Student ratings are not precision tools that produce a measurement compared to a known standard. Some evaluators over-interpret slight differences as indicative of a problem, a decrease in quality, or an indication that one faculty member is materially better than someone else. In reality, a faculty member can teach different sections of the same course and receive different student ratings results.
Aggregated and viewed over time, student ratings provide just one piece of evidence documenting teaching effectiveness. See the Documenting Your Teaching pages for additional means of collecting information from the instructor, students, and peers.