As part of a study designed to better understand the impact of forced migration on fertility, mortality, violence and traumatic stress among Sudanese nationals living in southern Sudan and Ugandan nationals and Sudanese refugees living in northern Uganda, we interviewed 3371 individuals from 1842 households in the Ugandan and Sudanese populations in the West Nile. Interviews were structured and were administered in the native languages of Lugbara or Juba Arabic. The study's design involved a multi-stage sampling design.
The full training of the interviewers took two months. The project objectives and the rationale behind the structure of the survey instrument as well as that of each question in the questionnaire were discussed in detail. Great attention was also paid to issues such as initial contacts, maintaining a professional attitude while in the field, avoiding influencing the respondent, and reducing interviewer and courtesy biases. The importance of collecting information by means of standardized questions so that the same question was asked to all respondents is stressed and questioning and probing skills were developed. Supervisors were instructed separately on data collection guidelines, their roles and their responsibility to ensure data quality. Keeping in mind the sensitive nature of some of the questions regarding violence and trauma and the fact that the team members were from the study population and probably had experiences similar to the respondents, a workshop on sexual and gender-based-violence was conducted by a consultant to the UNICEF office in Kampala, before the survey. The aim of this workshop was to increase awareness and sensitivity of the team towards respondents and their experiences. Another consultant to the project reviewed the team's interviewing skills and the project's data quality control measures just before the start of the survey. Problem areas were identified and remedied.
Data were complete and analyzed for N = 3179 respondents: 2,540 (75 %) of the respondents were women (15–50 years of age) and 831 (25%) were men (20–55 years of age). Details of the sampling, translation and assessment procedures, as well as the socio-demographic characteristics of the populations, have been described elsewhere [15 (link)].
Traumatic events were assessed using a checklist consisting of possible war and non-war related traumatic event types (i.e. witnessing or experiencing injury by a weapon or gun, beatings/torture, harassment by armed personnel, robbery/extortion, imprisonment, poisoning, rape or sexual abuse, beatings, abduction, child marriage, forced prostitution/sexual slavery, forced circumcision, etc.). The checklist was compiled after interviews with key informants (security personnel, doctors, community leaders, women's representatives) and 30 respondents from all three populations about their personal history of stressful events. Following these interviews, the single events obtained in these studies were rated as being potentially traumatic by experts. The following pilot checklist was pre-tested among further 44 Ugandans and Sudanese in areas not selected for the survey and modified according to the suggestions of the respondents. A primary item analysis based on inter-item correlations led to the exclusion of some events that were obviously not directly related to traumatic histories, e.g. the experiencing of witchcraft. Events included 19 experienced events and 12 witnessed events. Respondents were asked for each event type if they had experienced or witnessed such an event ever (i.e., lifetime experience) and if it happened in the past year. PTSD in respondents was assessed using the Posttraumatic Stress Diagnostic Scale (PDS), modified for assessment by trained lay interviewers [16 ]. The PDS is a self-report measure widely-used in industrialized countries as a screening instrument for the diagnosis and severity of PTSD based on DSM-IV Criteria.
Confidentiality was assured and it was explained that researchers were not working for any UN or Ugandan government organization. Informed consent was obtained using a standardized form explaining the potential risks of participation and explaining that no compensation would be provided. Informed consent forms were signed by the respondent and a witness; fingerprints were taken from illiterate respondents. No financial incentives were provided and respondents were informed that no improvements in living conditions were to be expected as a result of participating in the survey. Respondents were provided with referrals to counseling services provided by NGOs where available.
The full training of the interviewers took two months. The project objectives and the rationale behind the structure of the survey instrument as well as that of each question in the questionnaire were discussed in detail. Great attention was also paid to issues such as initial contacts, maintaining a professional attitude while in the field, avoiding influencing the respondent, and reducing interviewer and courtesy biases. The importance of collecting information by means of standardized questions so that the same question was asked to all respondents is stressed and questioning and probing skills were developed. Supervisors were instructed separately on data collection guidelines, their roles and their responsibility to ensure data quality. Keeping in mind the sensitive nature of some of the questions regarding violence and trauma and the fact that the team members were from the study population and probably had experiences similar to the respondents, a workshop on sexual and gender-based-violence was conducted by a consultant to the UNICEF office in Kampala, before the survey. The aim of this workshop was to increase awareness and sensitivity of the team towards respondents and their experiences. Another consultant to the project reviewed the team's interviewing skills and the project's data quality control measures just before the start of the survey. Problem areas were identified and remedied.
Data were complete and analyzed for N = 3179 respondents: 2,540 (75 %) of the respondents were women (15–50 years of age) and 831 (25%) were men (20–55 years of age). Details of the sampling, translation and assessment procedures, as well as the socio-demographic characteristics of the populations, have been described elsewhere [15 (link)].
Traumatic events were assessed using a checklist consisting of possible war and non-war related traumatic event types (i.e. witnessing or experiencing injury by a weapon or gun, beatings/torture, harassment by armed personnel, robbery/extortion, imprisonment, poisoning, rape or sexual abuse, beatings, abduction, child marriage, forced prostitution/sexual slavery, forced circumcision, etc.). The checklist was compiled after interviews with key informants (security personnel, doctors, community leaders, women's representatives) and 30 respondents from all three populations about their personal history of stressful events. Following these interviews, the single events obtained in these studies were rated as being potentially traumatic by experts. The following pilot checklist was pre-tested among further 44 Ugandans and Sudanese in areas not selected for the survey and modified according to the suggestions of the respondents. A primary item analysis based on inter-item correlations led to the exclusion of some events that were obviously not directly related to traumatic histories, e.g. the experiencing of witchcraft. Events included 19 experienced events and 12 witnessed events. Respondents were asked for each event type if they had experienced or witnessed such an event ever (i.e., lifetime experience) and if it happened in the past year. PTSD in respondents was assessed using the Posttraumatic Stress Diagnostic Scale (PDS), modified for assessment by trained lay interviewers [16 ]. The PDS is a self-report measure widely-used in industrialized countries as a screening instrument for the diagnosis and severity of PTSD based on DSM-IV Criteria.
Confidentiality was assured and it was explained that researchers were not working for any UN or Ugandan government organization. Informed consent was obtained using a standardized form explaining the potential risks of participation and explaining that no compensation would be provided. Informed consent forms were signed by the respondent and a witness; fingerprints were taken from illiterate respondents. No financial incentives were provided and respondents were informed that no improvements in living conditions were to be expected as a result of participating in the survey. Respondents were provided with referrals to counseling services provided by NGOs where available.
Free full text: Click here
