Formative Assessment: a summary
By Shirley Clarke
Over the years there has been a great deal of confusion about what formative assessment actually is, with some people, even countries, seeing it as any kind of assessment that helps the learning or something a teacher might ‘give’ to a child in order to get some formative information.
The true conceptual framework describes the teaching and learning process minute by minute as creating independent learners who can self-assess, make improvements and know how to learn. It is the central structure of John Hattie’s Visible Learning Program, not surprisingly, as all the features of formative assessment have significantly high effect sizes.
The key elements:
- A learning culture, where pupils have self-belief and know how to learn and teachers have high expectations and belief that all pupils can succeed
- Pupil involvement at the planning stage
- Pupils knowing learning objectives and co-constructing success criteria
- Discussion about what excellence looks like
- Effective questioning
- Talk partners and classroom discussion
- Effective self, peer and teacher feedback
The following pages outline these elements. These will be particularly helpful if you are new to formative assessment and want to get the most out of watching the clips.
My book ‘Outstanding Formative Assessment’ groups the elements in the following ways, for greater coherence:
- The Learning culture: a) developing a growth mindset, b) understanding learning, c) establishing talk/learning partners, d) involving pupils at the planning stage
- Lesson starts: prior knowledge engaging openers
- Learning objectives, co-constructed success criteria and knowing what excellence looks like
- Feedback: ongoing questioning, visualizer stops to model success and improvement, peer coaching, self and peer cooperative improvement, marking after the lesson
One of the ways you can filter the clips is by formative assessment strand, as above, explained in detail here.
- A learning culture
- a) The growth and fixed mindsets
I have drawn on the work of Carol Dweck who is an authority on self-esteem. She established a simple framework (2000) which has given us access to the years of research about how children feel about themselves and their learning. What matters the most, in terms of motivation, is whether we see ability as ‘fixed’ (an entity learner) or ‘growth’ (an incremental learner). In short, people with a ‘fixed’ mindset will only tackle tasks which they know, in advance, they will succeed at.
People with a ‘growth’ mindset not only willingly tackle difficult tasks, but thrive on them. I have given examples of the two mindsets below in terms of their characteristics and the repercussions. Our aim, of course, must be to develop a growth mindset for ourselves, for all adults involved in working with children, for parents and all our pupils.
The ‘fixed’ mindset
|Characteristics of a ‘fixed’ mindset||Repercussions|
|My intelligence is a fixed trait – I have a certain amount of it and that’s that.||I worry about how much intelligence I have and it makes me interested in looking and feeling as if I have enough. I must look clever and, at all costs, not look stupid.|
|I feel clever when things are easy, where I put in little effort and I outperform my peers.||Effort, difficulty, setbacks or higher performing peers call my intelligence into question, even if I have high confidence in my intelligence, so I feel stupid.|
|I need easy successes to feel clever.||Challenges are a threat to my self-esteem so I won’t engage with them.|
|I don’t want to have my inadequacies and errors revealed.||I will withdraw from valuable learning opportunities if I think this might happen.|
|Even if I’m doing well initially, I won’t be able to cope with a problem or obstacle.||I readily disengage from tasks when obstacles occur.|
The ‘growth’ mindset
|Characteristics of a ‘growth’ mindset||Repercussions|
|Intelligence is something I can increase through my own efforts.||I am keen to work hard and learn as much as I can.|
|I acknowledge that there are differences between people in how much they know and how quickly they master things.||I believe that everyone, with effort and guidance, can increase their intellectual abilities.|
|I love to learn something new.||I will readily sacrifice opportunities to look clever in favour of opportunities to learn something new.|
|I am excited by challenge.||Even if I have low confidence in my intelligence, I throw myself into difficult tasks – and stick with them. I set myself goals and make sure I have strategies to reach them.|
|I feel clever when….||I am fully engaged with a new task, exerting effort to master something, stretching my skills and putting my knowledge to good use (e.g. helping other pupils learn).|
People with a fixed mindset need to constantly prove their ability, showing that they are special or even superior, whereas people with a growth mindset believe that intelligence can be developed through learning, something which brain research has proved to be true.
Strategies for developing a growth mindset – for teachers, parents and all involved in education
Modelling a growth mindset
We need to model our own growth mindset and love of learning by emphasising processes of learning, the importance and excitement of meeting challenges, putting in effort and using strategies which help us learn. We need to teach children that intelligence can be developed.
We need to transform ‘difficulty’ into ‘new or deeper learning’ and avoid sympathy when children encounter failure or difficulty. We need to show enthusiasm about challenging tasks and welcome mistakes or misunderstanding as opportunities for feedback and further learning.
Teachers with a fixed mindset often give lower achievers less demanding work in order to preserve their self-esteem, making sure they succeed, telling them how clever they are and dooming them to fall further behind. This approach also ensures that they will only feel successful when they can do things easily.
With a growth mindset, you tell pupils the truth. If they don’t have skills or knowledge or they are underachieving, this is not a sign of something shameful, but a sign that they need to work harder or be helped to find new strategies. By giving children greater access to tasks (i.e. increasing the level of support within the task itself), for instance, children instantly have greater access to the success criteria used in formative assessment.
Praising effort and achievement rather than ability or personal attributes
Praising children’s intelligence harms their motivation and their performance. Children love to be praised for their intelligence and talent, but if this is the norm, the minute they encounter an obstacle their confidence drops. If success means they are clever, than failure can only mean they are not.
Examples of praise comments which focus on effort and achievement rather than ability and help create a ‘growth mindset’ culture are as follows:
Well done! You are learning to…..
Fantastic! If it makes you think it means you are learning.
Mistakes help us learn.
Remember –you don’t know yet!
Every time you work hard you are growing your brain a little more
- b) Understanding learning
Meta-cognition is the term used to describe learning about learning, or what learning consists of. The Sutton Trust research gives its use a potential gain of 8 months and ‘high impact for low cost’ as the verdict.
When tasks are more complex for a pupil, the quality of meta-cognitive skills rather than intellectual ability is the main determent of learning outcomes.
Over some years, teachers in the learning teams have found ways of helping children identify different learning ‘muscles’ or ‘powers’ so that the learning skills we are using can be discussed and developed in the same way that we would develop an academic skill.
The growth mindset gives children the appropriate attitude and self-belief, but meta-cognition gives them the tools to be able to talk about and understand their learning, giving them a shared language and understanding. It is not enough to talk to children about effort, for example, without making it clear what it means to put effort into a task.
I have taken three well known educationalists whose individual lists of learning dispositions together seem to make a comprehensive whole, and synthesised these into one ‘ideal’ list of learning powers. The sources are Guy Claxton’s learning dispositions, Art Costa’s ‘Habits of Mind’ and Chris Quigley’s ‘Secrets of Success’.
The final list is as follows, but could be added to as appropriate:
Get lost in the task
Do one thing at a time
Break things down
Plan and think it through
Draw diagrams, jot down thoughts or things which help you think
|Don’t give up||Work hard
Try new strategies
Ask for help
Take a brain break
|Be cooperative||Listen to others
Say when you don’t understand
Be kind when you disagree
Explain things to help others
|Be curious||Ask questions
Look for patterns and connections
Think of possible reasons
Ask ‘What if..?’
|Have a go||Have a growth mindset
Don’t worry if it goes wrong
Learn from mistakes
Be excited to try new things
|Use your imagination||Be creative
Let your imagination go
Think up new ideas and questions
|Keep improving||Keep reviewing your work
Identify your best bits
Improve one thing first
Try to be better than last time
Don’t compare your self to others, only yourself!
Take small steps
|Enjoy learning||Feel proud of all your achievements
Feel your neurons connecting!
Imagine your intelligence growing by the minute!
Use what you have learnt in real life
Know you can do it if you have input and you practise
Teachers have used these categories in different ways, but always aiming for a ‘split screen’ approach, where the focused learning power has equal status to the knowledge or skill learning objective of the lesson.
One of the most successful strategies has been to attach a ‘character’ to each of the 8 categories, write a story for the character in which the various elements are explored then use this story to introduce the particular learning power to the class. Teachers usually focus on one story a week, displaying the characters and their breakdown of skills, until all the categories are known.
Once the dispositions are known, teachers usually take one of the posters/characters/lists displayed to place on the whiteboard alongside the learning objectives of the lesson, day or week and that is referred to during the lesson. Some teachers ask children which learning power they think best fits the skill they are focusing on for that lesson.
A high school teacher from Kentucky described her approach:
I gave the students a copy of the Successful Learning sheet with the 8 learning powers. We discussed each and how they relate to learning. Students nominated characters to represent each, mainly Disney characters, which seem to fit every category!
We then had an election to choose the representative for each. Next, we chose one power to focus on for the remainder of the year. Because of our continuing work with talk partners, we chose to focus on being cooperative. Each time students work with partners, in groups or sharing of any kind, I reminded them of the success criteria for being cooperative.
I would hear them sometimes reminding each other when necessary. I noticed students made progress in the area of communicating with their partners. They also got better at explaining to one another, using vocabulary and diagrams etc.
Many students who did not normally speak were much more comfortable talking to a partner. Some were more confident in their explanations and ability.
Rita Messer, Washington County High School, Kentucky
- c) Talk/learning partners
The emergence of talk partners was a direct result of the studies which showed that not enough ‘wait time’ (Rowe, 1974) was given for children to answer questions and that ‘dialogic talk’ (Alexander, 2004) was a missing component in the classroom.
The ‘hands up’ culture was also excluding many children from thinking. Let alone answering. With random talk partners changing every week, teachers have also been able to move towards mixed ability learning, with children learning from each other, linking with 20 years of research which shows that grouping children has very little impact on their learning and causes damage to children’s self-esteem (Sutton Trust and Hattie, J. 2009).
To ensure quality talk in the classroom:
- Pupils need to have thinking time to answer a question but discussing with a talk partner during that time or using mini whiteboards makes the thinking time more productive.
- Talk partner discussions need to be very focused and not too long (e.g. 30 seconds to come up with one thing you can see in this writing/ 1 minute to think of a good simile for a cat/ 2 minutes to decide what has gone wrong in this calculation) to avoid pupils losing momentum and going off task..
- Teachers need to avoid asking for ‘hands up’ because the same few children are always first with their hands up, do most of the answering and most of the class opt out of listening and thinking as a consequence.
- Random talk partners is the most effective organisational device (techniques follow in the feedback section) which need to change either weekly or fortnightly. Pupils appreciate the fairness factor and get to appreciate the rich variety of social and learning experiences they encounter because of the frequent change.
- Strategies need to be put in place to ensure quality talk, such as sharing of the rationale and surveying opinion regularly about the impact of talk partners from pupils’ point of view; using ice breaking activities when partners change; generating success criteria for good talk and good listening; using these to discuss how well pupil talk is developing and finding ways for pupils to self and peer evaluate their paired talk.
- Teachers need to avoid asking too many closed recall questions and ask more worthwhile questions which will extend pupil understanding and begin lessons in a more productive way, also revealing misconceptions which can then be taken account of.
- Responses to children’s responses need to be sensitive and respectful to establish an ethos of confidence to give one’s opinion, whether that is right or wrong.
- d) Pupil involvement at the planning stage
The tradition has been that teachers plan lessons away from the children, sometimes asking them what they already know about the subject matter but paying lip service to their involvement. By involving children at the planning stage, their interest and motivation is increased and their achievement is greater because of this ownership.
The key elements necessary for quality involvement are:
- What they already know/can do
- Immersion in the subject matter
- Presenting them with the skills coverage
- What they would like to learn
Successful strategies for establishing prior knowledge include:
- Breaking down the topic into various headings and asking what they already know about each part (e.g. for a study of minibeasts what do they know about the lifecycle of a butterfly, the habitat of ants, what different minibeasts there are, what insects eat etc. etc.)
- Giving children resources to explore to see what they know (e.g. electrical components/magnets and metals) or vocabulary cards to discuss
- Showing a picture with a statement (e.g. a picture of a Stone Age house and a statement ‘This is a Victorian house’ Do you agree or disagree? Give reasons)
- Children’s discussions and responses should give a clear idea of prior knowledge and impact on not only content but also pitch.
Immersion is usually an afternoon of various input, often laid out in the hall, for children to get a broad idea of the content, giving them greater chance to think about what they would like to learn (e.g. for a topic on India: video, food tasting, saris, music etc. For a science topic they might be given materials to explore or a difficult task, which will lead to them seeing what skills they need to be taught (e.g. can you make a plant grow sideways?)
Presenting skills coverage is an important stage, because otherwise you can end up with lots of random ideas rather than links with the skills you want to cover. Once children have these, either in child speak or learning objective form, they can match their ideas, co-constructing activity ideas with the teacher.
- Lesson starts
An effective start to a lesson is a feature of good formative assessment: capturing interest and immersing children immediately in the subject matter, establishing how much they know at the beginning of each lesson. A range of questioning templates has emerged as teachers have experimented with this, including:
- The range of answers (What does a plant need to grow? Soil, light, chocolate, sand, water, coke. Discuss)
- The statement (Goldilocks was a burglar. Agree or disagree? Say why)
- Right and wrong (Why does this bulb light up and this one not?)
- Odd one out (Which of these shapes is the odd one out?)
We need to plan worthwhile questions which will deepen and further pupil understanding, rather than ask them to recall a simple fact.
- Learning objectives, co-constructed success criteria and knowing what excellence looks like
All learners need to know learning objectives in order to have a chance of succeeding. Two things seem to matter: learning objectives should be decontextualised (e.g. write an account rather than write an account of an underwater world) and authentic (what you really want them to learn).
Teachers often stick to the curriculum language when maybe their real learning intention for a lesson is slightly different. Once the learning objective is clear, success criteria and everything else follow much more easily. An unclear learning objective, for instance, might be ‘to learn how to make a cake’ if what is really intended is ‘to be able to write instructions’.
As learning objectives became the norm in the UK there appeared a myth that the first words uttered should be the words of the learning objective and it should always be written on the whiteboard before the lesson starts.
Although the learning objective might be appropriate at the beginning of the lesson (often in mathematics), its appearance before children’s interest is captured can kill their interest.
Success criteria are a breakdown of or ingredients of the learning objective. For closed learning objectives they are often chronological and are always compulsory (e.g. the steps in a mathematics algorithm or the ingredients needed for instructional writing).
For open learning objectives they can be compulsory elements, such as the aspects needed in a science conclusion or they might be things that you could include (e.g. the possible elements in a good characterisation). Using success criteria has had a major impact on both teaching and learning, but mainly in equipping pupils with the tools to be able to self and peer assess.
Broad key skills produce broad success criteria so it can often be necessary to take each of the success criteria in turn and make those the focus of a lesson or series of lessons. ‘Persuasive writing’ would be a good example of this, where each element is worthy of a number of lessons:
|To write a persuasive argument||Letter to local MP||
We could take ‘striking up empathy’ for instance, present pupils with two contrasting examples of persuasive letters, one which empathises well and one which doesn’t and get them to analyse the pieces in order to generate success criteria for empathy. We might find ‘flattery’, ‘mentioning something the recipient is personally connected with’ ‘appealing to his/her better nature’ and so on.
In order to have maximum impact, success criteria:
- Need to be known, in a basic form, by teachers first.
- Should be the same for all learners in a class – differentiation by access should be sought via the amount of support provided within the activity.
- Must be co-constructed with pupils, or they have little meaning and less impact on learning.
- Can be used across the curriculum, including social skills, thinking skills etc.
- Need to be constantly referred to by pupils and ticked off for closed skills.
- Do not guarantee quality for open skills (e.g. writing) so seeing what excellence looks like is key.
Pupil generation of success criteria
We have learnt that success criteria must be generated by pupils to have maximum impact. There are now several very high quality techniques for not only getting children to generate success criteria, but also to help them understand what excellence looks like for the learning objective in focus.
- Doing it wrong at the visualizer/document camera – they will want to correct you!
- Presenting something incomplete (e.g. an incomplete invitation)
- Presenting something incorrect (e.g. a mistake in a calculation) for children to discuss
- Presenting one excellent product and asking children to identify the features (in writing not secretarial features)
- Presenting one good and one not so good product for children to compare. Vital to show more than one excellent example to avoid children being constrained to one style
- Eavesdropping their discussions about what should be included in a …. and writing them up as you listen
Once success criteria have been generated by pupils, they can be made into cards, sheets in folders, posters on walls, stored on Interactive Whiteboards/in SMART notebooks etc. and used whenever that skill recurs. Pure, decontextualised learning objectives lead to generic success criteria, which can be used in any context, so this should be the aim when generating them, otherwise their shelf life is too short and pupils do not see the vital link between and within subjects when skills are transferred.
In writing, success criteria provide a basic framework, but should not detract from what good writing consists of. Looking beyond the wow words and similes is the route to children understanding what makes a good piece of writing, focusing on author’s intent and impact on the reader.
In mathematics, some teachers have found that getting children to create their own individual success criteria for a skill once they feel they know it has been extremely successful. Random examples can be discussed at the visualiser and children asked to improve their success criteria.
This has led to us seeing that the most important time for feedback in mathematics is at the individual pupil generation stage, where their misconceptions can be clearly visible and able to be discussed and improved.
By comparing two contrasting examples, and further excellent examples for close analysis, children get to see what a good one looks like, another essential component of formative assessment. Excellence is shared before the children work independently, enhancing their chances of success. Magpieing ideas and words is encouraged.
Questioning individual children throughout a lesson, making them articulate their understanding helps give feedback to the teacher about children’s understanding. Key questions include:
- What are you going to include?
- What do you mean by….
- Tell me more about that
- Give me one example
We know that the more immediate the feedback the better, so we aim for continual review during lessons, stopping lessons at intervals to take one random child’s learning and show it under the visualizer/document camera for all to see. Children look for the successes and where an improvement can be made, then use this modeling to help them make their own improvements in their work as they go along.
Instead of children swapping work, cooperative improvement is more effective, with two children focusing on one book between them, discussing together, with the author holding the pen and reading their work aloud and penning the discussed improvements.
All the elements of formative assessment come into play as the feedback happens. The learning culture of a growth mindset is vital for children to want to strive for improvement and look forward to their work being critiqued by the whole class. The success criteria are the key focus for discussing best bits and improvement needs. Talk partners discuss the visualiser work and their own work cooperatively, articulating their thoughts and learning from each other.