The Independent and Sex Selective Abortion: Not Even Wrong

According to The Independent:

Pressure is growing for the Government to curb the way that ultrasound scans can be used to inform pregnant women about the gender of their unborn babies following an investigation by The Independent suggesting that female foetuses are more likely to be aborted within certain ethnic groups living in Britain.

But how do we know that this alleged pressure is indeed growing?

Well, its because The Independent has published articles on this particular issue for three days running, and in the parallel universe that newspaper editors and proprietors inhabit that’s all the proof they need. No newspaper will ever willingly admit, after flipping over into campaign mode, that their efforts to stir up a shit storm have been a complete waste of everyone’s time, including their own. The pressure must be growing because the newspaper says so and that’s an end to it.

You’ll notice, of course, that The Independent refers here to its own ‘investigation’ which, so it claims, suggests that female foetuses are more likely to be aborted in some ethnic groups living in the UK. In response to this repeated use of the word ‘investigation’ I can do no better than to quote the great Inigo Montoya:

You keep using this word. I do not think it means what you think it means.

If you’re new to this particular story then I should point out that I personally took a very close look at the data that The Independent used to carry out its ‘investigation’ and very quickly realised that the statistical methods that newspaper adopted are not just wrong, they are not even wrong. The data set that the newspaper obtained from 2011 Census data compiled by the Office for National Statistics is wholly inappropriate for the purpose to which it was put. One simply cannot reliably identify anomalous variations in male-female sex ratios at birth from population data on families with dependant children. There are far too many variables and too many potential sources of confounding, particularly when looking at data relating to families where one or both parents are first generation migrants, to link any apparent anomalies in the male-female sex ratio in that data set back to the use of sex selective abortion even if you obtain all the data necessary to make an accurate assessment of current sex ratios in the migrant populations you’re interested in, but the newspaper didn’t even manage to do that.

One of the key assertions that the newspaper made in its first article on this subject, that there were some sex ratio anomalies that persisted in the data across families of all sizes in which the mother was born in either Pakistan, Bangladesh and Afghanistan rest squarely on set of calculations that simply cannot be performed using the data set that the newspaper was working with. In fact it’s not even possible to calculate the male-female sex ratio for all families included in the data set let alone for all families sizes, such was the incompetence of the journalist who requested the data using the Freedom of Information Act.

What the newspaper has carried out here is not an investigation at all, in fact the best description I’ve come across for The Independent’s efforts came from a Twitter follower who described it as ‘amateur stat mangling’.

As for the newspaper’s claim that the Department of Health has launched an investigation based on their own ‘investigation’, here’s what the Department of Health had to say when it was contacted by the newspaper:

“We can confirm that officials have been in touch with Imperial College. We continue to monitor birth ratios on a yearly basis and will consider any new studies as they are published.”

That’s not really an investigation – again, The Independent keep using that word without demonstrating any concrete understanding of which it actually means.

What the DoH has actually said here amounts to nothing much more than ‘Okay, send us your figures and we’ll have a look’, and from my own past experience of working with medical statisticians in a local public health department I can be pretty confident then when the newspaper’s data arrives what will follow will be one of the shortest ‘investigations’ in the history of the Department. Medical statisticians are no mugs and will have no more difficulty in spotting the numerous manifest flaws in the newspaper’s methods than I experienced, which is to say none at all.

Given the length of my main article looking at the newspaper’s data there is one specific issue I highlighted which some people may not have picked up because it’s towards the very end of the article, making it one that’s worth reiterating in this shorter piece.

It is, of course, obvious that the three migrant communities that the Independent most clearly identified as exhibiting, in their view, suspicious-looking sex-ratios, those originating Pakistan, Bangladesh and Afghanistan, have one thing in common. They are all primarily Muslim communities.

They also have something else in common, which is much less obvious. None of these migrant communities, nor the countries from which they originate, are amongst those communities/countries where there is, or even has been, any tangible evidence that sex selective abortions are carried out.

The countries where we know this to be, if have been, an issue are China and India – and in both of these countries the current male-female sex ratio at birth is estimated to be around 1,120 males to 1,000 females – and South Korea, where the current ratio is around 1,070 males to 1,000 females, which is within what is generally assumed to be the natural range for human populations of 1,030-1,080 males to 1,000 but only because of a relatively recent fall in the numbers of male births in that country. By way of comparison, Pakistan and Afghanistan have current male-female sex ratios at birth of around 1,050 males to 1,000 females, which is near enough identical to that of the UK (1,052 males to 1,000 females) and for Bangladesh the current male-female ratio at birth is lower, at 1,040 to 1,000, than our own.

There has never, beyond perhaps the occasional unverified anecdotal claim, been any clear reason to suspect that sex selective abortion is used within any of The Independent’s three target communities beyond the fact that members of those communities do happen share a marked physical resemblance to members of migrant communities originating from India, where this practice is known to take place – and when I say ‘marked physical resemblance’ I mean, of course, that they all, for the most part, have brown skin, which has traditionally been more than enough for some people to treat this very different nationalities and cultures as if they all the same.

Now, when I was looking at the Independent’s data, such as it is, one thing that I did notice was that, in terms of overall size and composition, the data provided for first generation migrant families from Afghanistan was actually very similar in most respects to that provided for first generation migrant families from China, not least in terms of their apparent male-female sex ratios where these could be calculated. Yes, the data set itself is entirely worthless but one can at least see from it whether or not there were any other migrant populations with very similar looking ‘anomalous’ sex ratios to those that the newspaper made a point of mentioning in their article that didn’t get the same treatment and this is precisely what has happened here with the data for families with parents born in Afghanistan and China.

Afghani families are singled out by the newspaper for allegedly having suspicious looking sex ratios across, supposedly, all family sizes but Chinese families aren’t, even though the apparent male-female sex ratios in the dataset for both group are not only very similar but, in some, skewed even further towards males in the figures for Chinese families.

I know some people will look at that and think ‘Hmm… Islamophobia..?’ – that, I think, is unlikely.

I suspect that it’s rather more a matter of The Independent seeking to cynically exploit an existing prejudicial climate to lend credibility to their story far in excess of anything it might actually merit or, to put it in much more straightforward manner, the newspaper knows perfectly well that much larger proportion of the British public will be inclined to buy into the idea that Muslim communities may be engaging in morally and legally dubious practice than would the case were they to point the finger at the Chinese community, a position that is helped along to a considerable extent by public awareness of the fact the practice of sex selective abortion in China is, to a very large extent, driven by the Chinese state policy on population control.

The default assumption there is, of course, that the current situation in China is artificially generated by country’s ‘one child policy’ and that the current  imbalance in the male-female sex ratio would therefore not exist were Chinese families permitted to have as many children as they wanted, hence there would be not reason to suppose that a similar imbalance in the male-female sex ratio would be found in Chinese families living elsewhere in the world, well beyond the reach of the Chinese state. However, in the case of Indian and other South Asian communities the assumption that much of the public is likely to make is that the imbalance in the male-female sex ratio evident in India – but only in India, as there is no evidence of such an imbalance for Pakistan, Bangladesh or Afghanistan – is purely down to cultural values that migrant communities from that part of the world will bring with them when they come to the UK.

In reality, of course, it’s nothing like that simple. The preference for male children over female children we see in China is still a cultural preference for all that it is one that is currently being activated by state policy to the obvious detriment of female children, so there is no reason for us to suppose that first generation Chinese migrants entering the UK are any less likely to bring with them that cultural preference for male children than first generation migrants from the Indian sub-continent.

That being the case, The Independent’s decision to point the finger at first generation Afghani migrant families but not at Chinese families of the same ‘vintage’ is perhaps the only genuine anomaly in this entire story that actually merits further investigation. It is not the Department of Health or NHS that has awkward question to answer, rather it is the journalist who wrote up the story and the editorial staff at the newspaper who passed it for publication.

Ultimately we are back in a place where we have been several times before. This is by no means the first occasion on which irresponsible media reporting has stirred up a short-lived moral panic surrounding the practice of sex selective abortion nor indeed is this the first occasion on which the response to such stories has been a call for the government to impose some sort of ban on informing women/families of the sex of a foetus, if it can be reliably identified during the course of a routine ultrasound scan, for all that such a ban would, in all likelihood, have absolutely no impact whatsoever on the practice itself.

If you ban NHS doctors, midwives and radiographers from informing women/families of the sex of a foetus following an ultrasound scan, which the government could do without the need to enact any specific legislation, then any women/families that are absolutely intent on obtaining this information, for whatever reason, will go off and obtain the exact service they require from the private sector, and if you legislate to prevent the private sector offering that service then there’s no great shortage of private clinics on the continent for these women/families to turn to.

None of this is to say, of course, that sex selective abortion never happen here in the UK and that there is simply no problem here to address at all. Rather it is to point out that the problem of sex selective abortion, such as it does exist, is extremely small in scale relative to the annual number of births, and indeed abortions, carried out every year in the UK and that it is not an issue that can be readily, let alone effectively, addressed by making changes to abortion law or issuing new practice guidelines to doctors or, least of all, by creating a culture and climate amongst health care professionals in which pregnant South Asian women and their families are routinely treated as objects of suspicion.

It’s time to cut through the bullshit and call things for what they really are, this is not ultimately about abortion at all, it is about domestic abuse. Forcing any woman to have an abortion against her wishes, for whatever reason, is a form of domestic violence, and the way you tackle that is the way you tackle any other type of domestic abuse, by getting your hands dirty and putting in the work at the grass roots level to identify and support victims, to give women facing pressure from a partner or from family a safe route out of that situation and, by working within communities to challenge and, ultimately. change the culture which permits such abuses to happen.

As for The Independent’s risible ‘investigation’, its statistical methods and its proposed ‘solution’, these are, as the great Wolfgang Pauli might have observed, not even wrong.

  • JuliaM

    “…then any women/families that are absolutely intent on obtaining this information, for whatever reason, will go off and obtain the exact service they require from the private sector…”

    I son’t think the main problem with Rajendra Kale’s bizarre suggestion is actually the ease with which it could be circumvented – it’s more a case of just why should all women be ‘punished’ by having to forgo a medical advancement simply because a few backward cultures would like to misuse it…

  • Durga

    Sex selection in China is actually illegan, and where it happens is certainly not ‘driven by the Chinese state policy on population control’.

    Like most abortions in the UK sex selective abortions (when they occur) are allowed through legal loopholes created by vague wording (technically only ‘special exceptions’ are allowed in the UK). If those loopholes were tightened up it wold prevent widespread abuse of the abortion law and return it to status as a ‘third line of defence’ only in cases such as genuine mental suffering (ie, diagnosed by a professional, not simply “I don’t feel ready yet”).