When running a text comparison, there are a few things to consider:
- The use of a template file. If there is a chance that the students copied and pasted some legitimate content into their test files, such as the questions to be answered, you should include this content as a template file. This can remove a lot of noise from the actual test results.
- The scope. By default, each file is compared to all other files contained in the same task; this is appropriate to detect copying among the students in the same class. In addition, the files in the current task can also be compared to those to specified previous tasks; this would be appropriate to detect copying from a previous crop of students that were assigned a similar homework before.
- The threshold for the extent of overlap. By default, the minimum overall number of matching words is 50. This may be adjusted in keeping with the typical length of the test files.
The overall result of a comparison can be displayed in three different views: Grouped by Files, by Matches, or in a Tree view. The first two contain tables that can be re-sorted by any category by clicking on the head of any table column. The tree view is not sortable. It is also not particularly accurate; its purpose is only to quickly highlight some similarity relationships.
In any of the above three views, click on the name of any individual file to view the duplications it contains and the names of overlapping files. To view an overlapping file side by side, click on the number overlapping words between the current file and the second file in question. This side-by-side display is also accessible directly from the global Matches view.
Matching lines are highlighed in blue, whereas ones that overlap the template file are shown in gray. Click on the arrows preceding any match to scroll the corresponding text passage in the other file into view.