Tag Archives: testing

Assessing without levels

We recently held the first in a series of voluntary curriculum conferences for mid-leaders to share their ideas about what might influence the design of our new post-levels curriculum. Ideas that were shared during our first meeting:

  • Designing a new English curriculum and post-levels assessment system from scratch (which you can read all about here)
  • An Ethic of Excellence (which you can read all about here)
  • Using cognitive science to inform curriculum design (which you can read all about here)
  • Assessing without levels

assessing without levels 1 The chance to break free from using National Curriculum levels for assessment offers us real opportunity:

  • The opportunity to provide our students with formative feedback that means something
  • The opportunity to create an “Ethic of Excellence” – where excellence is expected and everyone can improve and aim for excellence
  • The opportunity to develop a curriculum that instils a growth mindset – no glass ceilings or self-labelling by students, e.g. “I’m a level 5”
  • The opportunity to develop a curriculum that goes “beyond” the traditional programme of study, to provide real stretch and challenge
  • The opportunity to develop more meaningful reporting for students and parents

In February of this year, a number of schools submitted applications to the DfE for Assessment Innovation Funding to develop post-levels assessment systems that could be shared with other schools.  Durrington High School were one of the 9 schools that were successful, and whose ‘Growth and Thresholds’ model is very closely aligned with our own thinking.  Much of the initial ideas on ‘assessing without levels’ shared at our first curriculum conference were informed by this model.

 

Thresholds and progression

New thresholds based on mastery of core knowledge and skills will need to be determined.  Our English department’s recent work with David Didau saw the development of our initial threshold names.  It also saw the birth of the “Beyond” threshold, which we are keen to adopt in all subjects to ensure we ‘keep the bar high’ and extend our students beyond the traditional confines of their current Key Stage and into the next. Progression to the current and new GCSE thresholds could then look something like this: Slide4

 

Organising concepts and threshold knowledge & skills

Each subject area will need to determine their own ‘organising concepts’ or ‘big ideas’ as well as the key knowledge and skills for each of the six thresholds.  An example for Science is shown below: Slide6 Assessing only the key knowledge and skills that continue into Key Stage 4 and beyond, will allow our students to develop a deeper understanding of concepts.  It will also provide our teachers with an opportunity to give focused and meaningful formative feedback to students and parents on how to progress further.

 

Baseline determination

A range of data could be used to determine a baseline for each student, including, for example:

  • KS2 English + Maths test scores
  • KS2 Teacher Assessment and dialogue with feeder schools
  • MidYIS / CAT3 ability testing
  • FFT estimates
  • Internal tests on entry
  • Reading ages

baselines Professor Robert Coe’s blog has also been useful in developing our understanding of what makes a ‘good’ test.

 

Achievement focused tracking and reporting

Once a baseline has been established for each student, progress could then be measured relative to this.  For example, a student with a baseline of “Excellence” would be making Expected progress if assessed as achieving the “Excellence” threshold, Good progress if they were assessed as “Beyond” and Less than expected progress if assessed as “Confident” etc. Slide8 This model supports our aim that all students can move through the thresholds, aiming for the “Excellence” and “Beyond” thresholds, providing they have demonstrated the required mastery of key knowledge and skills at those thresholds. Tracking could also incorporate numerical values broadly aligned to the new GCSE thresholds. Using the previous example, a student with a baseline of “Excellence” would be allocated a score of 9.  If the same student was assessed as achieving at the “Confident” threshold (allocated a score of 8), their achievement would be -1 for example. Slide9

 

Postscript

There was wide support during the meeting for agreeing a set of common thresholds to be used by all subjects.  Ideas were then discussed in subsequent department meetings, prior to pooling them via subject leads using a padlet wall. thresholds consultation We then decided upon the following thresholds to be used in all subjects: agreed thresholds

 

Preview

We have since held the second of our curriculum conferences for mid-leaders.  In the next blog in this series, we will share our ideas on Designing a post-levels curriculum and assessment model from scratch, including more detail on:

  • why we are moving away from levels
  • suggestions on how to move from organising concepts to a lesson by lesson overview
  • curriculum planning tools that could be used to adopt a common planning framework
Advertisements

Using cognitive science to inform curriculum design

We recently held the first in a series of voluntary curriculum conferences for mid-leaders to share their ideas about what might influence the design of our new post-levels curriculum.

Ideas that were shared during our first meeting:

  • Designing a new English curriculum and post-levels assessment system from scratch (which you can read all about here)
  • An Ethic of Excellence (which you can read all about here)
  • Using cognitive science to inform curriculum design
  • Assessing without levels

Super Glue

Why do students struggle to retain information from one week to the next?  What can we do to help make things stick?

Head of Maths Neil Siday, shared his thoughts with us on how cognitive science might help us to achieve this by planning smarter.  Much of Neil’s thinking has been informed by reading Joe Kirby and David Fawcett’s brilliant blogs on cognitive science and memory, as well as the work we did recently with David Didau.

Getting the content and challenge right

Cognitive scientist Daniel Willingham states that your memory is a product of what you think about and not what you want to remember – in other words, if your students aren’t actually thinking and making meaning then it won’t be learnt.

Memories are created by the release of chemicals, like dopamine.  If we pitch the challenge just right, we create an emotional response that releases dopamine.  Too little challenge offers too little reward, too much challenge and students won’t engage emotionally.

Memory is also thought to be domain specific – which means that we need to fill it with meaningful subject content.  Making sure that tasks are designed to provide opportunities to actually think and solve problems are therefore key to retention.  Time should be spent building up structural knowledge with practice to achieve automaticity first, before extending to deeper learning.  Sharing worked examples, modelling, building time for students to think and asking questions that encourage students to think is important.

Willingham’s simplified model of the mind

Slide1

Working Memory

  • deals with ‘the here and now’
  • is used to process and filter what we teach, make meaning and form our understanding
  • has fixed, limited space, which is easily overloaded with distractions or irrelevant information, which leads to misunderstanding
  • is key to transferring information to our long-term memory

Long-Term Memory

  • provides background info to working memory to help make sense of info
  • is almost unlimited
  • is where retention occurs

When the working memory is dealing with new information it calls upon the long-term memory to help make sense of it.  This retrieval process in itself aides long-term retention.  The information needs to be worked on in the working memory for it to be retained.  It is therefore paramount that we plan tasks so that what “sticks” is what really matters.

Storage and retrieval

Memories have a storage strength and a retrieval strength.

Retrieval strength is basically how easy it is to recall information at a later date.  This decreases over time, which is why you struggle to recall some things that are “on the tip of your tongue”

Storage strength is basically how well information has been learned.  Deeper learning = greater storage strength.  With low storage strength, retrieval strength decreases quickly.

It is therefore desirable to have high storage and retrieval strength, although even when information is buried, it can quickly be re-mastered.

storage centre

Spacing and interleaving

Robert Bjork’s New Theory of Disuse describes how making learning easier increases retrieval strength.  However, without the deeper processing that encourages long-term retention, this retrieval strength quickly diminishes.  Contrary to our intuition, it is thought that forgetting is actually key to increasing our storage strength.

Hermann Ebbinghaus first introduced the world to his forgetting curve and the spacing effect back in 1885.

Slide15

The graph shows how the amount of newly acquired information we retain declines over time without any attempt to retain it. To increase retention over time Ebbinghaus thought that spaced repetition could help.  Spacing works on the idea that we learn better when information is spaced out in intervals over a longer time span rather than when information is repeated without intervals (massed presentation).  Each repetition is thought to increase the length of time before the next repetition is required.

Bjork also argues that spacing reduces the accessibility of information in memory and in doing so fosters additional learning of that information.  Building in opportunities to revisit information at the point of ‘almost forgetting’ for students is good, as it means they are more receptive to learning new information.

Spacing may well be one of the most effective ways to improve learning, but what do you do in between repetitions?  Bjork argues that an effective strategy might be to interleave our study, which requires learners to constantly ‘reload’ or retrieve information, allowing them to extract more general rules that aid transfer.

With careful curriculum design, interleaving multiple topics allows us to space them out, rather than blocking them together and gives us an opportunity to revisit and build on prior learning.

Slide2

Massed presentation

  • rapid improvement
  • performance
  • poor retention

Spaced presentation

  • sustained improvement
  • learning
  • improved retention

Activating prior knowledge

Knowing things makes it easier to learn new things.  When designing and mapping out our curriculum it is important we:

  • Build on prior knowledge as connections are built between the prior knowledge that is in long-term memory and new knowledge
  • Plan to return to, and draw on previous knowledge (build retrieval strength)
  • Make links / connections explicit

Connecting_the_Dots

Testing Vs re-study

Frequent testing is also thought to help us remember.  Testing does far more than assess knowledge or skills – in fact it provides opportunities for learning. The very act of retrieving information from memory makes it easier to recall in the future.

Practice testing has been shown to outperform re-study, where 4 blocks of study with practice tests outperformed 8 blocks of study without practice tests.  In this way, shorter, more frequent (e.g. once per week) testing would appear better than testing once per Half Term, as more retrieval from long-term memory occurs.  When mapping out a curriculum, building in plenty of opportunities for students to practice may be advantageous.

Pre-testing is also thought to aid long-term memory – even when students perform poorly on them.

Types of testing

We also need to think more about low stakes/high impact testing and other ways can we get students to demonstrate their understanding, apart from traditional test questions in traditional test conditions, e.g:

  • A quick pre-unit quiz – which has the potential to set up triggers and create a ‘cognitive buy in’ for students, who are more likely to want to know the answers
  • Multiple choice questions – which means more decisions and therefore more thinking as more potential incorrect options are opened up to them
  • Cumulative knowledge testing – e.g. questions from units 1 + 2 also appear in a unit 3 test

Key messages

  • Get the challenge right
  • Avoid overloading the working memory and focus on meaningful content
  • Activate prior knowledge, build concrete content and develop applied thinking
  • Build high storage and retrieval strength
  • Plan for spacing, interleaving and practice
  • Utilise low stakes, frequent, cumulative knowledge testing

Daniel T. Willingham’s book “Why Don’t Student’s Like School” is available from our Teaching and Learning library

Why don't students like school, Willingham