The choice of test type to use in immigration processes is not straightforward. Policy makers need to listen carefully to the advice of professional developers
The outcomes of language tests have important consequences for test takers and for society. This is particularly true when tests are used in migration. For this reason, decision to use tests, either to support learning or as an instrument of policy, should not be taken lightly.
A first and fundamental question is whether or not tests are really necessary or if another method of assessment would be more appropriate.
If policymakers decide to use language tests, it is vital that the tests that they select meet the highest quality standards. They should also ensure that there is an effective mechanism for monitoring any negative consequences that may arise from the policy and the use of the tests. At all times, fairness to test takers should be given priority.
When approaching testing, it is all too easy to suppose that a single solution will apply in many cases, but policymakers need to be made aware of the complexity of the issues they are entering in to. That is why it is important to have a clear understanding of how test results will be used from the outset.
In many countries, policymakers use the metaphor of "the journey to citizenship" in their presentation of immigration processes. In France policy is presented under the slogan: "Acting at every step in the integration process", while in the UK newcomers will be asked to start demonstrating their engagement with society soon after arrival. In the words of immigration minister Phil Woolas: "British citizenship is a privilege which must be earned, providing mechanisms to speed up or slow down the journey towards settlement."
But in the UK there has been a poor understanding of the "stages of the journey". If high quality tests are to be developed that are fit for purpose and with the potential for positive impact, it is important to have a better understanding of this "pathway".
In most destination countries this pathway follows six typical stages: pre-entry; arrival and entry; extension of stay; permission for long-term settlement; application for naturalisation, and granting of citizenship. Through careful test development and validation more effective tests can be developed that meet the policy needs at each stage of this journey.
For example, a settled migrant studying English might choose to take a test at the end of the course as a record of achievement, perhaps to move up to a higher class or as a qualification for use in the work place. Such a test should fit with educational objectives in keeping with life-long learning and basic skills development.
If the test is needed to determine whether a newcomer has the English language skills to enter the country
to practice medicine, other considerations are likely to be more important. The test will need to be widely available outside the country and be supported by systems to prevent cheating to gain unfair access.
Investigating impact is an integral part of establishing the usefulness of an assessment system – its fitness for specific purposes and contexts of use and to guide future developments. Examination providers need to collaborate with the various stakeholders who use of their examinations and to listen to feedback about assessment practices.
For examinations boards this means collecting relevant data and, when necessary, taking appropriate action to improve both the tests themselves and the uses to which the test results are put.
Some of the key questions we need to ask are summarised, right. To these we can add: who is teaching towards the examination and under what circumstances and what kinds of courses and materials are being designed and used to prepare candidates? We also need to understand what effect the examination has on public perceptions generally and how it is viewed by students, examination takers, teachers, parents, and the wider public outside education.
On the basis of information collected (the evidence), informed decisions can be made about the fairness of the assessment regime as a whole and on how improvements can be made.
KEY QUESTIONS
What is the purpose of the test and how does this influence the level, content, administration and use of results?
It makes a difference whether a test is intended to motivate the learners (to help them use and improve their current competence in the target language), to ascertain whether their competence is sufficient for participation in well-defined social situations (for study or work), or to make decisions that affect their legal rights, such as their right to remain in the country or acquire citizenship. It is very important for the test takers to know the purpose, so that they can prepare accordingly. Other members of society should also understand the intended purpose so that the results can be interpreted and used correctly.
What level of language proficiency is required for the stated purposes?
An appropriate level of proficiency in the target language should be chosen to reflect the stated purpose of the test and use of the results. Test development and validation procedures must be employed to ensure that this level remains stable and consistent whenever the test is used.
Who is responsible for test administration to ensure that standards of conduct are met?
Standardised procedures are necessary for the accreditation of suitable test centres and the administration of the tests. There must be adequate provision of information and support systems for the test takers and other test users.
What are the procedures for monitoring test outcomes and what evidence is collected to demonstrate fairness and that the test does not lead to discrimination?
Data should be collected and analysed regularly to ensure that the test is valid and reliable and fulfils its intended purpose effectively, and to monitor for unintended consequences. Changes in the groups taking the test, as well as the effect of general changes in the contexts in which the test is being used, can only be detected and acted on when there is a proper system of monitoring the results of the test for the various groups of test takers
• Questions by Nick Saville and Piet Van Avermaet