thinning out

If a learner cannot regularly independently and accurately answer over 70% of their assignment questions then it means too much too hard work has been set to the learner and the timely practice can't do its job - embedding learning; thinning out is the process of fixing this.

Signs thinning out is needed

If over 20% of the assignment is regularly given the star code then "thinning out" is probably required.

If over 30% of the assignment is regularly given the ~ (oops) or star code then "thinning out" is probably required.

What it means

The phrase "thinning out" comes from gardening it means - removing some of the "too many" seedlings, to give the other seedlings a chance to grow well. The seedlings removed will ideally be the weaker looking ones, but sometimes they are just randomly chosen.

The gardening analogy works well as if a teacher gives a learner feedback on "too many" questions, the learner will get overwhelmed and won't remember (or worse muddle) the help the teacher gives. (In the gardening analogy no seedlings can grow well as they are too squashed together, there is not water, light, soil for them to grow well).

Reducing waste

Wasting time giving help ineffectively (both the learner's and teacher's) is a waste in itself  - but worse it will also have a negative impact on all the other layers the learner is learning; better to remove some layers. In other words help timely practice do its primary job "embed learning" rather than its secondary job "provide the opportunity to give small amounts of targeted tutoring to teach what hasn't been learned yet".

Remember that without timely practice much of the learning it preserves would have turned into forgetting. Any layers that merit a star would almost certainly have been lost to forgetting without timely practice. By deleting some layers, we enable timely practice to help more layers to become embedded learning. Unlike in the gardening scenario where seedlings thinned out just die, in the timely practice scenario the layers can be taught later and next time teaching will be on  "on firm foundations".

Thinning out is rarely needed

Thinning out is rarely required, and generally only on the first round of "finding and firming learning foundations" teaching. Reasons for thinning out include

  • a very inaccurate learner (almost always this will turn out to be because the learner has learning gaps and is trying to do work which is a "bit too hard for them")
  • an overly optimistic teacher or
  • a learner getting secret help (almost always this will turn out to be because the learner has anxiety - so a good remedy is to ensure learners are practising and learning work within their proximal learning zone, which conveniently enough requires removing work that the learner finds "hard" for now)

An "overly optimistic teacher" is likely to

  • add too many layers via "teach" (that is adding more than layer to "teach" per topic more than occasionally) 
  • keep "too many layers within timely practice" by thinking "I can help this learner with this when I feedback" too frequently 

This will mean there will be too many layers in the learner's timely practice assignment, but in actual fact the learner still needs more teaching not just practice. 

How to do thinning out

The remedy regardless of the cause (or causes) is the same, delete some of the layers, but which ones? ... possible strategies include

  • changing every star symbol to a delete symbol (this is only available for layers which are still fragile learning). The teacher may need to do this on more than one assignment.
  • selectively changing some star symbols to a bin symbols (this facility is only available for layers which are still fragile learning) within the assignment. The teacher may need to do this on more than one assignment. The teacher can choose layers to delete that
    • have been pre assessed but not yet taught
    • the teacher is "not looking forward to helping the learner with" 
    • have a high proportion of star and ~ responses (see the right hand end of the blue question number bar)
    • have a lower depth of learning (see the right hand end of the blue question number bar, if the layer is no longer fragile, this information will not be shown)
  • asking the learner. The teacher can say to the learner that they (the learner) has too much "needing help" work they are trying to learn, so ask the learner to choose 1 or 2 questions which you will help them on, the other questions you will remove from their practice now and teach them that work later in the year.
    • advantage: the learner probably has a better sense of what they have "nearly learned" than you the teacher
    • disadvantage: the teacher can't mark the assignment until after the lesson, so the learner won't get to practise the "kept" questions when they should.
  • selecting from the Learner's Progress Page. On the learner's progress page, the teacher can see all the layers in timely practice for the learner. The teacher can choose any layers not in an assignment to best learned later - LESLEY update. The best layers to choose are
    • fragile layers (depth of learning ≤ 5) these layers have few successful practices, 
    • when there are several layers in a single topic which are fragile - pick the higher numbered layers to best learned later - LESLEY update - without other similar questions to practise, feedback on just the easiest layer may be successful,

however layers that are in the current assignments (the top left corner of the square turned over) cannot be best learned later - LESLEY update from the learner's progress page.


Sometimes thinning out is not required, as there are other - short term - causes of a learner making many errors in an assignment e.g.

  • if the learner is returning from an absence, the teacher could (if the teacher has time) give the learner more help than normal just to help their maths brain "back in gear". If the absence was long it could be wise to set the learner "homework" rather than "classwork" assignments - so the learner can catch up first on their timely practice on robust learning - and then decide which of the fragile learning layers to delete or reteach.
  • if the learner "got out of bed the wrong side that morning" 
    • the teacher and learner might agree that the learner will have a fresh start on their new assignment and the learner and teacher should not worry about correcting or giving feedback on "just this assignment"
    • the teacher might ration the learner to say help on just 2 questions (the learner may chose which). While the learner is deciding which questions they want help on,  the learner could "have a go at finding and correcting some questions themselves" - and for the rest of the questions, both the teacher and learner can rely on the app to "get another practice question" soon.