Measuring the agency of women in ICCO's STARS project : the development of an interview guide

(1)

Measuring the agency of women

in ICCO’s STARS project

The development of an interview guide

Laurien van Houten

(2)

Bachelor Thesis: International Development Studies Course code:734302042Y

Social Geography

University of Amsterdam

1012 WX Amsterdam, the Netherlands

Bachelor Thesis: Measuring the agency of women in the STARS project Supervisor: Fred Zaal

Second reader: Carolina Frossard Date: 03-08-2019

(3)

Preface

This thesis marks my final work for my bachelor Future Planet Studies with a major in Social Geography at the University of Amsterdam. In this thesis, I combine my interest in gender equality with sustainable and international development. My interest in the topic of gender equality, which I consider to fall under the umbrella term of (intersectional) feminism, has developed itself because I find it to be a fascinating topic due to its complexity and the way it is intertwined with other topics such as sustainable development. Sustainable development has been of concern to me from a young age. I am aware that I have been born in a lot of privilege. Ever since I had the opportunity to specialize in International Development Studies at university I have been looking for ways to help others who do not experience such a privilege. The

importance of development to be sustainable gives the concept extra depth and layers that need to be considered before working with it. In my opinion, these extra layers are crucial for making an impact that will last long term.

What I have enjoyed very much about writing this thesis is that I had the ability to actually present it to the STARS project and the ICCO Cooperation itself. Therefore, I felt that my work could maybe contribute to the ways that STARS navigates gender inclusivity in their program, which gave me a sense of purpose while working on it.

Writing this thesis has been quite the experience and it has taught me a lot. I have learned about clear communication, discipline, disappointment, and the fact that things are not always in my control. Actually, most of the time they are not. However, as a result of this, I am now able to be much more flexible in my work than I was before.

I want to thank Fred Zaal for being my supervisor while I was writing my thesis. He helped me a great deal with remaining positive and somewhat calm, even in times when I thought this thesis would never come to an end. I would also like to thank Marco Dekker for giving me the

opportunity to write this first part for the development of an interview guide. Also, I would like to thank Mees for motivating me in times I was struggling to continue my writing process. My colleague Gerrit, who took over my shifts at work for me to have time to study, also deserves a big thanks. And last but not least, I want to thank the respondents for my interview and Martha Dakisa, who executed and transcribed the interviews for me.

I hope you enjoy reading my thesis. Laurien van Houten

(4)

Abstract

The STARS project by the ICCO Cooperation was implemented in four countries in Sub-Saharan Africa in 2016. The project aims at facilitating access to financial and agricultural markets for smallholder farmers. The development of this interview guide works towards an interview guide that will create awareness into the ways in which women participating in STARS experience agency and empowerment. Eventually, this interview guide will give ICCO more insight into the gender aspect of their STARS project and might serve as a starting point to adjust certain interventions of the project in such a way that they are more accessible for all genders. The content and logistic aspects of the interview were tested by doing a semi-structured pilot interview with four women in the town of Holeta, located in central Ethiopia. Additionally, approaches that could benefit the development of the interview guide were elaborated upon. This was done as a means to give a structured insight into the possibilities for the continuation of the development of this interview guide.

The data that came out of the pilot interview was analyzed by doing a thematic analysis according to the six-phase plan for thematic analyses by Nowell et al. (2017). Notable results that were derived by looking at the content of the data were that in all cases most decisions regarding work and finances were made by the husband and wife together. Most decisions are made with their possible benefit to the household kept in mind. Also, it became apparent that out of custom the ownership of land and equipment is shared. Private ownership was not

deemed relevant by the participants of the pilot interviews. Results that stood out relating to the logistic side of the pilot interview were the importance of delving into the cultural context on beforehand and taking into consideration local circumstances before conducting such an interview.

In the conclusion section the main findings that came out of the literature study combined with the pilot interviews were summarized in an organized manner. The components that were considered valuable for the further development of this interview guide were named and described. In the discussion section the approaches used and additional approaches that could benefit the development of the interview guide were reviewed and recommendations for the future development were given.

(5)

1 Introduction

Strengthening African Rural Smallholders (STARS) is a five-year project initiated by ICCO

cooperation in 2016 (ICCO Cooperation, 2019). The original intention of the project is to provide access to financial markets for African farmers (ICCO Cooperation, 2019). The STARS project aims to ultimately reduce food insecurity and poverty in four countries in Sub Saharan Africa (Ethiopia, Rwanda, Senegal and Burkina Faso) and thereby strengthening value chains. “To achieve gender equality and to empower all women and girls” is the fifth Sustainable Development Goal (SDG) that was adopted by the United Nations Member states in 2015. It is stated in the SDGs 2018 report that even though a lot of progress has been made in the field of women empowerment, women still find themselves in situations where they do not have equal rights to men (Goal 5 Sustainable Development Knowledge Platform, 2018). Achieving women’s empowerment, therefore, can be seen as a crucial part of international development.

In line with SDG 5, ICCO International values women’s empowerment and gender equality as vital parts of international development. Therefore, the organization takes necessity in tracking all progress that is made within this domain in ICCO’s projects all over the world. ICCO

Cooperation in its implementation aims to include these values in all its work. In the STARS project, half of the farmers participating in the project are women for instance (ICCO

Cooperation, 2019) . In the Midterm Evaluation of the STARS project in Ethiopia in 2018, it is described that ‘it is important that women should not face barriers that men do not, simply because of their gender’ (STARS MTE Report, 2018).

In order to get an accurate insight into the ways in which women participating in the STARS project experience agency and empowerment, an interview guideline was developed. The guideline stipulates the implementation of a pilot interview. Additionally, multiple additional research options are elaborated upon and reviewed for their costs and benefits, which in the future could serve to complement the data that was obtained by doing the pilot interview. The content of the pilot interview is based upon a combination of the Pro-Women’s Empowerment in Agriculture Index (Pro-WEAI) and the Theory of Change (ToC) for Ethiopia by STARS, which were discussed in the theoretical framework. Also, different paradigms concerning microfinance and women’s empowerment are discussed after which they will be applied to STARS. The collection of data will be elaborated upon in the methodology section and consists of a

combination of literature study and four pilot interviews. The interview data was analyzed by making use of a thematic analysis in six phases according to Nowel et al. (2017). Finally, in the conclusion section, an overview of the findings was given that review this first step in the development of an interview guideline on agency and women’s empowerment in STARS.

(8)

Additionally, the recommendations and discussion section elaborates upon optional alterations to the approach that was used in this instance. Also, possible additional methods of collecting and analyzing data will be named.

(9)

2 Theoretical framework

2.1 Women’s Empowerment in International Development

Empowerment is described in the World Bank’s Sourcebook on Empowerment and Poverty Reduction as a concept that applies to all socially disadvantaged groups, of which women are a part (Rayan, 2002). Malhotra (2002) has specified three particularities that exist only for women’s empowerment. Firstly, women are a group that intersects with all socially excluded or otherwise disadvantaged groups throughout society. Women can be found within all of these marginalized crowds (Malhotra, 2002). Secondly, interfamilial relationships and household activities are at the core of women’s disempowerment. Therefore, policy proposals should be made with an awareness of their impact at the household level (Ibid). Lastly, it is argued that women’s empowerment does not only require a transition in any institution, but especially so in those that are based on patriarchal standards (Ibid). Women’s empowerment as part of

development studies is deemed necessary because first off it is a critical component of human welfare. Secondly, an increase in women’s empowerment has shown to have a positive effect on reaching other development goals related to it, such as poverty reduction and promotion of growth (Malhotra, 2002).

2.2 Women’s empowerment in Agriculture Index (WEAI)

The tool that will be developed in order to measure the agency of women participating in the STARS project is inspired by domains and indicators used in the project-level Women’s

Empowerment in Agriculture Index (Pro-WEAI). This Pro-WEAI is a tool that originates from the Women’s Empowerment in Agriculture Index that was published in 2012 (International Food Policy Cooperation, 2018). The tool is used in order to track empowerment, to measure gender equality, women’s inclusion in the agricultural sector and agency. The original WEAI inspired the development of different versions of the WEAI, of which the Pro WEAI is one (Ibid).

The Pro-WEAI is currently being developed in order to assess women’s empowerment in agricultural development projects (WEAI, 2019). The information about the Pro WEAI tool used here may differ from the final Pro-WEAI instrument since the latter is still in its testing phase. The pro-WEAI, as well as the WEAI, are based on Kabeer’s (2005) framework of empowerment. Kabeer (2005) describes empowerment as: ‘a process of change on the interrelated dimensions of resources, agency, and achievements’. She underlines the importance of assessing agency or ‘one’s

(10)

ability to make strategic life choices’. The concept of resources is relevant because resources form the means through which agency is expressed in a society. Lastly, achievements are described as the degree to which an individual is able to realize their potential (Kabeer, 2005). These three concepts were dissected into smaller, more detailed components in order for them to be suitable to form a base for the development of the WEAI and later on the different versions of the WEAI (International Food Policy Cooperation, 2018).

The Framework of Empowerment by Kabeer (2005)

Figure 1. source: Kabeer (2005)

The Pro-WEAI concept of agency is measured by twelve indicators that are divided into three types of agency: intrinsic agency, instrumental agency and collective agency. The main types of agency and the indicators that they consist of are listed down below.

• Intrinsic agency

• Autonomy in income

• Self-efficacy

• Attitudes about intimate partner violence

• Respect among household members

• Instrumental agency

(11)

• Ownership of land and other assets

• Access to and decisions on financial services

• Control over use of income

• Work balance

• Visiting important locations

• Collective agency

• Group membership

• Membership in influential groups

Since the STARS project aims at the facilitation of access to financial markets and agricultural services, not all indicators of the Pro-WEAI are considered to be relevant for the interview guide. The indicators that are deemed important all relate to decision making power and are in the field of finance and ownership. These indicators are marked below.

•

Intrinsic agency

• Autonomy in income

• Self-efficacy

• Attitudes about intimate partner violence

• Respect among household members

•

Instrumental agency

• Input in productive decisions

• Ownership of land and other assets

• Access to and decisions on financial services

• Control over use of income

• Work balance

• Visiting important locations

•

Collective agency

• Group membership

• Membership in influential groups

These indicators and the ways they were operationalized will serve as a direction for the questions asked in the interview guide. The method used to measure these indicators, however, will be different from the original Pro-WEAI tool since we aim to develop an interview guide that

(12)

will be a qualitative measuring tool, instead of a more quantitative tool such as the original Pro-WEAI survey (PRO-Pro-WEAI, 2019).

2.3 Strengthening African Rural Smallholders: outputs and impacts

The STARS Project was initiated in 2016 by ICCO Cooperation and ICCO Terrafina Microfinance together with the Mastercard Foundation. The program has three main objectives: The first one being access to finance through microfinance institutions. Secondly, STARS aims at value chain development and lastly monitoring, evaluation and learning are at the core of their program. The STARS program is located in four Sub-Saharan countries: Burkina Faso, Senegal, Rwanda, and Ethiopia. The development of this interview guide was focused on the specific context of

Ethiopia. However, the interview guide was set up in such a way that adaptation in order for the guide to be useful in another STARS country is possible. When implemented in another country, the cultural context should be taken into consideration in order to compose a question list that suits the local reality of people living in this country.

The development of this interview guide was centered around the local contexts of Ethiopia for a variety of reasons. Firstly, due to the scope of this interview guide Ethiopia was the most

straightforward country to pick since the majority of the country’s population speaks or

understands the English language. The interviews were still done in local language Oromo Afaan since it was expected for the interviews to take place in a more natural fashion when the

participants were able to speak in their mother tongue. However, for the enumerators to speak English fluently made it easier to brief them on how to interpret interview questions and execute the interview. Secondly, even though there are laws in Ethiopia that are aimed at protecting women’s rights, in reality it is still men who are in charge of decision making on the allocation and use of land (ICCO International, 2019).

The development of this interview guide was aimed at measuring the way in which women are empowered and experience agency while participating in the STARS program., The output goals for STARS in Ethiopia as they were articulated in the Theory of Change (ToC) are listed below in order to shape the questions for this guide in such a way that they accurately measure the topics STARS aims to have an impact on. It is important for the development of this guideline to note that when smallholder farmers are named in the ToC, this includes smallholder farmers that are women. The fact that there are women that are smallholders is not specifically mentioned in the ToC, but for this interview the focus was put on them. Whether or not this translates back to reality and whether women come across hindrances that men do not in the STARS program and how they experience this is what will be measured with the final interview guide.

(13)

This way of interpreting gender in the STARS program derives from the fact that the STARS program is not a gender transformative program, it is a gender inclusive program (M. Dekker, personal communication, 2019). This entails that STARS does not aim at transforming gender roles or other cultural norms relating to gender. However, STARS does take high value in the fact that everyone is able to participate in their program and that their program is therefore gender inclusive (Ibid).

The part of the outputs of the STARS project that were deemed relevant to (partially) include in the interview guide are marked yellow. These yellow-marked aspects were considered to be relevant for the fact that they relate to the ways in which women are included in the STARS program and the ways in which they experience agency and empowerment.

The short-term outputs by the STARS program in Ethiopia:

• Microfinance Institutions (MFI’s) gain knowledge on agriculture and work on the development of financial products suitable for smallholder farmers.

• Producer Organizations (PO’s) are linked with large buyers and have access to finance and agri-inputs. PO’s offer appropriate services and products for smallholder farmers.

• Smallholder farmers have access to agricultural loans and savings accounts. Smallholder farmers have access to seeds, fertilizers, chemicals and storage facilities. Smallholder farmers have access to skills training on agricultural practices and market information and equipment.

• The collaboration between actors in this value chain improves through their communication in a sector platform.

The long-term outputs by the STARS program in Ethiopia:

• MFI’s take agricultural risk assessment tools in use (A CAT) and credit risk in agri-finance is decreasing. Agri-agri-finance becomes sustainable and profitable for MFI’s.

• PO’s gain trust with their members which leads to little to no side-selling and PO’s are now able to independently build a sustainable supply of produce.

• Farmers improve their yield and quantity and quality of produce because of their access to agri-finance, inputs and markets. They now have a strong and profitable position in the value chain.

(14)

To summarize, the points relevant for the development of this interview guide were all found in the short term outputs by the STARS program in Ethiopia: MFI’s being accessible and open to all smallholders including women, PO’s being adjusted and inviting for all smallholders including women and for all smallholders including women to have access to various resources relating to agriculture such as equipment and fertilizers. The reason for still including the long term

outputs in this report is to present a more complete picture and build context around the changes aimed for by STARS in Ethiopia.

2.4 Paradigms in microfinance for women’s empowerment

This section serves to put the theory that is used for the theoretical framework of this thesis into perspective. This was done by reviewing different paradigms in the field of microfinance or microcredit programs that aim for the empowerment of women. Microcredit programs are defined by the Microcredit Summit Campaign in 2009 as ‘programs that extend small loans, and other financial services such as savings, to very poor people for self-employment projects that allow them to generate an income, allowing them to care for themselves and their families’. STARS facilitates access to suitable financial products for smallholder farmers by partnering with microfinance institutions (Icco Cooperation, 2019). Therefore, the organization falls under the definition of a program working in microfinance. Three different paradigms will be

discussed here: feminist empowerment paradigm, poverty alleviation paradigm and financial self-sustainability paradigm (Mayoux, 2006).

Feminist empowerment paradigm is built upon the concepts of equity, empowerment and equality (Drolet, 2010). In this paradigm microcredit is viewed as an entry point for the economic, social and political empowerment of women (Mayoux, 2006). Also, women’s empowerment is seen as a crucial component of a larger transformation of social power relations. This social transformation targets socio-political empowerment by addressing unequal class relations and marginalization, a topic that was already described in section 2.1 of this theoretical framework (Ibid). Microcredit programs that are based on a feminist

empowerment paradigm perspective are often not valued in terms of financial success. Rather, it is evaluated what these programs have contributed to gender development and awareness on gender issues.

Poverty alleviation paradigm uses microfinance as a means to combat poverty, especially to combat poverty among the poorest households (Mayoux, 2006). Within this paradigm the provision of loans for consumption and production supplemented by the encouragement of small savings are viewed as a valuable method to alleviate poverty. This type of microfinance

(15)

programs is mostly executed by NGO’s targeting the poverty alleviation of communities (Ibid). The motivation behind also aiming at women within this paradigm is that women are often viewed as the ones responsible for a household to do well and also because levels of female poverty exceed those of male poverty (Ibid). Drolet (2010) states that ‘It is assumed that increasing women’s access to microfinance will increase women’s decision making in the household and access to income for both production and consumption and improve women’s status in the community.’ However, within the poverty alleviation paradigm it is stated that usage of the term women’s empowerment can best be avoided, since it is deemed ‘too controversial’.

The objective of the financial self-sustainability paradigm is increasing the access of as many poor people as possible to financially self-sustainable microfinance programs. An increase of income is the goal of accessing these microfinance programs (Drolet, 2010). The access of these microfinance services targets the ‘productive poor’: poor people with the ability to contribute to a free market economy (Mayoux, 2006). Many organizations with a large global influence support these self-sustainable microfinance programs. Examples of these organizations are the World Bank, the United Nations Development Program and United States Agency for

International Development (Ibid). Women are included in these self-sustainable microfinance programs because of the expectation of women’s economic activity to contribute to economic growth. Additionally, in general the female repayment rate for loans is high (Ibid).

Neoliberalism is often viewed as an underlying paradigm for the financial self-sustainability paradigm (Drolet, 2010; Mayoux, 2006). Key values of neoliberalism are economic growth and free market capitalism, values that can be found within the financial self-sustainability paradigm as well.

After reviewing these three paradigms in microfinance for women’s empowerment, the question where to place STARS within one of these paradigms arises. However, the STARS program carries characteristics of multiple paradigms and cannot solely be fitted into one of them. When it comes to core values of the female empowerment paradigm and those of STARS it is clear that these do not align. As described in section 2.3 STARS does not identify as a gender

transformative program. Rather, the program aims at accessibility for everyone. It can therefore be stated that STARS does not operate within this paradigm. The poverty alleviation program does have commonalities with the STARS programs in terms of approach and ideology. Both aim at the reduction of poverty by supporting the implementation of microfinance programs.

Additionally, STARS fits within the poverty alleviation paradigm for being a non-governmental organization which is the most common kind of organization to work within this paradigm. Lastly, similarities can be found in the approach to including women within these microfinance

(16)

services. Both STARS and the poverty alleviation paradigm acknowledge that including women will support the poorest group in most societies and is vital for the financial well-being of

households. STARS fits within the financial self-sustainability paradigm because their program is available only for smallholder farmers. Smallholder farmers are part of the productive poor, which is also the target group for the financial self-sustainability paradigm. However, the main aim of the financial self-sustainability paradigm is to let microfinance services contribute to an increase in capital within a neoliberal economy. Even though STARS does work on the

development of value chains in a free market system and explores ‘a wealth of opportunities for agricultural growth, innovations and varied cultivation’ (Icco Cooperation, 2019), the core motivation behind the program does not align with that of the financial self-sustainability paradigm.

(17)

3 Methods

3.1 General methodology

The aim of the interview guidelines on women’s empowerment in the STARS project was to develop them using a participatory approach. The reason to put together the interview guide in such a participatory way was based on the assumption that the people knowing best what is needed or relevant for research in a local area are the local farmers themselves. By working according to this approach the firsthand experiences of locals on the ground formed the base for the development of this interview guideline. To give a voice and listen to the input of local farmers and rural smallholders is very important for STARS, since they form the starting point from which the STARS project developed itself.

In order to set up the interview guide in a participatory way, there are several approaches that can be used and combined. However, due to practical constraints of this research, the only approach that was executed here is a pilot interview, which is further described in the following section. The additional approaches that can be used in order to develop an interview guideline on women’s empowerment in the STARS project were listed down below. All of the approaches were described, mentioning the advantages and disadvantages to take into consideration when adopting such an approach. In the discussion section of this thesis, a recommendation will be given for the continuation of the development of this interview guide in the future.

3.2 Module 1: The Pilot Interview

Firstly, topics that were deemed relevant for these specific interview guidelines were selected based on the Pro-WEAI. These topics were elaborated upon and then matched to similar topics that were found in the Theory of Change (TOC) for Ethiopia by the STARS project. By combining the Pro-WEAI and the TOC of Ethiopia as sources, a comprehensive topic list was created. This topic list was then divided into interview questions in a way that was well structured, culturally appropriate and understandable for the participants.

Finally, a pilot interview was executed in the form of semi structured interview with a number of four women. For the women in the sample, a certain degree of homogeneity was assumed. Homogeneity was assumed for the following variables: religion, ethnicity, education, and relationship status.

(18)

The aim of this pilot interview was to firstly get an insight into the suitability of the content of the interview in the local context. It was thereby reviewed whether the selected topics and questions align with the realities of the women participating in the interviews. Secondly, the pilot interviews helped fine-tune the best ways to execute them process wise. The

understandability of the topics, the length of the interview, the order in which the questions will be asked, and other practical matters were tested here.

3.2.1 Semi structured interview

The pilot interview that was conducted in order to obtain data for this research was a semi-structured interview. A semi semi-structured interview consists of an interview guide with topics and questions that must be covered. However, the interviewer does have the freedom to ask the questions in their own words or words that they deem to be suitable for the context. Also, the interviewer gets to choose the order in which the questions are asked (Harrel & Bradley, 2009). Often, a semi-structured interview, therefore, gets the form of a conversation that flows

naturally even though it is steered slightly by the interviewer (Ibid). In this case that meant that the interview was build up from a set of pre-determined questions and that additionally the respondents did have room to give their interpretations and opinions on the topics of the interview (Bryman, 2015, p 470). The motivation behind using a semi structured interview for this research was that a semi structured interview is very suitable to discover underlying thought patterns and explore idea’s that would not reach the surface when making use of a structured interview (ibid). A semi-structured interview gives the researcher the ability to get to the core of a subject and to entirely understand the answers that were given by the respondents (Harrel & Bradley, 2009).

3.2.2 Literature study

In order to understand the general context tied to the location in which the pilot interview took place, a literature study was executed. This literature study focused on the following topics:

• Location and geography

• Culture, language, and religion

• Gender

(19)

The aim of the literature study was not to delve deep into the topic and understand every underlying narrative of the culture in Ethiopia. The literature study more so serves as a way to create a more complete picture of the participants and their living environment for the pilot interview. Also, this literature study can serve as a starting point for further literature research in case there might be topics mentioned in the interview that aren’t suitable for the local context.

3.3 Additional approaches

3.3.1 Module 2: Interview with a control group of women not participating in STARS or

similar projects

In order to get an insight into the ways in which the STARS project contributes to the agency and empowerment of women, a control group of women that are not influenced by STARS or a program similar to STARS can be used. The women will be selected on homogeneity in other variables such as education, religion, ethnicity, and relationship status in order for the effect of participation in the STARS project to be as clear as possible. The groups will be asked the same interview questions, since STARS is not named in the pilot interview there is no need for

alterations in the interviews for the control group. Finally, the answers coming from both groups will be compared.

The benefit of adopting such an approach is that this approach gives the ability to pinpoint the aspects of agency and women empowerment that are measured to a different degree with the group of women that is a member of STARS compared to the group of women that are not. This indicates the impact that STARS has on the empowerment and agency of women.

A possible limitation for this approach is that in certain areas it might be challenging to find women that are not participating in any program, because there are many organizations focusing on smallholders and value chain incorporation active in multiple regions in Sub Saharan Africa (M. Dekker, personal conversation, 2019). Also, there always is the possibility of additional variables being at play, which will be further discussed in section 3.4.3. An additional limitation of this approach is that this particular research centers mainly on the experience of the women that are participating in STARS, their experience is what determines the degree to which they feel empowered and feel that they have agency. Therefore, a control group of other women will not be that relevant of a factor in this research if the angle remains to be focused on the point of view of the women.

(20)

3.3.2 Module 3: Interview with a control group of men

In order to get a better understanding of the role that gender plays in the empowerment and agency of women, a control group of men could be used. Besides the difference in gender, the sample will be selected on a similarity in the variables religion, ethnicity, education, relationship status and membership of STARS or a similar program in order to get the best possible insight into the influence of gender on empowerment and agency. The men will be asked the same questions as the women.

This approach can be advantageous because it gives the ability to compare the degrees of

empowerment and agency in both groups. Also, this approach gives an insight into differences in perception of empowerment and agency between the groups, where one of the parties might experience a great amount of agency power the other party might have an experience that is entirely different. By interviewing both men and women in a household a more complete image can be created of how topics such as decision making are allocated between the two.

A disadvantage of this approach is that this research focuses on the way women experience their agency and empowerment. This is what was guiding for rating their degree of both the concepts, not the ways in which others might experience them. Therefore, this addition might not be considered to be of the same importance for a research with the same angle. Additionally, this approach has the possibility of additional variables influencing the research variables. This is further discussed in section 3.4.3.

3.3.3 Module 4: Focus Group Discussion

A focus group discussion could be held in order to get an accurate view of the topics selected for module 1 in a local context and the ways in which they are relevant to the women participating. The topics that could be addressed in this focus group discussion are based on the three pillars of agency as they are described in the Pro WEAI: intrinsic agency, instrumental agency and group agency. For every pillar the women will be told a small story, a vignette, to open up the conversation to then have enough room for the women to share their own experiences, such as similarities or differences from the women that they were told about, within these topics. At last,

(21)

the women will also be asked whether there is anything that they would like to add to this conversation and research.

The women that participated in this focus group discussion will be selected by project staff based on their expected knowledge or opinions on the relevant topics, since the aim is to have a back and forth conversation that gives the women a legitimate say in the topics to be included in the guideline.

The advantage of doing a focus group discussion is that the participants have a conversation and therefore have the ability to argue and keep each other accountable for their statements

(Bryman, 2016, p. 203). Additionally, the researcher gets the opportunity to see the ways in which the group creates collective meaning and sometimes comes to a social consensus of certain phenomena. The result of this is that the information gathered by doing a focus group discussion is often closer to reality than the information gathered in an interview (ibid). Lastly, since a focus group discussion is not under as much control of the moderator the participants have the ability to discuss topics that they deem to be relevant in relation to the research topic (ibid).

Limitations for focus group discussions are that transcribing them takes a lot of time since it needs to be taken into account who is speaking when and simply because focus group

discussions often have a longer duration (Bryman, 2016, p. 518) Also, the moderator needs to make sure that multiple participants are not speaking at the same time and that speaking time per person is distributed in a way that is as equal as possible. The formerly named problems that could occur are ones that are near impossible to totally avoid in a focus group discussion. Lastly, there might be a group dynamic in which participants do not feel comfortable to express the way they feel about certain topics. In this case, an individual interview is a more suitable option (Bryman, 2016, p. 516).

3.4 Measuring the influence of STARS: limitations

The STARS project commenced in 2016 and although research has been done in order to determine baseline data on gender and the wellbeing and agency of women, this project wants to research the ways in which women participating in the STARS project experience the

influence of STARS on their empowerment more systematically. There are a few difficulties that might occur here which will be discussed in this section.

(22)

3.4.1 Generating comprehension

Firstly, it was assumed that the comprehension of the women of what the STARS project entails is low. The way to work with this was to not mention the project by its name as such. Instead, descriptions of typical activities or interventions implemented by the STARS project were given in order to then ask respondents about whether they were able to participate in this and if and how it influenced their empowerment. The assumption that it was the STARS project that brought about these possible changes in the way the women experience empowerment was made.

3.4.2 Anchoring vignettes

In this research, anchoring vignettes were used when doing the pilot interview as a way to increase the understanding of the questions of the respondents. It is also assumed that women could be reluctant to disclose intimate information, in such situations vignettes can help because with a vignette they do not have to share their situation. They can simply respond to someone else’s story and give their opinions on it. It is up to them whether they also reveal anything about how they would navigate such a situation personally. Donald et al. (2016) state that: ‘Anchoring vignettes are short descriptions of hypothetical individuals or situations meant to convey

complicated concepts and ensure that different respondents understand questions similarly across cultures and contexts. They have been generally shown to increase ease of response’. In practice, this came down to the person executing fieldwork reading hypothetical situations, relevant to one of the indicators that were used in the pilot interview to the respondent. After the vignette, a few questions regarding the same indicator were asked.

3.4.3 Multivariate relationships

Thirdly, it might not be clear whether it was the STARS project causing a change in

empowerment and agency or another factor. Since this is a qualitative research the main aim is not to prove certain relationships and causalities between variables, the focus is more so on the exploration of the content and depth of the data that was gathered. Also, since this interview guideline is still in its testing phase the possibility of additional variables being at play was not the main focal point when executing the pilot interview. However, in order to be aware of these possible additional variables and the different ways in which they occur the three multivariate relationships that occur most often will be described shortly here. This is done in order to get an insight in the ways to work with them for the continuation of the development of this guideline in the future.

(23)

Firstly, there might be a spurious relationship between variables. According to Bryman (2016), this occurs when: ‘there appears to be a relationship between two variables, but the relationship is not real: it is being produced because each variable is itself related to a third variable.’

Secondly, an intervening variable might be at play. An intervening variable occurs as an extra step between the two variables that are being analyzed (Bryman, 2016, p. 345). This means that the relationship between those two initial variables is not a direct one (Ibid). An example of this could be that in this research on of the focuses is on the relationship between gender and ownership of land. An intervening variable could be finishing high school here. In some cultures, access to education is influenced by gender and whether an individual finished high school might have an influence on their ownership of land. Therefore, an individual’s gender has an influence on whether they finish high school or not which in turn influences the extent to which they own land.

Lastly, the relationship between the initial variables might be moderated by a third variable (Bryman, 2016, p. 345). An example could be the relationship between finishing high school and income, which is moderated by gender. Thus, whether someone earns a different income

depending on whether or not they finished high school is moderated by their gender (Ibid).

3.4.4 Response bias

Lastly, there was the difficulty of response bias because it was a given that the pilot interviews were set up by STARS staff. Therefore, respondents might answer only with a positive attitude regarding the program. As this is initially a test phase in the methodology this was not a major issue

3.5 Qualitative data analysis

The usage of qualitative research has gained increasing recognition and popularity in research literature (Attride-Stirling, 2001). However, the analysis of qualitative data as part of qualitative research can be seen as the most complex part of the entire research process (Bryman, 2016; Thorne, 2001). According to Bryman (2016), researchers can easily lose their focus in

qualitative data because of the large amounts that are often obtained and the richness of the data. This generates difficulty in obtaining meaningful and useful results coming out of the data (Bryman, 2016; Nowell et al., 2017).

(24)

When doing a qualitative data analysis, it is important for the researcher to be transparent about the analyzing methods used (Malterud, 2001 as cited in Nowell et al., 2017). This can be done by using a systematic approach. The readers of the research should be able to understand the ways in which the researchers analyzed their data and what assumptions were underlying that analysis (ibid).

In this research, a thematic analysis was used. A thematic analysis is a systematic approach that is useful when trying to summarize the core components of a dataset in an organized manner (Nowell et al., 2017). Braun and Clarke (2019) have a similar definition and state: ‘Thematic analysis is a method for systematically identifying, organizing and offering insight into patterns of meaning (themes) across a data set’. In order to do a thematic analysis, it is required of the researcher to work in a structured way, which will eventually lead to a clear and well formulated end report (Ibid). Nowell et al. (2017) describe six phases to run through when doing a thematic analysis before coming to a final product. These six phases will be described and elaborated upon below.

3.5.1 Phase 1: Familiarizing Yourself with Your Data

Qualitative data often consists of a great volume of the source material, a variety of formats and a lot of complexity without any uniform structure (Nowell et al., 2017). However, in order to create a comprehensive analysis, all of the data should be taken into account. In order to do so, the researcher should immerse themselves with the data, reading it through over and over again while looking for patterns and meanings (Ibid).

3.5.2 Phase 2: Generating Initial Codes

In this phase, codes or labels are attached to parts of the text in order to apply structure and recognize the different components that makeup issues or themes within the text. A code identifies and produces a label and tags it to a part of the data that is potentially relevant to the research question (Braun & Clarke, 2019).

The codes applied in this data analysis were based upon the theoretical framework and the way it translates into the pilot interview. This is the case because the interview was set up in such a way that everything was understandable and clear for the participants by not using too much

(25)

jargon relating to gender studies or international development. The codes that were applied here relate to the content of the interview as well as logistical issues concerning the interview. The ones that are attached to the content of the interview relate to ways in which the women experience agency and empowerment in their work as smallholder farmers. The codes applied to practical issues relate to the relevance to the local context of the questions, the ways in which questions are formulated, whether the questions are understood well and the duration of the interviews. The coding of the pilot interviews was done manually.

3.5.3 Phase 3: Searching for themes

In the third phase, different codes will be combined and attached to an overarching theme. A theme consists of different parts that do not have a lot of meaning on their own (Nowell et al., 2017). A deductive approach will be used in this thematic analysis. That means that in case a combination of codes forms a theme, this is mostly dependent on whether or not they relate to the overall theoretical framework and research question (Braun & Clarke, 2006; Nowell et al., 2017).

3.5.4 Phase 4: Reviewing themes

In this phase, it is reviewed whether the individual themes are valid and whether the codes that make them up to form coherent patterns. Any defects in the initial coding and themes will be recognized to then be readjusted to securely display the meanings derived from the data (Nowell et al, 2017). This could entail that some themes do not have enough data to support them, while others need to be broken down further in order for them to be sufficiently specific. The themes are changed until they form a manageable set that accurately summarizes the significant parts of the text (ibid). Braun and Clarke (2006) state that: ‘It is to be expected to have to recode the data since coding is an ongoing organic process’.

3.5.5 Phase 5: Defining and naming themes

In the fifth phase, it is determined which parts of the data capture the topics selected for the initial research (Braun & Clarke, 2006; Bryman, 2016). At this point, it is considered in what way

(26)

each theme fits into the overarching narrative that glues together the entire set of data, in relation to the research questions (Braun & Clarke, 2006). Some overlap might occur within the themes because parts of the data might be used for multiple themes. However, since themes are often interrelated this is an inevitable commonality (Nowell et al., 2017). The selected themes will finally by elaborated upon in a comprehensive analysis that describes their core topic (Nowell et al., 2017). Moreover, a title will be given to each theme to accurately describe to the reader what it consists of.

3.5.6 Phase 6: Producing the report

The last phase of the thematic analysis consists of writing up the final analysis once the themes are fully established (Nowell et al., 2017). At this stage, the data is written up in a logical, concise and interesting manner in order to bring across the themes that were found within the data. Additionally, the methodological processes and logical thinking paths that were used to come to these findings will be outlined clearly (ibid). Direct quotes are used in order to generate a better understanding of certain parts of interpretation and to demonstrate where the themes were found in the original data. In the final analysis, the themes will be constantly referred to in order to build a valid argument. Additionally, there will be references to literature on which the theoretical framework was built, that complement and confirm the described findings (Ibid).

(27)

4 Data analysis

4.1 Local context

The findings described in section 4.1 on local context are all based on findings coming from literature studies.

4.1.1 Location and Geography

The pilot interviews were conducted in a town named Holeta Genet, which is approximately 40 kilometers away from the main town Addis Ababa. Figure 2 gives a visual representation of Holeta and the way it is located in as compared to Addis Ababa. Figure 3 gives a visual representation of Ethiopia as a whole, the location of Holeta can be found at the red marking point.

(28)

Figure 3. Map of Ethiopia source: (google.com/maps)

4.1.2 Culture, language, and religion

The interviews were executed in the language Afaan Oromo. Afaan Oromo is not only spoken in Ethiopia, but a number of approximately 50 million people also speak the language in Egypt, Ethiopia, Somalia, and Kenya. Oromo is, therefore, Africa’s third most spoken language (Ibid). Afaan Oromo is a language that is spoken by an ethnic group that goes by the name ‘Oromo’. The Oromo make up the most populous ethnic group or nationality within Ethiopia (Erena, 2019). The area that is predominantly inhabited by the Oromo bears the name Oromia.

Three main religions that are practiced by the Oromo are Islam, Orthodox, Protestant and Catholic Christianity, and Waaqeffanna. Waaqeffanna is the indigenous religion of the Oromo and only six percent of Oromo still lives according to this religion. However, Waaqefanna does form the base for many rituals that are still performed by most Oromo people (Hussein, 2014). Waaqeffanna is a monotheistic religion where they believe in the Supreme Being of Waaqaa as the creator and source of all lives (Aga, 2016).

The Oromo originally organize themselves in groups based on their age range. Each group has different socio-political functions and responsibilities for them to take care of (Hussein, 2004). The age groups, called Gada, are organized according to a strict and elaborate procedure. Also, the promotion to the next Gada is something that is celebrated extensively (Ibid). There are differences in rights between youth, adulthood, and seniority. Additionally, women have

(29)

different privileges within polygamous marriages, where the first wife has the most rights and the ones coming after her do not get to have as many. The least amount of rights goes to the mistress; she is not married to the man in question (Ibid).

It must be stated that this section is a very broad overview of culture language and religion of the Oromo. Therefore, there will inevitably be some inaccurate, incomplete or flawed

information due to the generalization of the used literature. However, the purpose of this section also was to give a broad overview of general trends that can be found on the topics described in order to provide context for the conducted pilot interview. The pilot interview then will dive deeper into the specificities or the subjects.

4.1.3 Gender

As in many other African countries, there is a clear and patriarchal distinction between behaviors that are deemed suitable for men and those that are deemed suitable for women in Oromia. These behaviors encompass a generalized outlook on the assumption of roles, access to property and participation in social activities (Hussein, 2014). There also is a clear division of labor, certain tasks are assigned to women only and others to men. Additionally, because women are actively withheld from the age sets that were described in the previous section, they are profoundly dependent on men for ‘most political-ritual services and for all activities connected with the defense of living camps, wells, herds, and shrines.’ (Hussein, 2014).

In terms of religion, women do have a key position in a ritual called Atete. Atete is originally a ritual to invoke rainfall in times of drought, but it is also put into action in case of other natural disasters. Men would not be allowed or able to carry out this ritual. The Oromo believe that women have a close link to nature and fulfill the role of being an intermediate figure between Waaqaa and the physical world. Whenever there is an emergency or a certain hazard, natural disaster related or not, the men have to ask the women to put their powers into action in order to settle down the situation. In this sense, men are dependent on the women.

In this section, it is important to note that this is a generalized summary of literature on the topic of gender and womanhood in the Oromo culture. Therefore, some parts might be incomplete or inaccurate when they are compared to the actual situation. However, this section serves as background information for the actual pilot interview, which will give a more realistic insight into the gender roles and women’s empowerment of Oromo women.

(30)

4.1.4 Politics

The country of Ethiopia is currently under the governance of a female president called Sahle Work Zewde (BBC, 2019). She is Ethiopia’s first female president and has spoken out about gender issues multiple times. The prime minister of Ethiopia is Abiy Ahmed. Abiy Ahmed is the first Ethiopian prime minister whose ethnicity is Oromo (BBC, 2019). Only in the first months of being appointed as prime minister, Ahmed has achieved a declaration of peace between Ethiopia and neighboring state Eritrea. The conflict between Ethiopia and Eritrea had been causing unrest since 1993 (BBC, 2019). Both leaders have been elected in 2018.

Abiy Ahmed is part of the Ethiopian People’s Revolutionary Democratic Front (EPRDF). The EPRDF came to existence in 1991 and since then operated in an authoritarian manner (BBC, 2019). However, the current prime minister has promised more transparency from the state. Currently, violence by security forces, corruption, terrorism, and media control are very common in Ethiopia (BBC, 2019; Freedomhouse.org, 2019). Ahmed is planning on improving this situation by editing the current repressive law in such a way that fair elections and media freedom are included (Ibid).

(31)

4.2 Content of the pilot interviews

4.2.1 Phase 1

The pilot interviews were analyzed according to the six phases of data analysis proposed by Nowell et al. (2017). In line with these phases, the author firstly familiarized themselves with the data by reading through the interviews extensively. This was done to make sure that the author was aware of what the exact results of the data gathering entailed and not even the smallest detail or peculiarity in the data would go unnoticed.

4.2.2 Phase 2

The second step of the data analysis by Nowell et al. (2017) is to generate initial codes. The initial codes that were generated are altered throughout the process of analyzing the data. Therefore, it is not deemed useful to write down the initial codes here. However, the codes as they were attached to the themes in the third step and then reviewed in the fourth step of the thematic analysis were written down in table (x).

4.2.3 Phase 3

In the third step of the thematic analysis by Nowell et al. (2017) themes were generated that were deemed relevant to the theory that formed the fundament for the development of this interview guide. Therefore, the generation of themes here is done through a deductive approach (Bryman, 2016). The themes selected are displayed in table (x), accompanied by a citation coming from the pilot interviews. The first names of the four women that participated with the pilot interviews are Tewabech, Fikirte, Bizuyehu and Asefelech. The citations are attached to the name of the woman they came from. The purpose of displaying the themes and examples of text that are part of these themes in such a way is to demonstrate in what way they were

constructed. Additionally, the citations are written down in the own words and ways of

expression of the women. To include the ideas of the women participating in the pilot interviews in their own words is considered to be a valuable part of this thesis. It is important to note that for purposes of structure and the scope of this thesis, the amount of examples that were used were kept to a limited amount. For all of the themes that can be found in table (x) there were more examples in the text that were in line with them, but they are not in the table for the aforementioned reason. The themes are displayed in the order in which they were found in the pilot interviews. In the fifth step of the thematic analysis they will be ordered and connected to the theory they were based on.

(32)

Table 1

Themes

Citations

Holeta Research Center

Tewabech: ‘I get training from Holeta

Research center […] Actually, this is the

core of my ideas being accepted by my

husband’

Shared decision making

Tewabech: ‘I prefer to do decision making

together with my husband. What, where

and when to sell produce should come

from the both of us.’

Fikirte: ‘He can bring ideas and I can too.

Then we discuss what is best for us and

what is demanded in the market. We look

into last season price situation of a given

cash crop, then discuss on how the price

of the crop will look like next season.

There is no one-way message to be

effected in our household.’

Men should not be trusted to do decision making on what to spend money on

Bizuyehu: ‘If he is the one to sell, he might

spend some good portion of money

without my knowledge and in

consequence, I might be strained or

constrained to manage the family on food

consumption. If men get that kind of

chance, they might drink and spend

money extravagantly.’

Family land

Tewabech: ‘No I don't ́t have land of my

own. Even though I got land from my

family, I don´t see it as my own land but

(33)

as the family land. My husband ploughs

on the land and in the future that land will

belong to our children. That is how I see

the faith of this land’.

Women have ownership of livestock

Tewabech: 'I got cows at home. The

income generated by selling milk and

butter, will belong to me.’

Men sell livestock on the market

Fikirte:‘[…] but he is the one who goes to

the market and sell the cow. He brings the

money and give to me.’

Women sell crops on the market

Fikirte:‘[…] in fact in my household it is I

who go to the market to sell any crop

produced except animals in the local

market. He doesn´t know how to

negotiate and sell crops, sometimes if we

go together, he sells at lower prices than

the market offers. For he thinks that these

people are poor who are feeding their

children. So we decided for any crop, I

should sell. I have this strategy, before

setting any price on the produce, I go

around in the market and ask how many

people are selling the same produce, then

I set my own price.’

Large expenditures are decided on by the man and woman together in a household. Smaller household expenditures can be decided on individually.

Tewabech: ‘If it is food consumption, yes,

it is I who decide what to spend the

income on. If it is something that needs a

lot of money, I will consult my husband

and we make the decision together.’

Fikirte: 'If it is for food consumption and

other small expenses, I do by myself. If it

is a large amount of money, for sure I will

(34)

discuss with him. Then we decide

together whether we should spend or

not.’

The group membership of a woman is discussed with her husband before deciding to join a group.

Bizuyehu: ‘After the committee of the

cooperative approached me, I asked him

whether that is a good idea or not. He told

me that it is a good idea and advised me

this kind of gathering helps me to

improve my social relation with the

society.’

Group membership provides women with shared knowledge and gives them a sense of community

Tewabech: ‘Being a member to a group

helps women to share their ideas and also

gain additional knowledge. We even

shape with one another and provide

advice on how to communicate and

convince our husbands.’

Getting a loan is a decision that is made together in a household

Fikirte: ‘If I hear about a good loan (low

interest rate) and need to take, I first

raise and discuss the issue with him. I

can´t take the decision without discussing

with him. He will do the same, if he wants

to take a loan.’

In general, women are not part of governmental groups

Fikirte: ‘It is usually men who will be

member to governmental groups. I don´t

have time to go there and be part of that

as I have a lot of activities at home.’

Women would be able to get an individual

loan if they wanted to.

Tewabech: ‘I can decide if I want. But

most of the time I discuss with him and

try to convince him. I tried once to take a

loan and raised this issue to him and he

was not in agreement with it. I tried my

best but he

(35)

refused to agree. Then I decided to take

the loan. He then kept quiet.’

Individual property is not relevant

Tewabech: ‘There is no segregated

property that belongs to one person only.’

4.2.4 Phase 4

The fourth step of the thematic analysis as it was proposed by Nowell et al. (2017) is reviewing the selected themes. This is where some of the themes that were previously selected are deleted or adapted in a way that they fit with the theory that is useful for this thesis. This is a step that the author revisited until finishing the analysis, since every insight into the results can give a more comprehensive look on the themes. This gives the author the ability to make an even more effective selection with every time they revisit the previous selected themes. The final themes are the ones that are displayed in table 1.

4.2.5 Phase 5

In the fifth step of the thematic analysis by Nowell et al. (2017), the themes are defined and named. In this thematic analysis this is the section in which the chosen themes are discussed and connected to the previously used theory. It will become clear in this step that some themes relate to more than one component of the used theory and might therefore be overlapping. The theory that was used before is divided in the three types of agency as they were described by Kabeer (2004): intrinsic agency, instrumental agency and collective agency.

Intrinsic agency:

Autonomy in income

(shared decision making; women sell crops on the market; men sell livestock on the market; men should not be trusted to do decision making on what to spend money on; women have ownership over livestock)

Firstly, intrinsic agency, which in the pilot interviews was measured by asking for the autonomy in income of the women. The first theme relevant to intrinsic agency is that of ‘shared decision making’. It became clear out of all of the interviews that it is deemed to be highly important for the women to make decisions together with their spouse. Therefore, there is some autonomy in income present in the sense that the women do have a say in decision making regarding income. However, they do not get to decide on their own what to do.

(36)

The second theme that is relevant to autonomy in income here is ‘women sell crops on the market’. Within all the interviews it became clear that selling crops on the market is the domain of the woman in the household. She is the expert on the quantities and prices of crops to be sold. Most women reported that the money that was made by selling the crops usually goes to a shared bank account or is used for household expenditures. Household expenditures are usually her responsibility, which gives the women some autonomy in income.

‘If it is food consumption, yes, it is I who decide what to spend the income on. If it is

something that needs a lot of money, I will consult my husband and we make the decision

together.’ Tewabech

The next theme connected to autonomy in income is ‘Men sell livestock on the market’. The reason why this is the case did not become clear on the basis of the interviews. It seemed as if this was something that had been in this way for years and therefore it is now part of the culture. However, women do receive the money that the man earned by selling livestock on the market. Therefore, this is a mixed case of autonomy in income. Women do not get to go to the livestock market themselves, but they do get the money that was earned by their men by going there.

The theme ‘Men should not be trusted to do decision making on what to spend money on’ is one that has a large impact on the autonomy in income of women. There seems to be a distrust that the women have regarding their husband’s ability to spend money in a way that benefits the family. The women participating in the pilot interviews had all enforced that they were the ones in charge of at least the household money.

The last theme that is connected to autonomy in income here is that ‘Women have ownership over livestock’. The reason for connecting this to autonomy in income is that the women also report that since they have ownership over the livestock, the income generated by selling their products also belongs to them.

Instrumental agency

Input in productive decisions:

(Holeta Research Center, shared decision making)

Instrumental agency was measured making use of the following indicators: Input in productive decisions ownership of land and other assets, access to and decisions on financial services and

(37)

lastly control over use of income. When it comes to input in productive decision this can be connected to the themes of Holeta Research Center and shared decision making. The first one because the women stated how the Holeta Research center equipped them with knowledge. This made it more likely for their husbands to take their input in productive decisions more seriously. The latter, shared decision making, gave them the opportunity to have a say in every decision made. This includes productive decisions.

Have you noticed a change in the way you were allowed to make decisions related to

production in the last years? (what are the reasons for that) what kind of changes came to

bring this shift of decision making?

‘Yes, particularly, after I joined this Burkeme women primary cooperative and getting

various agriculture related training which is provided by Holeta Research Center. After

that, I support him not only with physical work but also with ideas too wherein he starts

embracing. From that time on, I am the one who has the say in the farming activities.’

Bizuyehu

Ownership of land and other assets

: (family land, individual property is not relevant)

Ownership of land and other assets is connected to the themes ‘family land’ and ‘individual property is not relevant’. It became clear out of the pilot interviews that all the women participating own land they have inherited from their families together with their husband. Owning any property by themselves is something they do not see the use of since their land serves mainly to feed their family, directly as well as indirectly, anyway.

Acces to and decisions on financial services:

(Getting a loan is something that is decided on together in a household, women would be able to get an individual loan if they wanted to).

Access to and decisions on financial services is an indicator that is linked to the themes ‘Getting a loan is something that is decided on together in a household’ and ‘Women would be able to get an individual loan if they wanted to.’ All of the women reported that they would always have a

conversation with their husband before taking a loan. However, they also all reported that in case they wanted to do so, they would be able to get a loan by themselves.

Measuring the agency of women in ICCO's STARS project : the development of an interview guide