UN–Water and its Role in Global Water Governance Appendix 1 : Methodology Interviews

Abstract

Creswell noted that in qualitative research, " the intent is not to generalize to a population, but to develop an in-depth exploration of a central phenomenon " , which is best achieved by using purposeful sampling strategies (2005:203). A random sampling strategy would be inappropriate for the exploration of the central phenomenon of this study because the purpose here is not to generate a representative sample and then generalize the results to other coordination mechanisms or other contexts, but rather to learn from people who are 'information rich' and can best help to understand the specific interest of this research, UN-Water and its role in GWG. For the expert interviews, a sampling strategy has been chosen that combines elements of the maximal variation and snowball sampling procedures. The snowball strategy is a form of purposeful sampling in qualitative research that " typically proceeds after a study begins and occurs when the researcher asks participants to recommend other individuals to study " (Creswell 2005:206). The researchers' initial unfamiliarity with the topic and the complexity of the central phenomenon at hand were the decisive factors behind the choice for the snowball approach. The sampling process was initiated by an independent expert who was not interviewed, but exclusively served as 'trigger' to get the different snowballs rolling (c.f. Figure A1). However, once this process was underway, it needed to be steered in certain directions in order to give voice to experts from different backgrounds who might have different perspectives on the central phenomenon. The maximal variation approach allows for obtaining this diversity and thereby accounting for the complexity of the problem at hand. It is " a purposeful sampling strategy in which the researcher samples cases or individuals that differ on some characteristic or trait " (Creswell 2005:204). The characteristic of interest here is the expert's perspective(s) on UN-Water, which can basically fall in one or several of the four categories listed below 1 : • Members: experts of UN-Water member organizations. • Partners: experts of UN-Water partner organizations. • Affiliates: experts working for UN-Water or one of the affiliated programs. • Observers: GWG experts with no direct organizational link to UN-Water. After consideration of the limited time availability for the research process of this MSc thesis and the relative weight of the expert interviews in relation to the literature and document review, a sample size of a total of ten interviews, each between 30 and 45 minutes, has been deemed appropriate. 1 This categorization is of course not mutually exclusive but it is exhaustive for the experts under consideration (people who are unfamiliar with UN-Water are not considered as experts here). A2 Data Collection and Sample Size Several data collection methods exist in qualitative research and interviews are among the best suited and most commonly used instruments (Kumar 2005; Nohl 2009). Kumar noted that " [o]n one hand, interviewing can be very flexible, when the interviewer has the freedom to formulate questions as they come to mind around the issue being investigated; on the other hand, it can be inflexible, when the investigator has to keep strictly to the questions decided beforehand " (2005:123). A number of approaches can be distinguished on the spectrum between the two extremes of improvisation and determination but the one thing they all have in common is the fact that they do not give any specifications or limit the participants' freedom in answering the various interview questions. For the exploration of the central phenomenon of this research, a semi-structured anonymous interview design with open-ended questions was deemed most appropriate. This choice was based on the following considerations: • The semi-structured design gives the participants ample time and scope to express their diverse views and allows the researcher to react to and follow up on emerging ideas and unfolding events (Nohl 2009). • Results obtained through semi-structured interviews can be compared among each other since all participants are required to express their views about the same general themes (Nohl 2009). • Semi-structured interviews allow not only for assessing the participants' opinions, statements and convictions, they also allow to elicit narratives about their personal experiences (Nohl 2009). • Open-ended questions allow the participants to freely voice their experiences and minimize the influence of the researcher's attitudes and previous findings (Creswell 2005). • Anonymity was guaranteed in order to give the participants the opportunity to freely express their views and encourage them to also address politically delicate issues. A list of guiding questions was compiled and used to guide the expert interviews in order to make sure that all respondents address in the interview process the issues that are of interest for this study. However, this list was not used for standardizing the data collection procedure, it merely provided a frame for the discussions and was intended to trigger and guide the experts' narratives. Contact to experts was initiated with a personalized email request for a recorded, anonymous phone interview with a short description of the research purpose and central phenomenon attached. Interviews were then conducted individually over the phone 2