2011 Census: Detailed analysis - English language proficiency in England and Wales, Main language and general health characteristics

• English¹ was the main language for 92 per cent (49.8 million) of usual residents aged three and over. Of the remaining eight per cent (4.2 million), who had a different main language, the majority (3.3 million) could speak English well or very well

• Of the top 10 main spoken languages other than English¹ with the highest proportions who could speak English well or very well, four were native to countries where English was an official language, and the remaining six were Nordic or Germanic and native to countries where English language learning is compulsory at school

• Less than 5 per cent of the population aged 3 to 15 in all local authorities had a main language other than English¹ and could not speak English well or at all. The five local authorities with the highest proportions were all in London

• Around 300,000 residents aged 3 and over in England and Wales had ‘Not Good’ general health and could not speak English well or at all

• People with a main language other than English¹ who could not speak English well or at all had a lower proportion of ‘Good’ general health (65 per cent) than those with English¹ as their main language (80 per cent), or those with a main language other than English¹ who spoke English well or very well (88 per cent)

Notes for key points

1. English or Welsh in Wales.

Detailed characteristics for demography and families for MSOAs and wards can be found on the Nomis website.

Two questions⁴ on main language and proficiency in spoken English were included for the first time in the 2011 Census following the content consultation for the England and Wales census form. The questions provided information on the main language of UK residents, and their proficiency in English if English¹ was not their main language. For residents who lived in Wales, the English¹ category included those whose main language was Welsh, in order to acknowledge that Welsh and English are both official languages of Wales.

Question 19⁴ on the census form offered respondents the choice to classify their proficiency as being able to speak English ‘very well’, ‘well’, ‘not well’ or ‘not at all’.

The aim of this analysis is to investigate how for different main languages English proficiency varied and how the self-reported ability to speak English relates to general health, for England and Wales in 2011. The story initially covers different main languages at the England and Wales level and then goes on to look at English proficiency in local authorities for the 3 to 15 age group, then at England and Wales and local authority level by general health by proficiency in English language.

For the purposes of this analysis, people’s ability to speak English has been collated into the following groups:

• Main language was English¹

• Main language was not English¹: Could speak English very well or well (‘Proficient²’)

• Main language was not English¹: Could not speak English well or at all (‘Non-proficient³’)

English¹ was the main language for 92 per cent (49.8 million) of usual residents aged three and over. Of the remaining eight per cent (4.2 million), who had a different main language, the majority (3.3 million) were ‘Proficient²’ in English, while 863,000 were ‘Non-proficient³’. Language in England and Wales has previously been investigated, alongside other pieces of analysis of the 2011 Census. The availability of the detailed characteristic (cross topic) tables now allows further analysis to expand on what has been produced.

Of the 3.3 million who were ‘Proficient²’ in English, 1.7 million could speak English very well and 1.6 million could speak English well; and of the 863,000 who were ‘Non-proficient³’, 726,000 could not speak English well and 138,000 could not speak English at all.

Notes for introduction

1. English or Welsh in Wales.

2. This includes people who selected ‘very well’ or ‘well’ for question 19 on the 2011 Census form. This is grouped as ‘Main language is not English: Can speak English very well or well’ within published tables.

3. This includes people who selected ‘not well’ or ‘not at all’ for question 19 on the 2011 Census form. This is grouped as ‘Main language is not English: Cannot speak English or cannot speak English well’ within published tables.

4. Main language and proficiency questions as they appeared on the 2011 Census questionnaire:

The 2011 Census classified 88 main languages⁴ (excluding sign languages) other than English1. Using this information, it is possible to investigate proportions of people who have a main language other than English¹, who were ‘Proficient²’ and ‘Non-proficient³’, by their main language.

Of the 4.2 million usual residents aged three and over with a main language other than English¹, sign language accounted for around 22,000, with less than a third being ‘Proficient²’. As many adept in sign language have hearing difficulties, it is possible they had a reduced opportunity in developing speaking skills, but they might have had the ability to understand English, which was not recorded as part of the census. Therefore, for the purposes of comparing the proficiency of those with a main language other than English¹, those reporting ‘sign’ have been excluded from the analysis in this section.

Over 70 per cent of people were ‘Proficient²’ in English for 78 of the 88 main languages other than English¹ categorised by the 2011 Census, and over 90 per cent of people were ‘Proficient²’ for 35 of those languages. Table 1 shows that the distribution of the 88 main languages other than English¹ was skewed, with a high number of languages with high percentages of those who were ‘Proficient²’ in English. At the other end of the scale there was only one language where less than 50 per cent of people were ‘Proficient²’ in English.

For Table 2, a minimum population of 300 was chosen to avoid ascribing meaning to findings based on very small population sizes. This number was chosen as this is approximately the average population within an Output Area, the lowest geographical level at which census estimates are provided. For the other sections of the analysis, language groups with populations under 300 people were included in the data. The ‘Welsh (in England only)’ category describes those usual residents who have Welsh as their main language and lived in England at the time of the 2011 Census. All usual residents who had Welsh as their main language and who lived in Wales at the time of the census, are included in the English¹ category.

The prevalence ‘Proficient²’ of 99.4 per cent for the Afrikaans language may be explained by the possibility that most of this population could have been born and spent a significant amount of time in South Africa, where English, as well as Afrikaans and other languages, is one of the official languages. Similarly, Shona is a native language to Shona people, who predominantly reside in Zimbabwe and Zambia, and Tagalog and Filipino are native languages to many residents of the Philippines. In all of these countries, English is an official language, therefore these main language speakers could have needed to speak, and been taught to speak, English from a young age. The high prevalence ‘Proficient²’ observed for some of the Nordic and Germanic languages, including Swedish, Danish, Northern European Language (which includes Norwegian), Finnish, German and Dutch, could be due in part to Sweden, Denmark, Norway, Finland, Germany and the Netherlands having school pupils compulsorily beginning to learn English at a relatively young age.

Main languages other than English¹ with the lowest prevalence of proficiency in English were Gypsy/Traveller languages, Pakistani Pahari (with Mirpuri and Potwari), Vietnamese and Cantonese Chinese with 37.5 per cent, 55.2 per cent, 58.3 per cent and 61.0 per cent ‘Proficient²’, respectively (Table 3). Linguistic distance, that is a measure of how different one language is from another, could also be used in part to explain these lower proportions. For example, the linguistic distance between Asian languages and English is greater than that between English and the Nordic and Germanic languages (e.g. Chiswick and Miller, 2005).

Table 4 shows that the main language other than English¹ with the largest population was Polish. Around 72 per cent (396,000) of those with Polish as their main language were ‘Proficient²’ in English, with 28 per cent (151,000) ‘Non-proficient³’. Of the top 10 main languages other than English1 with the largest populations, French had the highest proportion ‘Proficient²’ at 94 per cent, followed by Spanish (90 per cent) and Arabic (82 per cent). These 10 main languages combined accounted for over 60 per cent of all those ‘Non-proficient³’ in English in England and Wales.

Notes for English proficiency of people with a main language other than English

1. English or Welsh in Wales.

2. This includes people who selected ‘very well’ or ‘well’ for question 19 on the 2011 Census form. This is grouped as ‘Main language is not English: Can speak English very well or well’ within published tables.

3. This includes people who selected ‘not well’ or ‘not at all’ for question 19 on the 2011 Census form. This is grouped as ‘Main language is not English: Cannot speak English or cannot speak English well’ within published tables.

4. Main languages as defined as part of 2011 Census output tables, as opposed to individual languages collected.

Identifying the number of residents aged 3 to 15 year that had a language other than English¹ as their main language who could not speak English well or at all (‘Non-proficient³’) is important for planning and identifying education needs. This is particularly relevant at a local authority level.

In England and Wales as a whole, there were 8.5 million residents aged 3 to 15 in 2011. Eight million (94 per cent) of these had English¹ as their main language, 0.4 million (5 per cent) had a language other than English¹ and were ‘Proficient²’ in English, and 78,500 (1 per cent) were ‘Non-proficient³’. All local authorities had less than 5 per cent of their population aged 3 to 15 that had a main language other than English¹ and were ‘Non-proficient³’ in English.

Table 5 identifies the top 10 local authorities with the highest percentage of ‘Non-proficient³’ 3 to 15 year olds in relation to their population at that age range. Of these local authorities, all of the top five, and seven of the top ten, were in London.

In terms of actual numbers, Birmingham had the largest population aged 3 to 15 who were ‘Non-proficient³’ in English, at 3,300. This does not appear within Table 5, as it only represents 1.7 per cent of Birmingham’s total population aged 3 to 15, due to this local authority having the largest overall population in this age group (195,000).

Notes for local authority analysis for English language proficiency in 3 to 15 year olds

1. English or Welsh in Wales.

2. This includes people who selected ‘very well’ or ‘well’ for question 19 on the 2011 Census form. This is grouped as ‘Main language is not English: Can speak English very well or well’ within published tables.

3. This includes people who selected ‘not well’ or ‘not at all’ for question 19 on the 2011 Census form. This is grouped as ‘Main language is not English: Cannot speak English or cannot speak English well’ within published tables.

The 2011 Census asked people to rate their general health as ‘very good’, ‘good’, ‘fair’, ‘bad’ or ‘very bad’. For the purposes of this analysis, we have dichotomised people’s general health into the following groups:

• ‘Good’ health – encompassing those who rated their general health as ‘very good’ or ‘good’

• ‘Not Good’ health – encompassing those who rated their general health as ‘fair’, ‘bad’, or ‘very bad’

Those in ‘Not Good’ health who are ‘Non-proficient³’ in English may require translation services to effectively communicate the details of their health concerns to healthcare providers. Whether these translation services are provided in the form of a friend or relative, or by the local service provider, they could be necessary for the provision of quality healthcare to the ‘Non-proficient³’ patient. Around 300,000 usual residents in England and Wales had ‘Not Good’ health and were ‘Non-proficient³’ in English.

A further potential challenge for healthcare providers is that some of those ‘Non-proficient³’ in English who are in ‘Not Good’ health, may have difficulty in making initial contact in order to seek medical help. Furthermore, if and when contact is established, people may feel uncomfortable reporting sensitive conditions to a medical professional via a third party. However, an increasing number of translation services and information sources are being made available by the NHS to facilitate better access to services for those who are ‘Non-proficient³’ English speakers. For example, the NHS Direct interpreter service and the NHS Local: Languages and minority ethnic communities information webpage.

Overall in England and Wales, 81 per cent had ‘Good’ health, which was a similar level to those with English¹ as a main language. This is due to the ‘main language is English¹’ category encompassing 92 per cent (49.8 million) of the aged 3 and over usual resident population in England and Wales. Those who were 'Proficient²' in English had a higher percentage, while 65 per cent of those with a language other than English¹, who were ‘Non-proficient³’, reported ‘Good’ health (Figure 1). It is possible that some of those in ‘Not Good’ general health may have less opportunity to improve their English speaking skills, due to, for example, being less mobile.

The following sub-sections consider the extent to which variations in general health may be due to different age and sex distributions within the proficiency categories.

Variations by sex and age

For the main language English¹ and the ‘Proficient²’ in English categories, the ‘Good’ health prevalence differences between the sexes were small, at 2.2 and 0.6 percentage points respectively (Table 6). However, for the ‘Non-proficient³’ category, there was a clear difference between the proportion of males and females reporting ‘Good health’, at 72.4 per cent for males and 60.2 per cent for females, a difference of 12.2 percentage points.

Overall, as age increases, the proportion of people with ‘Good’ health becomes lower, for all of the proficiency groups. However, in all age categories, those ‘Non-proficient³’ in English had a lower proportion of people reporting ‘Good’ health than those in the other proficiency categories (Figure 2). The amount by which those ‘Non-proficient³’ in English had lower proportions of good health than the other proficiency categories was largest for the age groups of 50 years and over.

Figure 2: General health by proficiency by age, England and Wales, 2011

Source: Census - Office for National Statistics

Notes:

1. English or Welsh in Wales.
2. This includes people who selected ‘very well’ or ‘well’ for question 19 on the 2011 Census form. This is grouped as ‘Main language is not English: Can speak English very well or well’ within published tables.
3. This includes people who selected ‘not well’ or ‘not at all’ for question 19 on the 2011 Census form. This is grouped as ‘Main language is not English: Cannot speak English or cannot speak English well’ within published tables.

Those ‘Proficient²’ in English had a higher proportion of ‘Good’ general health than those with English¹ as their main language, and than those ‘Non-proficient³’ in English (Figure 1). This could be explained by their younger age profile (Table 7).

Across all proficiency categories the proportion of ‘Good’ health decreased overall as age increased. However, the lower proportions of good health in the 'Non-proficient³' in English category compared to the other categories were most apparent in the older age groups; those aged 50 and over (Figure 2).

Table 7 revealed that those ‘Non-proficient³’ in English had a younger age profile than those with English¹ as their main language. Therefore, age profile differences do not account for the lower proportions of general health found in those ‘Non-proficient³’ in English.

Proficiency by general health: Local authority level analysis

Identifying the local authorities with high proportions of those ‘Non-proficient³’ in English in ‘Not Good’ general health may be a useful indication for helping to identify any requirements for translation services and potential challenges local areas may face in providing access to healthcare services for these usual residents.

Table 8 ranks local authorities with the highest proportion of people with ‘Not Good’ Health who were ‘Non-proficient³’ in English. Eight out of 10 of these local authorities were London boroughs.

In terms of actual numbers, Birmingham had the highest number of usual residents aged 3 and over who were ‘Non-proficient³’ in English and in ‘Not Good’ health at around 21,000. This represented 2.1 per cent of the overall aged 3 and over population of Birmingham (around 1 million). Bradford had the second highest number ‘Non-proficient³’ in ‘Not Good’ health at around 10,000, representing 2.0 per cent of its total population aged 3 and over (around 500,000).

Notes for variations in general health by English language proficiency

1. English or Welsh in Wales.

2. This includes people who selected ‘very well’ or ‘well’ for question 19 on the 2011 Census form. This is grouped as ‘Main language is not English: Can speak English very well or well’ within published tables.

3. This includes people who selected ‘not well’ or ‘not at all’ for question 19 on the 2011 Census form. This is grouped as ‘Main language is not English: Cannot speak English or cannot speak English well’ within published tables.