| Since Bruner's (1961) call for research into discovery-based learning, controversy has surrounded the efficacy of such a constructivist approach to instruction (e.g., Tobias & Duffy, 2009). For decades, research has investigated to what extent discovery-based instruction enhances learning tasks or conversely, detracts from them. Research has included wide varieties of domains and discovery-based instructional approaches. Samples have included both children and adults and both novices and experts within their specific domains. It seems that what the field needs is a definition of discovery learning from a practical perspective because a review of the literature reveals that although there might be an implied sense of what discovery learning is, the methodologies employed vary greatly. Furthermore, the characteristics of effective discovery methodology(s) need to be examined with careful consideration of the domain involved, the age of the sample, the comparison condition, and the outcome assessments. Therefore, two meta-analyses were conducted using a sample of 164 studies: the first examined the effects of unassisted discovery learning versus explicit instruction and the second examined the effects of enhanced and/or assisted discovery versus other types of instruction (e.g., explicit, unassisted discovery, etc.). Random effects analyses of 580 comparisons revealed that outcomes were favorable for explicit instruction when compared to unassisted discovery under most conditions, d=-.38 (95% CI=-.44/-.31). In contrast, analyses of 360 comparisons revealed that outcomes were favorable for enhanced discovery when compared to other forms of instruction, d=.30 (95% CI=.23/.36). The findings suggest that unassisted discovery does not benefit learners, whereas feedback, worked examples, scaffolding, and elicited explanations do. |
