Today’s Guest post is written by international Formative Assessment expert Shirley Clarke. Clarke is located in London, England. It is Part 1 of 2 in a series of posts from Clarke.

Part 1: Setting up talk partners

“If you have to talk you have to think. I get to learn things from other people I didn’t know.” 5 year old child (East Lothian)

In my first blog I talked about one of the key aspects of formative assessment - effective questioning. Today I am outlining the whole issue of talk and talk partners. Part 2 of this blog will look at teaching in mixed ability - a logical pathway.

First let me set the scene again: what is formative assessment and where do the pieces fit? Formative assessment, through its unfortunate mislabeling, has sometimes been misunderstood to be continuous summative action or something a teacher does to students to get information. Formative assessment is, in fact, a number of elements which enable rather than measure progress, and result in students becoming assessment literate. As Cowie and Bell (1999) stated ‘Formative assessment is the process used by teachers and students to recognize and respond to student learning in order to enhance that learning during the learning.’

My framework for formative assessment processes:

1. Laying the foundations, in which the conditions for pupils to be active learners, constant reviewers and self-assessors are set up: a learning culture, involvement in the planning and talk partners with mixed ability.
2. Effective starts to lessons, in which questioning strategies and a range of excellent examples are used to establish prior knowledge, capture interest, co-construct success criteria and discuss excellence.
3. Developing the learning, in which feedback from student to teacher, student to student and teacher to student are key, establishing and helping students articulate their understanding so far and focusing on constant review and improvement.

Setting up fantastic talk has an effect size of .82 in Hattie’s Visible Learning database - with the hinge point of one year’s input for one year’s growth being 0.4. Talk partners, response partners, learning partners, snowballing are some of the terms which describe pupils discussing together, planning together, cooperatively improving each other’s learning together. The greatest resource students have in any classroom is not the teacher, but each other. Watch a student having something explained to them one on one by the teacher and the student is unlikely to speak, for fear of being judged, for fear of wasting the teachers’ time, for incurring a sigh and so on. Watch a student explaining to another and you are more likely to see interruptions. As Dylan Wiliam said:

‘We need to be activating learners as instructional resources for one another’ (2011).

The emergence of talk partners was a result of the studies which showed that not enough ‘wait time’ (Rowe, 1974) was given for students to answer questions. The ‘hands up’ culture also excluded many students from thinking and reinforced fixed mindsets. Many schools started the ‘turn and talk’ strategy as the supposed answer to the problem.

I first started to question this around 15 years ago when I had the chance to watch some video footage of a 5^th grade class, on the carpet, engaging in paired talk. While I was in the classroom the whole effect was impressive: students all seemingly engaged in focused dialogue for 30 seconds. When I watched the playback I saw what really happened: 4 students were not talking to anyone. The students on either side of them had looked both ways and chosen their preference. One rather shabby child was rocking unhappily. The camera then focused on a boy who we see turning to his right to see a girl, turning the other way to see another girl, then noticing that in front of him is a boy, who is engaged in discussion. Nevertheless, the boy taps the boy on the shoulder, who, interrupted, turns round as the teacher says ‘OK everyone, eyes this way.’

This footage had quite an impact on me and it is one of the reasons I urge teachers to video their lessons, work in pairs, do Lesson Study, anything that enables you to see what is really happening rather than what you think is happening. So, I set up 14 teams of 30 teachers to experiment with different ways of setting the pairs, such as friendships (a disaster), similar achievement levels, different achievement levels (with a growth mindset you never say ability levels as it implies permanence and as we now know intelligence can be expanded, achievement is more appropriate) and random pairings. They also experimented with different lengths of time to be with each partner. By far the most effective was random pairing, changing weekly (every 3 lessons for secondary students).

Talk partners allow students not only time to think after a question is asked, but a chance to articulate their thinking. A common consequence is that students, not surprisingly, write more after having more time to rehearse their thinking out loud. Although the social impact of talk partners is significant, the learning impact, although more subtle, is greater. If the partners are chosen randomly, students have a rich diet of different learning partners over the course of a year.

The random pairing has further benefits: apart from the much appreciated fairness, students are often paired with someone the teacher would not believe would be a best fit. This takes students out of their comfort zone and challenges them to work on their weaknesses. Two shy students, for instance, have to start to talk and two talkative students have to learn to listen. Students learn tolerance, negotiation and cooperation skills, all vital life skills. Students for whom English is a second language are superbly supported when placed in a three and students with speech and language difficulties can have their first breakthroughs as a result of talk partners:

‘One autistic boy has gone from barely being able to tolerate one person, to really enjoying being included in changing partners and meeting his social targets.’ (Special school teacher)

Practical strategies

Choosing the pairs
Random pairing is usually carried out, quickly, at the end or the beginning of the week. Devices used are named popsicle sticks in a pot or bag, photographs stuck on a wall display or a computer randomizer, like ''The Hat’ from www.harmonyhollow.com. To avoid the same students being paired too often, some people create two sliding scales which ensure that all combinations take place before the pairings start again. Students then sit in their pairs for the week, usually placed by the teacher.

As with everything in education, flexibility is necessary. Some teachers like to change partners more often, sometimes twos can snowball to makes 4s, and sometimes a three must be made because of absences. Students can still be drawn off for a guided reading, a writing session or a mini-lesson, while the remaining students pair up. Young students can have a ‘magic spot’ on the carpet with their partner.

Choosing who answers
Teachers usually have a number of ways of deciding who will answer a class question, whether by popsicle sticks in a pot, name generators on the computer or names on the backs of chairs to name but a few. This avoids the damaging ‘hands up ‘culture, where the same students tend to be first with their hands up, distracting and stopping others from thinking, and makes a more inclusive learning environment. I need to emphasise that choosing students randomly to answer without first giving them a short time to discuss with their talk partner is not productive and can reinforce a fixed mindset. It is the articulation of the thinking, and being privileged enough to hear someone else’s ideas that enable students to be able to confidently answer at random.

Talk partner success criteria
Creating class talk partner success criteria sets the scene for discussion about what makes a good talk partner. Once displayed, these can then be referred to during lessons. The most popular method for co-constructing these is to role play, with another adult, a hilarious talk partner discussion in which the teacher demonstrates how not to be a good talk partner. The subsequent discussion with the class draws out the ingredients or success criteria for being a good talk partner, which can then be used for frequent self and peer evaluation.

Training students to talk and listen is key, with strategies such as a) frequently asking students to tell you what their partner said and b) using a timer for As to speak and Bs to listen then reverse it. With young students using picture cards of children facing each other or the front, to cue them each time, is very effective in making sure the partners engage.

Extremely popular is some kind of ‘complement slip’ on which students write a positive comment to their partner, saying why they enjoyed learning with them, kept in each child’s special folder.

Last word from students

Connect with Shirley Clarke on Twitter.