the Quality of Life in Detention (MQLD) survey in IRC Campsfield House and IRC Colnbrook. From August 2010 — June 2011, working together with Blerina Kellezi, she further refined it in IRCs Yarl’s Wood, Brook House and Tinsley House where it was administered to 158 men and women. This article refers to that data.6 As is standard practice with survey administration, the respondents were anonymized and their responses were not independently verified. Not only would it have been difficult to check without betraying the identity of the participants, thus breaching their confidentiality, but, for much of the information — like time in the UK (at least for the undocumented), contact with family and friends, medical concerns etc — there would have been no independent consistently reliable sources in any case. Though efforts were made to obtain a wide-ranging and random sample, we do not claim that the participants were statistically representative of the whole detained population. Indeed, we are aware that, given that the majority of surveys were completed in English, nonEnglish speakers are underrepresented. On other parameters, however, for example in terms of the proportion of ex-prisoners, or in the numbers who had at some point claimed to have applied for asylum, the sample reflects the overall distribution of the total population. In the future we hope to translate the survey into high-frequency languages and to make greater use of interpreters. The first half of the MQLD records a number of self-reported demographic variables including age, nationality, marital status, history of imprisonment, immigration status and addiction. It asks respondents to disclose whether or not they are currently under an ACDT plan or have been previously and whether they have any health problems. This part of the questionnaire includes a measure of depression in an abbreviated form of the Hopkins Symptom Check-List (HSCL-D).7 The second part of the questionnaire measures views of the ‘quality of life in detention’. This section is divided into 12 dimensions addressing detainee perceptions of humanity, staff decency, immigration trust, immigration procedural fairness, relation to other detainees, care for vulnerable, relationships, healthcare, communication, isolation, distress, and drugs. It includes individual statements measuring perceptions of regime, racism, and visits as well as some open ended questions asking the respondents to list the three best and worst aspects of their life in the current removal centre. In less than one third of the total cases, one member of the research team read the questionnaire to the participants allowing her to clarify the questions if needed. This approach was taken to address the residents’ low literacy rates and their mixed levels of proficiency in English. The remaining participants preferred to read the questionnaire themselves next to the researcher or in the privacy of their own rooms and at their own time. Overall, the questionnaire took between 45-60 minutes to complete. The questionnaire had a number of spaces where the answers to the open questions could be recorded. Prior to completing the questionnaire, all participants were given an information sheet and a consent form to read, or had these read aloud to them by the researcher. Detainees at this stage were informed that if they told us of any plan to self-harm or harm others that we would pass that information onto staff. All participants were given the option to sign the consent form though no attempts were made to persuade the participants to sign it if they were hesitant to do so. Verbal consent was obtained from all participants. It includes individual statements measuring perceptions of regime, racism, and visits as well as some open ended questions asking the respondents to list the three best and worst aspects of their life in the current removal centre 6. 7. Part of the data was collected using open-ended questions like: ‘How does this removal centre compare to others you have experienced in UK?’ or ‘What are the 3 most positive things for you about life in this removal centre?’ Such data was coded into communal themes and analysed using content analysis. The aim of content analysis is to describe absence or presence of certain ‘words, phrases or concepts’ in a text or written data. The remaining data was analysed using a number of inferential statistics as appropriate including correlations, ANOVA, Chi-Square and regression. Internal reliability and Principal Component Analyses were conducted on the health scale and quality of life questionnaire suggesting that both measures can be used with this population. That measure is a self report checklist that aims to detect symptoms of anxiety and depression in a 4 point Likert-type scale ranging from 1=’not at all’ to 4=’extremely’. The items included ‘Crying easily’ and ‘Blaming yourself for things’. The original checklist has 25 items and the one used in this study had 14. The items were chosen due to their appropriateness in the context, and because the participants were already completing a lengthy questionnaire. The 14 items retained in this study measured depression. Issue 205 Prison Service Journal 11