Where did all the children go?
The 2011 census data has been coming out in drips and drabs and I thought I would continue my exploration of demographics of the East End with a look at population and age. What I found was a slightly odd lack of school age children.
The census data I’m using (from the GLA) is reported at a ward level and is broken down into bands of five years and if plotted we get the chart below. For reference the data here described as Wapping relates to the ward of St Katharine’s and Wapping. What is interesting is that 6% of the Wapping population is age 0-4, but that this drops to 3-4% for the ages of 5-19 before increasing suddenly to 8% for those age 20-24. This is an interesting observation as it goes directly to the need for schools in the area.
A few months ago I was at a Network Wapping meeting and some of the founders of Wapping High School were there and introduced me to the term ‘Nappy Valley’. I had never heard this term used before, but was told that it was in common parlance amongst those with children to describe families leaving Wapping once their child approached school age. Whether this exodus is a factor of the environment, the availability of school places, or just the size of properties that causes this phenomenon, it is very apparent effect. That is of course unless we assume that fertility shot up massively in 2006.
Clearly the next question is, does this indicate that building a high school would lead to more families staying in the area, or is this proof we don’t need a school because families would leave anyway? The data doesn’t provide answers, but as I trawl through more data sets, the picture may become clearer.
I’ve plotted Wapping and the rest of LBTH (to remove any influence from Wapping) on the chart below.
If we look at the right side of the graph, there is, visually at least, a generally strong correlation between Wapping and LBTH. However, if we look at the frequency distribution for ages groups under 25, we can see that the are a disproportionately small number of children in Wapping, around 30-40% fewer.
You can still see a dip in LBTH, but nowhere near as large as the effect in Wapping.
I’m not sure though about the suddent increase in the proportion of the population once we reach the age range 20-24 and the difference between Wapping and LBTH. I am assuming that in both Wapping and LBTH there is an influx of graduates (this may be spurious), but I’m not sure why it would be higher in the rest of LBTH than Wapping- perhaps Wapping is too expensive, or has relatively few properties suitable for house shares, or that young folk are locating near trendy Brick Lane or close to Canary Wharf.
The two trends identified-children disappearing and a large number of young professionals are particularly evident when you compare Wapping to London as a whole. We can see there is a London-wide dip after the age of 5 but nowhere near as great as experienced in Wapping. Similarly, the post university influx is not as steep. We can also see on the right hand side of the graph that the older population in Wapping has a similar structure to London as a whole.
Comparing LBTH to London, we can see that LBTH is a young borough compared to London, with this difference driven by a large population aged 20-34. What is clear here is that LBTH is very similar to London for the proportion of the population made up by children. We can see that there is a dip after the age of 5 but nowhere near as much as in Wapping.
For completeness Wapping, LBTH and London are shown together below, which highlights that the ‘Nappy Valley’ effect is not present in LBTH or London as a whole, but that a spike in young adults is evident in both Wapping and LBTH but not the rest of London.
But is it statistically different?
I wondered whether or not these perceived differences were statistically significant or not and for the first time since I wrote my dissertation (where I famously proved that 18th century indentured mill apprentices were more likely to come from illiterate families than the population at large) I have performed a statistical significant test.
In conducting the test I used a chi square test as a measure of goodness of fit as I don’t have the raw data to calculate variance etc and looked at the distribution of all age bands.
Comparing the population distribution of Wapping to both LBTH and London as a whole gave a chi-square result of 1040 and 1902 respectively, which with 18 degrees of freedom, gives a p- value of zero (well, unless you round to over 200 decimal places). Essentially this means there is no probability that these differences occurred through chance. Interestingly I ran the test for LBTH as a whole against London excluding LBTH and achieved a chi-square value of 1985, indicating that LBTH is more different to London than Wapping on its own is.
However, because I was looking at differences across all age bands, we aren’t capturing the specific effect I am interest in, the ‘Nappy Valley’ effect. So rather than comparing each age category, I prepared the data into two categories: School age and not School age (school age defined as 5-19) and re-tested the significance. Again, the p value was zero, indicating that the difference is definitely significant.
I had a quick look for a national data set, though I couldn’t find one with 5 year bands, so my chart below is a bit more lumpy. What we can see is that Wapping is far more heavily populated with young professionals and a third to half as many children as might be expected. Also of interest is the 65-74 category- Wapping doesn’t appear to have an appropriate number of fresh retirees, suggesting a departure to the country pile on retirement?
I don’t know where the kids have gone and whether the logic underlying the Nappy Valley moniker is true, but I sincerely hope Wapping isn’t being preyed upon by a pied piper or worse, the Chity Chitty Bang Bang child catcher!