Some weeks back, I signed up for a new module in Coursera, Model Thinking, taught by Scott E. Page, Professor of Complex Systems from the University of Michigan. Let me begin by saying this isn't the course I enjoy the most because it is difficult. However, the reason I've stuck through it is because the models are so important and I knew that it would change my perception into a deeper and more insightful one. Scott has been wonderful teacher and explains the models very simply such that it's easy to grasp. But I enjoy this course about as much as climbing a mountain. It's a painful but very rewarding journey.
The next series of posts is for me to journal my thoughts about some key learning points I've gathered from the models introduced. Assessment in this module is composed of weekly quizzes, a mid-term and final examination; all of which are multiple choice questions that test for understanding and technical application of the models. The posts serves as a platform for me to internalise and to translate the models into education.
The Big Coefficient vs. The New Reality
This week's lecture was primarily about linear and categorical models. One of the big takeaways was learning how the relationship between two different variables could plotted on a graph and expressed as a mathematical formula. This is powerful as it is a tool to quantify how various factors affect an outcome.
Schools possess an immense amount of data of students and teachers but sadly, this data is not being analysed and evaluated to inform practice and policy. Making sense of data can do a lot to help remove redundancy and improve existing practices. It is provides an objective measure on the effectiveness of a particular policy or practice. Current measures are far too dependent on qualitative analysis.
Among the several techniques and methods introduced, R squared was among my favourite. It is a simple formula to see how a well particular model - categorisation - is at explaining the outcomes. In school, the variation in results are only examined along classes and teachers. However, I've always questioned the validity as it ignores a myriad of other factors that could play a more influential role in the student's performance. Review that depends on poor categorisation not only ignores the real problem but focuses energy and effort into tackling problems that are either insignificant or do not exist at all. Given that each school has an extensive profile of each student, it isn't too difficult to identify the significant factors that needs to be addressed.
It may seem like I'm advocating for data-based pedagogy but that would be inadequate and inaccurate. There is a place for good evidence and data to guide decisions on policy and practice in school. However, the my current stand is that schools use too little of it too poorly. We can afford to introduce more rigorous and statistically sound methods to make use of the data already available in each school.
Yet, we need to be aware of the limits of Science and Statistics. Such tools are primarily descriptive and can only explain for what is measured. To assume that the trend or relationship extends beyond the observable isn't a wise practice. Scott explains this simply as the tension between The Big Coefficient and The New Reality. I understand this to be the perennial tension between the Science and Art in teaching and learning.
We all need good Science to accurately describe our practice and its effects on students. However, it is also important for good Art to dream of new possibilities to explore in teaching and learning. It is these new possibilities that challenge our assumptions forcing us to reexamine 'truths' in teaching and learning. A good example is ICT enabled pedagogies.
Technology has and continues to change teaching and learning. It might be argued that technology is a product of Science which is a stand I beg to differ. Instead, the technology we see today is the product of Art - dreams and imagination - informed by Science and made possible through Technology.
Education has much to grow in terms of the Philosophy, Art and Science. Art will take a lifetime to master and Philosophy a generation to get it right. But for now, getting the Science right would make a great start.