Developing dialogue through guided reading

Gillian Russell, Coppice Primary School 

Project rationale 

As a teacher of year 6, at the start of the year I identified a group of six underachieving boys with high prior attainment (i.e. they were capable of achieving Greater Depth in the KS2 SATs Reading test, but were currently working at Expected level). 

My initial Guided Reading session with the group of six students showed that they struggled to verbalise their response to a text. However, they were showing considerable confidence in the Accelerated Reader online reading diagnostic test (Table 1). In addition, it was also clear that the students were not showing consistent growth when re-tested in spring year six (Table 2). I wanted to test my belief that being able to use PEEL (Point, Evidence, Explanation, Link) to answer questions about a text verbally in a small group setting, would significantly increase their ability to answer 3-mark written questions more effectively in the KS2 SATs Reading test. I also believed that adopting a discussion role in group sessions would provide students with a sense of purpose, and an understanding of what was expected of them during these sessions. The research question I investigated was: 

To what extent are discussion roles, implemented in guided reading for three months, an effective strategy for improving reading outcomes among underachieving boys in Year Six? 

Table 1: End of Year 5, Accelerated Reader, online Star Reading test outcomes 

Student  NRSS  PR  RA 
Student A  124  93  13:01 
Student B  106  70  10:10 
Student C  107  72  11:03 
Student D  110  78  10:01 
Student E  110  78  11:06 
Student F  141  99  14:04 

(NRSS: Normed Referenced Standardised Score; PR: Percentile Rank; RA: Reading Age)

Table 1 shows that students were well above average when tested on reading by an online programme. However, they were not showing the same level of skill when responding to written tests. 

Table 2: End of spring term Year 6, Accelerated Reader, online Star Reading test outcomes 

Student  NRSS  PR  RA 
Student A  120  90  13:04 
Student B  116  87  11:09 
Student C  104  66  11:04 
Student D  101  58  10:05 
Student E  107  72  11:07 
Student F  119  90  12:06 

(NRSS: Normed Referenced Standardised Score; PR: Percentile Rank; RA: Reading Age)

Table 2 shows that not all of the students were demonstrating consistent progress/growth when tested two school terms later. Therefore, their outcomes in KS2 SATs Reading test, were not secure. 

Intervention & intended impact

I hoped to show that if students were not allocated specific discussion roles, then they would feel no responsibility to contribute and play a part in the discussion. I kept my own participation to a minimum, so that students were able to demonstrate their current skills in group discussion. I also hoped to show that the real impact would come in their ability to use PEEL effectively to answer 3-mark questions on the KS2 Reading SATs paper which would allow them to achieve Greater Depth.

Over a three month period, I held a weekly Guided Reading session with the group. We read two texts, The Unforgotten Coat, and Millions, by Frank Cottrell Boyce. Students read sections from the text at home in advance, and considered prepared questions which were then discussed in the group. For the first four sessions I did not allocate discussion roles, but allowed the students to contribute in any way they wished. I observed that one student acted as Instigator (appointed by me); the remaining five students adopted roles as Builders, with some Challengers. No student acted as Prober, Clarifier or Summariser. 

Students did have prior knowledge of these roles from class work, but they were clearly not embedded. I tracked all sessions on a Harkness Discussion template. Initially, most of the sessions were dominated by four students; the other two students were largely silent.

At the start of the fifth session, I introduced the roles, and invited students to answer a questionnaire on which roles they would find easier to adopt, and those which they would not. All students said that they would find the role of Builder easier, and would try to avoid Prober, Challenger Clarifier and Summariser. Instigator was seen as a role which would be allocated by the teacher. 

For the remaining sessions, I allocated roles to students on a two session cycle. I specifically allocated roles which the students had said they would wish to avoid.

Research Method:

I used three methods to record students’ progress during the project:

  1. I filmed a Harkness Discussion where the students had prior knowledge of the question but no allocation of roles.
  2. Students completed a Discussion Roles questionnaire at the beginning and end of the project.
  3. I recorded sessions on Harkness Discussion templates
  4. I recorded student voice at the end of the project.

Impact data

The impact was immediate. Once students understood that they had a role, the balance of discussions was transformed. All students contributed with an increased sense of purpose; some students started to take notes so that they could challenge more effectively (they had experience of this from Debate lessons). There was also an increased use of PEEL, which meant that students were giving extended answers to a key question on a text, e.g. Regardless of what might await them, is it always right to send refugees back to their home country? The Unforgotten Coat, Frank Cottrell Boyce. Students were beginning to include evidence and explanation for their point, and began to make links back to their own, and other’s points. This in turn allowed opportunities for probing, clarifying and summarising. Students were more animated and the body language changed: students were leaning into the group and gesturing animatedly to reinforce their views. 

Their student voice in the last session was particularly interesting to hear. Some of their comments were as follows:

  • ‘I used PEEL naturally, without noticing it. I didn’t have to try and remember.’ (Student A)
  • ‘I used to just be a Builder, now I know I can be a Prober and ask better questions.’ (Student C)
  • ‘PEEL has helped me to make a point with evidence and now I can explain what I mean.’ (Student B)
  • ‘Roles helped me to talk because I felt I could speak and I didn’t want to keep it to myself.’ (Student F)
  • ‘I am not sure if it helped me. I am not that social, I tend to try being social.’ (Student D*)
  • ‘I try to talk in conversation. I have the feeling to not talk. I felt I wanted to talk when I had a role.’ (Student E)

* NB: Student D has a diagnosis for Aspersers. There was a noticeable difference to his participation when role cards were introduced. Without role cards, he was inclined to interrupt other students while they were speaking. He also dominated the group by shouting out responses before taking time to consider how to respond effectively. The change in the style of his participation following the introduction of the role cards was marked. He understood that his role was one of several, and that his participation was linked to the contributions of the other students in the group. This was a surprise to me as I had not anticipated this outcome from the project. This was a surprise to me as I had not anticipated this outcome from the project.

Research Ethics

I did not have any ethical concerns. I had explained what I was doing to the students, and the project did not cross any boundaries which were personal or private. I also checked with the Safeguarding Lead whether any of the students were likely to be affected by the issue of illegal immigration, an issue raised in The Unforgotten Coat; no students were likely to be affected.


I met a number of challenges. The availability of effective recording equipment was an issue: the sound didn’t work on my recording so I had to make notes retrospectively. One students had a poor attendance record. Occasionally a student had not read the text for the session in advance.

The full impact of the project cannot be fully measured until the results of KS2 SATs are published in July 2019

Next steps: 

I will share my project with colleagues and ensure that students are provided with the knowledge and skills to adopt different roles during group sessions throughout the curriculum. I intend to do this through increased modelling in Discussion Assemblies, CPD and in class support.


Reninger, K. and Wilkinson Ian .A. (2010) Using Discussion to Promote Striving Readers’ Higher Level Comprehension of Literacy Texts. International Reading Association

Soter A, Wilkinson, Ian A, Murphy P. Karen, Rudge, L. Reninger, K. Edwards, M. What the discourse tells us: Discussion and indicators of high-level comprehension. International Journal of Educational Research 47 (2008) 372–391

Soter A, Wilkinson, Ian A. Connors Sean P., Murphy, P Karen. and Fu-Yuan Shen, Vincent Deconstructing “Aesthetic Response” in Small-Group Discussions about Literature: A Possible Solution to the “Aesthetic Response” Dilemma. English Education, V42 N2, January 2010 

Harkness Model –

Discussion Roles –

Featured image: Pixabay

Leave a Reply