Getting older quicker

By

At Datalab we write about half a dozen blogposts in a typical month. Some we slave over and others we knock out in a matter of minutes, usually with one eye on the football or Paw Patrol. Being unashamed data wonks, we then pore over Google Analytics to see which of our posts have left the biggest impression. We are often miffed by how this seems to be inversely correlated with the effort we put in.

Earlier in the week we had a lot of positive feedback on Twitter for the chart below, which we had lying around as a by-product of some other work we had been doing. So we thought we would write it up for posterity.

The chart shows the percentage of pupils in the 2016 Key Stage 4 cohort (state-funded mainstream schools only) who achieved the expected standard at Key Stage 1, Key Stage 2 and Key Stage 4 by month of birth. We only included pupils who were assessed at all three Key Stages. The indicators used were:

  • % achieving Level 2B or above in reading, writing and maths at Key Stage 1
  • % achieving Level 4B or above in reading, writing and maths at Key Stage 2
  • % achieving five or more GCSEs at grades A*-C (or equivalent) including English and maths at Key Stage 4

As luck would have it, the averages of these three indicators are broadly similar for pupils we are looking at, at 56%, 58% and 59% respectively.

% achieving expected standard by month of birth and key stage, 2016 KS4 cohort

The chart shows that August-born pupils close the gap as they get older but remain behind September-born peers by the end of KS4. However, colleagues at the Institute for Fiscal Studies have previously found no evidence of long-term differences in outcomes among adults [PDF].

The average Progress 8 score for August-born pupils is 0.18 grades per subject higher than September-born pupils indicating (as the chart above suggests) that they close the gap slightly between Key Stage 2 and Key Stage 4. However, in terms of raw attainment (Attainment 8), they remain 0.3 grades per subject behind. On average, September-born pupils achieved 52.4 points compared to 49.5 among August-born pupils.

The cohort used in the chart above completed Key Stage 2 in the days of Levels. To bring things more up-to-date, in the following chart we look at the cohort which completed Key Stage 2 in 2016, when scaled scores had come into effect. As the expected standard at Key Stage 2 is higher, we have redefined the Key Stage 1 standard and include the Foundation Stage:

  • % achieving a good level of development in the Foundation Stage
  • % achieving Level 2B or above in reading, writing and maths (with at least one Level 2A or above) at Key Stage 1
  • % achieving the expected standard or above in reading, writing and maths at Key Stage 2

Although we don’t have data on achievement in Year 4, we have included it so that each series is two years apart and we have taken a punt on what we think the results would show for September-born and August-born pupils if we did have it.

% achieving expected standard by month of birth and key stage, 2016 KS2 cohort

Though crude, the chart suggests that the rate of catch-up for August born pupils was slightly quicker between Key Stage 1 and Key Stage 2 than between Foundation Stage and Key Stage 1.

The upshot

What is the upshot of all of this?

Most secondary schools are large enough not to suffer too much from year-on-year changes in the age profile of its cohorts.

But most primary schools are not.

Performance can be affected by an increased proportion of summer-born pupils (primary school performance is somewhat volatile in any event).

(For those who use FFT Aspire, age is one of the pupil characteristics used in contextual value added progress measures. Term of birth is also to be added shortly to the self-evaluation part of Aspire.)

We have long argued that age, gender and other contextual factors should be considered when evaluating a school’s performance.

Want to stay up-to-date with the latest research from Education Datalab? Sign up to our mailing list to get notifications about new blogposts, or to receive our half-termly newsletter

By | 2017-11-13T14:38:52+00:00 March 3rd, 2017|Pupil demographics, School accountability|

About the Author:

Dave Thomson is Chief Statistician at FFT with over fifteen years’ experience working with educational attainment data to raise attainment in local government, higher education and the commercial sector. His current research interests include linking education and workplace datasets to improve estimates of adult attainment and study the impact of education on employment and benefits outcomes.

7 Comments

  1. steve March 3, 2017 at 1:50 pm - Reply

    Whilst the birthdate of students is unlikely to have any notable impact upon a secondary school at full cohort level it could if we are looking at the performance of a sub-cohort.For example a school I know has around 15 or so PP students per year group – it is quite possible that there could be big swings in PP P8 from one year to the next purely as a result of differences in their dates of birth.

  2. Duncan Byrne March 3, 2017 at 4:28 pm - Reply

    My question is what schools should do to support pupils with summer birthdays?

  3. Tim March 6, 2017 at 9:56 pm - Reply

    After school and lunch time revision sessions based on star sign, no doubt.

    • Jon Barnes March 15, 2017 at 9:18 am - Reply

      Age makes a real difference to attainment. If two pupils start school in the same year group but one is born in September (beginning of the school year) and the other in August, the September-born child is almost exactly a whole year older: the older child turns five and remains five for a whole year whilst the other remains four almost up until the next year. Roll that back to when they’re younger and one will have been walking, playing, talking (and even been alive!) for a year longer than the other. It’s a gap that, as Dave’s chart shows, is still wide at the end of primary school.

      This is called the relative age effect, and it’s still prevalent at the end of education, as was proven in research into the draft age of players into the NHL. Players born in the first quarter of the year make up a far greater percentage of draftees than those in the final quarter because, relative to the draft season process, they’re older and thus, developmentally, bigger and stronger individuals, ideal for making them stand out to agents as ice hockey players when surrounded by people born later in the year. You can find the research in PlosONE: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0057753#s2

    • Jon Barnes March 17, 2017 at 2:22 pm - Reply

      Also, you’re right, it is related to star sign, as star signs are calendar-based.

      • Tim McLardy March 20, 2017 at 10:47 am - Reply

        I should apologise for making a flippant comment that added little to the article really. My criticism was directed at our leadership rather than the very real issue raised. That and the ineffectual, knee-jerk strategies employed to improve student progress.

        “Here’s a problem within the system… I tell you what, rather than change the system, we’ll see if the teachers have a spare minute to set extra homework for the summer-born.”

  4. Charms March 7, 2017 at 3:53 pm - Reply

    The whole education system needs a big review so it can support all children based on their learning levels before they start school.
    This would adjust the cutoff for the youngest and eldest children per year group, to cater for mixed age groups by up to 1 year at least.
    Opting for a later school start like the Finnish do at aged 7 years would enable children to be fully ready to learn in an academic style.
    The UK is behind many other European models of education and have better attainment and higher achieving pupils who are happier and keep their desire to learn.
    This is in opposition to UK pupils who get labelled with ADHD etc, due to beng forced into school at 4 years old and summer born pupils who lose interest in learning as their confidence and self esteem is undermined, when they find they can’t keep up or do the work as easily alongside their peers.
    Children of varying abilities clearly need to be treated as individuals and not robots in our failing and very archaic education system.
    If and when a new admission code is created for summer born children it will attempt to re dresss the balance that autumn and spring children have always enjoyed, but I believe much more can still be done to support the next generation so they are fit and ready for the jobs market and higher education.

Leave A Comment