Sammanfattning av publikation

van Veldhuizen, Horselenberg, Landström, Granhag & van Koppen, 2017 🔗

Interviewing asylum seekers: A vignette study on the questions asked to assess credibility of claims about origin and persecution

Year: 2017

Type of text: Vetenskaplig Artikel

Published by:  Journal of Investigative Psychology and Offender Profiling, 14 (1), 3-22

Language: English

Author: Tanya van Veldhuizen, Robert Horselenberg, Sara Landström, Pär-Anders Granhag & Peter J van Koppen

Pages: 20

Available at:

*** start with reading methods part after the short description of the textt***

Short description of text 

“The credibility assessment, in turn, is often at the core of the decision to grant or reject international protection. Thus, in the asylum interview, information is sought that can be used to discriminate truthful from fabricated claims, and the questions asked should match that objective. In order to evaluate whether current interviews are likely to be successful in this respect, we study the style, type, and content of questions formulated by Swedish asylum officials in response to fictitious yet realistic vignettes.” (5)

“This study is the first to systematically evaluate the style, type, and kind of questions formulated by asylum officials to assess the credibility of a claim about origin or persecution.” (17)

Most important results

“according to guidelines by the UNHCR, statements should be plausible, sufficiently detailed, and consistent over multiple interviews with statements of other applicants and witnesses, and with other available information about the country of origin in order to be credible (UNHCR, 2013). Even though these indicators seem to be endorsed in practice (Granhag, Strömwall, & Hartwig, 2005), they are disputed. Scholars argue that statements that are vague, inconsistent, or lack detail may be a result of a failing memory as opposed to deception (e.g., Juliet Cohen, 2001; Graham, Herlihy, & Brewin, 2014; Herlihy, Jobson, & Turner, 2012; Herlihy, Scragg, & Turner, 2002; Herlihy & Turner, 2006; Herlihy & Turner, 2009). The UNHCR also recognizes that, when using the credibility indicators, asylum officials should take into account the possibilities and limitations of human memory and the personal background and circumstances of the applicant (UNHCR, 2013).” (4)

“The large majority, 92%, of the 325 formulated questions was posed in an information‐gathering style [open-ended]. Only 2% (n = 8) of the questions was accusatory in style [yes/no questions], with most accusatory questions (n = 4) being formulated in response to Vignette 4.” (10) 

“There was a considerable overlap in the kind of questions that asylum officials formulated when the origins of the asylum seeker were questioned (see Figure 1). Both in response to Vignettes 1 and 2, the majority of the questions (72% and 69%, respectively) were about the life of the asylum seeker in the country of origin, identity documents, or the flight from the country of origin to Europe.” (11)

“On average, the asylum officials were relatively confident that their questions would aid the credibility assessment, with a mean of 67.08 (SD = 18.23) on a scale from 0 to 100. Our participants also reported that the questions were based on knowledge rather than intuition (M = 66.80, SD = 17.32), and that they expected their colleagues to ask similar questions (M = 71.88, SD = 19.54).” (12)

“In general, participants found the vignette stories plausible, with mean ratings ranging from 60.92 to 74.29.” (12)

“There was, however, a significant difference in the extent to which participants were inclined to grant a refugee status to the asylum seeker, F(3,61) = 9.95, p < .001, ƞp2 = .33. Post hoc comparisons showed that participants who responded to Vignette 4 were significantly less likely to grant asylum than participants who responded to the other vignettes (all p’s < 0.003).” (13)

“Two main conclusions can be drawn. First, Swedish asylum officials at least know that they should predominantly ask open questions in an information‐gathering style to elicit information from asylum seekers. Second, when the origins of the asylum seeker are assessed, Swedish asylum officials seem to rely primarily on questions that assess knowledge about life in the country of origin, identity documents, and the flight. Such a thematic overlap was not found when the interview was held to assess the credibility of claims about persecution.” (13)

“Thus, to be more correct in the interpretation, one could say that the results indicate that asylum officials mostly ask open and information‐gathering questions at the start of the interview and when the applicant’s story seems relatively plausible at first sight.” (15)

“Questions that communicate the burden of proof implicitly emphasize the different positions of the interviewer and interviewee. Thereby, the questions may negatively influence the rapport between the interviewer and the interviewee; one of the important requirements for effective investigative interviews (e.g., Fisher et al., 2011; Memon, Meissner, & Fraser, 2010; Vrij et al., 2014).” (15)

“Even though we do not know whether the same proportions of open and information gathering would be found in full interviews, the results do indicate that Swedish asylum officials at the very least know how a careful interview should be set‐up.” (17)


“In cooperation with the Swedish Migration Agency (Migrationsverket), 100 asylum officials equally divided over four different locations in Sweden were invited to participate in this study. A total of 74 case officers entered the survey and agreed to participate.” (6)

“two vignettes described cases in which evidence for the origin of the applicant was absent, and two vignettes described cases in which evidence for the persecution of the applicant(s) was absent. We designed two different vignettes within each type of case to examine whether there is a thematic overlap in the questions formulated in response to vignettes that had the same investigative focus (i.e., credibility of origin vs. persecution story) but were otherwise diverse. All vignettes differed from each other in terms of nationality, sex, and age of the applicant(s). Furthermore, in the origin, vignettes we alternated between an urban (Vignette 1) and a rural background (Vignette 2), and in the vignettes focusing on persecution, we changed the ground for fear of persecution from membership of a social group (Vignette 3) to political views (Vignette 4).” (7)

“Vignette 1, in which the origin of a southern Sudanese woman fleeing with her children was questioned; Vignette 2, in which the origin of an Eritrean young man was questioned; Vignette 3, in which the persecution story an Afghan Sikh family fleeing victimization and extortion by the Taliban was questioned; and Vignette 4, in which the persecution story of a Turkish young man who flees after participating in protests against Erdogan was questioned. To make sure that the vignettes were realistic, they were first checked by an experienced Swedish case officer who did not further participate in the study.” (/)

“17 participants read Vignette 1; 18 participants read Vignette 2; 18 participants read Vignette 3; and 12 participants read Vignette 4. Immediately after the participants had read the vignette for the first time, they were asked to rate on a scale from 0 to 100 to what extent they thought the story of the asylum seeker was plausible. The plausibility rating was followed‐up by a task to encourage deeper processing of the story. The asylum officials had to read the story a second time and highlight 5 aspects of the story that they wanted to elaborate on in the interview (see Appendices A–D for the phrases that could be highlighted). Now, the participants were asked to formulate five questions that they would like to ask the asylum seeker to assess the veracity of the claim about origins or persecution.” (7-8)

“This study was an imitation of a real asylum case. Our set‐up is void of one essential factor: the interaction between the applicant and the official. The consequence of the absence of interaction in the current study is that asylum officials have unlimited time to think about what questions they want to ask. They do not have to respond quickly to what the asylum seeker has said or think about follow‐up questions while listening to the answer. Asking open questions is difficult and cognitively demanding (Memon et al., 1994). It may be the case that in real asylum interviews, asylum officials lack the time or cognitive recourses to formulate open questions and instead ask more closed questions.” (15)

The responses were coded and analysed statistically. 

Suggestions for further research

“To examine whether the asylum officials’ knowledge [of how to best ask questions]  is also systematically incorporated in practice, it would be valuable to study real asylum interviews.” (15)

Summarised by: Josefin Åström