ABSTRACT
In this article, the authors report results from two pilot studies (n = 15 teachers total) investigating the effectiveness of the Online Content-Focused Coaching (CFC) programme for increasing classroom text discussion quality. Online CFC is comprised of a six-week online workshop followed by individual remote coaching sessions. Teachers in surveys and interviews responded positively to the different components of the programme, and reported that they saw significant improvements in their practice and quality of their students’ participation in discussions. Results based on analyses of teachers’ videoed class discussions likewise showed that teachers grew in their implementation of Questioning the Author techniques from baseline to the end of the workshop (ES = 2.03) and this effect was maintained through the coaching. Growth estimates for the rigour and interactivity of student contributions to discussions also demonstrated significant increases during the workshop (ES = .62), as well as across the coaching phase of the intervention (ES = .85).
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. Of the eight teachers who participated in the full intervention, one teacher was on leave during the pre-survey so the comparison is based on the seven teachers who do have data.
2. Given the CKT-R was the last part of the survey and two teachers did not finish the survey, our data yielded 13 teachers for this comparison.
3. It is important to note the relatively short duration for examining change during our pilot investigations – less than a half of a school year in study 1 and approximately one school year in study 2.
4. To calculate composite scores we simply calculated the mean of the dimension scores listed under each heading in Table 2.
5. For this analysis we used all of the videos from the 15 teachers across both studies. We examined two different types of growth models – we began by examining two-level univariate growth models, but we also examined a multivariate three-level growth model. Given the fact that both produced essentially the same growth coefficients and that our purpose here is primarily to describe the observed growth parameters for all teachers across studies, we present the simpler two-level growth models. In our analyses we centred time so that the intercept was the estimated IQA dimension score for teachers at the beginning of the study.
6. While the unconditional model presented here includes all three variance components, we decided to remove the random effect of r2i for implementation of Questioning the Author because there was no variation between teachers in their growth over the coaching intervals.
7. For effect sizes we report Cohen’s d for within-groups designs (dz) as discussed in Lakens (Citation2013). For example, dz for implementation of Questioning the Author was calculated using the mean difference from the HLM estimates (3.94 – 2.32 = 1.62) divided by the standard deviation of the difference scores (.80), resulting in an ES = 2.03.
Additional information
Funding
Notes on contributors
Lindsay Clare Matsumura
Lindsay Clare Matsumura is a Professor of Learning Sciences and Policy at the University of Pittsburgh’s School of Education, and a Research Scientist at the Learning Research and Development Center.
Richard Correnti
Richard Correnti is an Associate Professor of Learning Sciences and Policy at the University of Pittsburgh’s School of Education, and a Research Scientist at the Learning Research and Development Center.
Marguerite Walsh
Marguerite Walsh is a doctoral student in the Learning Sciences and Policy Program at the University of Pittsburgh’s School of Education.
Donna DiPrima Bickel
Donna DiPrima Bickel is a Fellow of the Institute for Learning (IFL) at the University of Pittsburgh’s Learning Research and Development Center.
Dena Zook-Howell
Dena Zook-Howell is a Fellow of the Institute for Learning (IFL) at the University of Pittsburgh’s Learning Research and Development Center.