bucks vr not a good test

[Hope]

As a parent with 2 children in Bucks and I work in a school, I don't have a problem with an exam being the main way of working out if a child is suitable for grammar school. What I do have a problem with is the type of exam. I feel that it would be better to have a paper that is not so biased towards vocabulary. VR has its place but as others have said it does not suit all children, maybe we could have a VR, Maths based and English based paper, this would hopefully catch all children. How it would be marked I don't know, would you have a % to gain in each paper? If children were tutored it would help their overall learning too. You will always get children who are no good at exams or who have problems at home etc which prevent them from getting the results they should and an appeal panel would still need to be used.

As for teacher assessments I think that puts the school/ teacher in a very difficult position it leaves them open for parents causing problems within the school if they do not agree with the assessment given. I know its a little like that now but only when a child needs to appeal.

I am not sure we could ever get a perfect system but I do think we could have a much better one than we have now.

[Sally-Anne]

Hi Tree

A formal exam is taken in november or maybe even earlier for children (private and state) who wish to apply for grammar school this contains sats type papers in english maths and science and cats type papers in a v wide range of tests of vr nvr etc etc (a bit like the ep reports we see)

this would have some particular advantages firstly as there is such a wide range of assessments children that are particularly good or bad at vr would not have a particular advantage/disadvantage and it will be much harder to coach v intensively hopefully reducing any coaching bias

I am no great fan of the Bucks VR test, but I am afraid that I can see this playing into the hands of richer parents who would simply start to coach much earlier than they do even now.

The current system already favours children who are taught at high-performing schools, because they will have a higher level of both literacy and numeracy. And that is before tutoring.

Personally I favour the introduction of an NVR test to balance out the VR test. NVR is far more difficult to tutor for - it's much more "either you've got it or you haven't". I know that Bucks CC has considered this at times but I know that there are issues with that solution as well.

Sally-Anne

[Tree]

I suppose its a matter of balence having multiple test papers as i suggested may well be too difficult and expensive although you would save some on less appeals. i think an vr and nvr + maths and english and i think science would be better, i'm not sure i completly agree with sally anne (probably a first) i think if you have a broader spectrum of assessments i think it will be more difficult to coach for and allows for the different strengths of children to come through.

The issue of coaching and the advantages for richer parents is a whole different thread and historically usually leads to it being closed after about 12 posts! i think this debate is starting from the position of believing in a selective system and trying to match children with a school that suits them the best. My feeling that the bucks approach based on 1 type of assessment is inherently unfair it advantages children who are naturally gifted at VR and probably those who can afford (time and/or money) alot of coaching

[andyb]

The grammar schools need to be made aware of this - however, a recent conversation with head of a HW grammar would suggest they are more than aware!!!

I know a couple of grammar school teachers (different subjects) and they have both said, independently of each other, that they can spot the unsuitable tutored child within the first couple of weeks of starting the school. It makes their job more difficult because they have to cater for 1 or 2 children in a class that are just not in the same "league" (for want of a better word) as the other 30 children. Everybody loses - except for the parent who can say their DC is at grammar school.

[Tree]

i missed Etienne's point about my 20% of parental and school reports i agree this is a minefield and maybe it's unworkable however there are some possible answers to this possibly this could be something that was filled in after the results of the exams were known so if the child got really good marks the school would be able to support the child and if they were v low they could agree unsuitability but it would given the school the option in some specific cases to modify the result either way in other words thay may think that some children who have done poorly are still suitable and also some children that have done well are unsuitable (although i can see this would a brave ht who said this)

however maybe the nest approach would be for this 20 % to only be factored in for the borderline cases

[Etienne]

I suppose its a matter of balence having multiple test papers as i suggested may well be too difficult and expensive although you would save some on less appeals.

I'm not so sure about that. If there's a statutory right of appeal, you cannot restrict the number of appeals.

My feeling that the bucks approach based on 1 type of assessment is inherently unfair it advantages children who are naturally gifted at VR and probably those who can afford (time and/or money) alot of coaching

A lot of people will agree that more than just VR should be tested, but it has to be manageable, and ideally less susceptible to tutoring. There has been discussion elsewhere about Durham's so called "Tutor Proof" CEM exams. I'm not sure there is such a thing as a "tutor proof 11+", but no doubt some formats would be less easy to prepare for.

What would, IMHO, be unworkable is 20% of marks for "other evidence" such as reports. The 11+ must be objective and consistent, so it really wouldn't be possible to have individual schools "modifying the results" at their discretion! Borderline and special cases are best dealt with by appeal or review.

[Sally-Anne]

There has been discussion elsewhere about Durham's so called "Tutor Proof" CEM exams. I'm not sure there is such a thing as a "tutor proof 11+", but no doubt some formats would be less easy to prepare for.

It is interesting to search for posts on the Forum about the supposedly "tutor-proof" CEM. This thread is not untypical:

http://www.elevenplusexams.co.uk/forum/ ... hp?t=12586

On there "Optimist" wrote:

Best thing is to trawl the internet and try bits of everything. CEM is Verbal Reasoning (VR), Non-Verbal Reasoning (NVR) and Numerical Reasoning (NR). In short, it is something you can teach your DC, yourself.

I am presently trawling the internet for materials, and using them on a friend's Year 5 DS. Friend's DS will sit the Warks 11+ in November 2010.

VR is heavily dependent on a wide vocabulary - both in knowing the words, and what theyactually mean - especially homonyms. It also will include a cloze exercise - there are many on the internet to look at. Alternatively, copy paste a paragraph from say Times Online, or First News, and remove key words, such as prepositions. Ask your DC to complete them. There will be questions on synonyms and antonyms, again, vocab is important. Comprehension will be multi-choice, so Try the Eng Bond Books and the Comprehension Bond Books. ABOVE ALL, DEVELOP THE VOCAB!!!

NVR, is exactly what it is. Use the Bond Books, and the two CD's on the Eleven-plus forum.

NR - This covers arithmetic and problem-solving. For these, best to get your son to used the KS2 publications in Arithmetic and Problem-Solving, by Schofield & Sims. They are ideal preparation.

The only reason that the CEM appears to be "tutor-proof" is that Durham refuse to publish any practice papers, and the one familiarisation paper apparently bears very little resemblance to the real test.

So, it is only tutor-proof until someone gets hold of a copy of the real test. I understand that has already happened in some areas, so round we go again!

In short, I fear that Etienne is right - there is no such thing as a tutor-proof exam.

Sally-Anne

[Etienne]

Personally I favour the introduction of an NVR test to balance out the VR test. NVR is far more difficult to tutor for - it's much more "either you've got it or you haven't". I know that Bucks CC has considered this at times but I know that there are issues with that solution as well.

My sentiments too. There's no perfect solution, but this would probably be a better system.

[Sally-Anne]

From the Durham CEM website - the bold type is mine:

Every year the test content changes to reflect our ongoing research and development and to ensure the tests remain novel and engaging for the children who take them. There is little or no advantage to be gained from using commercially available test materials to prepare for these tests. As such we believe that they offer a fair assessment for all children, whatever their educational background, as far as this is possible. The tests are not language-free, however, and directly target proficient readers of English. It is our policy not to issue any practice materials.

Back to the drawing board?

[Tree]

ok so i agree with etienne the 20% is probably a non starter and should be reserved for appeals

i think that the wider the assessment the more tutor proof it is; it's gonna be difficult to tutor your child in maths science english and nvr and vr to anything like the level possible for a single exam in vr and so children with natural abilities will be more reflected in the result.

the other area where it is possible to improve any bias is to have a flexible inter school transfer system at years 7,8 and 9 so children that are clearly working at grammar school level at non selective schools can swap with children struggling at the grammar schools which should address some of the over coached children or at least keep them on their toes