Ethnicity and the five-year-old
It is sensitive territory but, if we are serious about trying to offer equality of opportunity to all in our multi-racial society, it strikes me as important to venture there.
By the age of five, ethnicity already appears to have a significant impact on a child's health, behaviour and cognitive abilities.
The latest batch of data from the (MCS) has been published following the lives of more than 15,000 children who were born in the two years after the millennium.
But it also involves their families. And through the reams of data we begin to get a picture of how different people live.
The MCS - set up by the Economic and Social Research Council and co-funded by government - is a remarkable piece of longitudinal research that has already published data on children of the new millennium aged nine months and three years.
In the latest batch of data, the comparisons which jumped out at me relate to five-year-olds from different ethnic backgrounds as they begin primary school - disparities which mean some groups are always playing catch up.
Separate figures show that at GCSE some (but not all) of the minorities are still performing less well.
On the five-year-old's first day in reception class, the cohort study already found "substantial differences in adjustment...between ethnic groups".
It may feel uncomfortable, but identifying areas of weakness means support and resources can be targeted appropriately.
One aspect the study looks at is problem behaviour among our 21st century five-year-olds.
It is measured by asking parents whether their child has conduct problems, hyperactivity, emotional symptoms, peer problems and so on. The higher the score, the more problems reported.
PROBLEM BEHAVIOUR
Indian: 6.5
Black African: 6.5
White: 6.6
Mixed: 7.4
Other: 8
Black Caribbean: 8.2
Bangladeshi: 8.2
Pakistani: 9.2
Source: Millennium Cohort Study
A separate measure of a child's cognitive abilities also finds marked ethnic variation. The British Ability Scales (BAS) measures a five-year-old's facility at naming vocabulary, picture similarities and pattern construction. The higher the score, the greater the cognitive development.
COGNITIVE DEVELOPMENT
White: 102.2
Mixed: 101.3
Indian: 98.3
Black Caribbean: 98
Other 95.7
Black African 91.4
Bangladeshi 88.6
Pakistani 87.4
Source: Millennium Cohort Study
"Bangladeshi and Pakistani children appear to show lower levels of cognitive ability at school entry, and their mothers report more behaviour problems", the researchers note, although they are quick to insert important health warnings.
"Data on minority groups are subject to greater sampling error", they say.
"The response rates of some non-white groups, cultural expectations, the importance of other languages spoken in the home, and the number of siblings are just some of the factors that have to be considered."
We do have to be careful, certainly before assuming cause. Poverty, parental education, language may all play a role. But the findings are so consistent that it appears we are looking at something real.
For instance, the Pakistani and Bangladeshi children are behind in their physical development, creative development and mathematical development too.
One possible factor which emerges from the data is the role of parents in different ethnic groups.
Take the role of fathers. On average,1.3% of dads say they never play with toys or games with their five-year-old child. The figure for Pakistanis and Bangladeshis is 9%.
The study finds, on average, 2.5% of dads never play sport with their five-year-old. Among Pakistanis and Bangladeshis it is almost 12%. (With Indian fathers it is 2.4%).
A big difference is found among mothers too. Just 1% of mums never read to their child. Among Pakistanis and Bangladeshis, the figures is 7.1%.
Illiteracy rates may play big a part in this, as may differences in the way ethnic groups answer such questions.
Or the explanation may simply be that different cultures rear children in different ways.
It is not just in their mental development that we see big variation between ethnic groups. There are already apparent health disparities among the five-year-olds arriving at school.
For example, black children are three times as likely to be obese as a classmate from an Indian background and twice as likely as a white youngster.
OBESITY
Average: 5.3%
Indian: 4.1%
Other: 4.8%
White: 4.9%
Mixed: 7.1%
Pakistani/Bangladeshi: 8%
Black: 12.5%
Source: Millennium Cohort Study
The data does not distinguish between black children of African or Caribbean origin but the researchers conclude that "dietary differences are likely to be important here, and genetics may also play a role".
Indeed, there is evidence of a generational legacy.
Like their children, Afro-Caribbean mothers are almost three times as likely to be obese as an Indian mother.
These findings suggest public health campaigns might benefit from further targeting their effort towards different ethnic groups.
Other health issues affect ethnic groups in different ways.
The cohort parents were asked whether their five-year-old had any longstanding health problem. On this measure, it is white and mixed race youngsters who are reported to have the greatest level of chronic illness by their parents.
LONGSTANDING HEALTH PROBLEM
Average: 19.5%
Indian: 13%
Black: 14.5%
Pakistani/Bangladeshi: 16%
Other: 18.3%
White: 19.8%
Mixed: 23.9%
Source: Millennium Cohort Study
Why? Well, one important factor may be the proportion of mums who smoke.
MOTHERS WHO SMOKE
Bangladeshi: 4.1%
Indian: 5.2%
Pakistani: 7.5%
Black African: 7.8%
Other: 11.3%
White: 29.8%
Black Caribbean: 34%
Mixed: 38%
Source: Millennium Cohort Study
One sort of knows it, but when you look at the bald statistics it is quite startling to see the difference between ethnic groups.
Seven times as many white five-year-olds have mothers who smoke compared to their Bangladeshi class-mates.
I was also struck by the difference in smoking rates between black African and Caribbean mothers.
The researchers conclude that the relatively low level of longstanding health problems reported among Asian children "may reflect their lesser exposure to smoking in pregnancy and the higher rate at which they were breastfed in infancy".
Everyone has possible lessons to learn from the data.
A final observation: the differences in how ethnic groups regard themselves as parents. Asked how they felt they did as a parent, here are the proportions for mums who thought they did a "very good" job.
VERY GOOD MOTHER
White: 29.7%
Mixed: 33.6%
Indian: 39.3%
Other 42.3%
Pakistani / Bangladeshi: 47.8%
Black: 50.4%
Source: Millennium Cohort Study
And dads who thought they did a great job?
VERY GOOD FATHER
White: 34.2%
Mixed: 36.7%
Indian: 45.2%
Pakistani / Bangladeshi: 47.7%
Other 51%
Black: 55.5%
Source: Millennium Cohort Study
This is, I think, a warning we should heed before anyone leaps in and lectures parents from whatever ethnic background on how to bring up their children. No-one has a monopoly on wisdom.
I'm 
Comments
or to comment.