In its 2008 Communication, the EC refers to the many ’national, regional, minority and migrant’ languages spoken in Europe ’adding a facet to our common background’ and also ‘foreign languages’, used to refer principally to both European and non-European languages with a worldwide coverage.

In the context of the LRE project, we have used the above language types with the following definitions:

• National languages: Official languages of a nation-state.
• Foreign languages: Languages that are not learnt or used at home but learnt and taught at school or used as languages of wider communication in non-educational sectors.
• Regional or minority languages: Languages that are traditionally used within a given territory of a state by nationals of that state who form a group numerically smaller than the rest of the state’s population.
• Immigrant languages: Languages spoken by immigrants and their descendants in the country of residence, originating from a wide range of (former) source countries.

For similar perspectives, we refer to McPake and Tinsley (2007). In this context, we want to express our awareness of the deliberate inclusion of immigrant languages as part of the European repertoire of languages, while at the same time in this first round of data collection on multilingual policies and practices for as yet little reference is made to sign languages.

Domain 1 explores the availability of nationwide or regionwide official documents and databases on language diversity in each of the participating countries/regions. The availability of such documents and databases may contribute significantly to the awareness of multilingualism in a given country/region and can inform language education policy.

Domains 2–4 of the survey focus on education for non-adult learners provided by the state. The key distinction between organisation versus teachers is widely used in the European context (see, for example, Eurydice 2008). The responses in these sections are based on publicly available data as well as from official sources.

Responses in domains 5–8 are based on collected and reported data in the urban contexts of three cities per country or region (see Section 1.5 for details). Domain 5 explores language provision in a small sample of further (vocational) and higher (university) education institutions. Domain 6 focuses on languages in the audiovisual media and the press. Domain 7 concentrates on languages in public services and public spaces in terms of institutionalised language strategies, oral communication facilities and written communication facilities. The focus of domain 8, languages in business, is on company language strategies, internal communication strategies and external communication strategies. In each country/region a sample of 24 companies was aimed at.

The collection of this primary data took place in three cities in each country or region prompted by the following considerations:

• multilingualism is most prevalent in urban settings as newcomers tend to congregate there in search of work
• cities reinforce national dynamics in responding to language diversity
• large further and higher education institutions are present in cities (domain 5)
• the international press, cinemas and television stations are most commonly found in cities (domain 6)
• as a result, city administrators and urban planners need to create local policies on multilingualism (domain 7)
• the headquarters of many businesses are located in cities (domain 8).

The selection of cities was identical for countries 1-14 in Table 4. Here the focus was on the two cities with the largest population size plus one city where the regional/minority language with the highest status, vitality and/or number of speakers in the country is spoken. Countries 15-18 presented a challenge as they do not fit the above model.

Country 15, Bosnia and Herzegovina, has three national languages: Bosnian, Croatian and Serbian. The cities chosen for primary data collection were Sarajevo, where Bosnian is mainly used, Banja-Luka, where Serbian is mainly used, and Mostar, where Bosnian and Croatian are mainly used.

Country 16, Switzerland, comprises 26 cantons and has four official languages: German, French, Italian and Romansch. LRE research in all domains took place in three sample cantons: one German-speaking (Zurich), one French-speaking (Geneva), and one Italian-speaking (Ticino).

Country 17, Spain, comprises 17 autonomous communities and two autonomous cities. LRE research has been conducted for domains 2-4 in three autonomous communities – Madrid, Andalusia, Valencia – and two ‘historic nationalities’ – Basque Country and Catalonia. Three profiles have been created: a combined profile for Madrid, Sevilla and Valencia and two separate profiles for Basque Country and Catalonia. Basque Country has two official languages: Basque and Spanish. Catalonia has three official languages: Catalan, Spanish and Aragonese.

Country 18, the UK, comprises four countries that have separate governments and education systems. For the education domains (2-4) data has been collected on policies and common practice in each country/region. For domains 5-8 the cities in Wales and Scotland were chosen on the basis of population size. In England, after London the city of Sheffield was chosen for practical reasons. It has not yet been possible to research a further city, but it is hoped that this data will be available soon. In Northern Ireland it has so far only been possible to include Belfast in the survey.

The selection of the three cities and the proposed regional/minority (R/M) languages to focus on were decided upon in advance in co-operation with all participating national or regional teams on the basis of municipal statistics for the first two cities and regional/minority language/group statistics for the third city. Table 4 gives an overview of the cities surveyed per country (minus Germany).

Table 4. Three-cities approach for all participating countries/regions

Nr	Countries with one national language	Largest city	Second/Third largest city	Additional city	Dominant regional/ minority language in additional city
1	Austria	Vienna	Graz	Klagenfurt	Slovene
2	Bulgaria	Sofia	Plovdiv	Shumen	Turkish
3	Denmark	Copenhagen	Aarhus	Aabenraa	German
4	Estonia	Tallinn	Tartu	Narva	Russian
5	France	Paris	Marseille	Corte	Corsican
6	Greece	Athens	Thessaloniki	Xanthi	Turkish
7	Hungary	Budapest	Debrecen	Pécs	German
8	Italy	Rome	Milan	Trieste	Slovene
9	Lithuania	Vilnius	Kaunas	Klaipeda	Russian
10	Netherlands	Amsterdam	Rotterdam	Leeuwarden*	Frisian
11	Poland	Warsaw	Krakow	Gdansk	Kashubian
12	Portugal	Lisbon	Oporto	Miranda do Douro*	Mirandese
13	Romania	Bucharest	Iaşi	Cluj	Hungarian
14	Ukraine	Kiev	Kharkiv	Lviv	Russian
	Other countries	Largest city	City in region 2	City in region 3	Official language(s)
15	Bosnia & Herzegovina	Sarajevo	Banja-Luka	Mostar	Bosnian/Serbian/ Croatian
16	Switzerland	Zürich	Genève	Lugano	German/French/Italian
17	Spain Catalonia Basque Country	Madrid Barcelona Bilbao	Valencia Tarragona San Sebastian	Sevilla L’Hospitalet Vitoria-Gasteiz	Spanish Catalan Basque
18	UK: England Wales Scotland	London Cardiff Glasgow	Manchester Swansea Edinburgh	Sheffield Newport Aberdeen	English Welsh/English English/Scottish-Gaelic
	N.-Ireland	Belfast	-	-	English

*Absence of university leading to absence of university-based data

National or regional profiles are based on primary data collection for the 23+22+22=67 cities referred to in Table 4. As can be derived from Table 4, most dominant regional/minority languages in the chosen additional cities have the status of national language in adjacent countries. The focus of primary data collection for language domains 5-8 in each of the 24 participating countries/regions is summarised as follows:

• For language domain 5, the focus is on language provision in different types of adult education provided by the state. Two complementary sectors are addressed; language provision in vocational education for (young) adults aged 16-plus and language provision in academic/university education.
• For language domain 6, the focus is on language provision in audiovisual media, including public radio and television broadcasting, the largest cinemas, and in the press at the largest train stations and city kiosks in the cities surveyed.
• For language domain 7, the focus is on language provision in public services and public spaces at city level, more particularly on institutionalised language strategies, oral communication facilities and written information facilities at city (council) level in the cities surveyed.
• For language domain 8, the focus is on four different business sectors – supermarkets, construction businesses, hotels and banks. Researchers were asked to collect samples distributed as evenly as possible across multinational/international (M/I), national (N), and regional or local (R/L) businesses. In practice, this ambition turned out to be difficult to realise across all countries/regions.

In Table 5, a summarising overview of language domains and targets for primary data collection per city (3x) is provided.

Domains and targets for primary data collection per city

Nr	Language domain	Targets per city (3x)
5	Languages in vocational and university education	Largest institution for vocational education and training (VET) with language provision Largest public university
6	Languages in the media	Language provision in radio and TV programmes, at the cinema, as described in the best-selling newspaper Language provision in press at the largest train station and city kiosk
7	Languages in public services and spaces	Institutionalised language strategies, oral communication facilities and written information facilities at the central city level
8	Languages in business	Small-/medium-sized and large multi-/international, national and regional/local supermarkets, building construction businesses, hotels and banks

• 2010 - Initial questions and scoring proposals for all multiple answer options were developed by Tilburg University, the British Council and the Migration Policy Group in Brussels. The business domain was developed by CILT in London, using the ELAN survey (2006) as the starting point, and then further refined by the French research team. Advice for the public services domain was given by the Metropolitan Police languages team and other London public service providers.
• The draft version of the LRE Questionnaire was pre-tested in three pilot studies in Poland, Spain and the region of Catalonia in early 2011. The pilot studies were aimed at testing the content and construct validity of the LRE Questionnaire by taking into account variation in language policies and practices both between and within countries.
• On the basis of the pilot outcomes, the LRE Questionnaire, the Field Manual for researchers, and the scoring procedures were further adapted and then scrutinised by the LRE Steering Group and external experts. The final LRE Questionnaire was sent out in autumn 2011 to all national and regional teams for data collection.
• Different versions of the questionnaire were created for Bosnia and Herzegovina, Spain, Switzerland and the UK (see Table 4).
• Researchers were sent a detailed Field Manual explaining the background to the project, and how data collection was to be conducted. In addition, there was a two-day face-to-face meeting to discuss the methodology, and e-mail exchanges and phone calls took place with each research team.
• Once the national and regional teams had provided all answers to all questions, all the data obtained was peer-reviewed independently to ensure a double-checked and consistent interpretation.
• Subsequently, all peer-reviewed data was processed, analysed and reviewed by the LRE team at Tilburg University.
• Through the process, it became clear that some questions had been interpreted differently by different researchers, while others had not been fully understood. The process of clarifying these, standardising responses, and agreeing final interpretations was completed in early 2012. It was decided that some questions would not be scored due to unfeasibility of gathering the data. Data for questions on book collections in languages other than the national language in public libraries and bookshops proved impossible to collect in some countries. Questions on the languages required for undergraduate and postgraduate studies proved ambiguous, and have not been scored.
• The results for each country/region were sent back to all researchers and a further opportunity for feedback was given. The results were presented initially at the whole domain level, but subsequently it has been decided to present them at the more detailed question level in order to capture countries/regions policies and practices in more detail.