The 0Maths blog

Response screening

We use the data we gather to give you a heads up to potential underlying difficulties each pupil may be having. Certain answering patterns (as below) are correlated with certain specfic issues but the lack of a warning here is not conclusive.

The purpose is to alert you to pupils that may benefit from more comprehensive screening. We only have maths answers to work with, and those only with what they've covered in the last number of days (set at top of screen - normally 7) which may be a small data set where the noise overwhelms the signal.

From answering times

We check each pupils recent answering times to see if they increase in approximate proportion to the numbers in the question. These can be indicative of solving the maths by counting on (ie the absence of a strategy), which in turn can be a sign of dyscalculia , especially if the method persists up the school.

We don't flag it at all for puils in the first year of school (ie it is expected at that stage) and don't flag it as dycalculia in the second year. We only look at right answers less than 100. The absence of a strategy (what we're really flagging) is inevitable at various points in their education, such as when they've newly learned to add beyond 10 or to multiply by repeated addition.

Clicking the dyscalculia warning will give one of the following icons:

1. _+↷↷
   Addition times proportionate to second addend (ie adding by counting on and not using comutativity)
2. ↷↷+_
   Addition times proportionate to lowest addend (ie using comutativity but still adding by counting )
3. _-↶↶
   Subtraction times proportionate to subtrahend (ie subtracting by counting backwards)
4. ↷↷+↷↷
   Addition times proportionate to total (ie adding by counting on from zero)
5. __×↷↷
   Multiplication times proportionate to either multiplicand (ie multiplying by repeated adding), or division times proportionate to divisor. Calculations involving either the 10 or 11 times tables are not included.
6. ↷↷×↷↷
   Multiplication or division times proportionate to total (ie multiplying by repeated adding, where the adding is done by counting on). Calculations involving either the 10 or 11 times tables are not included.

We flag the strongest correlation where it is greater than 0.4 (statistically this would be classed as a moderately strong correlation). The red bar displayed under the icon indicates the strength of the correlation, where 0 (black) is not at all correlated (so we wouldn't show you), and 1 (fully red) would be where they count on with the uncanny rhythm of an atomic clock (ie it'll never actually be 1).

We flag these correlations even if answers are not slow - some children have lightning fingers. Early intervention will make a big difference later on.

From errors

To paraphrase Tolstoy "Happy answers are all alike; every unhappy answer is unhappy in its own way".

As we are collecting typed answers, incorrect (ie amended) answers are a potentially rich source of data for learning interventions.

We screen amended answers for the following 'interesting' errors, which are flagged if an error is made 2 or more times within the timeframe (by default, 7 days):

1. ⇝↭
   Wrong rote learning eg, a pupil often answers 54 for 7 × 8. If the same wrong answer occurs repeatedly, it is likely coming from memory. First learning dominates, so spending time on related questions can be helpful (ie 56 ÷ ? = 8)
2. ↑
   1 more than correct answer. in the case of addition this is likely because the pupil is counting on from the first given number, rather than the number after.
3. ↹
   Transposed digits - This could be a place value understanding issue, another sign of dyslexia, or simply a typo.
4. ↓
   1 less than correct answer In the case of subtraction, the pupil is likely counting back from the digit before the digit they started from.
5. Dyslexia
   Possible dyslexia A combination of the previous two - the pupil's answers for simple arithmetic are often one more or one less than the correct answer. This is often seen in dyslexic pupils though is not normally a part of dyslexia screening. (ie this is an indicator, not a diagnosis.)
6. 10↑
   10↓
   10 more or less than correct answer - likely a carry digit error if not seen alongside 1 more / 1 less.
7. -?
   Subtracting highest from lowest for each place Given a sum like 67-28 a pupil answers 41 (60-20=40✓ and then 8-7=1╳). They have learned that for a subtraction problem they should take the lowest number from the highest, and so they apply this method to each place value.
8. +?
   Place value error: more than 10 in each place Given a sum like 101+109, a pupil answers 2010 (hundreds:1+1=2 , tens:0+0=0 , units:1+9=10 = 2 0 10). They have learned to add each digit but have squeezed a 10 into one digit's space instead of carrying.
9. 0-0
   Vision issue - Often adds when seeing the obelus (division sign, ÷), for instance.
10. AD(H)D
    Possible ADD or ADHD - The pupil tends to use the wrong operator. Again, this is common in ADD and ADHD pupils but is not in itself normally part of screening.

The colour of the indicator relates to the curriculum strand it was discovered in. Clicking the icon tells you which question type (and which specfic questions(s)) the error was found in.

This is a beta feature. We'd welcome your feedback to make it more useful.