Who do you think we are? Part 1



The headline story to emerge from the 2011 Census data published at the end of January was that Polish is now the second most common language in England and Wales. But buried in the data are a host of insights into the way in which neighbourhoods have changed in the last 10 years.

Data can now be sifted for each of Birmingham’s 40 wards. So, with an hour or two to spare, I thought I would make a start on Handsworth/Handsworth Wood. In time I will post data for three wards – Handsworth Wood, Lozells & East Handsworth and Soho – and for a range of variables. But for now I will deal with just one ward, Handsworth Wood, and two of the most important variables: ethnicity and religion.

Before I get into the numbers, however, a word or two about the variables.

Ethnicity and religion

An ‘ethnic group’ question was first included in the England and Wales census in 1991, with the primary aims of enabling organisations to monitor equal opportunities and anti-discrimination policies and to allocate government resources more effectively. But what an ‘ethnic group’ is and which of them should get their own coveted tick-box on the Census form are contentious issues.

According to author Kenan Malik (The Meaning of Race), Huxley and Haddon first suggested that the category ‘race’ should be replaced by ‘ethnic group’ in their book, We Europeans,’ published in the 1930s. At the time ‘race theory’ and right-wing eugenics were fast approaching their apogee: the triumph of Naziism in Germany and the resulting murder of millions of Jews, Roma and homosexuals. Ethnicity was posited as a way of continuing to discuss and research race without the inevitable political connotations associated with it at the time.

However, as Malik points out, ‘Like race, ethnicity is a term that is used in a fairly promiscuous way, without there ever being a consensus as to its meaning, and there continue to be fierce debates among sociologists and anthropologists as what exactly ethnicity is.’ If there is a general sense pervading these different terms, he adds, it is that ‘race’ describes differences created by imputed biological distinctions, whereas ‘ethnicity’ refers to differences with regard to cultural distinctions.

This uncertainty about ‘ethnicity’ extends to the Office of National Statistics (ONS), the body responsible for the Census. In guidelines published in 2003 it explained, ‘Collecting data on ethnicity is difficult because of the subjective, multi-faceted and changing nature of ethnic identification and there is no consensus on what constitutes an “ethnic group”.’

It is no surprise then that when ONS consults on which questions should appear on the Census form and their specific wording it is bombarded with submissions, many from those seeking their own ‘ethnic group’ tick-box. The main argument of many of these groups is that they are worried about missing out on their rightful allocation of government resources – the very thing the ‘ethnic group’ question aims to facilitate. But in some cases there is another objective: the belief that achieving ‘ethnic group’ status is a step towards broader recognition of a group’s ‘national identity’.

This is why of the 606 responses ONS received to its 2006/07 consultation for the 2011 Census over two-thirds (441) came from ‘experts, community and special interest groups’. Many of these were repetitions. As ONS explains: ‘Some community organisations and special interest groups organised for copies of their response to be sent in by individuals as well, resulting in some duplication. This occurred particularly among organisations representing Sikh, Kashmiri and Cornish interests.’

Despite these write-in campaigns very few changes were made to the ‘ethnic group’ questions on the final Census form and only two new tick-boxes were included. The reasons given for rejecting more changes include lack of available space on the Census form and comparability (the fact that most official bodies want to be able to compare data from 2011 with data from 2001 so they can identify trends that may affect service provision.) In the end it was ‘Gypsy or Irish Traveller’ and ‘Arab’ who were found to be the two groups with the highest priority for their own tick-boxes. The reason? According to ONA, ‘There was user need for both groups in relation to policy development and service delivery. Information could not easily be collected about these groups from write-in options or other questions on the census questionnaire…’

One group who will have been bitterly disappointed at this outcome is Sikhs. Despite a concerted national campaign they failed to get their own tick-box (it is different in Scotland, but that’s another story). One obvious reason for this is that ‘Sikh’ already exists as a choice elsewhere on the Census form: in the question on religious identity or affiliation.

This voluntary question was introduced in 2001 and simply asks, ‘What is your religion?’ This too has its critics for a number of reasons: it is voluntary; it fails to distinguish between the religion a person is brought up in and what, if any, religion they actually practice; and it doesn’t provide tick-boxes for many minority religions that fall outside the main categories (Rastafarianism being a case in Handsworth/Handsworth Wood).

The first of these concerns are reflected in two submissions from Sikh organisations to the 2006/07 consultation and subsequently quoted by ONS:

‘If this question is made mandatory it would help to show the actual number of Sikhs in UK and their distribution. However, under the current system for collating statistics and monitoring, a dedicated Sikh ethnic group box is the only way to ensure protection of Sikh rights as a distinct British ethnic minority.’ (Sikh Education Welfare and Advancement Network UK and Sikh Women’s Alliance)

‘Strictly speaking for Sikhs to have a “religious identity” one needs to be fully practising, so the question may have been misinterpreted by a sizeable number of Sikhs in the 2001 Census. One option to make the question more useful would be to make it mandatory and explain it is simply asking: “With which religion do you associate through your upbringing?” ’ (Akal Sikh Group and Awaze Quam International)

In the event – and for reasons very similar to those of the ‘ethnic group’ question – no changes were made to the question on religious affiliation.

The numbers

So now we know how the data is arrived at what does the 2011 Census tell us about the changes in Handsworth Wood ward in the 10 year period from 2001 to 2011. Here are a few conclusions.

  • The overall population has increased by about 8.5% from 25,559 to 27,749. This is lower than the 9.8% increase for the city as a whole. There are probably several reasons for this increase. Among them cultural factors such as family size, occupancy levels and the conversion of large houses into flats.

  • ‘Asian/Asian British’ remains the largest of the broad ethnic categories and now accounts for over half of the ward’s population: 51.5%, up from 46.5% in 2001. However, there are subtle but significant changes taking place within this category. The percentage of those ticking ‘Indian’ is down from 37.6% in 2001 to 32.6% in 2011, whereas the other four ethnic groups in this category all show growth. The percentage of ‘Other Asian’ is 6.0% (up by 4.1%), ‘Pakistani’ is 8.7% (up by 3.4%) ‘Bangladeshi’ is 3.5% (up by 1.9% ) and Chinese is 0.7% (up by 0.5%).

  • Overall, the White population has declined the most, by just over 10% (from 33.0% to 22.9%). However, there is also significant changes taking place here. The percentage of those from ‘English/Welsh/Scottish/Northern Irish/British’ has fallen by 13.3% (from 29.2% to 15.9%). ‘Irish’ is also down by about 1.0% while those from ‘Other White’ has increased by 4.2% (from 1.6% to 5.8%). This reflects the recent influx of White European immigrants from Eastern Europe.

  • There has been a small increase in the ‘Black/African/Caribbean/Black British’ category, from 16.1% to 16.8%. But here again there are some unexpected trends. The percentage of ‘Caribbean’ residents has fallen from 13.4% to 11.3% whereas the percentage of ‘African’ and ‘Other Black’ have both increased (from 1.0% to 2.9% and from 1.7% to 2.6% respectively).

The figures for religious affiliation further clarify some of the above trends.

  • There has been a 7.3% fall in the number of people identifying themselves as ‘Christian’ although it is still the largest group in the ward at 33.5%.

  • The next largest group is ‘Sikh’ (25.6%) followed by ‘Muslim’ (15.7%), ‘No Religion’ (8.3%), Hindu (7.1%), ‘Religion Not Stated’ (6.4%), ‘Other Religion’ (2.5%), Buddhist and Jewish (both less than 1%). As a percentage of the total all groups have declined except for ‘Muslim’, ‘Sikh’, ‘Other Religion’ and ‘No Religion’, which have grown by 7.9%, 0.6%, 2.0% and 1.0% respectively in the 10-year period. The largest increase, which is for ‘Muslim’, reflects among other things the growth in the number of Pakistani and Bangladeshi residents.

Coming soon: Lozells & East Handsworth