In the ASCD Journal Educational Leadership, Jan, 2008 issue (Vol 65, No 4) is a section "What Research Says About..." by Jane David, director of the Bay Area Research Group, and it's topic is Classroom Walk Throughs. She makes it clear that walk throughs are "...not intended to evaluate individual teachers or principals or even to identify them by name in postobservation reports. rather, the goals of walk-throughs are to help administrators and teachers learn more about instruction and to identify what training and support teachers need."
With that purpose in mind, focusing on a single behavior and gathering objective data as opposed to the checklists typically used will help relieve the fears about bias of the observer. Gathering the same data in all classrooms (Class Learning Time, Wait Time, Level of Question, etc) can present a much more useful picture of the instruction in the school, and provide the basis for professional discussions.
Checklists, even when they are focused on observed behavior, are nearly always a record of the observer's impression of what is happening in the classroom. For example, walk through checklists often include an item on student engagement, and the choices are some form of 'poor, good, great' or 'none, few, some, most, all'. The item is checked based on the observer's impression or estimate of the level of engagement.
However, it's quite simple to use a data collection tool to do time sampled data collection of each student's level of engagement by doing sweeps of the classroom and recording the on/off task behavior of each student at that instance. In a classroom of 30 students, a sweep can easily be done in 90 seconds; the total time for the recommended 5 sweeps is less than 8 minutes. The results are an objective picture of the on/off task behavior of the entire class. When this is repeated for the entire school, the results can form a basis for decision making regarding goal setting and teacher support AND serve as the basis for determining actual change over time.
A series of 'impressions' is of much less value than a series of data based observations.
Be sure that the teachers are clear regarding the use of the data (not for evaluation, no names recorded), and that they are involved in the interpretation of the results. La Paz Middle School in Salinas, CA organized a data-collection day (it actually took a couple of days) where teachers gathered data in each other's classrooms using five laptops. Since the task is not one of judging or ranking behavior, but simply gathering data the potential for engaging the full staff will go far toward building a professional community engaged in a common endeavor.
Posts
