Transcript PPT
Measuring Violence Against Women : Operationalization of surveys: Key points Workshop on Gender Statistics and Human Rights Statistics Nadi, Fiji 4-8 August 2014 Charlotte French, Statistician Demographic and Social Statistics Branch - UNSD Introduction • Guidelines for Producing Statistics on Violence against Women: Statistical Surveys • Three key areas (focus of presentation) – Questionnaire (Annex VII) and dedicated survey (Chapter IV and V) – Ethical considerations (Chapter V) – Special interviewer training (Chapter V) Dedicated survey Survey on violence against women Reasons for conducting: – – – – – To raise awareness about the problem To monitor trends To contribute to indicators at a global level To compare between countries To understand more about violence, the associations, risk and protective factors Dedicated survey • Allows for multiple opportunities for respondents to disclose their experiences with violence • Designed to enable the interviewers to establish a rapport with the respondents • Generally, yield higher prevalence rates • Considered to be more accurate reflection of the true prevalence of such acts of violence Considerations when inserting a module into an existing survey – Ascertain that the sample design does not systematically exclude subgroups of the population – Avoid surveys on unrelated topics – Check the suitability of the vehicle survey – Interviewers – special training – How to place the module for good flow – Protective measures for the interviewers and respondents – Additional support and referrals Risk higher levels of non-disclosure resulting in underreporting of violent experiences overview Dedicated Modules Costs High Reduced Questionnaire Design Sample size Broad/flexible Tailored Adjusted to main survey Could be too small Interviewer selection Specifically selected Could be challenge Interviewer training Specific training developed Need special training Data analysis Adaptable to obtain nec measures Restricted/limited Questions • Be explicit in operationalization of concepts • Clearly defined question “have you been attacked?” vs. “ have you been slapped?” • Cue respondents to consider a variety of different settings (home, school, work, etc) Model questionnaire UNECE has produced a model questionnaire, which can easily be adapted to an individual countries context. • ECE Model questionnaire (Annex VI) – Questions on demographic characteristics of respondent – Question on partner violence • Types of violence covered: psychological, economic, physical, sexual – Question on violence by others • Types of violence covered: physical and sexual Ethical considerations Ethical Considerations • WHO recommends that all VAW surveys adhere to the following ethical principles. These are an essential component of planning and undertaking surveys on this topic. – The safety of respondents and the research team is paramount and should guide all project decisions – Protecting confidentiality is essential to ensure women’s safety and data quality. – All research team members should be carefully selected, receive specialized training and ongoing support. – The study design must include actions aimed at reducing any possible distress caused to the participants. – Fieldworkers should be trained to refer women requesting assistance to sources of support. Where few resources exist, short-term support mechanisms can be produced – VAW questions should be incorporated into surveys designed for other purposes only when ethical and methodological requirements can be met. Safety • The survey should have a safe name, that does not reveal the nature of the survey to reduce risks, e.g. “Women’s Health and Life Events Survey” • Interviews held only in a private setting • Allow participants to reschedule if they feel unsafe • Only one woman per household • Train interviewers to handle interruptions Confidentiality • Pledge of confidentiality • Interviewers should not conduct interviews in or near their own community • Questionnaires should never include names or other identifying information • Questionnaires and/or data files should always be kept in a secure location and data files should be anonymised • Access to and control of data files protected • Handling of findings, documentations Informed consent • Needs to be reiterated throughout the process. • Need to obtain consent but signed consent forms pose potential risk to the respondents: – May allow breach of confidentiality – May affect a respondent’s willingness to disclose violence and bias the results Minimizing emotional distress • Survey questions might elicit negative emotional reactions among both respondents and interviewers. • So, interviewers need to be trained: – How to deal with distress – When and how to terminate interviews • Interviewers should have access to counselors and should not do too many interviews so as to avoid “burn-out” Referrals • Have information on referrals ready to offer to all respondents – Include info on a paper or card that is small enough to be hidden • Discretion is of utmost importance Training of interviewers Training of interviews • All project team members must receive specialized training • Ensure that interviewers are able to perform their duties across a variety of scenarios and possible outcomes • Training of field supervisors and their role to understand all aspects of the interviewing and data collection processes Training materials from UNECE model questionnaire • Contains a “question by question” explanation • Training program for interviewers with training facilitators manual • Interviewer manual including ethical and safety considerations Training of Interviewers • Specialised training for VAW surveys must ensure interviewers understand: – the extreme sensitivity of the topic – violence against women and its impacts on victims – societal myths about violence against women and how these affect victims and interviewers – gender issues at a personal and community level – the goals of the survey or module of questions on violence against women – ethical requirements of surveys on violence against women, including importance of and strategies for addressing confidentiality, safety and support for respondents – skills needed for interviewing on this topic including encouraging participation in the survey and creating a climate that promotes disclosures of sensitive survey questions – interviewing techniques for building rapport with respondents – skills to detect when respondents are at risk of being overheard and re-schedule interviews accordingly – how to identify and respond appropriately to emotional trauma by referring respondents to resources in the local community and avoiding emotional involvement or counselling – how to identify emotional reactions in themselves that result from working on this topic (such as traumatization due to reliving own experiences or due to hearing traumatic stories day after day) and develop skills to manage and reduce stress Selection of interviewers • Female interviewers and supervisors • Selection process very important – criteria, base selection on attitudes, motivation as well as competency • Over sample for interviewers Interviewer training • Length of training: 2-3 weeks – e.g., the WHO Multi-country study had a three week training programme • Use multiple participatory-based training techniques Take home messages • Dedicated surveys vs module • Importance of ethical considerations • Specialized needs for training interviewers