A scientific take on teaching improvement

Guest Post by Dr Kylie Murphy, Academic Development Group, College of Science Engineering and Health.

A group of students dissect a fish.

©2012 RMIT. All rights reserved.

As the College of Science, Engineering and Health’s new Inclusive Teaching and Assessment specialist, it’s perhaps fitting that I lean towards science-based teaching and learning advice; that is, advice based on experimentally derived empirical evidence. This post is not a lesson in scientific method and I am not arguing that scientifically validated knowledge is the only valuable type of knowledge. I would however like to highlight a few practices that the best available empirical research in education has shown, repeatedly and consistently, improves student achievement — including in mathematics and the sciences. Drawing heavily on Professor John Hattie’s (2009) meta-analytic work, I’m offering only the bare bones here. If your interest is piqued, you’ll need to pursue the fleshy details for yourself!

Not wanting to get bogged down in statistical jargon, it’s worth noting that to be deemed a large ‘effect size’, the improvement in student achievement needs to be greater than the effect of ‘typical teacher’ practice. Large effect sizes usually indicate that, compared to the norm, a greater proportion of students progressed and they progressed considerably.  Now, the bony synopsis…

Classroom climate

In the category of classroom environment, the largest positive ‘effect’ on student achievement comes from environments that are welcoming of students and their errors. This is presumably because emotionally safe climates promote effort and risk-taking.

Curriculum design

On the matter of curriculum design, the findings are very consistent and transcend subject boundaries. The most important attribute is ‘balance’ between learning objectives that focus on increasing ‘surface knowledge’ (e.g., memorisable definitions and facts) and those concerned with ‘deep understanding’ (e.g., where the focus is on finding relationships between concepts, inferring hypotheses, identifying patterns and themes, generating arguments, and exploring useful applications). Not surprisingly, focusing too much on one and too little on the other has a minimal effect on student learning.

Students in an RMIT lab

©2012 RMIT. All rights reserved.

Specific teacher practices

There are many practices known to increase student achievement. These include having high expectations (i.e., not prejudging students), building and maintaining positive teacher-student relationships, undertaking regular professional development, paying genuine attention to learning objectives, formulating clear success criteria, setting challenging tasks, providing clear explanations and examples, providing opportunities for guided practice followed by constructive feedback, and continually seeking feedback from students as to the effectiveness of one’s teaching.

The biggest contributing factor is…

The biggest contributors to greater achievement appear to be not what the teacher does but what they get their students to do. For example, in more effective mathematics classes students are encouraged to be active in exploring potential solutions for most of the time (with as much or as little teacher assistance as is necessary), to go beyond the successful solution to the problem to include the interpretation of the solution, and to be diligent in frequently checking the quality of their work.

Students measuring water quality

©2012 RMIT. All rights reserved.

You may be wondering what measures of achievement were used in the 800 meta-analyses (i.e., thousands of individual studies) on which the above generalisations are based. You may wonder how the ‘effect sizes’ were calculated. You may have any number of bones to pick! For the answers, and a few concessions, grab yourself a copy of Hattie’s (2009) ‘Visible Learning’ published by Routledge, and read Chapter 2.

In conclusion

What’s important to know is that a great many studies converge to support the above principles of effective teaching. While even more effective strategies may be yet to be scientifically studied, and more scientific research on effective teaching is sorely needed, it is noteworthy that teaching practices with the largest effect sizes in individual studies tend to be similarly effective across studies involving different subjects, student ages, and student demographics.

Whether we look to experimental evidence or other forms of evidence about what makes teaching effective for most students, universities must invest in the pursuit of this evidence and invest in their teachers. More than ever before, universities need to provide evidence-based training, resources, and support to enable academics to maximise learning and achievement for the increasingly diverse students entering our increasingly competitive higher education sector.


Hattie, J. (2009). Visible learning: A synthesis of over 800 meta-analyses relating to achievement. New York: Routledge.

More information about Hattie can be found at http://www.findanexpert.unimelb.edu.au/display/person428067 and our own Tom Coverdale’s terrific discussion paper on Inclusive Teaching, which draws on Hattie’s work, can be found at http://www.rmit.edu.au/browse;ID=s1ket45y2qrb

Share your thoughts about improving the quality of learning and teaching in our comments section!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s