The principal's favorite isn't the only bias that creeps into observations and teacher evaluations. My favorite is the teacher who brings cupcakes to the staff lounge will be rated as a better teacher. The actual research regarding bias identified the 'well liked' teacher, the person who brought cupcakes, remembered birthdays, sent nice notes, etc, as being perceived as a better teacher by observers.

These bias, often unintentional, is why, after 25 years in education, I created the Data-Based Observation Method and supporting software. eCOVE Software (not required to implement the method, but easier) is a set of tools that gather objective data on teacher and student behavior in the classroom. The tools include such things as Class Learning Time, Level of Question, Response to Misbehavior, Attention to Gender (or other groups), Teacher Talk/Student Talk, Student Question Type, Time on Task, and many more.

Of course, the observer can fake the data, but that's very rare. In most cases, the bias in hidden, even from the observer. By gathering objective data (which can be confirmed by another observer), the bias are greatly diminished.

The DB Observation Method suggests providing the data to the teacher and asking "Is this what you thought was happening in your classroom?" That starts the reflective process and empowers the teacher to make changes based on the data, and not the whims or bias of the administrator. There is more about the method on my blog Data-Based Classroom Observation, and the software can be downloaded on my site eCOVE Software.
John Tenny | Homepage | 06.26.08 - 12:59 pm | #