more aware that meaningless assignments so often result in meaningless copying, are changing strategies and changing assessments, making their teaching more meaningful, making learning more meaningful, personal, and relevant, in schools which do not use Turnitin as well as in schools which do.

It might be a number of factors.

And, as suggested, it might not be Turnitin at all.

(To be continued)

1 thought on “Studies in statistics

  1. Pingback: Studies in Statistics 2 | Honesty, honestly…

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.