Developing formative assessment in the classroom; using action research to explore and modify theory

How can classroom assessment be used to support children's learning?

Most pupils, and probably many adults, see assessment as simply another word for testing, designed to find out if they know something. This study, however, looks at the learning and developmental aspects of assessment. The authors argue the importance of formative assessment as a process rather than as an end in itself, a process which can foster and promote learning environments for children. The paper summarised here does contain some data about classroom activity but more extensive data are contained in an earlier piece of research. (Torrance and Pryor, 1998) and some of this has been included in the digest presented here.

Assessment or testing?

The authors suggest that the traditional view of assessment regards it as seeing if a child knows and understands a predetermined thing, which can usually be identified as a learning objective in a syllabus; or can do a particular task. ‘Wrong’ answers tend to be contrasted with ‘right’ ones instead of using mistakes to provide opportunities for teaching and learning. Assessment of this sort is teacher led and very much focused on a line by line approach to the curriculum. The authors claim in particular that it usually involves judgmental evaluation which runs the risk of making weaker pupils feel stigmatised. This type of assessment is often referred to as ‘summative assessment’ and may take the form of tests and examinations, or of continuous assessment conducted during the course of study.

The authors suggest that for pupils the connection between summative assessment and learning is not always clear. Assessment or, in their eyes testing, is often seen as something which is quite separate from learning. So ingrained is this attitude to testing that many pupils will actually expect a summative mark or grade for their work as a matter of course after every task they complete. While acknowledging that this form of assessment has a role to play, the authors see assessment in a much wider sense as an interactive and social component of teaching and learning.

For a summary of a review of all research on assessment and illustrative teacher case studies see ‘Raising Standards through Classroom assessment.’

How are children assessed in the classroom?

In their previous research (Torrance and Pryor, 1998) Click to where can I found out more? Page 10) the authors identified and characterised two approaches to formative assessment - convergent and divergent.

According to this model, convergent assessment includes:

closed or pseudo-open questioning and tasks;
a literal and linear view of the curriculum;
judgmental or quantitative evaluation;
finding out if the pupil knows or understands something set out in the curriculum;
something done to the pupil by the assessor;
a measure of past or current achievement; and
a behaviourist view of learning.

By contrast, divergent assessment emphasises what the learner knows, can do or understand. This form of assessment is characterised by several features including:

open-ended questioning and tasks;
a more flexible view of the curriculum;
descriptive rather than purely judgmental evaluation;
finding out what the pupil knows, understands or can do;
a focus on finding out how pupils think rather than on whether they are right or not;
an emphasis on shared activities of the learner and the teacher;
providing a platform for future development; and
a view of learning in which pupils work together to create or construct new ideas.

The authors suggest that whilst convergent assessment is summative and therefore judgmental, divergent assessment lends itself to formative assessment. In their view, divergent teacher assessment provides more opportunities for the teacher to teach in Vygotsky’s ‘zone of proximal development’. (Vygotsky, 1986 Click to Where can I find out more? Page 10) In Vygotsky’s work on educational theory, aspects of social psychology are integrated with aspects of individual and cognitive psychology. The model developed by Vygotsky is based on the idea that children can be assisted, by adults or more experienced peers, to learn in the conceptual zone just beyond what they can reach alone. The ‘distance’ between their original and new, potential level of development was termed the Zone of Proximal Development (ZPD) by Vygotsky.

For more information about the ZPD, click on http://www.gtce.org.uk/researchofthemonth (‘Really Raising Standards - Improving Learning through Cognitive Intervention’ and also Torrance and Pryor,(1998). Click to Where can I find out more? page 10.

How did teachers change and develop their practice?

In the interventionist, second phase of the project, teachers sought to change their practice, using the theoretical framework established earlier. The main aims of the intervention were to:

make task and quality criteria more explicit;
develop questioning techniques;
make observation more effective; and
make feedback a central part of learning.

The authors report that the teacher researchers felt that task criteria should relate to both the task and to the social rules of the class during the activity. Ensuring that pupils knew how to go about their task was seen as extremely important in helping pupils understand the purpose of the task. By making the task more explicit the teacher is better able to focus and structure the assessment activities such as feedback.

The teacher researchers exploited the use of questioning. In particular they developed their questioning techniques in three main areas:

finding out what a pupil knows, understands and can do;
clarifying what pupils have done already and what they have to do next; and
fostering thinking-about-thinking (metacognition) by asking pupils how and why they did what they did in the activity.

In their approach to changing their questioning techniques, the teacher researchers worked in a dynamic and interactive way with their pupils. They used their research journals extensively to record the details of classroom interactions as seen on video and then wrote in their reflections about what had happened. Following this reflection the teacher researchers then modified their technique in subsequent lessons. For example, one teacher researcher wrote after observing one such interaction:

...This clearly showed the problem of pupil and teacher having to find a common understanding of the language they use in a mathematics lesson. In the next stage of my research, I intervened in the process of children’s learning...I asked the children to talk through the mathematical problems...to think their thoughts out loud. I found this useful in that it enabled me to see the pupil’s misunderstandings and choose the best time to intervene effectively.

The teacher researchers were concerned to use feedback in a way which avoided communicating overall judgment of pupils’ work. Instead they tried to use it to help focus pupils’ attention on learning goals rather than on performance and to lay the steps for proceeding to the next task.

One teacher wrote in her journal:

Try to make marking comments relevant to the task criteria...If this feedback is given orally I wonder whether it is necessary for a marking comment....I have experimented with the children evaluating their own work against the success criteria we agreed at the outset of the task.

And another commented:

I think I do less marking or I place less importance on the marking as I feel I am involved with the children as they write. The ‘learning’ and ‘assessing’ bit is done during the interaction.

As the project progressed, the teacher researchers’ understanding of the complexities of pupil-teacher interaction developed. One teacher researcher commented:

I found it really interesting. It taught me a lot about the children in the class and my teaching...Until you’ve analysed your own data, you can’t really take it in, even though you’re reading about someone else doing something.

And another said, “it’s not until you’ve done the action research and sat and found out what’s going on that...It is clearer.”

Observation was considered to be the way to gain information about what children knew, understood and could do:

I gain most of my formative data observing and talking to children about their work.

Observation and questioning were closely linked and teachers sought to change the quality of their questions in order to gain more from observation.

How was the research designed?

This was a collaborative project which involved 2 university – based researchers and 7 teacher-researchers from primary schools in the county of Sussex.

When the researchers conducted research into their own classroom practices, they used a number of different methods to collect data about what they were doing. These included:audio and video recording of classroom activities;

diaries in which they recorded classroom interactions in detail and their own reflections on them; and
interviews with some of the other teacher researchers.

During this first phase, the teachers were supported by the university researchers who provided them with a theoretical framework for the description and analysis of their experiences. This analytical framework comprised a set of descriptors of the sorts of communications made by the teacher with their pupils. By analyzing video recordings and their own records of classroom interactions the teachers were able to see how far their communications had the intended effect on the pupils.

For example, one such descriptor (labeled G) was:

Teacher questions student about how and why specific action has been taken

the intention of which is to enable the teacher to:

gain an understanding of why/how the student has approached or achieved the task and to promote a deeper understanding

and the potential effect of this on the student was:

articulation of thinking-about-thinking (metacognition) and deepened understanding.

There were fourteen descriptors in all, covering a range of teacher intentions including observing, judging, clarifying, enhancing quality and developing insight.

In the second phase of the project the researchers began to put into operation the changes they had identified as desirable from the first phase. Having highlighted aspects of classroom assessment they wished to improve, they focused on those categories in the framework that might help them to do so.

Implications

In completing this digest its authors began to ask the following questions about implications for practitioners:

has formative assessment featured as a topic for INSET within the school's professional development plans and should it do so?
how can teaching and support staff share best practice in questioning techniques - would peer observation or video be helpful here? Would a focus on ‘open’ as opposed to ‘closed’ questions be a useful starting point?
would it be helpful for teachers to think about how much 'convergent' and 'divergent' assessment goes on in their classroom and whether more divergent approaches to assessment could be developed?
would parents value a session looking at formative assessment to help them understand the school’s approach and to help them support their children’s homework?
could formative assessment be used to complement summative testing in monitoring programmes?
would pupils benefit from greater clarity of communication when teachers are telling them what to do and what constitutes doing it well?

What did the researchers set out to do?

The research aimed to do two things. First, the study set out to investigate current assessment practices in classrooms. Teacher-researchers collected data about how they framed and managed classroom activities. They explored how they used ideas about ability and ability grouping, independence of children, task setting, the impact of rewards on learning, what they meant by quality, what goals they set for their lessons. During this phase of the project the researchers analysed and characterized their actions and responses in terms of the convergent-divergent model discussed earlier.

In the second stage of the study the researchers set out to improve those practices. They focused on a number of features of classroom interactions including:

developing more open-ended forms of questioning;
placing a greater emphasis on observation – identifying what activity to observe and why – as a foundation for effective feedback; and
clarifying criteria for both task and quality.

The need to establish learning goals and criteria for judgement and to communicate them effectively to students was conceived as a continuing dynamic and interactive process.

What did the teachers find out about their own classroom practices?

Teachers found that:

their teaching frequently closed down opportunities for exploring pupils’ understanding;
they did not make clear the purpose of classroom activities to pupils;
they gave little explanation of what represented quality in pupils’ work as distinct from task completion;
in many cases the teachers found little focus on learning goals in their communications, as opposed to behavioural goals and classroom management;
helping questions were often found to reveal more about what a child knows, can do or understand than a testing question; and
their feedback was often about presentation rather than content.

A particular concern of the teachers was their style of questioning. One teacher commented, “...I don’t think I was very good before at asking open-ended questions of children…I was sort of testing what they could do all the time.”

From this opening phase, two issues were highlighted for development in the second part of the project. They were:

the need to make the purposes of activities and what would count as doing them well more explicit; and
the need to respond more flexibly to students in the classroom, including developing different approaches to questioning, observing pupils and feedback.

Where can I find out more?

Adey, P. & Shayer, M. (1994). Really raising standards: cognitive intervention and academic achievement. London: Routledge. A summary of this study can be found on the General Teaching Council (GTC) web pages under the title- Improving learning through cognitive intervention.

Black, P. & Dylan, W. (1998). Inside the black box: raising standards through classroom assessment.. London: School of Education, King's College.
A summary of this study can be found on the General Teaching Council (GTC) web pages under the title - Raising standards through classroom assessment.

Torrance, H. & Pryor, J. (1998). Investigating formative assessment: teaching, learning and assessment in the classroom. Philadelphia, PA: Open university Press.

Vygotsky, L. (1986). Thought and language. Cambridge, MA: Harvard University Press.

	Developing formative assessment in the classroom; using action research to explore and modify theory.

	Torrance, H. and Pryor, J., University of Sussex British Educational Research Journal, Vol. 27, No. 5, 2001, pp.615-631
	How can classroom assessment be used to support children's learning? Most pupils, and probably many adults, see assessment as simply another word for testing, designed to find out if they know something. This study, however, looks at the learning and developmental aspects of assessment. The authors argue the importance of formative assessment as a process rather than as an end in itself, a process which can foster and promote learning environments for children. The paper summarised here does contain some data about classroom activity but more extensive data are contained in an earlier piece of research. (Torrance and Pryor, 1998) and some of this has been included in the digest presented here. Assessment or testing? The authors suggest that the traditional view of assessment regards it as seeing if a child knows and understands a predetermined thing, which can usually be identified as a learning objective in a syllabus; or can do a particular task. ‘Wrong’ answers tend to be contrasted with ‘right’ ones instead of using mistakes to provide opportunities for teaching and learning. Assessment of this sort is teacher led and very much focused on a line by line approach to the curriculum. The authors claim in particular that it usually involves judgmental evaluation which runs the risk of making weaker pupils feel stigmatised. This type of assessment is often referred to as ‘summative assessment’ and may take the form of tests and examinations, or of continuous assessment conducted during the course of study. The authors suggest that for pupils the connection between summative assessment and learning is not always clear. Assessment or, in their eyes testing, is often seen as something which is quite separate from learning. So ingrained is this attitude to testing that many pupils will actually expect a summative mark or grade for their work as a matter of course after every task they complete. While acknowledging that this form of assessment has a role to play, the authors see assessment in a much wider sense as an interactive and social component of teaching and learning. For a summary of a review of all research on assessment and illustrative teacher case studies see ‘Raising Standards through Classroom assessment.’ How are children assessed in the classroom? In their previous research (Torrance and Pryor, 1998) Click to where can I found out more? Page 10) the authors identified and characterised two approaches to formative assessment - convergent and divergent. According to this model, convergent assessment includes: closed or pseudo-open questioning and tasks; a literal and linear view of the curriculum; judgmental or quantitative evaluation; finding out if the pupil knows or understands something set out in the curriculum; something done to the pupil by the assessor; a measure of past or current achievement; and a behaviourist view of learning. By contrast, divergent assessment emphasises what the learner knows, can do or understand. This form of assessment is characterised by several features including: open-ended questioning and tasks; a more flexible view of the curriculum; descriptive rather than purely judgmental evaluation; finding out what the pupil knows, understands or can do; a focus on finding out how pupils think rather than on whether they are right or not; an emphasis on shared activities of the learner and the teacher; providing a platform for future development; and a view of learning in which pupils work together to create or construct new ideas. The authors suggest that whilst convergent assessment is summative and therefore judgmental, divergent assessment lends itself to formative assessment. In their view, divergent teacher assessment provides more opportunities for the teacher to teach in Vygotsky’s ‘zone of proximal development’. (Vygotsky, 1986 Click to Where can I find out more? Page 10) In Vygotsky’s work on educational theory, aspects of social psychology are integrated with aspects of individual and cognitive psychology. The model developed by Vygotsky is based on the idea that children can be assisted, by adults or more experienced peers, to learn in the conceptual zone just beyond what they can reach alone. The ‘distance’ between their original and new, potential level of development was termed the Zone of Proximal Development (ZPD) by Vygotsky. For more information about the ZPD, click on http://www.gtce.org.uk/researchofthemonth (‘Really Raising Standards - Improving Learning through Cognitive Intervention’ and also Torrance and Pryor,(1998). Click to Where can I find out more? page 10. How did teachers change and develop their practice? In the interventionist, second phase of the project, teachers sought to change their practice, using the theoretical framework established earlier. The main aims of the intervention were to: make task and quality criteria more explicit; develop questioning techniques; make observation more effective; and make feedback a central part of learning. The authors report that the teacher researchers felt that task criteria should relate to both the task and to the social rules of the class during the activity. Ensuring that pupils knew how to go about their task was seen as extremely important in helping pupils understand the purpose of the task. By making the task more explicit the teacher is better able to focus and structure the assessment activities such as feedback. The teacher researchers exploited the use of questioning. In particular they developed their questioning techniques in three main areas: finding out what a pupil knows, understands and can do; clarifying what pupils have done already and what they have to do next; and fostering thinking-about-thinking (metacognition) by asking pupils how and why they did what they did in the activity. In their approach to changing their questioning techniques, the teacher researchers worked in a dynamic and interactive way with their pupils. They used their research journals extensively to record the details of classroom interactions as seen on video and then wrote in their reflections about what had happened. Following this reflection the teacher researchers then modified their technique in subsequent lessons. For example, one teacher researcher wrote after observing one such interaction: ...This clearly showed the problem of pupil and teacher having to find a common understanding of the language they use in a mathematics lesson. In the next stage of my research, I intervened in the process of children’s learning...I asked the children to talk through the mathematical problems...to think their thoughts out loud. I found this useful in that it enabled me to see the pupil’s misunderstandings and choose the best time to intervene effectively. The teacher researchers were concerned to use feedback in a way which avoided communicating overall judgment of pupils’ work. Instead they tried to use it to help focus pupils’ attention on learning goals rather than on performance and to lay the steps for proceeding to the next task. One teacher wrote in her journal: Try to make marking comments relevant to the task criteria...If this feedback is given orally I wonder whether it is necessary for a marking comment....I have experimented with the children evaluating their own work against the success criteria we agreed at the outset of the task. And another commented: I think I do less marking or I place less importance on the marking as I feel I am involved with the children as they write. The ‘learning’ and ‘assessing’ bit is done during the interaction. As the project progressed, the teacher researchers’ understanding of the complexities of pupil-teacher interaction developed. One teacher researcher commented: I found it really interesting. It taught me a lot about the children in the class and my teaching...Until you’ve analysed your own data, you can’t really take it in, even though you’re reading about someone else doing something. And another said, “it’s not until you’ve done the action research and sat and found out what’s going on that...It is clearer.” Observation was considered to be the way to gain information about what children knew, understood and could do: I gain most of my formative data observing and talking to children about their work. Observation and questioning were closely linked and teachers sought to change the quality of their questions in order to gain more from observation. How was the research designed? This was a collaborative project which involved 2 university – based researchers and 7 teacher-researchers from primary schools in the county of Sussex. When the researchers conducted research into their own classroom practices, they used a number of different methods to collect data about what they were doing. These included:audio and video recording of classroom activities; diaries in which they recorded classroom interactions in detail and their own reflections on them; and interviews with some of the other teacher researchers. During this first phase, the teachers were supported by the university researchers who provided them with a theoretical framework for the description and analysis of their experiences. This analytical framework comprised a set of descriptors of the sorts of communications made by the teacher with their pupils. By analyzing video recordings and their own records of classroom interactions the teachers were able to see how far their communications had the intended effect on the pupils. For example, one such descriptor (labeled G) was: Teacher questions student about how and why specific action has been taken the intention of which is to enable the teacher to: gain an understanding of why/how the student has approached or achieved the task and to promote a deeper understanding and the potential effect of this on the student was: articulation of thinking-about-thinking (metacognition) and deepened understanding. There were fourteen descriptors in all, covering a range of teacher intentions including observing, judging, clarifying, enhancing quality and developing insight. In the second phase of the project the researchers began to put into operation the changes they had identified as desirable from the first phase. Having highlighted aspects of classroom assessment they wished to improve, they focused on those categories in the framework that might help them to do so. Implications In completing this digest its authors began to ask the following questions about implications for practitioners: has formative assessment featured as a topic for INSET within the school's professional development plans and should it do so? how can teaching and support staff share best practice in questioning techniques - would peer observation or video be helpful here? Would a focus on ‘open’ as opposed to ‘closed’ questions be a useful starting point? would it be helpful for teachers to think about how much 'convergent' and 'divergent' assessment goes on in their classroom and whether more divergent approaches to assessment could be developed? would parents value a session looking at formative assessment to help them understand the school’s approach and to help them support their children’s homework? could formative assessment be used to complement summative testing in monitoring programmes? would pupils benefit from greater clarity of communication when teachers are telling them what to do and what constitutes doing it well? What did the researchers set out to do? The research aimed to do two things. First, the study set out to investigate current assessment practices in classrooms. Teacher-researchers collected data about how they framed and managed classroom activities. They explored how they used ideas about ability and ability grouping, independence of children, task setting, the impact of rewards on learning, what they meant by quality, what goals they set for their lessons. During this phase of the project the researchers analysed and characterized their actions and responses in terms of the convergent-divergent model discussed earlier. In the second stage of the study the researchers set out to improve those practices. They focused on a number of features of classroom interactions including: developing more open-ended forms of questioning; placing a greater emphasis on observation – identifying what activity to observe and why – as a foundation for effective feedback; and clarifying criteria for both task and quality. The need to establish learning goals and criteria for judgement and to communicate them effectively to students was conceived as a continuing dynamic and interactive process. What did the teachers find out about their own classroom practices? Teachers found that: their teaching frequently closed down opportunities for exploring pupils’ understanding; they did not make clear the purpose of classroom activities to pupils; they gave little explanation of what represented quality in pupils’ work as distinct from task completion; in many cases the teachers found little focus on learning goals in their communications, as opposed to behavioural goals and classroom management; helping questions were often found to reveal more about what a child knows, can do or understand than a testing question; and their feedback was often about presentation rather than content. A particular concern of the teachers was their style of questioning. One teacher commented, “...I don’t think I was very good before at asking open-ended questions of children…I was sort of testing what they could do all the time.” From this opening phase, two issues were highlighted for development in the second part of the project. They were: the need to make the purposes of activities and what would count as doing them well more explicit; and the need to respond more flexibly to students in the classroom, including developing different approaches to questioning, observing pupils and feedback. Where can I find out more? Adey, P. & Shayer, M. (1994). Really raising standards: cognitive intervention and academic achievement. London: Routledge. A summary of this study can be found on the General Teaching Council (GTC) web pages under the title- Improving learning through cognitive intervention. Black, P. & Dylan, W. (1998). Inside the black box: raising standards through classroom assessment.. London: School of Education, King's College. A summary of this study can be found on the General Teaching Council (GTC) web pages under the title - Raising standards through classroom assessment. Torrance, H. & Pryor, J. (1998). Investigating formative assessment: teaching, learning and assessment in the classroom. Philadelphia, PA: Open university Press. Vygotsky, L. (1986). Thought and language. Cambridge, MA: Harvard University Press.