Comparing Kent Test (11+) to the Cognitive Abilities Test (CAT)

What are CAT scores, and how do they compare to the Kent Test (11+)?

At my child’s school, they sit a Cognitive Abilities Test (known as a CAT test)  in the spring term of year 5. Whilst not mandatory within schools, these tests are taken nationwide and are often used for Kent Test appeals or reviews. In my own child’s school, the CAT scores are used as an indicator on a child's suitability for the Kent Test (11+) exam- although they can also provide a guide (for the school) of the child’s wider cognitive strengths and weaknesses. However, from a parental point of view, there is minimal information provided about what the CAT tests/scores actually mean. Indeed, for child #1, the CAT score was buried in their school report- and I had little understanding at the time, of what the score meant. The following are some of the links, and my interpretation of them, that I have found useful in getting my head round these results.

What is the CAT test?

The CAT tests are made up of three elements: Verbal, Quantitative and Non-verbal- which have common threads with the Kent Test exam.  Each of these sections produces 5 separate marks

• The raw score
•  The SAS (Standard Age Score)- this is the standardised score (and is the core element of the CAT results). The standardisation process takes into account age and importantly allows benchmarking against other pupils of the same age. The score for an ‘average’ child is 100, whilst the highest possible score is 141.
• The Stanine- this places the child’s standardised score on a scale of 1(low) to 9 (high) and offers a summary overview of the SAS score.
• The NPR (Nationwide Parental Rank) – this relates to the  SAS. It provides details for how many other children achieved that score or below- for example, a NPR of 80 means that the score achieved was in the top 20%.

The following research document from GL Assessment provides a great overview of the standardisation process, and a detailed breakdown of each of the marks mentioned above. The graph on page 10 of this document shows a normal  distribution curve showing the relationships of Stanines, national percentile ranks (NPR) and standard age scores (SAS) to each other- and is a good visual indicator of where your child’s score fits compared to their peers.

http://www.gl-assessment.co.uk/sites/gl/files/images/Guide-to-Standardised-Tests.pdf

GL Assessment, who also provide the Kent Test papers, have a whole host of informative research papers. I’ve found the following link very useful, as it provides more detail on the CAT tests and an interpretation of the results they provide. The table on page 16 provides a good overview of the percentages of children that fall into each of the 9 Stanine groupings.

http://www.gl-assessment.co.uk/sites/gl/files/images/Files/Testwise/CAT3-guidance.pdf

The CAT and Kent Test comparison

With the quantitative (math), verbal and non-verbal elements of the CAT test having a similar feel to some of the Kent Test, and with the results also having a similar look and feel, there is the temptation to compare the two results. However it would be incorrect to directly compare these two standardised results- though I see quite a few sites and forum comments doing a direct comparison.

The following link (you will need to register to download the pdf itself) provides a really useful article on the comparisons (and differences) between the Kent Test and CAT scores. It also provides a statistical breakdown of the distribution of the CAT marks- pointing out that that the CAT scores have a mark of 100 to represent the ‘average/expected’ child, and a standard distribution of 15. Effectively 68% of the pupils taking the exam will have a standardised score of between 85 and 115. It has been a few decades since I last touched bell curves and standard deviations, so  I would read the article for more info on this rather than treating my interpretation as gospel!

http://www.elevenplusforum.org.uk/viewtopic.php?f=24&t=10

One paragraph from this article that is probably worth mentioning in its totality is the following-

“We would say if your child scores around 108 or higher then your child is a strong candidate for taking the Kent 11+ test. If your child is scoring significantly less than 108 this does not necessarily mean that your child is not a strong candidate for the Kent 11+. The CAT’s are a snapshot of an unfamiliar test on a single day, a lower score can be indicative of many things. A low score is certainly something that should be looked at closely and understood but it is by no means a sure sign that the child is not a good candidate for the Kent 11+ test.”

Effectively, like all these type of tests, the CAT score is just a one-off test that in isolation should be treated with caution. They are a snapshot of a performance on one particular day, and reading too much into one result from one exam (whether high or low), in complete isolation from everything else (teacher feedback etc), would be wrong.

My Personal Experience

 If your child has taken the Year 5 CAT test,  the school should be able to provide you (if not automatically, then upon request) an overview sheet that shows your child’s results. An example of this is shown in the image below (results for my child #2), and provides a good overview of not only the scores they achieved, but where they fit in with their peers. From this sheet you will see that all the previously discussed scores (raw, SAS, Stanine and NPR) are on it.

From a personal perspective, having had both my children sit these CAT tests in 2014 and 2015, my experience is (from their feedback, not mine!) that the CAT tests are easier than the 11+ practice papers (especially the GL Assessment 11+ Practice papers) and the Kent Test itself. However, to provide a little counterbalance to this point, over the past two years both of my children have taken these exams, and both their CAT and 11+ exam pretty much mapped to the correct percentile (to within 1% of each other) when the results were compared to peers. Whilst a swallow does not make a summer- this suggests that they do provide a reasonable indicator (though I must stress again that one should not place too much emphasis on just two results) to the end Kent Test result.

Finally, given that state schools are not allowed to coach for the 11+ exam, I think exposure to CAT testing (with its similarities to the Kent 11+ exam) is a good thing. It will not only provide the child with exposure to Non-Verbal and Verbal reasoning type questions, but they will experience (and so learn) from the pressures that time constraints within an exam can bring.

For those looking for more information:

http://www.elevenplusexams.co.uk/advice/standardised-scores-an-explanation - provides a good overview of the standardisation process within the Kent 11+ exam results.