In order for someone to be recognised as a refugee in Norway, they must apply for protection upon arrival. In the ensuing asylum process, applicants are obligated to provide as much information as possible and a credible account of their case (Ot.prp. nr.75 (2006-2007)). The immigration authorities, on the other hand, are obligated to examine the case before making a decision on protection, thus to ensure maximum possible elucidation of the case (Forvaltningsloven §17). Consequently, the authorities must collect sufficient reliable information to enable the Directorate of Immigration (UDI) to decide whether an applicant is entitled to protection or can be granted the right to remain on humanitarian grounds, The central proof in an asylum case is the asylum interview (UNHCR, 2013).
The way in which the asylum interview is conducted is therefore key for the type of information that emerges and whether it is considered reliable. Previous research has shown how various factors such as interview method (Walsh & Bull, 2010; Griffiths & Rachlew, 2018; Langballe et al., 2010), understanding of the legal obligations involved (Schjatvet, 2021) and the interpersonal encounter (Risan & Madsen, 2019) affect the quality of investigative questioning and interviews. However, no studies have previously been conducted in Norway based on direct observation of how UDI staff conduct asylum interviews in practice, where these factors can be studied empirically.
In its asylum interviews, UDI applies the investigative interview method. The objective is to collect as much relevant, reliable and accurate information as possible. The investigative interview method is based on open information-seeking, with alternative hypotheses and awareness of memory and cognitive limitations (Jakobsson, 2014). Open-ended questions are used, and the information provided is checked against available facts. The asylum interview is more complex than a police interrogation and unique in the sense that the investigative interview method must be applied across cultural and language barriers. Asylum interviews almost always require an interpreter, which makes communication difficult, and interviewers need to possess the necessary cultural competence and country information in order to conduct a good-quality interview. A key point in this evaluation was therefore the extent to which the investigative interview method has been adapted to account for the asylum setting, and whether there is a need for any further adjustments.
Research assignment and design
This report is the result of a project that Fafo has undertaken on assignment from the Directorate of Immigration (UDI). We were commissioned to investigate whether UDI’s implementation of asylum interviews is of sufficient quality. We have operationalised ‘sufficient quality’ by examining whether the asylum interviews meet UDI’s internal quality requirements (UDI 2019-015V1), and whether these requirements are in line with recognised quality standards for the investigative interview method. The report therefore addresses the following research questions:
- How the asylum interviews are conducted in practice and whether the implementation meets UDI’s internal quality standard.
- Whether UDI’s quality standards live up to those found in research literature on the investigative interview method.
- Which of the ways to conduct asylum interviews help collect reliable, sufficient and relevant information.
- To what extent the interview reports provide a good account of the asylum interviews.
- How the institutional framework around the asylum interviews has a bearing on the implementation.
The data material consists of 50 audio recordings of asylum interviews and 20 interview reports. The asylum interviews were conducted in late 2020 and early 2021, immediately before and after the COVID-19 pandemic hit Norway. The audio recordings were randomly selected, but with an even division between in-person and online interviews. The sample turned out to consist of 24 interviews that were held in person on UDI’s premises, while 25 were conducted online.
A quantitative analysis was undertaken of 49 audio recordings, where we coded each part of the interview according to whether the applicable quality standard was met. For example, we coded the type and number of questions, expressions of empathy, active listening, inappropriate behaviour and the length of each phase, as well as the number and type of questions per interview topic.
Ten interviews were analysed using a specially developed analytical tool, the Griffiths Question Map, a method that presents all the questions in the interview graphically in a chronological map of the entire interview. By reviewing a completed analysis, the interviewers can see the pattern of questions and topics that they or others have used, and thereby identify strengths and weaknesses by comparing with other maps and/or feedback from colleagues proposing better strategies.
Nineteen interviews with attached interview reports were analysed qualitatively. The qualitative analysis made it possible to describe the implementation of the interview in more detail, and to see how the interaction between the interviewer and the applicant affects the amount and quality of the information that emerges. To assess the quality of the interview reports we collated the audio recordings with 19 of the interview reports. We examined what is made available in writing for the further casework and in what ways, if any, information is changed when written down.
To obtain insight into the institutional framework that impacts on the implementation of the asylum interviews and to better understand the procedures that precede and follow the interviews, we conducted a group interview with case officers in UDI.
Main findings
It can be stated that in general, UDI’s interview model and quality standard largely live up to the quality standard found in the research literature. UDI’s methodological basis is also in line with what are referred to as ‘best practice guidelines’, developed by the European Union Asylum Agency (European Asylum Support Office, 2015). There are, however, a few minor points in which the quality standard deviates from or is less detailed than what is described in the literature. A quality standard should ideally better operationalise concepts such as ‘professional attitude’ or how credibility can be objectively and strategically identified. Research and the EUAA interview guide underscore the importance of the post-interview evaluation. This includes both an evaluation of the interview itself and a self-evaluation by the interviewer. The rationale for highlighting this aspect of the investigative interview method is that systematic evaluation promotes self-development and ensures that interviewers are able to maintain a professional attitude over time while also taking care of their own needs. UDI’s quality standard describes a number of aspects to be considered after the asylum interview, but evaluation is not included among these.
While the asylum interview method and the quality standard conform to the international level, the implementation of asylum interviews has a potential for improvement. In general, there seems to be large, undesirable variation in the implementation between the best and the poorest interviews. None of the interviews was so poor as to give grounds for concern, and we find no obviously inappropriate statements or behaviours among the interviewers. We nevertheless find substantial differences between the best and the poorest interviewers. Given these differences in interview professionalism and the way in which the interviews were conducted, we may assume that not all asylum seekers have an equal opportunity to present their statements, nor will all asylum cases be equally well substantiated. We have been unable to assess whether all asylum seekers are provided with sufficient opportunity to present their statements, since such an assessment would depend on the interviewers’ judgements and the outcome of the cases, which we have been barred from examining.
One main finding in our data material is the large variation in how asylum interviews are conducted in practice. We find variations in all parts of the asylum interviews: their length, how and how many questions are asked, and the extent to which the interviewer demonstrates empathy. Moreover, there are variations in the way in which rapport is established and in the amount of detail in the description of identity, and there are large variations in whether and how the required information is supplied. Not all variations can be seen as a sign of varying quality. The asylum seeker’s background, education and degree of vulnerability will affect the richness of detail and length of the free explanation. Some of the variation may also be due to the interviewers’ differing need for information in the different cases. We have been unable to check for this aspect, due to the lack of access to background documentation, assessments and case outcomes. However, some of the variation may be associated with the interviewer’s ability to establish a good rapport and elicit supplementary explanations through correct use of appropriate prompts, as well as to appear empathic when the asylum seeker shows obvious signs of stress and discomfort. We wish to emphasise that we have found no indications of inappropriate behaviour or language among interviewers. We can thus ascertain that the great majority of interviewers behave in a professional manner. Although case officers point out challenges associated with online interviewing, we find few noticeable differences in quality between in-person and online interviews.
The report is organised according to the various stages of the interview. The initial stage, where a safe setting is created and the contact between the interviewer and the asylum seekers is established, defines the premises for the further course of the interview. The interviewers and asylum seekers in many cases seem to connect well. There is warmth and empathy in the tone of voice, and many succeed in remaining informal and relaxed in an otherwise formal and serious setting. This gives a good basis for a conversation. Maintaining a calm pace, checking whether the asylum seeker has understood or has any questions, while explaining in a suitable manner what will happen during the day, all seem essential. At the start of the interview, a lot of time is also devoted to clarifying personal information. The information that has come to light in the police registration is reviewed, often leading to many errors being corrected. Reducing the time spent on clarifying personal data, either by scheduling this for after the end of the interview or ensuring that it is properly undertaken by the police, appears to be opportune.
The information-gathering stage consists of the free narrative, exploration of topics and handling of any inconsistencies as required. The information-gathering stage is the key part of the interview and tends to be the asylum seeker’s only opportunity to tell their story and describe their reasons for applying for asylum. It is essential that this stage functions optimally in terms of both interview technique and setting. We find considerable variation in the way in which interviewers facilitate the free narrative. Some interviewers are very good at communicating the objectives, information needs and processes associated with the free narrative, while others introduce this thematic part rather superficially.
In our analysis of the exploration of topics, i.e. how the interviewer checks the credibility of the narrative and explores various topics from the free narrative in more detail, we have focused on the way in which questions are asked, what topics are followed up and how, and whether and how the asylum seekers’ fear of persecution if they are returned is explored. In this part of the interview, we also find considerable variation in the length and number of questions. Whether a large number of questions is problematic or not, depends on their appropriateness. The review with the GQM method highlights the importance of adequate funnelling during the exploration of topics. Strategic use of question types and good funnelling not only generate more detailed explanations and exhaustive answers, but also help reduce the need to ask many questions.
Good and sufficient information is essential for creating a positive and safe setting and a shared understanding of the objectives of the interview. This will have an impact on the further course of the conversation. Although many interviewers are good at providing the necessary information, there is a large variation in terms of how much information is provided prior to and after the asylum interviews. We find that interviewers who combine information with establishing initial contact, and thereby devote much time to this, are able to establish a good rapport and make the asylum seeker feel at ease. When the interviewer spends no more than a few minutes providing the same information, the ensuing interview appears to be little suited to creating an appropriate setting for the further interview process. Saving time initially seems less appropriate if this comes at the cost of a clear framework and relationship of trust. We find that no information is provided at the closure of the interview, and interviewers take little time to elaborate and explain. Time constraints are often cited as the main reason. Very few asylum seekers in our data material are informed unprompted about their right to receive a copy of the report. None were informed about their right to also receive a copy of the audio recording.
One objective of the asylum interview is to help identify vulnerable asylum seekers. The guidelines state that any health-related or other types of vulnerabilities that require special facilitation of the interview should be investigated at the initial stage. This is very rarely done. Also, very little time is spent at the end of the interview to investigate health challenges. On the whole, the questioning is far less detailed and explanatory than what is described in the interview report. This is extremely problematic, since little inquiry is made into traumatic events, violence, torture and mental health during the thematic exploration (see Chapter 6). It is therefore questionable whether the current interview practice helps identify all the vulnerable groups among asylum seekers.
As regards the quality of the interview reports and whether they reflect the asylum interviews, we can ascertain that most of the reports we have analysed provide a precise account of the content of the interview. The level of detail is generally high, and in practice, all the asylum seeker’s statements are recorded in writing. On the other hand, questions can be raised regarding the reporting process itself. One particular challenge stems from the fact that the interviewer keeps writing throughout the conversation.
In light of research literature on memory and with reference to studies of the investigative interview method, we take a critical view of this practice. We recognise the interviewers’ need for taking notes. We nevertheless believe that this issue requires more attention, and so-called post-interview note-taking should in our opinion be tested in the UDI setting. The interviewers should not be writing during the interview. Not only is this time-consuming, it can also lead to an unconscious reluctance to include new details and hinder professional curiosity, since this will impose extra work on the interviewer. It is also risky to only ask for information that is special or that in itself will give grounds for protection. This information is intertwined in the memory with information on other, more mundane issues. The interviewer therefore needs to be able to sort the information retrospectively and consider what should be included in the report or not.
