|
Detecting and Determining Greyscales
Research integrity is the core focus area for the HEADT Centre. This includes three aspects:
- Plagiarism: copying of materials from other author’s works without an indication (such as quotation marks) and without a complete reference.
- Data falsification: creating data without actual research or manipulating data in order to achieve a particular conclusion.
- Image manipulation: creating or altering an image in order to achieve a particular conclusion.
One research goal is to develop metrics to help distinguish between
the various greyscale zones that detection tools reveal. The tables below
show a first attempt to develop a rating system based on the number of
contiguous copied words within a standard unit. These tables were updated and partly developed for the exhibition How Trustworthy?.
The following table shows an example for measuring data manipulation or fabrication analog to the plagiarism greyscale.
Another goal is to review detection tools and to supplement or replace the tools as circumstances permit.
This research strives to understand more clearly what constitutes appropriate scholarly behavior. That is important, since research integrity decisions today depend on human effort. Part of our research is to find out on what decision makers base their findings (e.g., guidelines or standards) and whether they consider grey zone issues. If so, which grey zones, and are they appropriate? The answer to these questions varies across different disciplines, but automating misconduct detection requires clear definitions.
Project Members
Prof. Michael Seadle (PhD)
Dr. Thorsten S. Beck
Melanie Rügenhagen (M.A.)
Wjatscheslaw Sterzer (M.A.)
Katharina Toeppe-Hudy