Peer instruction transforms the medical science classroom

Improved academic performance, engagement and attendance come out of peer instruction, where students discuss difficult concepts with their peers and interact with the whole class.

Ever increasing numbers of students. Poor student attendance and engagement.

These two very different problems, encountered by medical science module leads in two institutions, were addressed by the same novel flipped classroom strategy. 

The joint study across two universities compared the impact of Peer Instruction on academic performance between beginner and advanced undergraduate biomedical/medical students.

It also evaluated the overall effects of Peer Instruction on:

• academic performance;
• motivation; and 
• engagement. 

The study found that Peer Instruction greatly improved the learning experience of all students.

The study was jointly led by: 

• Dr Nephtali Marina-Gonzalez, Connected Curriculum fellow in the Arena Centre and module lead of Cardiovascular and respiratory function in health and disease (CRF) in year one of the UCL Applied Medical Science (AMS) degree 
• Professor Manuel Hernandez-Guerra and Professor Enrique Quintero, module leads of Diseases of the Digestive System (DDS) in year four of the Bachelor of Medicine degree, Universidad de La Laguna (ULL), Tenerife, Spain.

Growing numbers of students force a change in teaching strategy: the scenario at UCL

At UCL, the Applied Medical Science (AMS) degree was established based entirely on a flipped learning approach: an online teaching system and face-to-face teaching sessions in small groups (five students per group) that allowed students to learn at their own pace and explore topics in greater depth, encouraging qualities such as creativity and problem-solving skills.

However, this model became unsustainable due to the ever-increasing number of students and insufficient numbers of teaching staff. 

Teachers were having to deliver the same tutorial session six or more times a day. This caused a lot of anxiety among staff as it was generally perceived as a time-consuming and inefficient method that would inevitably have an impact on teaching quality.

A large-scale flipped classroom solution was clearly needed – where students work through the content in advance, coming to class pre-prepared to ask questions.

Poor student engagement and poor attendance: the scenario at ULL

At ULL, the main challenge was poor student attendance and engagement. 

The medical bachelor degree at ULL is largely delivered using a traditional approach of passive learning in large-class lectures.  

The cohort size in year four is currently 148 students but lectures are normally attended by just 30 or so students. This is due to timetable incompatibilities and low interest in traditional teaching methods that fail to promote participation among students.

The result has been increasing levels of dissatisfaction in both students and staff.

The Peer Instruction approach

Peer Instruction is a powerful teaching approach created at Harvard University* to address students’ struggle to apply threshold concepts** and to boost the effectiveness of in-class questions and student discussion. 

Peer Instruction is able to support active learning in all sizes of classes (especially important for the ever-increasing number of students).

It works on the theory that students at similar cognitive levels can sometimes explain content better in their own words, where educators may experience the so-called “knowledge curse” - a tendency to underestimate how long it will take another person to learn something new or perform a task. 

Peer Instruction was implemented simultaneously in both institutions in the winter of 2017.

The peer instruction method has various components:

1. A flipped classroom component, where students are provided with online lectures (with a duration of ~45 min) with embedded quizzes to do in advance. 
2. A motivational strategy, where 10% of the final mark is offered as a reward when students complete the online quizzes.
3. A face-to-face Peer Instruction session, which is done in the scheduled lecture or seminar. 

The structure of the face-to-face Peer Instruction sessions 

The face-to-face sessions were designed based on the method described in Crouch & Mazur, 2001*.

1. The instructor starts with a brief recapitulation of the topic (five minutes).
2. The instructor poses a conceptual multiple-choice question focused on one key concept (so-called ConcepTest). Clinical case studies were used to construct ConcepTests according to different levels of Bloom’s taxonomy.
3. Students are given one minute to think about the question in silence and vote for one of the choices.
4. The instructor checks the answers. If more than 70% of students answered the correct answer in the first attempt, the instructor then provides a final explanation and moves on to the next question. If less than 70% of students chose the right answer, students are instructed to turn to their neighbour and discuss the question and their answers.
5. Students are asked to vote a second time.
6. The instructor facilitates a class-wide discussion where students share and justify their answers, the instructor provides an expert opinion and confirms why the right answers are right and the wrong answers are wrong.

This can take anywhere from two to ten, or more, minutes, depending on the question, the answers, and the richness of the discussion. 

Creating online materials for flipped learning 

In UCL, the Cardiovascular and respiratory function in health and disease (CRF) module used to be delivered using the flipped classroom approach. The online material was, therefore, already available by the time Peer Instruction was adopted. 

It took approximately one month to create the ConcepTests for the face-to-face sessions. 

At ULL, the Diseases of the Digestive System (DDS) module used to be delivered using a traditional approach of lectures in a big classroom. 

In the spring of 2017, the module was flipped and online material and ConcepTests for the face-to-face sessions were created from scratch. 

In UCL:

• online lectures were made using Articulate storyline 
• embedded activities allowing feedback were made available to students using Moodle 
• interactive classroom response systems employed for Peer Instruction sessions included TurningPoint App and clickers. 

In ULL:

• online lectures and embedded activities allowing feedback were made with EdPuzzle 
• interactive classroom response systems employed for Peer Instruction sessions used wooclap in ULL.

The impact of Peer Instruction

This practice is a first for both UCL and ULL. 

We compared the effectiveness of Peer Instruction in three rubrics: 

1. in-class performance;
2. final exam scores; and
3. student satisfaction. 

At UCL, we discovered that Peer Instruction improves in-class problem-solving skills .

The exam scores revealed that the Peer Instruction group achieved similar results to those in the previous cohort who attended traditional flipped classroom tutorials.

This suggests that Peer Instruction has similar academic benefits to those obtained with small group flipped classroom sessions, but with the added benefit of facilitating these results in large classes.

At ULL, the exam scores showed that students achieved a higher number of correct answered questions and averaged scores with the Peer Instruction module, as compared to the traditional lecture module.

Year four students in ULL expressed higher student satisfaction than year one students in UCL, suggesting that transitional students coming to university for the first time expect to be taught by professors, and might feel uncomfortable being asked to teach each other rather than have a figure of authority teach them, whilst more mature students are more prepared for this academic environment.

Sharing the approach and broadening its use

Our next steps are to transform the entire curricula of Applied Medical Science at UCL and of the medical degree at ULL to incorporate Peer Instruction as the official method to deliver flipped classroom teaching to large classes. 

At ULL, we conducted the first practical workshop for faculty members of the School of Medicine in March 2018 in which we shared our experience with Peer Instruction and provided training and support for those interested in implementing the method.

Top tips for introducing this approach in your own classroom

1. Provide students with a slide showing the correct answer and a brief diagram explaining clearly why it is the right answer.
2. Do not attempt more than three-four ConcepTests per one-hour session.
3. Do not overload students with extensive information and numerous sessions. Better performance and student satisfaction can be achieved with one or two sessions per week over a lengthy period.  
4. For the in-class ConcepTests, you should always provide one correct answer, three wrong answers and one option with “I don’t know the answer”.
5. If you can’t make your own online lectures, you might consider using publicly-available material such as osmosis.org (subscription fees may apply).

The benefits of Peer instruction

Dr Marina-Gonzalez says about Peer Instruction:

• It transforms a traditional teacher-led classroom into a more-engaging, student-led classroom. 
• It builds a classroom community in which the students become the teachers and are able to freely speak up in class with their peers. 
• It improves engagement in the classroom lessons and promotes an atmosphere of respect and responsibility which help students develop lifelong communication skills. 

Prof Enrique Quintero, module lead, Disease of the Digestive System, Universidad de La Laguna, says:

“Peer Instruction could be a revolution in medical education. Implementing this method throughout both, basic and clinical curricula will have a profound impact on the quality of the medical schools in our country.” 

Feedback from students

“Peer Instruction helps you to apply the content and understand the concepts rather than remembering just the facts.”
Nawamin Panyapiean, first year undergraduate, UCL

“We can get more transferrable skills from Peer Instruction, such as teaching other people, having the courage to talk in front of people and being wrong.”
Trixica Kapdee, first year undergraduate, UCL.

*Crouch CH & Mazur E. (2001) ‘Peer Instruction: Ten Years of Experience and Results’, American Journal of Physics, 69(9), pp970–977

**Meyer JHF & Ray L (2005) “Threshold concepts and troublesome knowledge (2): Epistemological considerations and a conceptual framework for teaching and learning” 49 (3), pp 373–388