A new paper has been published in Educational Psychology Review, written by Lin Zhang, Paul Kirschner, William Cobern and John Sweller. It is worth reading in full and you can gain access to a pre-print via researchgate. However, here are my comments.
Firstly, the authors demonstrate the extraordinary popularity of inquiry learning with those who seek to reform school science curriculums. Zhang et al. allude to the fact that the recent review of the “Australian Curriculum: Science,” doubled-down on inquiry learning - every content strand in the draft began with either ‘explore’ or ‘investigate’.
Inquiry learning goes by many names and includes varying amounts of guidance. This has meant that when one version of it becomes discredited, a newly named version emerges with a perhaps slightly different emphasis. Advocates then effectively challenge critics to prove its ineffectiveness all over again. This is why I have lost patience with those who endlessly discuss definitions of inquiry learning. I no longer believe their motives are genuine. The burden of proof lies with those who advance a particular teaching method, not those who are sceptical of it.
Zhang et al. argue that two sets of U.S. science curriculum standards from 1996 and 2013, although using different terminology, involve a ‘pedagogically identical approach to science teaching,’ that emphasises, ‘teaching science through investigations’. Quite so.
The authors note that one type of evidence predominates in the sources of those who write these standards. This evidence is based upon entire teaching programmes. Typically, an organisation will fund an inquiry-inspired curriculum, incorporate a range of instructional approaches and spend time on intensive teacher development workshops. These programmes are then either evaluated on a before-versus-after basis or in comparison to a business-as-usual control group that is given none of the elements of the intervention. This means that we cannot be sure that any effect comes from the inquiry elements rather than, say, the teacher development workshops.
The authors describe two additional sources of evidence that, in contrast, do not support inquiry learning. The first is evidence from carefully controlled experimental trials in which only one factor is varied at a time, and the second is from large-scale correlational studies based on surveys such as those conducted by the OECD and the IEA.
Zhang et al. note that people often object to carefully controlled trials on the basis that they do not represent the complexity of everything that is included in a teaching programme. However, this is the point. It seems strange that it is necessary to explain to those who seek to teach science to schoolchildren, the necessity of varying only one factor at a time in order to establish a causal relationship.
Correlational evidence, in contrast, cannot establish causal relationships. It can only demonstrate that more inquiry learning is associated with poorer PISA scores, for example. It is always possible that a third factor causes both the teachers to use inquiry learning and the students to score poorly on PISA. However, when such evidence reaches the kind of volume described by the authors, and when read alongside evidence from randomised controlled trials that are capably of demonstrating causal relationships, it seems eccentric to ignore it and continue to refer only to programme-based studies.
The lack of scepticism around forms of inquiry learning and the readiness of anyone who thinks about education for five minutes to prescribe it as the solution to stagnating science performance seems, on the face of it, inexplicable.
In fact, what it demonstrates is the memetic power of bad ideas and just how much work there is to do in order to rebuild education on a rational foundation. Even when it is science and reason that we seek to teach.
In my occasional 'discovery learning' lessons, only one or two in the classroom discovered what was intended. The others just learned that they were not good at discovering.