The effect of professional development on in-service STEM teachers’ self-efficacy: a meta-analysis of experimental studies

International Journal of STEM Education

Table 5 Meta-regression model for substantive PD features

Moderator	Regression component	β	SE	95% CI
Active learning	Cognitive apprenticeship-based	− 0.94	0.79	[− 2.39, 0.71]
Active learning	Mixed	− 0.44	0.67	[− 1.76, 0.87]
Collaborative participation	Participated	0.71	0.83	[− 0.92, 2.33]
Coherence	Summer institutes	− 2.36**	0.85	[− 4.02, − 0.69]
	Separate training in semester	− 2.17*	0.93	[− 4.00, − 0.35]
	Mixed	− 2.22**	0.78	[− 3.78, − 0.66]
PD contribution to sources of efficacy	Vicarious experiences	0.52	0.76	[− 1.35, 3.82]
	Social persuasion	− 0.38	0.21	[− 3.04, 1.86]
	Emotional arousal	0.11	0.56	[− 1.82, 2.47]
	Mixed	0.21	0.58	[− 0.92, 1.35]
Intercept		2.41*	1.02	[.42, 4.41]

*p < 0.05, **p < 0.01, and ***p < 0.001. Reference group of coherence: job-embedded PD providing timing; reference group of active learning in PD: Constructivism-oriented instructional design; reference group of Collaborative participation: no collaborative participation; and the reference group of PD contribution to sources of efficacy: mastery experience