Research evidence on what makes CPD effective is accumulating. Yet, fundamental questions about the specific CPD features that lead to programme success remain. How teachers are supported to engage in the learning process, and more specifically the opportunities they have for active engagement, is believed to be a critical programme design feature (Desimore, 2009). Such a participant-centred approach to CPD (Patton et al., 2012) reflects a constructivist perspective on learning, which suggests that learning is neither linear nor straightforward and that it involves instead a process of knowledge construction, reconstruction and remaking (Dewey, 1938; Organisation for Economic Co-operation and Development – OECD, 2007). The implications for CPD are clear – CPD is maximised when teachers are active in the process of knowledge-creation.

Although contemporary theories of learning (i.e. social constructivism) offer the theoretical support for designing and delivering CPD with more opportunities for engaging actual practical activities, and research evidence suggests that teachers generally value such opportunities (Armour & Makopoulou, 2012), very little research has been conducted to assess these assumptions empirically. In other words, does an increase in active learning, or does an increase e opportunities for engaging in practical activities during structured CPD, enhance CPD outcomes? In short, is active CPD more beneficial CPD? 

To begin to answer these question, the study reported in this paper is based on an independent evaluation of a national CPD programme on inclusive Physical Education (PE) in England. The CPD was delivered to teachers and school staff in the form of a day-long, six hour course, on the principles and practices of inclusive PE. The two major objectives which underpin the data being presented in this paper are as follows:

(i)                   To capture elements of course tutors’ pedagogical practices in a systematic way (with a focus on their use of theory and practical activities as ways of facilitating professional learning);

(ii)                  To measure the effect of course attendance on participants in terms of self –efficacy change over time

Achieving these objectives, it was reasoned, would enable us to better understand the extent to which  the CPD course design and delivery features were linked to positive impacts. In other words, does variability in tutors’ practices (objective i) predict variability in participants’ perceptions aout programme outcomes (objective ii).

The context. Launched in 2013, the CPD Programme aimed to increase the competence and confidence of primary, secondary, and trainee teachers to deliver high quality Inclusive Physical Education (IPE). The ‘inclusion spectrum’ (Stevenson, 2009) provided the ‘theory of instruction’ (Wayne et al., 2008) for the programme. Research design and sampling. A case study design (Thomas and Myers, 2015) was adopted, whereby the case was identified at the level of individual CPD courses. To capture the anticipated variation in programme implementation, and given the ad hoc nature of course advertising and delivery across England, a cluster sampling procedure was used. Clusters were based on the nine geographic regions in England and CPD courses considered for inclusion in this study were the first two courses run in each cluster each year. Participants. Between October 2013 and September 2015, a total of 27 courses, delivered by 20 tutors across eight geographical regions were observed in their entirety (all six hours). From this course sample, and to measure the impact of the course on participants’ self-efficacy, a repeated measures design was adopted with data collected at three point in time: prior to (T1, n=427), immediately following completion (T2, n=379), and 4-6 months following CPD participation (T3, n=58). Data collection tools. To examine tutors’ practices in a systematic way (objective 1), both quantitative time-on-task data was recorded. We were particularly interested in measuring the proportion of course time that tutors dedicated to theoretical and practical components. The proportion of time that tutors dedicated to practical activities was therefore assigned as our primary independent variable. To examine participants’ perspectives on the nature and quality of CPD implementation, quantitative data was obtained via an ‘opportunities to learn’ inventory, distributed at the end of each course. To measure programme impact, a bespoke, ten-item, inclusive PE practice self-efficacy inventory was developed by drawing upon existing validated instruments (Block et al. 2013; Humphries et al. 2012). Participants’ perceptions about the quality of the course served as a secondary independent variable, whereas self-efficacy change served as our dependent variable. Data analysis. Data were analysed using hierarchical linear mixed modelling, which is the most appropriate strategy when individual-level data are nested within higher-order groups (i.e. when teachers are nested within CPD courses) (Tabachnick and Fidell, 2014). The final, best-fitting, model estimated the effect of practical learning opportunities during the CPD course on mean changes in participants’ self-efficacy over time (T3 minus T1).

Preliminary analysis. The intraclass correlation coefficient revealed that a substantial proportion (10%) of the variation in self-efficacy change over time was associated with course-level factors, thereby justifying the evaluation of practical opportunities within courses as a meaningful predictor of self-efficacy change. Main effects. The outcomes of the final model indicated that there was a substantial interaction effect between each of the predictors and whether or not the participant worked in a primary or secondary school. The final model revealed that primary and secondary school participants with below average self-efficacy entering the course benefitted more (0.67, ±0.28; effect size, ±95% confidence limits) (p < .0001). There was also statistically significant and substantial difference between primary and secondary school participants in terms of the effect of practical opportunities on self-efficacy change (1.12; ±0.99; difference in effect size between groups; ±95% confidence limits) (p = .03). In short, this means that opportunities for engaging in practical activities were more important predictors of self-efficacy change for participants from secondary than primary schools. There were also substantial differences for primary and secondary school participants in the effect of course size (0.88; ±0.92) (p = .06) and course quality perceptions (0.99; ±1.09) (p = .07) on self-efficacy change. However, these effects were not statistically significant and are therefore only possibly likely.

Armour, K., and K. Makopoulou 2012. “Great expectations: teacher learning in a national professional development programme.” Teaching and Teacher Education 28: 336-346. Desimore, L.M. 2009. “Improving impact studies of teachers’ professional development: Towards better conceptualisations and measures.” Educational researcher, 38: 181-199. Dewey, J. 1938. “Experience and Education.” Kappa Delta Pi. Hopkins, W. G., Marshall, S. W., Batterham, A. M., & Hanin, J. (2009). Progressive statistics for studies in sports medicine and exercise science. Medicine & Science in Sport & Exercise, 41, 3-12. Organisation for Economic Co-operation and Development (OECD). 2007. Understanding the brain: The Birth of a learning science. Paris: OECD publications Patton, K., M. Parker, and M.M.Neutzling. 2012. “Tennis shoes required”. Research Quarterly for Exercise and Sport, 83 (4): 522-532. Tabachnick, B.G., & Fidell, L.S. (2014). Using multivariate statistics. 6th Ed. Pearson.