Atteberry, A., Bryk, A., Walker, L., & Biancarosa, G. (2008, March). Variations in the amount of coaching in Literacy Collaborative schools. Paper presented at the annual conference of the American Educational Research Association, New York, New York.

Author: Atteberry, A., Bryk, A., Walker, L., & Biancarosa, G.



The purpose of this study was to use activity theory and hierarchical linear modeling (HLM) to examine differences in the amount of coaching that teachers received in 18 Literacy Collaborative schools over two years. The Literacy Collaborative builds upon 30 years of research and development grounded in the reading theories of Marie Clary and elaborated by Fountas and Pinnell. The overall goal is to improve the reading and writing achievement of all children in a school.

One-to-one coaching is seen as the most intensive component of the Literacy Collaborative model, and this study set out to examine how much one-to-one coaching was accomplished and what accounted for the variance. Using activity theory, factors that would seem to explain actions are hypothesized for the coach, the teacher, and the school context. This paper is interesting if only for its theorizing in each of these areas. Ideally, in Literacy Collaborative schools the suggested ratio is 1 coach to 12 teachers with a coach completing two coaching sessions per month with each teacher. The Literacy Collaborative holds that there are four important components to a coaching session: pre-brief, modeling, in classroom observation, and debrief. The research study examined the differences between this ideal and what actually happened in the eighteen schools.

The study illustrates well the links between hypotheses generated from theoretically perspectives (in this case activity theory) that are tested using well-developed measures that generate quantitative data and subsequently applying HLM. On average, teachers involved in the study received 3.12 one-on-one coaching sessions per eligible semester. Overall, the descriptive statistics suggest substantial variation among teachers in exposure to coaching. Interestingly, 70 percent of the teachers who received no coaching worked in the two largest schools in the study. The study found that the average coaching session took a total of 73 minutes with a standard deviation of 20 minutes and with only 4% of one-on-one coaching sessions containing all four components.

Using HLM, the study found, as predicted, that coaches tended to work more often with teachers who were more limited in their prior literacy training, but were committed to the school and had an active orientation toward their colleagues. Overall the pattern of school size effects suggests that coaches may be confronted with strategic allocation decisions in larger schools. Measures used in this study can be found at the project’s website.

Date Added: 11/14/2008


  • People
  • People » Administrators
  • People » Departments of Education
  • People » Literacy Coaches
  • People » Policy Makers
  • People » Reading Coaches
  • People » Reading Specialists
  • People » Teachers
  • Practices
  • Practices » Gr. PreK-5
  • Research
  • Research » Observation Instruments
  • Research » Potential Questions
  • Research » Research Design
  • Research » Research on Literacy Coaching