ThinkEd: Drunk with success

If the old song Ten Green Bottles was sung at an education conference it would certainly go like this:

 Ten green bottles hanging on the wall

Ten green bottles hanging on the wall

And if one green bottle would accidentally fall

There’d be 90% of the bottles hanging on the wall.

That’s fine. But the games start in the second verse:

 Nine green bottles hanging on the wall

Nine green bottles hanging on the wall

And if one green bottle would accidentally fall

There’d be 88.8% of the bottles hanging on the wall.

 Some might have thought that there would be only 80% of the bottles there now but no, in education we often tell the story by using percentages that are based on a more immediate sample rather than taking account of the starting sample. Perhaps you might say that the distortion in the above example is not great. But what about say the end of the fourth verse when there would only be six of the original bottles left. Would that be 60%?

Seven green bottles hanging on the wall

Seven green bottles hanging on the wall

And if one green bottle would accidentally fall

There’d be 85.7% of the bottles that got into the fourth verse hanging on the wall.

Now you see the pattern – we look pretty good at keeping green bottles. But in reality? This has been exercising my mind a little lately as I look at the three key markers of education that are likely to produce lasting advantage for your young people: access to two years quality early childhood education (15 hours a week), successfully completing secondary school and gaining a post-secondary qualification.

So an analysis of education success needs to consider the performance at each level in terms of the birth cohort – how many of 100 babies succeed at later ages. Also of importance is the fact that the success profiles of key groups should also be reflected and for this exercise Maori, Pasifika and Pakeha/Other are the three groupings chosen.

With early childhood education the picture is complicated by the fact that children get access for widely differing amounts of time. It seems inequitable to simply count a child as having access if they at some point enter the gate of a centre. On this basis New Zealand claims high levels of participation. But on the basis of who consumes the ECE resource the picture is interesting. We know that 28% of babies born are Maori but over the first five years of their lives they consume only 20% of the resource. The picture for Pasifika is not quite as good – 16% of babies will be Pasifika but they consume only 6.1% of the ECE resource. That means that Pakeha and other groups which make up 56% of the babies born consume 73.9% of the resource.

But regardless of their ECE access and experience, all young ones are allowed to start school so it is valid to return the group measure on starting school to 100% – the 100 babies born.

The second education advantage point – successfully completing secondary school – is one which does suffer from the ten green bottle syndrome. The results for success are often presented as success for the group that actually completes schooling. But we should go back to the birth cohort and factor in the disengaged (those who just drop out of the system for one reason or another) and we should also look at those who get success within the expected timeframe. For the purposes of this exercise I have determined to be achieving NCEA Level 2 in 12 years of compulsory schooling. This certainly is a degree of success that should enable a student to proceed to a tertiary qualification.

 This then produces some very interesting results: 13 of our 28 Maori babies will get NCEA Level in Year 12, 7 of our 16 Pasifika babies will succeed, and of the 56 Pakeha/Other babies, 36 will get there. On the basis of successfully completing secondary school on this measure, 45 of our 100 babies will get NCEA Level 2 in Year 12.

 Assuming that these 45 students go on to a post-secondary qualification we know that a very stable statistic is that half of students who start a post-secondary qualification actually complete it. So that tells us that we can expect success (on the basis of completing the qualification) for 7 Maori students of the original 28 Maori babies, 4 of the 16 Pasifika babies and 18 of the 56 Pakeha/Other babies.

Success is an outcome not only of the performance at a moment in time but also the cumulative processes that impact on different groups of students in different ways. People with greater statistical skill than me will see issues with the outline above. That is good; let those skills now be applied to more sophisticated and more accurate pictures of what happens to the ten green bottles. A good place to start would be with recognition that not all the bottles are green!

 And does the conventional approach to presenting results obscure a growing issue – the performance of Pakeha/Other in our system? Consider the above and the final result of 32% for Pakeha/Other and 25% for both Maori and Pasifika within a system performance of 29% overall. A community that cares has to care for each and every baby born.