Report | Doc. 13361 | 04 December 2013

Integration tests: helping or hindering integration?

Committee on Migration, Refugees and Displaced Persons

Rapporteur : Ms Tineke STRIK, Netherlands, SOC

Origin - Reference to committee: Doc. 12604, Reference 3784 of 20 June 2011. 2014 - First part-session

Summary

A growing number of Council of Europe member States have introduced integration tests for migrants, based primarily on knowledge of the language of the host country, but also sometimes on “citizenship” issues, including history, political institutions, society and democratic values

There is no doubt that an ability to master a language can contribute to successful integration. However there are serious concerns that some of these tests may in fact be hindering integration and leading to exclusion: pre-entry tests can be an obstacle to family reunification, while integration tests for migrants once in the host country can prevent them from enjoying secure residence rights, causing resentment in migrant communities and discrimination against certain groups, particularly people who may be illiterate or have low levels of education.

There are also concerns that the primary purpose – or at least a predictable consequence – of these tests is to cut down the number of migrants arriving or remaining in the countries concerned. If this is the case, they are a step backwards.

Member States should ensure that these tests are based on achievable attainment levels, that the tests and learning processes are financially supported, and that alternatives to testing are available (to reflect the fact that not everyone has the same linguistic capabilities or needs).

Furthermore, the Committee of Ministers should look at how to adapt the Council of Europe’s language proficiency reference levels (the “Common European Framework of References for Languages” (CEFR)), used by member States in many of these tests, but which were never designed as an instrument for measuring integration.

A. Draft resolution

(open)

1. Knowledge of a receiving society’s language(s) without doubt facilitates the successful integration of migrants. This is the foundation on which integration tests were introduced by a small number of member States, initially for citizenship purposes. They have now, however, been embraced by a growing number of States and applied not only for citizenship, but also for residence and even as a pre-entry requirement, primarily in cases of family reunion.

2. Not only has the use of tests increased significantly, but the standards required have also gone up. Furthermore, alongside language, a range of “citizenship” issues, such as facts about history, and knowledge of political institutions, society and democratic values, have been introduced in a number of countries.

3. There are two primary concerns over the use of these tests. The first is whether they promote integration or have the opposite effect. The second is whether they are being used not so much as an integration measure, but rather as a migration management mechanism to limit the number of migrants entering and/or remaining in the country concerned. A drop of at least 20% of people seeking family reunion in one member State and a drop of 40% of those seeking or granted permanent residence permits in another gives a clear indication of the effect, intended or otherwise, that the introduction of these measures can have. These decreases are a reason for concern, as family reunification and greater security in terms of residence rights are beneficial for the integration of migrants. Especially vulnerable migrants need to be supported in the integration process by granting them access to these rights, instead of excluding them.

4. Encouraging integration through language and other testing is not per se problematic, and it is a measure that many member States are likely to continue with in one form or another. It is, however, important to be aware of the limitations of such tests and ensure that they contribute to integration and do not become a barrier to it. Rather than promoting testing, offering language courses and possibly obliging migrants to participate in these courses may offer greater advantages and develop their language skills without running the risk of excluding migrants. They may also promote integration as a two-way process, requiring an investment by both the host society and the migrants. Furthermore, testing of knowledge does not, as such, improve language skills and can only be effective if it is offered at the final stage of a language course provided by the host State.

5. The Parliamentary Assembly is concerned that current integration tests in Europe are not as effective as they should be. In the first place, the standards of attainment required sometimes exceed what is reasonably attainable by many immigrants or candidates for immigration, leading to the exclusion of many people who would otherwise have no problems integrating. This raises human rights issues, notably with regard to the right to family life and protection from discrimination. It is particularly problematic in the case of family reunification and when dealing with people who are illiterate or with low levels of education, the aged, refugees and others. Furthermore, where integration tests are a barely veiled migration management measure, they inhibit and are detrimental to integration and they should be discontinued.

6. Member States of the Council of Europe are therefore invited to reappraise their approach to integration tests by evaluating their long-term effectiveness as a tool for efficient, sustainable and user-friendly integration, in order to ensure that:

6.1. the language proficiency levels set in these tests are attainable. This will require that:

6.1.1. the language levels be not set too high and that they be differentiated with regard to what is expected in terms of speaking and listening ability (not going beyond the A2 level of the Council of Europe “Common European Framework of Reference for Languages: Learning, Teaching, Assessment” (CEFR)), and writing and reading ability (remaining at the basic A1 level of the CEFR);

6.1.2. the discriminatory effect of testing be avoided by giving preference to graduated scales of attainment, reflecting recognition of effort rather than simply achievement. Furthermore, testing has to take into account the needs and abilities of those who do not have the same levels of literacy and education, or those who may be in a position of vulnerability or have other difficulties, such as the elderly and refugees;

6.1.3. tests should not be the only method. Rather than or in addition to tests, one should consider other ways of showing commitment to integration, such as civic commitment or progress, or by using mechanisms adopted in other countries, such as interviews with trained personnel to ensure fairness;

6.2. adequate financial support is provided by the State for preparatory courses. Wherever possible these should be free of charge for migrants as it has been shown that making migrants pay for courses, which can run to 400 hours of learning or more, acts as a major barrier and disincentive;

6.3. appropriate measures are in place to ensure that failure rates in the tests, which can be high, do not have a discriminatory effect and do not lead to exclusion or a state of limbo for those who do not succeed. Failure could have the consequence of requiring further efforts, but it should not lead to the denial of the right to family reunification, permanent residence rights or citizenship.

6.4. particular attention is paid to pre-entry requirements and the impact they can have on the right to family life as enshrined in Article 8 of the European Convention on Human Rights (ETS No. 5), and also family reunion as established in the European Union Directive on the Right to Family Reunification. In this respect, failure in the test should never be the sole reason for excluding migrants from family reunification when they fulfil all the other criteria;

6.5. alternatives to these tests are explored and monitored, and used where appropriate.

B. Draft recommendation

(open)

1. The Parliamentary Assembly refers to its Resolution … (2013) “Integration tests: helping or hindering integration?”.

2. The Assembly notes that knowledge of a receiving society’s language(s) facilitates the successful integration of migrants. This is the foundation on which integration tests were introduced by a small number of member States and why they have been embraced by a growing number of them. They are now applied not only for citizenship, but also for residence and even as a pre-entry requirement, notably for family reunion purposes.

3. Statistics and evaluation studies show that language and integration tests have led to a decrease in the number of applicants for family reunification, permanent residence permits and naturalisation. These tests can also have a discriminatory impact, depending on the gender, age, educational background and nationality of the people concerned. This impact raises serious questions as to whether tests which are connected to the granting of residence rights are the right instrument for promoting the integration of migrants. This calls for serious reconsideration of the policy of merely testing and demanding a certain level of knowledge, rather than promoting language skills and integration.

4. Not only has the use of tests increased significantly, but the standards required have reached higher levels, often using as a benchmark the Council of Europe “Common European Framework of Reference for Languages: Learning, Teaching, Assessment” (CEFR).

5. The Assembly notes that the CEFR offers reference levels which are widely used for benchmarking language competences and represents one of the many successes of the Council of Europe. It recognises, however, that this instrument was never established as a mechanism for establishing whether or not a certain language level was indicative of a level of integration. It is only a measure of linguistic ability.

6. The Assembly also notes the important activities carried out by the Language Policy Unit of the Council of Europe, and in particular its work on Linguistic Integration of Adult Migrants (LIAM).

7. In this context, the Assembly invites the Committee of Ministers to:

7.1. undertake, though its Language Policy Unit, further work in respect to the CEFR, in order to identify appropriate ways of drawing on its content in the integration process. In this respect, it may consider it appropriate to produce guidelines on how the CEFR can be used and the limitation on its use for integration purposes, or examine an alternative tool, based on the CEFR, which might be more appropriate for integration purposes than language proficiency levels;

7.2. put forward alternatives to integration/language tests as a means of promoting and measuring integration and improving the prospects for integration of migrants and would-be migrants;

7.3. promote further the work of the Council of Europe on Linguistic Integration of Adult Migrants.

C. Explanatory memorandum by Ms Strik, rapporteur

(open)

1. Introduction

1. Languages play a major role as indicators of belonging for migrants, and poor knowledge of a receiving society’s language can be a serious impediment to successful integration

and basic socialising. This is the foundation on which integration tests have been introduced, as a means of promoting integration, focusing in particular on language learning.

2. The origins of many national integration strategies go back to requirements for naturalisation. A common factor in these earlier requirements, both inside and outside Europe, was that the level of linguistic competence was rarely specified, and it was often possible for would-be citizens to meet the requirements with very low levels of functional language.

3. In recent years, however, there has been a steady but significant change in requirements for citizenship as well as an increase of requirements for residency, family reunification, etc. The main target group of the tests are family members of residents in the member States. This is the case for both the pre entry test and the integration requirements after admission.

4. Today, in many countries, a specified level of language competence has become a prerequisite for migrants to obtain not only citizenship but also permanent residency.

The latest phenomenon is the requirement for long-term migrants to demonstrate their commitment to integration by showing language competence before leaving for the destination country. This has led to a proliferation of linguistic or citizenship “tests”, particularly in European Union member States.

5. These tests have become increasingly sophisticated with “a sea change in membership requirements” and “fully defined criteria with standardised methods of assessment” being introduced.

As these tests have become increasingly formalised across Europe, one of the side effects has been a reduction in the possibility to take into account the individual circumstances and capacities of the persons for whom the tests have been designed.

There are likely to be other consequences too, including when there are heavy financial costs involved for those having to take the tests and undergo the pre-test courses.

The lack of flexibility in tests means that even where language programmes intended to help people pass them exist, they may not help them achieve anything linked to their life goals or integration, as the methodology, subject matter or level of attainment may not be appropriate for this. Pre-entry tests that are intended to assess someone’s employability may be inappropriate for someone not intending to work. For those seeking to enter for marriage, tests may either exclude them or, in the case of arranged marriages, lead to “matches” that may be less than ideal for the persons concerned. The costs of preparing for, and taking, tests may be prohibitive for some and therefore discriminate against those with low levels of income. Furthermore, the inability to pass, or difficulty in passing a test for permanent residence or citizenship may reduce the feeling of being integrated and cause resentment. Additionally, the need for protracted studies may inhibit the ability to work, which in itself is a key domain in integration.

6. This report seeks to chart the evolution of these tests and to uncover the issues around their introduction that should be addressed by the Council of Europe and its Parliamentary Assembly.

7. Therefore, among the many questions we need to look at, the most important is perhaps to determine whether these tests are a step forward or a step back and whether they improve the integration of migrants. It is also necessary to examine some of the other implications of using such tests in cases of family reunification, permanent residence and citizenship. Furthermore, the impact of these tests on people with little or no literacy in their own language, women or other vulnerable groups such as the aged and the sick needs to be examined.

8. In the preparation of this report, I have been greatly helped by a number of experts who attended a hearing on the issue in Brussels in March 2012, the Language Policy Unit of the Council of Europe

and also by Mr Christopher Hedges, former Chair of the European Committee on Migration (CDMG), who assisted me as a consultant.

2. The current situation in member States

9. It is very evident from the policies and practices relating to language and citizenship testing that I have examined in preparing this report that there has been a steady hardening over time. In order to illustrate this, I propose highlighting the United Kingdom and Dutch experiences as case studies in order to show how these integration tests have developed. These case studies can be found in Appendix 1.

10. It is also evident from many of the studies and reports I have examined that the Council of Europe “Common European Framework of Reference for Languages: Learning, Teaching, Assessment (CEFR)” is extensively used as a basis for assessing someone’s level of “integration”. Whilst it was not initially envisaged that the CEFR would be used as an assessment tool in these circumstances, overt reference in requirements for migrants to the proficiency levels described in the CEFR provides clear evidence of a steady increase in levels of language competency required by many States for the purposes of permanent residence or citizenship. The concern is that this upward trend will lead to many migrants being denied security of residence.

The Council of Europe “Common European Framework of Reference for Languages: Learning, Teaching, Assessment” (CEFR)

This is a reference tool widely used, amongst other things, in the assessment of language proficiency.

It has a sliding scale of proficiency at a number of levels arranged in three bands:

A1 and A2: basic user

B1 and B2: independent user

C1 and C2: proficient user

One of its advantage is that it can be used to define “profiles”, for example A2 level for speaking, but A1 for reading or writing, rather than homogeneous levels (A2 for all competences). A fuller explanation of this tool is provided in Appendix 2.

11. In looking at the comparative situation across member States I have divided the review into two areas, namely pre-entry tests and post-entry tests.

12. Although many countries have introduced tests for those seeking entry for employment, the examples given here focus primarily on the integration of migrant families.

2.1. Pre-entry tests

13. According to one academic study (the PROSINT study), the introduction of mandatory pre-entry language tests for family reunification is a clear case of countries learning from and using the policies of others. The phenomenon started in the Netherlands in 2006, but was then introduced by Germany (2007), the United Kingdom and Denmark (2010) and Austria (2011).

14. France has also introduced a test for family members wishing to enter the country, examining their knowledge of French and the French Republic's values, and must demonstrate sufficient proficiency in French or at least commit to learning it. The difference with other States is that failing the test does not result in the denial of admission to France, but only a delay. Family members who fail are invited to attend courses in the country of origin. They last no longer than two months (180 hours). If the family member passes the test a visa is issued, if they do not the visa is still issued, but further courses may be required once in France.

Furthermore, all non-EU migrants to France have to sign an “integration contract”. In Denmark, by contrast, the pre-entry test is taken in Denmark itself and applicants are given a temporary short-term visa to enter the country and to prepare for and take the test within nine months.

15. A common factor in all these countries is that the language level required is set at a low level, namely A1.

16. No State-sponsored support is offered by the Netherlands, the United Kingdom, Germany or Austria; indeed, in Austria, the post-entry courses that were previously provided were withdrawn on the basis that new migrants would not need them, as they would already have started their integration before entry. Germany, however, has a well-developed infrastructure through the Goethe Institute that can help potential migrants prepare for the test.

17. There is a distinct blurring of the aim of pre-entry tests for family reunification purposes between promoting integration and reducing numbers. This can lead to a degree of cynicism on the part of those involved and can consequently reduce the effectiveness of what might be termed “legitimate” integration measures. The Dutch Government has been most explicit in mentioning the likelihood of a reduction in such migrants as a goal or side-effect of this policy. It referred to its pre-entry programme as a “selection criterion” that would reduce the levels of “migrants incapable of integration” by something in the region of 25%. It mentioned specifically migrants from Turkey and Morocco as being problematic in terms of their weak starting position in the labour market because of their educational background. None of the other countries explicitly mentions limiting migration as a direct or indirect goal, although the United Kingdom has said that it anticipates a reduction of 10% in applications by spouses or partners. However, the reality of these policies has been a significantly higher reduction in arrivals of family migrants than was anticipated in the Netherlands (although in recent years these have risen again somewhat). In the United Kingdom, the reduction in the first year following implementation was in the region of 20%.

2.2. Post-entry tests for permanent residence/citizenship

18. Post-entry programmes differ widely in their requirements, in particular the level of language knowledge, the way of testing and the obligation to attend courses. The following table gives an overview of the main language requirements for permanent residence.

Country	Requirement	Language/Integration course	Language level	Civic knowledge
Austria	Test	Formally no, in practice yes	A2 within 2 years, B1 after 5 years	No
Czech Republic	Test	No	A1	No
Germany	Test	Yes	B1	Yes
Italy	Test	Yes	A2	Yes
Netherlands	Test	Only offered to refugees	A2 within three years	Yes
Spain	Interview	May be imposed by authorities	Case-by-case basis	Interview
Sweden	None	Individual plan	Case-by-case basis	No
Switzerland	Interview	May be imposed by authorities	Case-by-case basis, Cantonal	Interview
United Kingdom	Test	Yes, if English is below B1	B1 or shows progress up one level in course	Yes

2.2.1. Language tests

19. In a number of countries such as Austria, Germany, Italy, the Netherlands and the United Kingdom,

it is necessary to take a test offered by a specified test provider. In Spain and Switzerland, by contrast, the level of language knowledge is assessed by an interview with a civil servant who has a high degree of discretion in his/her decision. Language competence can also be assessed through an interview or conversation with an immigration officer (this happens in Croatia, France, Greece, Luxembourg, Malta and Turkey).

20. Along with the countries already referred to, others also requiring a test or language certificate include Bulgaria, Estonia, Finland, Iceland, Latvia, Lithuania, the Republic of Moldova, Norway, Portugal, Romania and Slovenia. Thus a large number of States now have explicit language requirements.

21. The range of language skills required varies from country to country.

At one time Denmark stood out as the country with the most restrictive level of language skills. There have been steady increases in difficulty over the years, from B1 in 2002 to B2 in 2006, with the number of marks required being raised in 2008. However, Denmark has recently changed its integration policy and has done away with the points system and the obligation to sit a test for family reunification, although it intends to strengthen actions to combat fraud and forced marriages. At the other end of the spectrum, there are a number of countries that have no formal language requirements.

2.2.2. Country knowledge (also referred to as “knowledge of society” or “citizenship” tests)

22. Requirements vary tremendously and besides the practical knowledge needed to function in a new country, knowledge of topics such as national social and political institutions, society, facts about history, the Constitution, democratic values and the parliament are also routinely examined. States that use these tests include Austria, Denmark, Estonia, France, Germany, Hungary, Latvia, Lithuania, Republic of Moldova, the Netherlands, Romania and the United Kingdom. It is interesting to note that the percentage of questions that must be answered correctly and the number of questions asked are generally higher in Europe than in traditional immigration countries such as the United States, Canada and Australia.

23. An alternative way of assessing country knowledge is through an interview (Slovakia and Switzerland being examples). France also requires (as part of its “integration contract”) a visit to the local “Mairie” for an introduction to French history and culture. In Greece, there has been an informal interview assessing language skills and knowledge of Greek history and culture since 2000. Luxembourg is unique in that, since 2008, an applicant for citizenship is required to take three two-hour civics courses on Luxembourgish history and society as a requirement for naturalisation.

24. It is, however, worth mentioning that the majority of the countries examined do not have any citizenship knowledge requirement.

3. How effective are the integration tests?

3.1. Measuring integration

25. The Council of Europe has identified several key domains of integration which could be used to measure the extent to which language and civics tests support integration. In its 1988 report “Measurements and Indicators of Integration” the Council of Europe refers to four key dimensions of integration – economic, social, cultural and political.

26. It will immediately be apparent that the ability to use the receiving society's language with a degree of proficiency is important in all these areas. For example, in terms of economic integration, the ability to understand the language of the workplace is paramount. In many countries, the right to vote in national and local elections is something that comes with the acquisition of citizenship and it is clearly in the interests of new participants in the electoral process to understand what they are voting for. The question, however, is to what extent does language and citizenship testing lead to better prospects of integration across the above domains?

3.2. What evidence is there of the effectiveness of tests in promoting integration?

27. There are few studies on this issue. However, among the research across Europe some examples can be given.

28. In the United Kingdom, the Centre on Migration, Policy and Society (COMPAS) at Oxford University has recently finalised a study on Citizenship and Integration in the UK.

The aim of the project was to understand the process of integration and acquisition of citizenship in order to improve the capacity of the United Kingdom to develop, implement, monitor and evaluate policies and measures for the integration of migrants.

29. Three key conclusions came out of the research: the first was that applicants for British citizenship were well integrated in British society and more likely to be engaged in the labour market, more active in civic life, feeling more British and more likely to have friendships across ethnic groups than either the foreign-born population as a whole or, significantly, the United Kingdom population as a whole. The second was that those who had had experience of the Life in the UK learning materials and test, language learning in the framework of the “path to citizenship” naturalisation measures, and citizenship ceremonies were broadly satisfied with the experience; and the third was that integration into British society could not be understood as a single, one-dimensional process, but rather as a multi-dimensional set of processes, occurring in different domains of life, with several possible pathways to integration.

30. Insofar as a general conclusion can be drawn, those involved in the study felt that the “Life in the UK” learning experience contributed to a sense of belonging in the United Kingdom, with legal citizenship and the English language as key foundations for integration in other areas of life. This suggests that the current “Life in the UK” system is broadly successful. The report, which is still to be published, will however emphasise the need to continue to provide language support as a means of enabling integration. Recent public expenditure reductions will, however, mean that, in practice, this recommendation will not be taken forward.

31. Turning to Germany, in 2006, the Federal Ministry of the Interior commissioned research into how integration courses might be improved. A key finding of the study was that, even after 600 hours of tuition, approximately 40% of all integration course participants were not able to achieve competence in German at level B1. A key recommendation from the report was therefore to introduce a more flexible system with variable teaching hours, depending on the assessed needs of learners.

32. In the Netherlands, as mentioned earlier in the report, a first evaluation of the new Civic Integration Act in 2010 highlighted some serious difficulties in the first year following implementation of the new Act, but significant improvements were made in 2008 and 2009. Numbers of participants in civic integration courses increased rapidly, and by December 2009 127 000 migrants had taken the test. Of these, 83 000 took part in courses, but a further 44 000 did not receive any formal support and it was assumed they had prepared for their integration test in another way. Since January 2010, passing the test has become a prerequisite for obtaining a permanent or independent residence permit. This requirement has caused a drop of 40% in the number of permanent residence permits applied for and granted. As these numbers are broken down by nationality, one can conclude that some nationals, like Moroccans and Turks, are more affected than others. Migrants who don’t manage to obtain permanent residence rights as a result of the tests have no option other than to remain on a temporary basis. In addition, as of 2013, family members are required to pass the test within three years of arrival. If they fail to do so, their residence permit can be withdrawn and they can be expelled with the family being separated as a result. Furthermore, since January 2013, integration courses are no longer offered by municipalities. It is left to the private market to develop them. Migrants who lack the financial means to participate can however be granted a loan of up to 5 000 euros.

33. By the end of 2009, about 20 000 migrants had participated in civic integration programmes on a voluntary basis. Almost everyone eventually passed the test and on average about 80% passed at their first attempt. However, little is known about the long-term effects on the participation in Dutch society or about the integration of those who were successful in the tests, whether post-entry and pre-entry.

34. Again in the Netherlands, a small-scale mainly qualitative analysis has been made of the effects of integration programmes on the participation in society of 29 skilled migrants.

The results of this analysis were mixed and very tentative. Some 50% of participants in the labour market oriented programme achieved sustainable employment in a reasonable period after completion of the programme. The programme for parent carers seemed to increase parent involvement with schools and voluntary organisations.

35. The report of the INTEC Project (a comparative study of integration and naturalisation tests and their effects on integration in nine European Union member States) has also provided some valuable insights.

The point is strongly made in this report that successful integration depends on many other factors besides knowledge of language and society. Furthermore, knowledge of the language of the member State is not always necessary to become integrated, especially if the migrant lives in an environment where another language, for instance English, is spoken.

36. A Belgian (Flanders region) evaluation showed that migrants who had successfully completed an integration course were more likely to obtain work than those who did not. This perhaps shows the potential of such programmes, provided they are properly adapted to meet the needs of participants. Danish research

found that a more efficient municipal integration policy resulted in more employment participation, but also highlighted that the necessary support can take different forms. However, the Belgian evaluation also revealed that participation in the course did not always result in a substantial improvement in language skills or in social contacts with Flemish-speaking Belgian citizens. And according to Austrian respondents, a receptive society is a crucial factor in their participation and integration in society and in avoiding a sense of isolation.

37. Some people agree that tests, such as those used in Germany for example, help women to overcome isolation and to resist being forced into marriage. The tests also facilitate exchanges and communication between people who find themselves in the same situation. If the process is embarked on while migrants are still in their country of origin, it encourages them to take an interest in the host country and the national language and thereby improve the conditions in which they will, in the future, be integrated.

3.3. Some questions about the appropriateness of the tests

38. Having said this, it is questionable whether such tests are really appropriate for predicting the degree of integration. As a rule, according to the INTEC study, tests tend to be an obstacle to integration rather than an aid. Amongst other things, the study found that knowledge acquired by migrants before arriving in the host country is often lost when they finally arrive because the tests are often held a long time after their arrival. Even a cursory examination of statistics shows that such tests have led to a drop in the number of visa requests, particularly in the Netherlands and Germany. It has also been noted that some people living in rural areas or areas far from the places where lessons are held have difficulty in following classes, and that it is often very costly for them to do so. Tests can often be particularly demanding for older people or for those who have not had the benefit of higher education.

39. This leads to the question: what is the appropriate level of competence that should be expected in such tests in order for them to support integration rather than becoming a barrier to it? Again, there is very little in the way of empirical research that can help, although many language experts suggest that, for someone with little formal education or limited literacy in their own language, something in the region of A2 in an additional language is the maximum they will ever be able to attain. Additionally, it is likely that this level will be reached only in speaking and listening, with a lower level in reading and writing.

40. In the United Kingdom, some language experts have reported a phenomenon they call “course blocking”. This is where students who have little or no formal literacy in their mother tongue cannot progress beyond approximately A2 in speaking and listening or A1 in reading and writing (namely the lowest levels), despite being very motivated to learn and attending classes regularly. The inevitable conclusion from these admittedly non-scientific reports is that, even where mandatory attendance and/or appropriate financial support is in place, some migrants may never achieve the required level. This in turn suggests that a scaled test is better than a one-level test, and that writing should be omitted or at least the level required should be basic.

Case studies

During the preparation of this report, researchers have provided some case studies that were collected and published by the Goethe Institute in Germany. I will cite two of these to indicate the practical difficulties that these tests pose.

The first example involved a Thai woman who wanted to marry a German man. The woman had to rent a flat in Bangkok to attend a course, organise all the administrative matters and pass the test. Although the woman had worked as a hotel manager in Ireland she had severe learning problems and especially with the A1 test, which was very time-consuming. She finally passed the test but in total it cost her in the region of €2 000.

The second concerned a woman from Cameroon. The woman had had only four years of schooling and had never learnt a foreign language before. In order to attend a course, which was only available in the capital, she had to leave her home town and her two children. She lived in the house of a friend, but for that she had to help on their farm. In order to save electricity she was only allowed to do her learning and homework during daytime.

3.4. What are the alternatives to tests?

41. It is interesting to note that some countries have taken a different approach and the level of linguistic competence required for permanent residence or citizenship in some is still fairly low. In Spain and Switzerland eligibility for citizenship is determined on the basis of an interview and the level of linguistic competence required is determined on a case-by-case basis. If the abilities of a particular individual are deemed to be too low, then the authorities might require that person to attend a language development programme.

42. In the United States – one of the largest recipients of inward migrants over the last century – knowledge of English is still tested fairly subjectively. Although the level of competence is not defined in US Citizenship legislation it is accepted to be somewhere at or below the level of A1. As in Spain and Switzerland, the applicant’s ability to speak English is determined by a US Citizenship and Immigration Service officer during an interview. To be successful, the applicant has to read one out of three sentences correctly to demonstrate an ability to read in English. To pass the writing test, the applicant must write one out of three sentences correctly. Also, during the interview, the officer will ask the applicant 10 questions on civic issues, of which they must answer six correctly. Of interest, however, is the fact that all the sentences and questions are available for applicants to study and memorise. Yet despite this apparently simple test, new US citizens are fiercely loyal to their adopted country and recent examples of “home-grown” terrorists there have generated considerable shock. One way of reading the testing experience in the United States is that it has contributed to a sense of achievement and belonging as opposed to a hurdle and form of exclusion.

43. Another alternative is that provided for in France, where the failure to pass the test does not lead to a denial of admission, but rather a delay in admission. This reduces the adverse impact on family life of the person failing the test and their families.

3.5. Can conclusions be drawn about the effectiveness of integration tests?

44. Whilst there are success stories, there are also stories of failure and of persons who have simply given up before starting. For many, the pressure of courses and tests prior to and after entry is often an insurmountable barrier that they cannot overcome. It is worth emphasising that the minimum level for citizenship in many countries – B1 in reading, writing, speaking and listening – is a secondary school leaving level for second languages in many European countries. In some countries it is also the case that many native citizens would struggle to pass a test at this level in their mother tongue.

45. As there is such limited information available it is difficult to reach any conclusions about the long-term effects of integration measures and tests on the four key domains of integration identified by the Council of Europe – economic, social, cultural and political. Such evaluations as are available are relatively short-term and tend to show mixed results. They do however highlight the need for a flexible, tailored approach in order to better meet the needs as well as match the capabilities of participants. Whilst participants are generally grateful for support, particularly in learning a new language, there is a degree of resentment when that support has to be self-financed. The inevitable conclusion is therefore that, if tests are to continue into the future, there is a pressing need for further comparative evaluation studies into their long-term effectiveness as a tool for efficient, sustainable and user-friendly pre- and post-entry integration measures. A list of questions will need to be addressed, including those contained in Appendix 3 of this report.

4. Effects of the tests: are human rights at stake?

46. Alongside the question of the effectiveness of the tests it is also necessary to look at the human rights implications of these tests, some of which have already been brought up earlier in the report.

47. First and foremost, there is a strong argument that the very existence of pre-entry tests for people seeking family reunification can breach their human rights.

Whilst people migrating for the purposes of marriage of course have a choice as to where they can live, often that choice is restricted by practical considerations. For example, would it be right to oblige a woman who had made a successful career to return to famine or poverty in her country of origin simply because of her desire to marry a particular man who could not pass a test? Effectively, in such a situation, the parties concerned are being denied the right to family life. This right, enshrined in Article 8 of the European Convention on Human Rights (ETS No. 5), is to be protected. States Parties to the Convention are only allowed to limit this right if this is strictly necessary in a democratic society. An even stronger, subjective right to family reunification is established in the Family Reunification Directive, adopted by most of the European Union member States in 2003.

The right to family reunification

The subject of family reunification is topical in that the EU Directive on the issue is currently being reconsidered by the European Commission. It has also been of long-term concern to the Council of Europe, whose stance is that family reunification must be considered as a fundamental right. The right to family life is of course enshrined in Article 8 of the European Convention on Human Rights and the right to marry in Article 12. Family reunification is also an essential component of successful integration. It has often been argued that the introduction of restrictions on inward migration, including tests that have to be passed, inhibit these fundamental rights. At the very least, such restrictions reduce the freedom of choice as to where families might wish to live. The application of pre-entry tests might also affect traditional patterns of marriage and could lead to resentment amongst existing long-term migrants. And on a practical level, integration is made easier when migrants do not have to worry about their family in another country.

In November 2011, the European Union issued a Green Paper (COM(2011)735) on the right to family reunification of third-country nationals living in the European Union. One of the questions asked in this paper concerned the optional clause (Article 7.2), which enables member States to require third-country nationals to comply with integration measures. This was one of the most controversial and debated requirements during negotiations on the present directive, and in its current form it does not give any precise indication what these integration measures should entail or how they should be applied. However, in the Green Paper the point is made that the admissibility of integration measures should depend on whether they serve the purpose of facilitating integration and whether they respect the principles of proportionality and of subsidiarity. Decisions on the application for family reunification in relation to passing tests should take into account whether there are available facilities (translated materials, courses) to prepare for them and whether they are accessible (location, fees). Specific individual circumstances (such as proven illiteracy, medical conditions) should also be taken into account.

The summary of stakeholder responses to the Green Paper highlights widely differing views and practices across member States ranging from no integration measures at all to compulsory participation in pre- and post-entry classes and the passing of tests.

A worrying trend is that some member States suggested changes to the rules on this issue with a view to giving member States the flexibility to introduce even more restrictive provisions (Austria, Germany and the Netherlands for example). I am quite sure that Austria and Germany wanted to make sure they could maintain their pre-entry test, but that they did not want the possibility to make them more restrictive. The Netherlands did. Other member States favoured a less restrictive approach with several opposing compulsory integration measures (Belgium, Finland, Portugal, Romania and Slovakia, as well as Turkey).

Many questions remain unanswered and further research is necessary. A list of some of these questions is contained in Appendix 3.

On the basis of this consultation round, Commissioner Malmström announced that she would not propose amendments to the directive, but would advocate the design interpretative guidelines for member States on how to apply the provisions. Furthermore, she announced that infringement procedures would be started vis-a-vis member States that did not comply with the directive.

Regarding pre-entry tests, however, the Commission has already clearly expressed its view. In 2011, the European Commission took the position in a procedure before the Court of Justice that a pre-entry test is not in accordance with the Family Reunification Directive if the requirement implies that family reunification can be denied for the sole reason that the applicant has failed the test. According to the Commission, Article 7.2 aims to promote integration, but cannot be used to undermine the objective of the directive to promote family reunification. This position does not leave any room for member States to maintain these tests. The Commission’s view has already had an impact in national courts. For example, in October 2012, the Administrative Court of Berlin asked the Court of Justice of the European Union (CJEU) whether the language test foreseen in the EEC-Turkey Association Law should be allowed and whether it was in compliance with Article 7.2 of the Family Reunification Directive. Shortly after this, the Dutch District Court fully endorsed the Commission’s position before the Court of Justice, indicating that a preliminary ruling was not necessary as the interpretation of the Commission was crystal clear.

48. The revised European Social Charter also protects the right to family reunion for migrants under Article 19.6, and the European Committee of Social Rights has looked at the issue of test requirements linked to integration policies. This examination has prompted the committee to state that a requirement that members of a migrant worker's family sit language and/or integration tests to be allowed to enter the country, or pass these tests once they are in the country in order to be granted leave to remain, discourages applications for family reunion because of its particularly stringent nature. As a result, it undermines the substance of the obligation laid down in Article 19.6 of the Charter. Such requirements are therefore, in the opinion of the committee, not in conformity with the Charter.

49. A further consideration is the fact that tests require a specified level of knowledge and skills. The effect of this is that they can be used as a means to exclude poorly educated migrants, resulting in discrimination against those who have not received the necessary level of education, even though they might be useful and productive members of society.

50. There is also the issue of access to rights and benefits in receiving societies. There is an increasing risk that many migrants will be left in a state of limbo. They may have to renew their periods of temporary residence as an alternative to a permanent stay which, in many countries, will be an expensive process. This may cause all sorts of anxiety and the question has to be asked: is it fair and reasonable to subject someone who has been working for many years, and has been productive and honest, to the uncertainty of a temporary and insecure situation? If migrants retain their temporary residence permits because they cannot meet the requirements for permanent residence, they will in general have fewer opportunities for integration. Migrants with a temporary residence permit, depending on the member State where they live, may be barred from full access to the labour market, from social benefits or from voting in local elections.

51. The most important consequence of not obtaining the right to permanent residence is the continuation of insecurity regarding the right to residency. If, for instance, an income requirement is no longer met, migrants may face withdrawal of their permit or expulsion. Undoubtedly, this uncertainty will affect their prospects for integration. Although the conditions set by member States are meant to promote integration, the opposite effect can be produced in the case of migrants who are willing to integrate but who lack the capacity to pass the tests. Furthermore, in some member States they may even lose their temporary residence permits if they do not fulfil the integration requirements (for example in Austria and the Netherlands). So, ultimately, is it appropriate to deny residence to someone who, measured against the Council of Europe domains of integration, is well integrated, simply because they cannot pass a test?

52. There are also particular issues in respect of refugees. According to the United Nations High Commissioner for Refugees, it has to be borne in mind that persons who are forced to leave their home and their country do not always have all the documents they need in their possession, and that the trauma and pressure they undergo may have an impact on their ability to learn a new language, despite their efforts and their willingness to do so. Also, language tests and examinations concerning the history and culture of the host country may be more difficult for certain categories of refugees and non-forced migrants, in particular the elderly or those with low levels of education or literacy. But the consequences of insecurity about a permanent stay are arguably greatest for refugees or migrants who have been granted a residence permit on humanitarian grounds. For them, there will be the constant fear of expulsion to a country where they risk persecution or arbitrary violence which has perhaps caused them physical and/or mental trauma in the past, and the memory of which they would rather put behind them.

53. Much more work needs to be done on examining the human rights implications of these tests, and, as with other parts of this report, some of these questions are included in Appendix 3.

5. Conclusion and recommendations

54. Tests were instigated by a small number of countries but have now been embraced by a significant number of States. That being the case, it is right and proper that their use and effectiveness should be critically examined, particularly where there are concerns that they breach human rights, where they have unintended consequences, where they exclude persons with a legitimate claim to entry, permanent residence or citizenship, and where they are actually functioning as control measures disguised as integration initiatives. The fundamental dilemma for Europe is therefore the extent to which tests promote integration and the point at which they become a barrier and actually begin to inhibit it.

55. There may be circumstances in which it is appropriate to start the process of integration prior to arrival. It is not unreasonable, and may indeed be beneficial, to give intending migrants some clear ideas about the reality of life in the country to which they aspire to migrate. Often contacts and indeed well-intentioned relatives and friends can paint a completely unrealistic picture of life in another country. Providing accurate and honest guidance can therefore go some way towards managing expectations and promoting integration. It could also be argued that some form of testing to ensure that individuals have actually learnt from the orientation and training that has been provided is also legitimate.

56. There are some good examples in which countries provide subsidised or free pre-entry training in the language and knowledge of the receiving society for the applicant. This practice should be encouraged as it clearly supports integration. However, if such programmes are linked to a formal test, this should serve only to indicate what further support might be needed post-entry and should not be the sole means of excluding migrants who fulfil all the other criteria for entry.

57. A related issue is the level of competence required to pass pre-entry language tests. The levels should be attainable and there should be strong support for learners. There should be alternatives to formal tests for those who do not have the necessary levels of literacy to take them, and under no circumstances should migrants who have failed tests despite their best efforts be denied family reunification. This means that making an effort should be a requirement but achieving a given competence level should not.

58. Turning to post-entry tests, these were originally used to check that persons seeking citizenship had adequate knowledge of the language (and latterly country knowledge) of the receiving society. There are some examples of good practice, where tests have been set at an appropriate level, with alternatives also available, and where effective support is offered to those preparing for them. However, these examples are increasingly rare, and of even more concern is the fact that some countries that had good models are now changing them and introducing tougher measures.

59. Of greater concern is the fact that tests originally intended for citizenship are increasingly being used to determine whether or not someone should be granted permanent residence. It is very clear that uncertainty about their future status and security of residence will inhibit the ability and motivation of migrants to invest in their integration. Furthermore, a temporary residence permit can undermine opportunities to integrate, for instance because of poorer prospects in the labour market, lack of access to social benefits or to a mortgage in order to buy a house, etc.

60. It is equally clear that setting levels of competence that for some are unattainable will prevent access to the full range of rights and responsibilities that these migrants should enjoy. The most vulnerable groups in this respect include older people and those with low literacy or little education.

61. Whilst countries have every right to seek to reduce or manage migration, using means that are described as integration measures in order to do so will surely devalue genuine and well-intentioned integration programmes.

62. A particular problem occurs in relation to family reunification. Tests that result in excluding persons who otherwise fulfil the requirements for entry or that have the potential to breach the human rights of the persons concerned should not be permitted.

We should also take steps to ensure that failure in a post-entry test does not exclude the person concerned from access to services, rights and benefits that they would otherwise have enjoyed.

63. Further research into the effectiveness of tests in promoting integration is required, for example to ascertain what the appropriate level of language competence should be and identify the best ways in which this level of competence can be developed. There should also be an exploration of the levels of competence that can reasonably be achieved by persons with limited literacy.

64. In 2012, the Council of Europe launched a website devoted to the Linguistic Integration of Adult Migrants (LIAM).

Its main aims are to offer a platform that enables member States to exchange experiences and reflect on policy and practice in this area; to provide assistance to them in developing coherent and effective policies (or in reviewing them) in keeping with shared Council of Europe values and principles; to provide practical support (including tools) for the effective implementation of policy and for assessment; and to encourage good practice and quality assurance in language teaching and testing.

65. The work already undertaken by the Council of Europe’s Language Policy Unit is therefore an invaluable tool in aiding member States to design language policies, courses and tests appropriate to the needs of migrants and indeed of society. Using the guiding principles outlined on the LIAM website, member States should be encouraged to:

define required proficiency levels in a realistic and flexible manner that reflects the actual needs and capacities of individual migrants;

ensure that formal tests, where used, conform to accepted standards of quality and are not misused to exclude migrants from society;

devise effective incentives rather than ineffective sanctions; these could include tangible rewards for language learning, such as speedier access to employment or social benefits in order to enhance motivation;

consider the advantages and disadvantages of different types of testing and focus on “best practice” in test design;

ensure that tests are taken under conditions which are equally fair for all.

66. Member States should be encouraged to look at alternatives to tests that might be fairer and less discriminatory for all or for specific groups of migrants. There is a pressing need for further comparative evaluation studies of the long-term effectiveness of tests as a tool for efficient, sustainable and user-friendly pre- and post-entry integration. In particular, this research should examine the extent to which the mode of entry (work, family reunion, asylum) should be taken into account. We should consider asking the Language Policy Unit of the Council of Europe to join in this work. The unit is well placed to advise on this and could also be asked to consider whether further work might be undertaken based on the CEFR to make it a more suitable aid to generic language testing, and to pilot some processes the unit has already developed.

67. Last, but by no means least, there should be a comprehensive, open and honest examination of the risks posed to integration by allowing certain migrants to reside only on a temporary basis because of their inability to pass a linguistic or integration test.

Appendix 1 – Case studies showing how integration tests have developed

(open)

The United Kingdom

In the case of the United Kingdom, it is particularly interesting to look at the evolution of requirements. In the British Nationality Act of 1948, the requirement was simply that an applicant had to have “sufficient knowledge of the English language”. In 1981, the requirement was changed to include sufficient knowledge of the English, Welsh or Scottish Gaelic language. In 2002, however, the Nationality, Immigration and Asylum Act introduced an objective standard for the first time. This was equivalent to what is known as level B1 contained in the Council of Europe Common European Framework of Reference for Languages (CEFR) scale.

There was, however, an alternative, which was to show progress from one language level to the next through taking a language course with a specially developed citizenship curriculum. The clear policy intention of this alternative route to citizenship was to accommodate those applicants for whom the achievement of the entry level standard would be difficult or impossible, perhaps through low levels of education or literacy in their own language.

In 2005, the “Life in the UK” knowledge test was introduced together with a handbook. The purpose of the (computer-based) test was not to alter the requirements but to provide a more pragmatic and cost-effective means of assessment. The earlier language-testing route however remained, so the knowledge test was simply an additional option.

In 2007, however, the test became necessary to acquire permanent residence (although it was not necessary to take it again for the purposes of acquiring citizenship). The objective of this policy shift was to ensure the integration of anyone seeking to stay permanently in the United Kingdom. The alternative “progress” route for persons whose language competence was too low to enable them to take the test, remained however. In 2010, the United Kingdom went a step further by introducing a pre-entry language requirement for spouses, which was similar to that introduced earlier by the Netherlands in 2006. Anyone now seeking entry as a spouse has to provide evidence of having taken a language test before an entry visa is issued. The requirement was set at a lower level than previously set for citizenship (namely A1 as opposed to the equivalent of B1 of the CEFR). Unlike the Dutch model, there was however no civics element. Again, the stated policy intent was to support integration.

The situation is set to change again in the near future with a substantial hardening of the requirements. As from October 2013, all applicants for permanent residence or citizenship will have to pass a new test based on a third edition of the “Life in the UK” handbook which will have British culture and history at its heart. The language level for permanent residence will be B1, rather than the lower A1, for all applicants and the alternative “progress” route will be removed.

The Netherlands

There was no real policy in the Netherlands until the beginning of the1980s. Migrants or “guest workers” as they were often referred to were expected to return to their countries of origin. This did not happen and many asked for their families to join them. In the 1980s, the government decided to take political responsibility for the immigrants, introducing a so-called “minorities” policy in order to provide them with greater rights and improve their integration. The Dutch Nationality Act made naturalisation a right rather than a “favour”, but one of the conditions was a language and integration requirement which required the newcomers to be able to have a short conversation with a municipal official.

Following an increase in the number of persons naturalised, some political parties considered that naturalisation had become too easy, with too little demanded of migrants. The government nonetheless adopted an integration policy in which immigrants’ own responsibility for integration was emphasised. In 1998, this policy started to change when the Newcomers Integration act was adopted. This obliged newcomers to participate in integration courses with a test at the end measuring language skills, with a goal of A2 of the CEFR. This level was not attained in 60% of cases. The courses and tests were free of charge and those dropping out could be fined. This rarely happened, however, as most newcomers found a job, and it became pointless and expensive to insist on them paying a fine and continuing with the courses when they had made the most important steps towards integration.

The Newcomers Integration Act had an important impact on the debate on Dutch nationality law. Some thought it was wrong to expect more from newcomers in terms of language and integration, than from future Dutch citizens. As a consequence, the Dutch Nationality Act was changed, and as of 1 April 2003, those applying for Dutch nationality were required to pass a “naturalisation test”, examining oral and written language skills at level A2, as well as knowledge of Dutch society. The cost of the test was set at €260. However, 40% of all those who registered for the test did not pass it. After its introduction, the numbers of naturalisations dropped by 50%. Surprisingly there was no parliamentary debate on the implications of this drop and on whether or not this was desirable and benefited integration.

The 9/11 terrorist attacks led to a new debate on immigration and integration in the Netherlands with the rise of Pim Fortuyn and his LPF party. It was put forward that integration had to start in the country of origin and then continue after arriving in the Netherlands. A new law entered into force in 2007. Passing an exam became a condition for permanent residence and naturalisation, as well as an independent residence permit. The level of the exam was maintained at A2 (even though past experience had shown this to be unachievable for many), it was no longer paid for by the State and immigrants were required to pay for courses for which they could obtain a loan (70% refundable on passing, up to a maximum of €3 000). Soon after its introduction it was clear that the new system did not lead to the desired results. Few immigrants followed the courses due to the cost. In the autumn of 2007, the government started paying for the integration of all immigrants. As a consequence, the numbers of immigrants following courses rose. Nevertheless funding for integration has been cut again. Today, only refugees have the right to reimbursement of the costs for their integration courses. This will almost certainly lead to a decrease in the number of persons taking these courses. At the same time, not passing the test within three years can lead to the withdrawal of the residence permit, potentially affecting all family members as they are the target group of the Civil Integration Act.

Appendix 2 – The Council of Europe “Common European Framework of Reference for Languages: Learning, Teaching, Assessment” (CEFR)

(open)

This reference tool aims at providing a transparent, coherent and comprehensive basis for the elaboration of curriculum guidelines, the design of teaching and learning materials and the assessment of language proficiency. As indicated by its title, the CEFR is a framework of reference, not a normative instrument.

The CEFR defines proficiency at six ascending common reference levels arranged in three bands:

A1 and A2: basic user

B1 and B2: independent user

C1 and C2: proficient user

It uses “can do” descriptors to indicate the user/learner’s proficiency at each level – what learners can do with the language(s) they are learning. The levels are not fixed points on an ascending scale of L2 competence, but are increasingly broad bands of proficiency.

A unique feature of the CEFR is its flexibility with further subdivisions within these levels. The CEFR uses a series of descriptors to indicate precisely a learner’s level of competence in each of the areas of speaking, listening, reading and writing (there are 34 scales, summarised in a “self-assessment grid”), as it is rare for someone to be equally competent in each of these areas.

The originality and usefulness of this process is that it does not lead to “closed” levels; it involves intermediary levels and in particular can be utilised to ascertain the specific areas on which individual learners might need to work in order to achieve their goals.

The CEFR can in particular be used to define “profiles”, for example A2 level for speaking, but A1 for reading or writing, rather than homogeneous levels (A2 for all competences). In adapting the CEFR levels for official purposes such as residence or citizenship, it is important to set realistic and achievable levels, bearing in mind that the majority of native speakers do not attain the higher CEFR levels.

Appendix 3 – Questions to be examined concerning integration tests

(open)

Questions on the impact of integration tests:

Do tests promote integration? Do they have other (perhaps unintended) effects?

To what extent does the mode of entry (work, family, refugee) make a difference?

Does the offer of education make a difference in outcomes?

How can integration policies that do not include tests be compared to those that do in respect of their effectiveness?

What indicators of integration other than tests might be used?

Do tests contribute to the integration of migrants, and if so, to what extent?

At what point do language or integration tests cease to be of benefit and become a burden?

For those people with little or no literacy in their own language, what is a realistically achievable level of competence?

As this has proven to be much more effective and productive, should migrants be encouraged and supported to take language lessons as soon as they arrive in the receiving country?

Would it be more appropriate to teach the language of the receiving country during the waiting period, once a visa application has been made?

Questions relating to family reunion:

What safeguards need to be put in place to ensure that families of persons accepted as refugees seeking family reunification are not disadvantaged by pre-entry or post-entry testing?

To what extent does the introduction of tests as a requirement for family reunification improve the integration prospect after arrival of the spouse?

To what extent does testing constitute an obstacle to family reunion?

What could be the unforeseen consequences of denying family reunification (long-term separation, fewer integration prospects for the spouse living in Europe, irregular migration?)

Taking into account the right to family reunification, to what extent may integration or language requirements be imposed on spouses who want to reunite?

Questions on the human rights implications of these tests:

Do formal tests merely restrict or delay family reunification, permanent residence permits or citizenship for those granted refugee status?

Do pre-entry tests infringe the right to family life or the right to marry and, if so, what might be the remedies?

Which categories of migrants are not able to pass the test, and therefore excluded from certain residence rights?

In what way does this exclusion affect their integration or integration prospects?

How can we guard against integration measures being used as a means to reduce migration?

If this is the perception, what impact is there on the credibility of integration initiatives?

What are the risks regarding integration and human rights of allowing certain migrants to only reside on a temporary basis, because they are not able to pass a language or integration test?

Would it be a breach of someone’s human rights to remove them from a receiving society or to administer penalties simply for failing to pass an integration test?