Does Peer Instruction Work in High Schools?


Post by Ryan Campbell with an introduction by Julie Schell, June 2012 

Many Peer Instruction Network members wonder if PI works in high schools and how. For example, Andrew from Tomball, Texas wants to know – What does PI look like in high schools and member Janet from Falls Church, Virginia asks, how can PI  be well-integrated in high school? 

We have a number of experienced PI users in PI-Network – such as  Steve who teaches high school math and statistics in Phoenix and Larry who uses PI in high school economics in Dallas. So, the short answer is yes – PI does work in high schools – but the how part Andrew and Janet are inquiring about require a more expert level response. 

This week, we have Part 1 in a 2-Part series guest post from high school PI expert – Ryan Campbell, an active PI-Network member who uses PI in his history courses (you can contact Ryan using the PI-Network, just sign up or in and search for his name).

Ryan has some some great tips for using PI in general and in the specific context of secondary education — Let’s turn to Ryan to learn more.  –Julie Schell 

The Ten Non-Commandments for adapting PI to the high school setting: Part 1 in a 2-Part Series 

Post by Ryan Campbell 

I shall begin with a confession à la Mazur; I’m an international schoolteacher and have been for ten years. I currently work at the fabulous British International School in Jakarta, which as a British curriculum school (albeit one that offers the International Baccalaureate Diploma) means that I teach all year groups from Grade 6 to Grade 12 with class sizes ranging from 27 to 7.

I began to experiment with PI properly this year and have developed a set of 1o tips based on a year of trial and error (with the emphasis on error) followed by some commentary. They are in no way intended to be prescriptive, hence the title of my series.Here are tips 1-5.

Ryan is using PI in a high school in Jakarta

1st Non Commandment: The learning objective and the ConcepTest were meant to be together. Don’t separate them.

Peer Instruction hints at a better way of teaching in our high schools. A path towards evolutionary rather than constant utopian revolutionary change, of incremental pedagogical improvements over time based on data supplied by the students themselves rather than endlessly recycling one ‘new latest teaching approach’ after another.

But if you’re going to go down this path then you have to know what is the end point you’re aiming for. This is where a well-written (or at least clear) learning objective comes into its own.

Once you have a learning objective—you will know exactly how to a) choose your be pre-class readings b) target your ConcepTests or clicker questions.

2nd Non Commandment: ConcepTests work equally as well for skills as they do for content.

I discovered this fairly early on and it makes PI even more of an asset at the secondary level as explicit teaching of skills is normally expected. Here’s an example of a ConceptTest I put together this year for a group preparing for  history exams:

As you can see source analysis, as a skill, lends itself perfectly towards ConcepTests. If you don’t teach history (and there are some poor souls who don’t) you will find your students master essay writing skills, such as working with sources, much faster if you use PI. This applies equally well to topic sentences and paragraph construction along with most other academic skills.

3rd Non Commandment: Plan for adolescent stuff getting in the way.

Despite the success of PI at the tertiary level, secondary education is an altogether different beast.  Therefore, if PI is to have the impact it deserves in high schools then some adjustments should be made.  After all, the students are younger (with all that implies), class sizes are smaller, external exam boards frequently set curricula for high schools, and often lesson observations expect a certain style of instruction which will need to be accommodated.

The next thing to consider when adapting PI to the high school setting is target audience.  What works brilliantly in a Harvard lecture hall with an audience of a couple of hundred college students will need adapting to a much smaller, younger and perhaps less eager cohort.  This different target audience will have implications for both the typical pre-class and in-class PI workflows.

Starting with the pre-class readings and questions, it is important to choose the input material carefully. These materials do not necessarily have to be readings. In fact, at the secondary level, I would make sure that at least one out of three at-home tasks was not a reading.

YouTube clips work very well here as one example. It would also be wise to choose the readings carefully. I have been experimenting with in-text summaries[1] and I believe that this approach, while time intensive, does have benefits in terms of comprehension of material. I would also use a mix of incentives to ensure that the pre-class work is done.

My own approach is beginning to coalesce on a combination of short answer quizzes interposed with discussion style tasks on the forums on our virtual learning management system. Another key advantage to this approach is that the forum answers will give you excellent material to use as options for your ConcepTest answer options.

4th Non Commandment: Cut the input segments from 15 mins to 5-10 minutes.

The rules for the in-class lecture segments also change with a younger audience. Take the 15-20 minutes guideline and make it 5-10 minutes at the very outside. It is also important to use a variety of inputs here. Try experimenting within the PI structure with video, readings, teacher led discussion, and peer presentations, not just mini-lectures led by you.

Feel free to experiment here, on one condition, if you do find something that works particularly well, let me and other PI-Network members know.

5th Non Commandment: You have a mixed ability cohort. Plan your ConcepTest answer response option with this fact in mind.

The ConcepTests themselves are the next element you will have to adapt. At its core, PI is based on the idea that the instructor is too far removed from the students to understand/remember the conceptual mistakes the students make when first learning a new concept. This distance between instructor and student can be exacerbated the younger the target audience is (using PI with middle school students is different again).

Constructing the different answer responses for the ConcepTest questions was something I was having problems with until I read a quote from Mazur about this. His solution was to ask the students an open-ended question and then plug in their responses as question options. This works perfectly and leads to a slight adaption to the PI in-class workflow. I will discuss this more fully next week, in tip 6.

NEXT  5 Join me next week to read about the next 5  Non commandments for adapting PI in high schools – RC 



[1] Check out for more details on in-text summaries


Comments RSS

What do you think?

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

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

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: