I have been teaching for a while now. It was clear early on that aside from the personalities of individual students each class has its own distinctive set of behavior patterns collectively. Most groups do although in all cases it takes experience and contemplation to discern them. With any group the temptation is to identify the group with the members of the group that are most vocal. They provide a good deal of data quickly about the group but it is by definition not representative . Ideally I get to work with a number of student outside of class and get to know them a bit even if they are quiet in class. That adds to the overall picture and helps it be better informed. The same process applies to larger groups, such as the subset of the faculty that attends our faculty meetings. If you don't talk to the non-speakers outside of the meetings you have a non-representative sample of their behavior if you just rely on what is said. And yet, it is not the speakers in faculty meetings...