This three part blog series looks back at how the Instructional Coaching Group developed and refined their model for instructional coaching through research spanning more than two decades. Missed part I or part II? You can read them here:
Part I
Part II

In 2007, Jake Cornett and I completed a more rigorous study of instructional coaching, in which 51 middle school teachers attended an after-school workshop on an inclusive planning and teaching strategy, The Unit Organizer (Lenz et al., 1993).

Observations showed that, compared to their colleagues who did not attend the workshop, teachers who received coaching were more likely to implement (87% vs. 33%) and taught with closer fidelity to the original model. Further, more coached than non-coached teachers told us that they continued to use the teaching routine (68% vs. 18%); and coached teachers also told us they would be more likely to use the teaching routine in the future (96% vs. 35%).


The Lean-Design Research Method

Coached by Barbara Bradley, my colleagues and I at the Kansas Coaching Project utilized design research (Bradley et al., 2013) to improve and refine our original instructional coaching model. These improvements, combined with the lean-research methods popularized by Eric Ries in The Lean Startup (2011), led to what I refer to as “lean-design research” in The Impact Cycle (Knight, 2017).


The refinements included:

  • coaches and teachers using video to get a clear picture of reality
  • a revised model for goal setting (PEERS goals)
  • identification of a bank of powerful coaching questions
  • identification of questioning skills
  • more effective ways to demonstrate teaching strategies
  • a streamlined, clearly defined coaching cycle: The Impact Cycle


Led by David Knight, my colleagues and I also conducted a multiple-baseline study to measure the impact of coaching on teaching and student engagement (Knight, 2018). Again, observation data showed that teachers used significantly more targeted teaching practices after they had been coached than before. Additionally, measures of student engagement showed significant gains in time-on-task behavior after coaching (d = 1.03).

Over the past two decades, we have conducted many other research projects, including a review of the research literature (Knight, 2008) and a study of the characteristics of effective coaches (Knight, 2010). Together, the studies described here and in various books related to instructional coaching have led to the articulation of a simple and powerful model for instructional coaching.

More recently, I’ve written about how those being coached can become better professional learners (Knight, 2018), about the role of autonomy within coaching (Knight, 2019), and the importance of putting student engagement (along with student achievement) at the heart of instructional coaching (Knight, 2019).

Our research continues today so we can continue to learn and improve how best to support all educators because above all, ICG remains dedicated to creating professional development for coaches, teachers, and leaders so students experience better learning, better lives.


Browse ICG’s 20+ years of research here >>


Bring an ICG consultant to your school or district >>


Register for an ICG workshop >>