When it comes to the reliability of scholarly publication venues in the context of research support at universities, the phenomenon of predatory publishing is a major challenge for all stakeholders involved. There are tools now, primarily checklists, for assessing journals. However, practice shows that in most cases a clear classification ("good" or "bad") is not possible. A group from subproject 4 "Predatory Publishing" of the AT2OA² project is working on creating a new testing model. In this model, the reliability of a journal is located on a seven-point scale on the basis of certain formal quality criteria. There is a graphical illustration for the result.