The National Association of British Arabs
REPORT ON THE 2011 CENSUS – MAY 2013
Arabs and Arab League Population in the UK
There had long been a recognised absence of accurate data on the number of Arabs living in the UK, together with numbers of first generation Arabs and mixed Arabs who identified themselves as such in the Census. This absence has become increasingly frustrating to community groups, to services being developed for Arabs living in the UK and to researchers who are now frequently addressing questions of Arab immigration and the views of first generation Arab and mixed Arab populations in relation to their opinions both on their new homeland and on the political situation in their original homelands.
This report has been prepared by the National Association of British Arabs and represents the work it has done with the Office of National Statistics in relation to the 2011 Census. A preliminary report was sent out recently by NABA; this is the second report which combines the first with additional information which has come out since then. It is not the final report as it does not include statistics from Scotland and Northern Ireland where the structure of the Census was different and the results will therefore reflect that.
The report is broken down into four sections. The first section deals with NABA’s involvement with the Office of National Statistics and the 2011 Census, the reasons for NABA’s involvement and the way in which information is traditionally collected; the second section highlights the methodology used by the Office of National Statistics for the 2011 Census; the third section gives the results based on the numbers of individuals who had ticked the newly introduced ‘Arab’ box of the Ethnic Origin category of the Census and the fourth section gives the results based on the country of origin and passports.
The only previous study of the numbers of Arabs as an ethnic
identity had been done by the Greater London Authority in 2005(Ref: 1).
This sought to enumerate the numbers of Arabs living in the capital only and using
the countries making up the Arab League (Appendix 1) as taken from the Census
(2001) (see full report on NABA website). This research has been the most
comprehensive to date and it was estimated that there were the following
numbers of ethnic Arabs living in the capital:-
Male 55,884 52.3 48.4
Female 50,949 47.7 51.6
A further report was prepared by the Atlantic Forum organisation in 2009. This was not a demographic study but, rather, sought the views of 355 Arabs living in Britain in relation to identify, politics and community (Ref: 2).
NABA’s role in 2011 Census
The issue of the size of the Arab population in Britain has been addressed by the National Association of British Arabs (Ref: 3) who lobbied the government since 2004 for recognition of Arab as part of the Ethnic Profiling used widely in all government/non-governmental documents. These figures were considered essential in order to influence government policies in health, education, discrimination, migration etc and also to deliver services that are appropriate to the needs of people from different communities and cultures. Their absence, particularly in relation to health and discrimination, was a serious obstacle to gauging many problems, including the numbers of victims of violence and specific health problems which were not identified accurately because of the lack of ‘Arab’ as an ethnic identity on either health or police profiles. It was well recognised that there was also a reluctance to use the existing facility of the ‘Other’ ethnic profile box as it was felt that this was both demeaning and irrelevant as a method of recoding ethnic origin.
In the absence of any concrete help from government bodies in identifying the Arab population in the UK for purpose of public services etc, NABA eventually approached the Office of National Statistics (ONS) to seek their assistance in gaining recognition of ‘Arab’ as a separate ethnic identity. This was done on the basis that ethnic profiles were taken directly from Census statistics. The ONS kindly allowed NABA representatives to join the Diversity Group preparing for the 2011 Census and NABA has worked with the ONS since 2005 to this end.
A census is held every 10 years in the UK (England, Wales, Scotland and Northern Ireland); this report deals with the results from England and Scotland. It is mandatory for everyone to complete a census form and fines are levied against non-responders. The response rate has been 94% for both the 2001 and the 2011 Censuses.[i] In drawing up the Census, the Office of National Statistics approaches and involves a variety of stakeholders to take part in the preparation of forthcoming censuses in order that it addresses as many concerns as possible.
According to a Census Stakeholders Consultation held in 2006/07, more than 80 per cent of all respondents stated that they had a requirement for an ethnic group question in the forthcoming 2011 Census. The reasoning behind such a request includes:
· To provide public bodies with a better understanding of the communities they serve and hence inform service provision;
· To enable organisations to meet their statutory obligations to the Race Relations Act of 1976 and Race Relations (Amendment) of 2000 where other sources are inadequate to provide sufficient data for small, geographically dispersed ethnic populations (Appendix 2);
· To inform policy development and monitoring, including the needs of vulnerable and disadvantaged communities.
Accurate, reliable data about ethnic groups, language and religion is an essential part of the planning process to help ensure equality of outcomes for all diverse communities.
However, there is only limited space available on the Census form which means that some communities and groups will have tick boxes whilst others must write in their answers in the ‘Other’ box. A key aim of the ONS consultation was to ensure that the questions asked were ‘acceptable to the greatest number of people from all communities and groups’. It has been found that response rates will be reduced if the questions are not acceptable, or people do not understand why they are asked.
In relation to information gathering, the Census approached the subject from several different perspectives based on the understanding that whilst many people considered themselves as fitting into one category, they should also be offered the opportunity to respond to all elements which they felt relevant to them. The ONS therefore revised the Census Form to allow maximum flexibility for all respondents.
The items covered in the 2011 Census were therefore:-
· Ethnicity[ii] – in this, respondents were able to tick one or more of a variety of boxes, including the newly agreed specific box for ‘Arab’, which they felt reflected their ethnicity. In addition, or instead of, they could go for a ‘write in’ response.
· Country of Birth – this was specific to the country of birth but, as will be seen later, there is no correlation between ethnicity and country of birth ie somebody might be born in an Arab country but not be ethnically Arab.
· National Identity – this allowed respondents to express what they felt their national identity to be regardless of any other nationality they held ie they could hold British nationality but at the same time hold an additional national identity which they saw as their original.
· Passports held – this allowed respondents to write in any primary or additional passports held. It should be remembered that the UK allows dual nationalities.
[i] The response rate is the total number of usual residents whose details were completed on a returned questionnaire, divided by the estimate of the total number of usual residents. Prior to the census, ONS set a headline target of achieving 94 per cent response for England and Wales overall.
In 2011 the UK Census questionnaire was sent out to every home and communal establishment in the UK. As with any self completion questionnaires, respondents sometimes make errors when recording their answers, resulting in data that were not valid for estimation and analysis. This is called 'item non-response'.
Reasons for not answering correctly can be unintentional, for example where a respondent misses a question or thinks they can tick more than one option, or intentional where a respondent either does not know the answer or does not want to provide the answer.
Some correctly recorded values were considered to be invalid because they were inconsistent either with other values on the questionnaire, or with auxiliary information or definitions. For example, if a person gave their age as ten years old and gave an occupation these values were considered to be inconsistent because by definition a ten year old cannot be in employment. This is referred to as 'item Inconsistency'.
The 2011 Census data were edited in two stages to correct for item non-response and inconsistencies. A limited number of edits were applied, followed by item imputation which corrected for non-response and inconsistencies. (Taken from Office of National Statistics ‘Response and Imputation Rates’) http://www.ons.gov.uk/ons/guide-method/census/2011/census-data/2011-census-user-guide/quality-and-methods/quality/quality-measures/response-and-imputation-rates/index.html.
The Office of National Statistics (ONS) Classification of
ethnic groups determined by the 2001 Census comprised seventeen categories
under five broad headings:
· White (White British, White Irish and any other White background)
· Asian or Asian British (Bangladeshi, Indian, Pakistani and any other Asian background)
· Black or Black British (African, Caribbean and another Black background)
· Mixed and Other have been combined (Chinese, all mixed groups and any other ethnic group)
· ‘Not Stated’(allowed for those who felt they did not fit into any of the above categories).
Given the range of inward migration to the UK, these categories were considered to be inadequate for the 2011 Census.
One example of this was that the ONS found that between 2005 and 2008, the proportion of records with ‘Not Stated’ values was an important measure of the usefulness and quality of the data. In England and Wales, 9.6% of live birth records had ethnicity recorded as ‘No Stated’. This category was not evenly distributed across local areas, varying from zero to 98%. This would signify that there were; (a) a number of people who did not find their own ethnicity recognised as part of the profiling, together with (b) a number of people of objected to responding to this question. The non-responders and those “Not Stated” are from a mixture of nationalities.
The East Midlands region has the largest proportions of ‘Not Stated’ with Leicester at 98%. In relation to infant deaths, the percentage of ‘Not Stated’ ethnicity was 10.5% for England and Wales ranging from 3.4% (North East) to 32.5% (East Midlands).
As an example of the need for such monitoring, gestational age and ethnicity are not routinely collected at birth registration but are important factors in understanding births and infant deaths. Gestational age is collected by the NHS Birth Notifications system as it is an important determinant of healthy outcome for babies. The NHS Birth Notifications system collects ethnicity to assist with organisations’ monitoring of their service delivery. Whilst ethnicity is usually self-defined, the baby’s ethnic group as stated by the mother is recorded in the system. Individuals may choose not to state ethnicity. In local areas with a very high proportion of ‘Not Stated’ records this ‘opting out’ may not be the sole reason. The ‘Not Stated’ response category for ethnicity may also indicate non-responses of ‘Not Known’, ‘Missing’, or ‘Not Asked’.
The results of the 2011 Census have been published in batches and can be found on the Office of National Statistics website (http://www.ons.gov.uk/ons/guide-method/census/2011/index.html). It should be noted that the Census results are for England and Wales only; Scotland and Northern Ireland publish separate results and do not have the same categories as those of England and Wales.
To date the majority of published results have related to the numbers of population in each district, numbers of second homes, income, etc. However, over the last few months results on ethnicity and national identity have come through. NABA are analysing the results as and when they come out and will publish them on the website. It is not anticipated that all results will be available until the end of 2013/early 2014.
The ONS examines the range of outputs that can be cross-tabulated, for example analysing ethnicity by language, country of birth, religion and national identity.
National identity information is used to measure identification within the different countries of the UK, as well as nations outside the UK. It was not included in the 2001 Census, but there was strong demand for information after 2001. When the national identity question was introduced, there was some evidence to show that British-born people from ethnic minority groups were more likely to answer the ethnic group question if a national identity question was asked first, as it still allowed them to state that they were British. This suggests it has a positive effect on the response to the ethnic group question.
As can be seen from Appendix (3), there were a number of respondents, whom we assume to be from mixed marriages (whether these occurred in the UK or in Arab countries), whose ‘written in’ responses reveal a wide variation in how they viewed themselves. Whilst we have the descriptions, we do not have the numbers in each category. Neither are we able to determine whether they have been included in the figures above (Appendices 4a, 4b, and 5).
The figures released show that the number of Arabs in England and Wales who chose to categorise themselves as ethnically Arab either by ticking the ‘Arab’ box or by writing in responses are as given in Table 1:
As can be seen, many people responded to the question in terms of a refined categorisation, and specifically those Arabs from North Africa and Somalia. We cannot be sure how many of the individual identities have also responded in the general ‘Arab’ box so there may well be some overlap in these figures. Neither do we know how many first generation Arabs are within these figures.
The figure of 366,769, and more specifically the figure of 240,545, was lower than we had anticipated as an overall figure of Arabs in the UK.
We therefore looked at the latest tranche of results from the Census (published 26 March 2013) which dealt with countries of origin. As can be seen, the numbers who have identified themselves as being born in Arab countries comes to a total of 404,207 individuals. This figure alone is in excess of the numbers of those identifying themselves as Arab/North African and the difference (37,439) could well be those who do not identify themselves as Arabs but were born within Arab countries.
However, as explained above, it has not been possible to get more definitive figures for the numbers of ‘mixed Arabs’ and therefore we believe it is true to say that the figures of 366,769 and 404,207 are not the most accurate figures available if we are to include ‘mixed Arabs’ in England and Wales although they do give us a ball-park figure. We must also bear in mind that we still do not have figures from either Scotland or Northern Ireland. As their Census did not include the ethnic category ‘Arab’, we must assume that we can only gain information if people have either written in their responses or have identified a country of birth from one of the Arab League countries. In relation to the latter however, we must bear in mind that these figures will also include some numbers who do not consider themselves as Arabs.
Country of Birth
Appendix 6 deals with the distribution of those born in Arab countries throughout England and Wales. It is not possible to know whether the respondents are all permanent residents of the UK. The figures however are a fair representation of the spread of Arab countries represented in the UK. There were no respondents from Comoros, Djibouti or Mauritania. (Figure 1)
These results detail the ‘write in’ individual responses which reflect what respondents felt to be their national identity. It is interesting to note from this section that very few of the Arab countries are actually represented in this section, indeed only Iraqi and Somali respondents appear to have filled in this section in addition to non-detailed ‘North African’ (Appendix 7). This table has been included although it is difficult to gain accurate information from it as the categories are so wide eg ‘Middle East and Asian’.
The findings outlined above are duplicated in those responders who wrote in the type of passports they held; again only Iraqi and Somali were detailed in addition to non-detailed North African. It could be that respondents were not aware that they were entitled to respond with more than 1 answer here as dual nationality is acceptable in the UK. (Appendix 8)
Figure 1 Numbers of Persons Born in the Middle East, North Africa and Somalia – England and Wales
(Reproduced from Table QS213EW of the 2011 Census Results)
t should be noted that a number of responders entered their
nationality so the numbers of those ticking the specific ‘Arab’ box may or may
not include those who responded with a ‘write-in’ response with their
nationality. We would re-emphasise that the figures are not therefore
definitive. Neither can we compare the figures with the 2001 Census when there
was not a specific response for Arab ethnicity. It is believed that many
British Arabs ticked one of the ‘Other’ boxes in the 2001 Census.
Whilst the Census strove to be as comprehensive as possible and to give the maximum amount of opportunity for respondents to identify themselves by ethnicity, national identity and passport, the results show a limitation of response from many respondents, particularly in relation to the question of ‘National Identity’ and ‘Passports Held’. This could be for a variety of reasons and might include the following:
1. Misunderstanding of the availability and acceptability of identifying themselves in several different ways without contradiction.
2. A reluctance to identify themselves too closely with their countries of origin.
3. Fear of admitting that they held more than one passport although this is legal in the UK.
It would appear however that as a brief to date we can conclude that:
1. Arabs make up approximately 0.4% of the population of England and Wales.
2. Numbers counted for the ‘Arab’ tick box are approximately 240,000. However it should be noted that this does not necessarily include those who have put in specific nationalities (see above).
3. The numbers of people bornin Arab countries living in England and Wales is 404,207 representing 0.5% of the total population. This does not include those born in England and Wales or outside the UK who consider themselves as Arabs.
4. London is the most ethnically diverse area with the highest proportion of minority groups. Two London boroughs had the highest proportion of people who identified as Arab, Westminster at 7.2% and Kensington & Chelsea at 4.1%.
5. Arabic is now the 7th most commonly spoken language in England & Wales with 159,000 people entering this category making it spoken by 0.3% of the population.
What we have not been able to ascertain to date is:
NABA is still awaiting the relevant results from Northern Ireland and Scotland. However we believe that the figures contained in this Report represent a relatively clear picture of the Arab population of the UK in 2011.
NABA would like to acknowledge and thank the Office of National Statistics for their considerable help in this work and for their ongoing support which has made this report possible.
Mrs M Jalili (Cllr)
National Association of British Arabs (The Secretariat)
2. Yafaie Y. British Arabs: Identity, Politics and Community. Results of an Exploratory Survey by the Atlantic Forum, Spring 2009. http://www.naba.org.uk/reports/BritishArabs_YafaiF.pdf
3. National Association of British Arabs. A study for consideration of inclusion of ‘Arab’ as an ethnic group on future census returns. November 2004. http://www.naba.org.uk/reports/arabPopUK_04.htm
We have included within this paper downloads of results to
date. These include: