Research in Action: The Productive Failure Approach to Learning Statistics at Millennia Institute

Research in Action: The Productive Failure Approach to Learning Statistics at Millennia Institute

Friday, 18 March 2016


Productive Failure (PF) was conceptualized by Associate Professor Manu Kapur, Former Head of Curriculum, Teaching and Learning Academic Group as well as Learning Sciences Lab at the National Institute of Education (NIE).

PF engages students in designing solutions to complex, novel problems prior to learning the concepts required to solve those problems. The PF research team has impacted more than 6,000 students and 100 teachers of primary to junior college (JC) levels from 13 Singapore schools.

Mr Jaron Pow, who was formerly the Head of Department (Math) at Millennia Institute (MI), worked closely with A/P Kapur and his team on PF in its earlier stages.

Discussions began in 2012, when a core team of five teachers including Jaron started implementing PF in the Math curriculum. The programme was piloted in 2013 with two Statistics units: Normalisation and Hypothesis Testing, and one pure Math unit of Mathematical Induction.

In 2014, the three topics were scaled up from a few classes to all classes across the level, and another one to two units were developed as well.

The challenge was to implement the PF approach in the classroom without taking up more time than if they had used conventional teaching approaches. But Mr Pow found that they were able to do it with no difficulty.

Mr Pow says the collaboration had given them “extensive exposure to consider how learning works, as opposed to just teaching the A-Level contents”.

“It was really a great project for us because, if I could say so, this is my most successful collaboration in terms of how much it affected the teachers’ competency, in all the work that I’ve done in my 3 to 4 years as Math HOD.”

According to him, Dr Kapur and his team came in with topic expertise in Math and also extensive knowledge of how learning happens. The research team also designed pre- and post-tests to measure students’ understanding, tabulated the data, and summarized the results for the teachers.

“If not for this collaboration I wouldn’t have seen the professionals do it on a first-hand basis,” says Mr Pow. ”So it served this instructional purpose for us in terms of how to evaluate pedagogy, and it also let us move forward with this new knowledge in the years to come.”

The students benefited from the research-school collaboration as well. “In PF, they (students) are forced to use their internal resources to figure out a problem, and when the teacher finally shares the solution, it will make their understanding better and therefore truly changing their learning”, says Mr Pow.

Recognizing the need for continuity and sustainability, Mr Pow describes how this knowledge was passed on from the core team to other teachers.

“We felt it was important to look further into the future. Developing a unit may be easy, but your teachers need to come on board along the way, it cannot remain within the core team of five people,” he says.

“So we rotated the people in the team. We also conducted a series of intensive training sessions for the teachers who were not in the team but had to conduct PF for their classes, just like Manu did for us when we first started.”

Mr Pow feels the positive impact that had resulted from MI’s collaboration with NIE on PF. “If I could have a decision on things, we would have a lot more of these collaborations,” he says.

“I really feel that it has made a sound difference in practice. These collaborations have so much room to change things in Singapore schools.”

Recognizing the alignment of PF’s principles to recommendations made in the current review of the A-Level Math curriculum, MOE has recently awarded a $450,000 grant to translate the PF design principles across key concepts in the A-Level Statistics curriculum.

This translation will be carried out by the research team and MOE’s Curriculum Planning and Development Division (CPDD)’s Mathematics Unit, so as to leverage on its existing processes and structures. Such a translational effort will help JC students to experience deeper learning of statistical concepts.