Sats Sats Sats!!!!
Moderators: Section Moderators, Forum Moderators
-
countrymum
- Posts: 686
- Joined: Tue Mar 08, 2011 10:16 am
Re: Sats Sats Sats!!!!
*like*Guest55 wrote:
Bright pupils should not be made to feel failures and should be able to enjoy their last year in Primary school.
Apologies if I come across a bit moany (not my usual style on here) but I so want to enjoy this year with DD, to climb trees, sing with the birds, walk in the (hopefully soon) sunshine. Of course we can do that anyway
but you get my point, everything just seems to be taken up with the sats. Homework is all sats, all the DC's seem to be talking about is sats. I just can't help think in some small way these children are being 'taught' to pass a test to make the school look good. Maybe that is a very basic way of looking at it. Moved if she wants to then yes I guess I should not take the opportunity away from her. She is aware what sats are and what they are all about and knows not many get the 6, but I suppose I should praise her efforts and enthusiasm in wanting to do them. But I know her and know she shall be downbeat not 'achieving' them. If she is teacher assessed a level 6, does anyone know what that means? I know some who do the level 6 papers but are teacher assessed a lower level? Is the TA a more realistic target and the level 6 taking is more just having a go?
Re: Sats Sats Sats!!!!
Her secondary school will use her test for her baseline targets. The league tables are compiled from them. The tests form a national baseline that is currently used to standardise GCSEs.
Last year's English results were always going to be lower than the previous year because their KS2 tests were lower!
Even if she scores 100% that will only be a level 5 - the papers do not provide questions to sublevel and the tests shouldn't be used in that way. There is a sophisticated statistical package that aids test construction and it doesn't involve sub levels. The tests are put together to give the greatest information at the threshold boundaries.
This does therefore mean that gaining the L6, which by no means implies a child is a secure L6, a school does have the indicator of exceptional primary performance. That is quite different from a L5, which could be a 5c.
Regarding her self-esteem, I would definitely emphasise that a test that doesn't have a single child in a whole county who succeeds is not a test that she should worry about not achieving.
For a very high attaining child failure is a very hard lesson to learn and sadly it is also something that requires practice.
I know of several children who are absolutely flying in their super selective grammars who didn't pass the L6 test, if this helps.
Edit: one other point is that many 'level 6' children in KS3 did not score highly enough on the 2012 tests to have scored the L6 and the teachers were very confident that the children were L6 and had scored L6 on KS3 papers. I know of a couple of schools who gave the tests to their children. The maths tests require a high level of using and applying. The English tests require a well articulated response with evidence of insight.
I don't think the tests indicate a level, but I do think they indicate a great deal of potential.
Last year's English results were always going to be lower than the previous year because their KS2 tests were lower!
Even if she scores 100% that will only be a level 5 - the papers do not provide questions to sublevel and the tests shouldn't be used in that way. There is a sophisticated statistical package that aids test construction and it doesn't involve sub levels. The tests are put together to give the greatest information at the threshold boundaries.
This does therefore mean that gaining the L6, which by no means implies a child is a secure L6, a school does have the indicator of exceptional primary performance. That is quite different from a L5, which could be a 5c.
Regarding her self-esteem, I would definitely emphasise that a test that doesn't have a single child in a whole county who succeeds is not a test that she should worry about not achieving.
For a very high attaining child failure is a very hard lesson to learn and sadly it is also something that requires practice.
I know of several children who are absolutely flying in their super selective grammars who didn't pass the L6 test, if this helps.
Edit: one other point is that many 'level 6' children in KS3 did not score highly enough on the 2012 tests to have scored the L6 and the teachers were very confident that the children were L6 and had scored L6 on KS3 papers. I know of a couple of schools who gave the tests to their children. The maths tests require a high level of using and applying. The English tests require a well articulated response with evidence of insight.
I don't think the tests indicate a level, but I do think they indicate a great deal of potential.
-
countrymum
- Posts: 686
- Joined: Tue Mar 08, 2011 10:16 am
Re: Sats Sats Sats!!!!
Thanks very much moved for your helpful and informative post. I feel so appreciative that there are people like you on here who know the ins and outs of these things. I'm 'just' a parent (albeit an enthusiastic one) and is sometimes hard to keep up with all the jargon. So, thank you 
You mention no sub levels in the KS2 Sats, would that mean that she would be given a level either 5 or potentially 6 then? I know when they have been doing practice papers at school they are told their sub levels on the paper they have done and so was of the thinking her Sats level when she has done all the papers in May would include her a final 'sub level' too?
You mention no sub levels in the KS2 Sats, would that mean that she would be given a level either 5 or potentially 6 then? I know when they have been doing practice papers at school they are told their sub levels on the paper they have done and so was of the thinking her Sats level when she has done all the papers in May would include her a final 'sub level' too?
Re: Sats Sats Sats!!!!
There are no sublevels. Since leaving teaching I have done a lot of research into different methods of assessment.
The 'SATs' are not at all designed to give sublevels.
What happens is that each question is trialled at least twice, often more. During these trials statistical data is collected. The data is used to see two things: how 'easy' a question is and how well it does its job to differentiate between children. A question that 50% of children get right, but those children are evenly split between level 3-5 children provides no useful data from which to level a child in the final test. Teacher assessment levels are used for the test data and performance is correlated against them.
When the test is constructed, the model uses as many questions from the threshold regions as possible, so that it is certain that a child has achieved the threshold of a level. In practice the tests are ramped in difficulty, so that there isn't a huge jump between the level 3 boundary to the level 4, etc. ideally, all the questions would focus on the threshold points.
The questions are grouped by curriculum level, but not all areas of the curriculum fall into a level. The questions are then grouped by difficulty. If you look at the programmes of study and the levels there are no sublevels.
A test could consist of questions that max out at level 5c or 5b as long as it ensures that the children have crossed the level 5 threshold it has done its job. 100% may only indicate that a child is a solid level 5c!
The 'SATs' are not at all designed to give sublevels.
What happens is that each question is trialled at least twice, often more. During these trials statistical data is collected. The data is used to see two things: how 'easy' a question is and how well it does its job to differentiate between children. A question that 50% of children get right, but those children are evenly split between level 3-5 children provides no useful data from which to level a child in the final test. Teacher assessment levels are used for the test data and performance is correlated against them.
When the test is constructed, the model uses as many questions from the threshold regions as possible, so that it is certain that a child has achieved the threshold of a level. In practice the tests are ramped in difficulty, so that there isn't a huge jump between the level 3 boundary to the level 4, etc. ideally, all the questions would focus on the threshold points.
The questions are grouped by curriculum level, but not all areas of the curriculum fall into a level. The questions are then grouped by difficulty. If you look at the programmes of study and the levels there are no sublevels.
A test could consist of questions that max out at level 5c or 5b as long as it ensures that the children have crossed the level 5 threshold it has done its job. 100% may only indicate that a child is a solid level 5c!
-
countrymum
- Posts: 686
- Joined: Tue Mar 08, 2011 10:16 am
Re: Sats Sats Sats!!!!
Thanks moved! Really helpful.
Re: Sats Sats Sats!!!!
This makes more sense to me and thank you. Although my ds is not getting 100% I believe that is because he is rushing through the paper( I have only given him 1 )moved wrote:There are no sublevels. Since leaving teaching I have done a lot of research into different methods of assessment.
The 'SATs' are not at all designed to give sublevels.
What happens is that each question is trialled at least twice, often more. During these trials statistical data is collected. The data is used to see two things: how 'easy' a question is and how well it does its job to differentiate between children. A question that 50% of children get right, but those children are evenly split between level 3-5 children provides no useful data from which to level a child in the final test. Teacher assessment levels are used for the test data and performance is correlated against them.
When the test is constructed, the model uses as many questions from the threshold regions as possible, so that it is certain that a child has achieved the threshold of a level. In practice the tests are ramped in difficulty, so that there isn't a huge jump between the level 3 boundary to the level 4, etc. ideally, all the questions would focus on the threshold points.
The questions are grouped by curriculum level, but not all areas of the curriculum fall into a level. The questions are then grouped by difficulty. If you look at the programmes of study and the levels there are no sublevels.
A test could consist of questions that max out at level 5c or 5b as long as it ensures that the children have crossed the level 5 threshold it has done its job. 100% may only indicate that a child is a solid level 5c!
With the previous one given in school in December. Finishing in half an hour and not struggling with the questions ..the ones he gets wrong he realises straight away where he has gone wrong without me showing him. The paper 5 is not presenting a challenge. I have looked at paper 6 and this would definitely be a challenge, even if he were not to get it he will have learnt new things that will be a help in high school. He did pass for grammar but as we are out of catchment and the school uses distance rather than score achieved it is not certain at all that he will get a place.
Debs
Re: Sats Sats Sats!!!!
They are subleveled though - in the KS2 to KS4 progress matrices pupils are tracked from their sublevel at KS2.
We get the actual mark not just the level, a low level 5 will perform at a different 'level' to someone scoring 95+ from day one.
We get the actual mark not just the level, a low level 5 will perform at a different 'level' to someone scoring 95+ from day one.
Last edited by Guest55 on Sat Feb 02, 2013 4:24 pm, edited 1 time in total.
Re: Sats Sats Sats!!!!
Not sure how schools use the data, but I am sure how the tests are put together. I have spoken in depth to the team who produce the papers.
Tracking within schools, however, has not formed part of my research. Scores from tests are often used for inappropriate things. A CAT score is given a numerical value, but it is only the centre point of a very wide range of marks that could indicate a child's ability. These scores are often used by schools to assess their pupils despite the fact that the score could easily be 20 points out. Teachers are often asked to justify why a child has got a score that is lower/higher than a previously comparable score.
Tracking within schools, however, has not formed part of my research. Scores from tests are often used for inappropriate things. A CAT score is given a numerical value, but it is only the centre point of a very wide range of marks that could indicate a child's ability. These scores are often used by schools to assess their pupils despite the fact that the score could easily be 20 points out. Teachers are often asked to justify why a child has got a score that is lower/higher than a previously comparable score.
Re: Sats Sats Sats!!!!
It's not schools that produce these matrices, moved, it's RAISEonline.
Re: Sats Sats Sats!!!!
Sounds like the raiseonline team and the year 6 test development team need to speak to one another. Nothing has changed at the blob.
