The following guidelines are designed to support teachers’ collaborative learning to enact Instructional Conversations for Equitable Participation (ICEPs).

We organize ICEP rubrics into domains, indicators, and behavioral markers (see Figure 1). Domains are the broadest level, and behavioral markers are the finest or most specific. ICEPs have four domains: (a) Contextualized Discourse, (b) Collaborative Activity, (c) Complex Ideas Using Everyday Language; and (d) Equitable Participation. Each domain consists of four to six indicators, organized into observation rubrics of teacher and student actions. Overview rubrics on pages 12 and 13 provide a snapshot of teacher and student behaviors associated with ICEPs, whereas domain-specific rubrics on pages 8 through 11 provide greater detail, nuance, and more precise terms to understand, observe, and improve equitable classroom talk.

The primary purpose of the ICEP rubrics is collaborative improvement. They are not designed for accountability or high-stakes hiring or promotion decisions. Observers should receive professional development before using ICEP rubrics to recommend instructional changes. We strongly suggest using the rubrics within the context of small, job-alike teacher learning teams (Ermeling & Graff-Ermeling, 2016; Gallimore et al., 2009) in which routines are already in place for teachers to talk with one another about their practice in generative rather than evaluative or overly-critical ways. These routines are essential for continuous improvement.

Observing. Observed teachers should make decisions about when, why, and what to observe. See the PDAR protocol for guidance about making these decisions and an observation protocol to gather descriptive notes during the observation. Peer observations should focus on one rubric at a time, based on the shared instructional aim or goal guiding the observation. Teacher teams can decide in which order to use the rubrics. However, after going through all of the individual rubrics, teachers may use the overview ones that incorporate all of the domains.

The observation should be at least 15 minutes long enough to detect instructional aim in its entirety but not so long that observers are overwhelmed with data to make decisions about evidence levels. The observation can take place “live” or by watching your team member’s video recording.

During the observation, observers should focus on describing rather than interpreting what they see. They should write descriptive field notes of student and teacher behaviors of interest within indicator- or domain-specific fields. Often, these descriptive field notes are the most useful piece of information to observed teachers in terms of continuous improvement. These notes can be made on an Observation Sheet or in GoReact using the comment function as you watch the video.

Gauging evidence levels. The evidence levels (Little Evidence, Some Evidence, Consistent Evidence) should be considered formative rather than evaluative. It should be understood that:

• Evidence levels assess observed teaching rather than comprise generalizations about the teacher;
• Evidence levels on the right end of the scale (Consistent Evidence) are not always feasible or desirable; and
• Evidence levels often vary by lesson content and activities.

Observers should immediately assign evidence levels after observing and use the full 1-5 continuum. For example, observed segments that fall between “sometimes” and “consistently” should be scored a 4. In addition to frequency or consistency, observers should also pay attention to quality or intensity distinctions in the rubrics when scoring.