The original meaning of ‘pedagogue’ was apparently a slave who escorted Roman children to school. The term ‘pedagogy’ is almost universally equated with constructivist theories of learning – certainly whenever I hear the term, I think of Piaget, Vygotsky, et al. To say that in recent years teachers have been slaves to constructivist ideas about learning is an exaggeration … but not such a great one.
However, with Ofsted finally taking a more open-minded view on what great teaching looks like, it has become an exciting time to be a teacher. The promise is that professional judgement, informed by better quality evidence, will allow teachers to take control of practice in a way that has not existed in the time I have been a teacher. Why then, given that evidence for the ineffectiveness of discovery learning has been around since the 1950s, did it take so long for constructivist claims to be challenged within the profession?
Science is messy; Pseudoscience is certain
One reason is that genuine science rarely gives a clear, unambiguous answer to a difficult questions. It is easier for simplistic and pseudoscientific ideas about learning to gain ground, because they are easy to understand and offer certainty, whilst the science that challenges them is complex and equivocal. The answer to ‘what works’ is rarely straightforward: A 2011 article looking at the effectiveness of discovery learning vs explicit instruction during learning illustrates this.
The abstract concludes that discovery learning does not benefit learners, whereas feedback, worked examples, scaffolding, and elicited explanations are more effective. However, beyond the headline, where they start to tease out the detail, the findings are a lot less straightforward.
For example, the article suggests that feedback was shown to be highly effective; but the interpretation of this can be problematic. For many teachers ‘feedback’ still means ‘AfL’. Yet many features of AfL strategies (e.g. peer assessment) appear to fall within a definition of unguided learning. Later in the paper, the authors say:
Although the type of unassisted-discovery task moderated trends favoring explicit instruction, unassisted tasks, tasks requiring invention, and tasks involving collaboration with a naïve peer were all found to be equally detrimental to learning.
Which implies that a core feature of AfL, peer-assessment, may not be especially beneficial when students lack expertise in a subject (I’m glad to see this slowly leading to a debate about the value of AfL).
Another example of the equivocal nature of science, is that the effectiveness of explicit instruction appears sensitive to the subject matter being taught:
Among the variety of different domains in which more explicit instruction was found to benefit learners, verbal and social learning tasks seemed to favor explicit instruction most, followed by problem solving and science.
It also appears age is an important moderating factor:
Adolescents were found to benefit significantly more from explicit instruction than did adults.
The moderating effect of age across the two meta-analyses did not follow the expected pattern of results. First, the adolescent age group was shown to benefit least from unassisted-discovery conditions, as opposed to the children, as had been predicted.
It strikes me that the significant age effects found in this paper help explain why the ineffectiveness of discovery learning still appears counter-intuitive to so many secondary school teachers. When we reflect upon our own experience as learners, we likely recall more recent instances (i.e. as adults) than our experiences in childhood. As experts in a subject (compared to children) our domain knowledge is likely to be sufficient for us to learn effectively with minimal instruction or guidance – and it’s easy to adopt a ‘common-sense’ notion that the same would be true for children. In addition, discovery learning may have marginally more benefit (or at least less detriment) at primary school than at secondary school and some of the difficulty that education has got itself into may be caused by inappropriately trying to apply strategies from primary school to secondary (and vice versa).
More messiness: It also matters what type of of explicit instruction is used:
Analyses of the types of explicit instruction in the comparison conditions indicated that worked examples benefited learners more than direct teaching and also indicated that feedback and providing explanations are useful aids to learning. The finding that worked examples evidenced greater learning than did unassisted discovery is expected given the worked-example effect (Sweller et al., 2007). However, the finding that worked examples benefitted learners to a greater extent than did direct teaching was unexpected.
So feedback and explanations are useful, but worked examples appeared to benefit most. Perhaps schools be pushing ‘responsive marking’ a little less and encourage a wider range of strategies?
The conclusion of the paper appears to be that constructivist approaches should be ditched in favour of explicit instruction. However, as usual in science, such simplistic conclusions are confounded.
In the second meta-analysis they compare the use of ‘enhanced discovery’ techniques – a constructivist approach that uses feedback and scaffolding techniques as a guide, but doesn’t give explicit instruction. They conclude:
Thus, the construction of explanations or participation in guided discovery is better for learners than being provided with an explanation or explicitly taught how to succeed on a task, in support of constructivist claims.
So, was constructivism right all along – should we all switch to enhanced discovery learning? Again, that simple conclusion is confounded by the fact that these benefits appears to be domain and age sensitive. For example:
Although enhanced-discovery conditions led to better learning outcomes for all age groups, adults seemed to benefit from enhanced-discovery tasks more so than children. Interestingly, the adolescents tended to benefit least and the adults tended to benefit most.
When science presents conclusions this messy and complicated – is it any surprise that teachers prefer certainties and pseudoscience?
Another explanation for the enduring and unquestioned status of pedagogy, to the extent that the industry regulator insisted upon seeing it applied in every lesson, is the false-consensus effect.
The post-hoc analysis of the meta-analysis gives a clue as to how this false-consensus has been maintained for so long. The authors note that positive findings for discovery learning were typically reported in second-tier journals, whereas first-tier journals tended to report positive findings for explicit instruction (An example of journal ranking for education research can be found here). Peer-review isn’t perfect, but typically more reliable studies get published in higher quality journals.
Most teachers don’t read much further than newspaper articles on research in education. The lack of peer review and the (typically) poor understanding of science amongst journalists allows pseudoscientific ideas and relatively weak research supporting discovery learning to get disseminated. When teachers are exposed to these simplistic ideas – and the more complex counterarguments are never made – it can easily lead to the false view that there is a valid consensus about the best way to teach. The reason constructivism was not more widely challenged, was because teachers, leaders, consultants, media, Ofsted constructed a false-consensus around pedagogy.
Will blogging eventually create a similar false-consensus, I wonder? It’s easy to see how our natural inclination to read people who agree with us could easily create a false-consensus among education leaders over time. Perhaps there’s hope in that claims can be challenged more quickly and democratically online; there is no bottleneck in the distribution of opinion like there is for newspapers and books. On the other hand, sorting the wheat from the chaff of online sources means that reliance on ‘trusted sources’ will eventually begin to influence what teachers read.
Avoiding new orthodoxies
The complexities of scientific findings and the ease with which false-consensus effects arise, presents a continuing risk that as we free ourselves from one pedagogy we become ‘enslaved’ by another. Where the genuine complexity of learning is not reflected in the leadership and regulation of schools, it’s easy to see how a new orthodoxy might arise: In the name of ‘consistency’ we might insist teachers apply inappropriate strategies because we fail to take into account that learning is different in different domains. In the name of ‘professional development’ we persuade teachers to adopt methods that simply don’t work well for the age group that they teach. In the name of ‘accountability’ we create oversimplified models of learning; things that can be easily observed and easy to measure, but do not actually reflect the complexities of learning.
As a profession we should prize informed professional judgement over simple solutions and quick wins. My hope is that the profession begins to recognise how messy learning really is and that different age groups, different subjects, even different topics and demands within subjects may require very different sets of strategies – and that informed professional judgement is the only way to tackle that complexity. We don’t need a new pedagogy, only an appreciation for the cautious conclusions and honest uncertainty of science.