Advanced search

An interesting trend with NC L5 at KS2 SATs

(54 Posts)
PastSellByDate Fri 02-May-14 10:29:28

Hello all:

My fondness of numbers/ statistics has lead me to consider some interesting tables the Guardian published regarding this past December's announcement of 2013 (May 2013 cohort) KS2 SATs results:

The article is an interesting read - but scroll down to the tables of overall results in England between 1997 - 2013 for English & Maths and there is a rather interesting trend:

English: 1997 20% achieve NC L5 and this steadily climbs until 2007 48% achieve NC L5 - and it stays in this general approaching 50% ball park until 2013 (48% that year).

Maths: 1995 - 13% attained NC L5 and this climbs to 31% in 2005 and then on up to 41% in 2013.


I'm sure there isn't a simple explanation - but if basically 50% of all English pupils are attaining NC L5 in English & 40% are attaining NC L5 in Maths - and there isn't a watering down of standards (which as a parent I can't really determine) - does this mean that NC L4 needs to be presented to parents as the minimum a child should achieve?

I just wonder if our school had been aiming for NC L5 rather than scraping NC L4 - if this entire process of preparing for KS 2 SATs would have been less stressful for the school (? less stressful for parents having to deal with kids suddenly getting endless photocopies of KS2 SATs busters books for homework after years of no homework at all/ and endlessly taking past KS2 papers for 'practise').

Our school seems to have thrown everything at DD1's Year 6 cohort in one last ditch effort to get >65% to NC L4+ in Maths/ English and I can't help but wonder if the pace of the curriculum and the standard of content hadn't been slightly higher if everybody wouldn't have had a more pleasant Year 6.

PiqueABoo Tue 13-May-14 22:46:58

I once got Excel to make some graphs out of RaiseOnline library (no logon required, anyone can go get that) KS2-KS4 transition data which had KS2 sub-level granularity, however it was KS2 data for the cohort who had just done their GCSEs so not exactly current. Their lastest KS1-KS2 transition data didn't have the sub-level granularity, just whole levels at KS2.

At KS2 it was normal distributions but with some significant distortion around 4c and 5c i.e. suggested 'boosting' to those levels. The top end was also clearly 'clipped' at 5a i.e. some of those children could have been more 'stretched'. GCSE grade graphs were much the same but the distortion was obviously for grade C.

AmberTheCat Tue 13-May-14 22:39:33

The results should be skewed towards the upper end if children have good phonic knowledge, but it should still be a normal bell curve shape. The massive jump on this graph at the pass mark, and the fact that it actually goes down on the number before the pass mark, is so pronounced that it's hard to see any other explanation than teachers giving kids who were nearly there the benefit of the doubt, whether consciously or subconsciously.

As I say, completely understandable, but given that different teachers will respond to the pressure to have as many children pass in different ways, I think it makes sense not to publicise the pass mark until afterwards.

ipadquietly Tue 13-May-14 22:30:52

I'm not sure this is meant to be a normal distribution.
Children are expected to know ALL the sounds in the phonics test by the end of Y1, having had two years of phonics training. Surely, that expectation, and the fact that many children have good phonic knowledge by the end of Y1, would surely skew results to the +30 end.

Wouldn't it?

AmberTheCat Tue 13-May-14 22:16:40

Not that I know of, but then they're largely externally marked. I'm a fan of teacher assessment, but, given the pressure schools are under to achieve results, I can completely understand the temptation to give children the benefit of the doubt when a pass mark or grade boundary is known, which I think is what that graph reveals.

pointythings Tue 13-May-14 22:04:47

Amber that does look scary... Any similar data for KS2 SATs?

AmberTheCat Tue 13-May-14 21:50:42

Far be it from me to defend the DfE, but there are good reasons for not publishing the pass mark for the phonics test in advance. Have a look at the distribution curve here:

pointythings Tue 13-May-14 21:33:59

I know why I'm cynical - 2012 English GCSEs, anyone? If they're grading on a curve, they need to be open about it.

ipadquietly Tue 13-May-14 17:27:21

As in the Y1 phonics test this year. They're keeping the pass mark secret until the 30th June. Cue manipulation of results.

I wonder why we're all so cynical. hmm

pointythings Tue 13-May-14 10:07:09

Whereas the real explanation is that the powers that be want the state sector to be seen to fail...
Honestly, if the people setting these tests are incapable of creating tests of equivalent difficulty year on year with only minor statistical variations then they should be sacked.

mrz Mon 12-May-14 21:22:44

The explanation given is that the threshold reflects slight variations in difficulty (25-34 slight?)

PiqueABoo Mon 12-May-14 21:18:03

can' think => can't think

PiqueABoo Mon 12-May-14 21:17:16

Can't say I'm impressed by the apparently political nature of the thresholds. I can' think of an obvious reason why they couldn't pitch successive L6 maths papers at the same difficulty level, at worst +/-1 mark from year to year. Was it a (futile) attempt to get secondary to take it more seriously?

spanieleyes Mon 12-May-14 20:36:13

Yes, 13 in the trial, 19 the following year and 22 the next!

Feenie Mon 12-May-14 20:21:27

Yes, that was the sample, wasn't it? Reading was similarly nuts, as I recall.

spanieleyes Mon 12-May-14 19:23:51

The 2011 level threshold for L6 in maths was 25 out of 50, the next year it was 34!

Feenie Mon 12-May-14 18:50:47

They are generated after papers are completed and vary according to who has an election year coming up/has just come into power/has a point to prove, etc.

Fairly sickening.

PastSellByDate Mon 12-May-14 18:38:07

Hi Feenie:

I am indeed talking about 2012 cohort (she applied for all this/ began campaign in 2011 - hence my thinking 3 years ago - also thinking about this as 3 school years ago - i.e. this played out whilst DD1 was in year 4, which is 3 school cohorts ago.).

Very interesting point - How do they come up with thresholds? Are these generated as a result of marking (so a normal curve is applied to the scores?) Or are these decided in advance?

No idea myself - but hopefully someone will be along soon.

ipadquietly Mon 12-May-14 17:50:49

Last year the threshold for level 5 was made 2 or 3 marks higher than the year before. If thresholds are moveable, the statistics become pretty meaningless.

Feenie Mon 12-May-14 17:44:10

Level 6 papers didn't exist three years ago. Your friend's ds could have been teacher assessed at level 6 - that never went away. Level papers disappeared in 2002 and reappeared in 2012.

PastSellByDate Mon 12-May-14 15:01:36

First off thanks all for interesting comments.


I can only judged by our school but 3 years ago a parent virtually had to threaten legal action to allow her son to sit a Level 6 paper. The school didn't want him to because basically he'd spent all of Year 6 supporting other students and they had no work to demonstrate they had been working with him at this level - I suspect.

Last year 2 students were put forward.

This year I know that 15 students were selected out of the main class for extension lessons and 'all were told if they worked hard, they could achieve NC L6'.

Not sure if all will be sitting the Level 6 test - but my impression is the school is far less concerned about repercussions of pupils failing L6 paper and prefer to encourage students to go for it.

I think my DD1 has a chance with maths but am slightly dubious about Reading/ SPAG at Level 6. But very flattering she is going for this - although I won't know for sure until I hear she has actually sat 2 papers today (or indeed Tues & Thursday).


BravePotato Mon 12-May-14 14:33:15

interesting thread.

I don't feel bad that my son will "only" get a 4b (probably) with all those level 5s and 6s. For him, that is 3 sub levels higher than predicted based on y2 SATs.

It does make me realise though that I may have been a bit complacent with his maths. He is apparently a comfortable level 5 and I have not pushed for level 6 test (as some parents did), as I did not see the point.

Now I can see why one might want to.

I like MN for getting alternative views, though the sheer number of kids on MN doing level 6 is a bit intimidating!

mrz Sat 03-May-14 20:55:08


Multiply normally, ignoring the decimal points
Then just count up how many numbers are after the decimal point in both numbers you are multiplying, then the answer should have that many numbers after its decimal point.

spanish11 Sat 03-May-14 20:40:19

Ok, in my primary school I learnt this way. Remove the decimal, for example, 0.13*0.0123 I will do 13*123= 1599 and then divide this number by 100.000= 0.001599.
I multiplied 0.13 by 100 y 0.0123 by a 10.000 the 100*10000= 100000.
in my school they told me that if a multiply a number with 2 decimal with a number with 3 decimal my result will be a number with 5 decimal (2+3).

mrz Sat 03-May-14 19:59:59

Level 4 would begin multiplying simple decomal by a single digit
Level 5 would use all 4 operations with decimals to 2 decimal places

TeenAndTween Sat 03-May-14 19:10:12

multiplying decimals becoming harder (I guess)

2 x 4.5

3 x 3.7

3.1 x 2.7

0.13 x 0.0123

Join the discussion

Join the discussion

Registering is free, easy, and means you can join in the discussion, get discounts, win prizes and lots more.

Register now