Paul A. Kirschner & Mirjam Neelen
One of the first studies on the best approach to learning, was a meta-analysis by Benjamin Bloom – yes, indeed, the guy from the taxonomy (knowledge, application, evaluation, etc.) – and his famous article ‘The 2 Sigma Problem’ (1984). By far, the biggest effect, and thus the best learning result, was achieved through one-on-one instruction, followed by positive reinforcement of learning by the teacher. All 5 ‘top 5 variables’ having the most positive impact on learning as listed by Bloom, are directly related to the instructor (number 3 is corrective feedback, 4 is providing cues and explanations, and 5 is promoting active participation in the classroom).
Next, in 2008 the impressive book on an even more impressive study was published: Visible Learning by John Hattie.
The book is a collection of more than 15 years of Hattie’s research and it includes 816 meta-analyses of research on how children and adolescents learn best. In total, he analysed 62,169 studies with over 83 million participants.
Back in 2003, Hattie already presented a synthesis of over 500,000 studies on the effects of six types of influences (the learner, home, school, school-principal, peers, and teacher). He concluded that giving feedback to learners was the most effective (also see our blog here on peer feedback in the workplace, which discusses Hattie’s feedback model), even more than the cognitive skills of the learner! Again, almost all items (8) from the top-10 influencers go back to the teacher, such as quality of the instruction, direct instruction, remediation, creating a positive learning climate, and challenging the learner.
In his 2008 book, Hattie discusses and judges 138 influences which he then ranges on effect size. Some striking results: as number 1 we find previous grades, number 26 direct instruction, only as 45 comes parent engagement, as 79 regular assessment, 106 class size (bummer for many) and as 118 problem-based learning.
Unfortunately, starting after the publication of the book and coming to a peak in 2014, Hattie’s research has been challenged (see, for example this sequence of blogs and this blog) because his statistics were supposedly flawed. The debate is mainly around the use of effect sizes, for example that it’s unclear if he has used absolute or relative effect sizes. Another point raised is that Hattie has summarised the effect sizes of the 800 meta-analyses using unweighted averages. For example, Arnold explains that
Small and large meta-analyses have equal weight, while I would assume that the number of studies on which a meta-analysis is based indicates its validity and importance. Instead I would have opted for weighted averaging by number of studies, students or effect sizes. At a minimum, it would be interesting to see whether the results are robust to the choice of averaging (p. 220).
Hattie himself admits that there were some issues with the statistics. He says (in this interview) that unfortunately, at the very last minute, he put in the CLEs as part of the confidence levels rather than the actual common language exercise.
Three years later when some students in Norway alerted Hattie to the errors, the researcher says that he has corrected the errors in the reprints of Visible Learning. Hattie also says that in the twenty or thirty years since he started working on the Visible Learning stuff “No one – not a single person – has critiqued the idea and come up with an alternative explanation for the data.”
We must admit that we’re not familiar with the details of the problems with Hattie’s statistics (besides, we’re not experts in statistics whatsoever so the whole debate is hard for us to judge) but for now, we assume that Hattie has fixed the errors and that the original conclusions still mostly stand and it’s these conclusions that count!
His conclusions in the chapter ‘Bringing it all together’ are crystal clear. Teachers have, without a doubt, the strongest influence on learning effectiveness. Second, if teachers want to have a positive influence on their learners, they need to be directive, authoritative, caring, and actively and passionately engaged with teaching and learning. Third, they need deep and solid domain knowledge to be able to give meaningful feedback in response to what the learner thinks and (thinks (s)he) knows. This also goes, fourth, for the intentions and success criteria for their lessons, so that they can determine if their lessons really work. Basically, Hattie argues that it’s evident that ‘it is what teachers know, do and care about which is very powerful in the learning equation’.
So, what about the individual teacher? Hattie also analysed the difference between expert teachers and ‘regular’ experienced teachers. It shows that these two categories differ in 16 ways. Of these, 3 make the major difference when it comes to learning effectiveness.
- set challenging goals for learners and give them difficult tasks;
- have deep conceptual representations of the learning content, the teaching of that content, and how learners learn. As a result of this, their knowledge is better organised, they’re better at seeing and explaining connections between (new) content and learners’ prior knowledge, they’re able to connect learning content to other topics in the curriculum, and so forth; and
- are better at monitoring problems from learners and give the learners more relevant and useful feedback.
These three characteristics really separate the wheat from the chaff; the experienced teacher and the expert teacher. So, if you want to make sure that learners really have a chance to learn effectively, you now know what needs to be done. Depending on your role or your responsibility (that is: are you an administrator/principal or a teacher yourself?), you need to either invest in high quality teachers or BE a high quality teacher yourself by doing the right thing and keep developing yourself so that you’ll become the expert that Hattie describes.
Coe and co (pun intended) (2014) published a study with the title “What makes great teaching”, in which they discuss questions such as
- what makes ’great teaching’?
- what kinds of frameworks or tools could help us to capture it?
- how could this promote better learning?
And their answers which look quite like Hattie’s criteria for expert teachers are:
- Deep knowledge of the subjects they teach AND high quality instruction including reviewing previous learning, providing model responses, giving adequate time for practice and progressively introducing new learning (scaffolding)
- Assessing teacher quality through multiple measures including measuring student gains.
- High quality teacher feedback.A good place to start when YOU want to make a difference for learners!
Arnold, I., (2011). John Hattie: Visible learning: A synthesis of over 800 meta-analyses relating to achievement. International Review of Education, 57, 219-221. Retrieved from https://link.springer.com/content/pdf/10.1007%2Fs11159-011-9198-8.pdf
Bloom, B. (1984). The 2 sigma problem: The search for methods of group instruction as effective as one-to-one tutoring. Educational Researcher, 13(6), 4-16. http://www.comp.dit.ie/dgordon/Courses/ILT/ILT0004/TheTwoSigmaProblem.pdf
Coe, R., Aloisi, C., Higgins, S., & Major, L. E. (2014). What makes great teaching? Review of the underpinning research. Retrieved from http://dro.dur.ac.uk/13747/1/13747.pdf
Haesler, D., (2015). Chatting with John Hattie – Pt. 1 [blog]. http://danhaesler.com/2015/02/16/chatting-with-john-hattie-pt-1/
Hattie, J. (2003, October). Teachers make a difference: What is the research evidence? Paper presented at the Australian Council for Educational Research Annual Conference on Building Teacher Quality, Melbourne. http://www.acer.edu.au/documents/rc2003_hattie_teachersmakeadifference.pdf
Hattie, J. (2008) Visible learning. London, UK: Routledge.