QDR Qualitative Data Repository
Not a member yet
110 research outputs found
Sort by
Data for: Exploring sources of insecurity for Ethiopian Oromo and Somali women who have given birth in Kakuma Refugee Camp: A qualitative study
No full textData consists of interview excerpts utilized for the analysis of a paper submitted for publication entitled "Exploring sources of insecurity for Ethiopian Oromo and Somali women who have given birth in Kakuma Refugee Camp: A Qualitative Study."
Project Summary
Background: According to the United Nations High Commissioner for Refugees, 44,000 people are forced to flee their homes everyday due to conflict or persecution. Although refugee camps are designed to provide a safe temporary location for displaced persons, increasing evidence demonstrates that the camps themselves have become stressful and dangerous long-term places—especially for women. However, there is limited literature focused on the refugee women’s perspective on their insecurity. This qualitative study sought to better understand the ways in which women experienced insecurity at one refugee camp in Kenya.
Methods and Findings: Between May 2017 and June 2017, ethnographic semi-structured interviews accompanied by observations were conducted with a snowball sampling of 20 Somali (n=10) and Oromo Ethiopian (n=10) women, 18 years and older, who have had at least one pregnancy while living in Kakuma Refugee Camp. The interviews were orally translated, transcribed, entered into Dedoose software for coding, and analyzed utilizing an ethnographic approach. Four sources of insecurity became evident: Tension between refugees and the host community, intra/intercultural conflicts between the refugee community, direct abuse and/or neglect by camp staff and security, and unsafe situations in accessing healthcare both in transportation and in mistreatment in facilities. Potential limitations include nonrandom sampling, focus on a specific population, inability to record interviews and possible subtle errors in translation.
Conclusion:In this study, we observed women felt insecure in almost every area of the camp, with no place in the camp where the women felt safe. As it is well documented that insecure and stressful settings may have deleterious effects on health, understanding the sources of insecurity that are faced by women in refugee camps can help to guide services for health care in displaced settings. By creating a safer environment for these women in private, in public, and in the process of accessing care in refugee camps, we can improve health for them and their babies.
Data Generation
Sampling:The population of the study was limited to Somali and Oromo women over the age of 18 with no upper age limit who had given birth at least once in Kakuma Refugee Camp. Due to both the infeasibility of collecting a random sample within the camp and the sensitivity of the topic, participants of interviews were selected through a snowball sampling approach through contacts used in previous research, including a hired mobilizer with previous experience in similar research. The mobilizer was responsible for recruiting women fitting the criteria outlined above who were willing to talk about their pregnancy experiences and stressors.
Interviews:The interviews lasted between 30-60 minutes each. They were conducted in a place of each participant’s choosing, typically their own homes, with the assistance of a translator. The interviews were all carried out by the same female researcher (AL) and the same female translator. The researcher was a trained interviewer with previous interview experience in rural settings and was enrolled in the Master of Science in Global Health Program at the University of Notre Dame during the time of study. The translator was a refugee of Oromo Ethiopian descent living in Kakuma Refugee Camp herself. Twenty interviews were conducted in total. After the initial interviews, twelve of the twenty interviews were identified that required further clarification. These twelve interviews were repeated with the corresponding interviewee to cross-verify that the relevant meaning had been captured and to expand details within the respondents’ interviews. Children were usually present during the interviews, and at times, men were also present during the interviews. The researcher asked the questions in English and the translator translated the question as close to verbatim as possible for the participant. Due to IRB constraints and in order to maintain rapport with the participants, interviews were not recorded; however, detailed and verbatim notes from the translator were taken during the interview and typed up within twenty-four hours.
Data Analysis
Analysis: One researcher (AL) developed a codebook organically through reading over interviews and notes. Typed and de-identified interview notes were uploaded to Dedoose a qualitative analysis program, and given codes and sub-codes from the aforementioned codebook. Codes included insecurity, health, pregnancy experiences, healthcare facilities, stressors, income, coping, and support. The subcodes of insecurity include general insecurity, host community, refugee community, healthcare facilities, issues regarding security, and suggestions for improvement.</p
Data for: From legal doctrine to social transformation? Comparing U.S. voting rights, equal employment opportunity, and fair housing legislation
No full textThis is an Annotation for Transparent Inquiry (ATI) data project. The annotated article can be viewed on the publisher's website.
You will need to use the Chrome browser with the Hypothesis extension installed to view the ATI annotations
Data Generation & Analysis
This article is a historical-comparative study of three major civil rights statutes (voting rights, equal employment, and fair housing) and is motivated by one fundamental question: why are some civil rights laws more successful than others? Consensus exists among civil rights scholars that voting rights was far and away the most successful of the three; that fair housing was a general failure; and that equal employment opportunity was moderately successful. Our study seeks to explain this specific hierarchy of civil rights outcomes across these three cases. As is the case with all historical-comparative research, qualitative data and qualitative analysis are essential because each case is constructed and compared holistically with respect to historical events and outcomes, the actors involved and the choices they made (and why they made them), and the consequences of those actions for civil rights policy success.
(NOTE: the following should be read as a shorthand summary of how we constructed our study and use of qualitative sources therein. The paper itself elaborates in greater detail each of the following bullet points)
This article is a historical-comparative study of three major civil rights statutes (voting rights, equal employment, and fair housing) and is motivated by one fundamental question: why are some civil rights laws more successful than others? Consensus exists among civil rights scholars that voting rights was far and away the most successful of the three; that fair housing was a general failure; and that equal employment opportunity was moderately successful. Our study seeks to explain this specific hierarchy of civil rights outcomes across these three cases. As is the case with all historical-comparative research, qualitative data and qualitative analysis are essential because each case is constructed and compared holistically with respect to historical events and outcomes, the actors involved and the choices they made (and why they made them), and the consequences of those actions for civil rights policy success.
The larger questions of which qualitative data sources were used; why they were chosen, and how they were interpreted/analyzed rests on one guiding principle: how do potential qualitative sources aid in the development and/or testing of the authors’ theoretical framework and/or hypotheses? Particularly when dealing with broad sweeps of history, the vast majority of available qualitative data surrounding a particular set of historical processes and outcomes will not likely be directly relevant to the authors’ objectives. In sum, we were guided by three basic considerations:
How do our core theoretical claims/hypotheses guide the selection of qualitative data?
How do our data sources contribute methologically to a historical-comparative design?
What is the reasoning behind our empirical analyses/interpretations of the qualitative data?
It is common in historical-comparative research to initially analyze qualitative sources in the study’s front-end theoretical discussion/literature review in order to build a comparative design and theoretical framework that will guide the remainder of the paper. Those sources may rely on secondary data, primary data, or a combination of the two. Either way, this typically involves first considering alternative theories/explanations for the historical outcomes under consideration. If none of these current explanations can adequately explain comparative outcomes across all cases, then we must seek other theories/explanations. In our study, we identified four such alternative explanations cited by other scholars: 1) formal enforcement authority; 2) bureaucratic infrastructure; 3) policy entrepreneurship; and 4) white resistance. Each of these factors is usually (but not always) constructed as a binary qualitative “variable” that is either “present” or “absent” in each case, and determined by the secondary literature and/or primary data. We considered all four alternative explanations for civil rights policy success, concluding that none of them could explain the specific hierarchy of civil rights outcomes across all three cases. Instead, we built our own theory of civil rights policy success (rooted largely in the sociology of law/law and society literature) and identified one central factor (or “variable” if you like) we claimed could explain observed outcomes: the extent to which each policy incorporated a “Group-Centered Effects (GCE) enforcement structure. As our guiding theoretical principle, GCE-based enforcement then guided our selection and interpretation of the qualitative data in our study’s formal empirical-analytic sections.
Logic of Annotation
(NOTE: Annotations 1-3 provide some context and examples for how qualitative data initially helps in constructing the comparative design in ways described above)
The vast majority of the annotations (annotations 4-19) in this project appear in the paper’s formal empirical-analytic sections, in which we try to show how our central GCE variable can explain observed comparative outcomes across all three cases. Thus, we explicitly chose qualitative data that invoked or was directly related to GCE factors. Potential sources included:
Secondary studies/analyses of civil rights laws
Archival government documents such as:
Memos and correspondences among officials from the Department of Justice (voting rights); the EEOC (equal employment); and/or HUD (fair housing)
Agency reports, published guidelines/regulations, etc.
U.S. Congressional hearings
Supreme Court rulings
Newspaper articles (especially the New York Times and Washington Post)
Drawing on a broad sampling of qualitative data from the above sources, we constructed and compared three analytic “narratives” for each respective case, keeping our analyses and conclusions tightly connected to our GCE-informed theoretical framework. Put simply, we chose and interpreted the data in ways that showed how voting rights built the most aggressive GCE approach to civil rights enforcement; fair housing built the weakest GCE approach; and equal employment built a moderate GCE approach.
Each of the 19 annotations for this project highlight one or more of the above themes relating qualitative data selection/interpretation to our theoretical and empirical objectives.
Some annotations include links to web pages so the reader can themselves view the cited primary documents; when that is not feasible, I have included a supplemental Word file with photocopies of the annotated documents.
All annotations include a corresponding “analytic note” in which I provide elaboration for why and how the highlighted qualitative source was chosen/interpreted.
</ul
Gender and participation in community based development: Evidence from the Decentralized Climate Funds program in Senegal
No full textProject Summary
The ‘Decentralizing Climate Funds’ (DCF) project, implemented in Kaffrine, Senegal and Mopti, Mali, is a project that aims to support locally led climate change adaptation. The underlying belief of the projects approach is that involving communities in project selection and implementation creates a greater impact and will enhance individual, household, and community resilience to climate change, relative to a top-down decision-making approach. DCF’s approach rests on the assumption that bringing decision-making to the local level will increase the voice and inclusion of the poor and marginalized within communities by giving them a greater say in decisions that affect their lives, ensuring that development initiatives are correctly targeted to the needs of all beneficiaries. This study seeks to evaluate the ability of the DCF project to encourage the voice, participation, and inclusion of marginalized communities by focusing on participatory development. In particular, given Senegal’s patriarchal context, the study aims to answer whether the DCF approach allowed women to increase their participation and voice their needs.
Data Overview
The study is based on in-depth focus group discussions and semi-structured interviews that explore the subjective and dynamic views of participation by different beneficiaries. Fieldwork was conducted in two villages and one city in the Kaffrine region of Senegal that had engaged in a community decision-making process and received funding for a public good from the DCF project. Recruitment for the focus group interviews was through the head of the village, who helped identify women who were part of the village-level deliberations on selecting an investment. While there are selection bias concerns when speaking to and recruiting these women as they are all actively participating in the development and decision making in their communities, it was essential to speak to them to understand the level of participation within group meetings. For the semi-structured interviews, participants were selected through a snowball sampling strategy. The researchers walked around the sites and found residents to speak with contingent on their willingness and availability. This approach was described to and approved of by the heads of the respective villages to ensure the team was respecting the norms of the community. The interview notes were thematically analyzed using an inductive approach that systematically identified themes and patterns within the data
Investigating and Archiving the Scholarly Git Experience
No full textProject Overview
This project contains the data that underlies the behavioral study done as a part of Investigating & Archiving the Scholarly Git Experience (IASGE), a funded project thanks to the Alfred P. Sloan Foundation. IASGE addresses the need to investigate how academics are using Git hosting platforms, how they can be adapted to academic needs, and how the scholarship hosted on them can be archived. The results of this project aim to inform the that scholarship on Git hosting platforms move from a phase where they are highly active and collaborative, to a state where they are stable, permanently citable, and under active, professional preservation.
This half of the IASGE project, the behavioral study, is a three-part, mixed methods empirical approach to understand the scholarly users of Git and Git hosting platforms (GHP) whose use patterns vary from beginner to advanced. Methods employed include focus group sessions, a survey, and interviews. The data used include, but are not limited to, user profiles, frequency of feature use, and common practices, all which inform how librarians can support scholars in sustainable and long-term research. Through focus groups, a survey, and interviews, the study considers two groups of scholars: those who actively use Git to conduct their research within the academy, and those who are aware of Git but have not fully adopted it into their workflow in a standardized way. While each scholar may have varying reasons and degrees to which they include Git throughout their workflows, they all belong to the community of scholars who have the potential to utilize the power of Git to optimize their research and scholarship output.</p
Data for: Implementation of home blood pressure monitoring among French GPs: A long and winding road
No full textProject Summary
Background: To explore the perception of home blood pressure monitoring (HBPM) by general practitioners (GPs) in everyday practice in order to identify facilitators and barriers to its implementation in daily practice.
Results: The first reasons given by GPs to explain their difficulties with HBPM (Home Blood Pressure Monitoring) implementation were the usual lack of time, material and human resources. However, all of these motives masked other substantial limiting factors including insufficient knowledge regarding HBPM, poor adherence to recommendations on HBPM and fear of losing their medical authority. GPs admitted that HBPM use could enhance patient observance and decrease therapeutic inertia. Despite this observation, most GPs used HBPM only at the time of diagnosis and rarely for follow-up. One explanation for GP reluctance towards HBPM may be, along with guidelines regarding hypertension, HBPM is perceived as being a binding framework and being difficult to implement. This barrier was more predominantly observed among aging GPs than in young GPs and was less frequent when GPs practiced in multidisciplinary health centers because the logistical barrier was no longer present.
Discussion: In order to improve HBPM implementation in everyday practice in France, it is necessary to focus on GP training and patient education. We must also end "medical power" in hypertension management and turn to multidisciplinary care including nurses, pharmacists and patients.
Data Generation
Procedure: Data for this project were generated by a qualitative study comprising the conduct of six focus groups between October 2016 and February 2017, gathering 41 general practitioners in primary care practice in Lorraine (North Eastern France), with thematic and comprehensive analysis. Each meeting was recorded and subsequently fully transcribed to gather all statements. Each focus group was composed of GPs using HBPM more or less regularly. Focus groups were led by a moderator and an observer (GD, LD). The observer’s role was to collect non-verbal behavior, noting silences and hesitations. The role of the moderator was to ensure that each of the GPs could speak on each theme. It was not a questionnaire but a semi-structured open discussion. The duration of the focus groups was about two hours.
Recruitment: Sampling was carried out among a population of general practitioners working in a primary care office setting. Variability in age, gender, type and place of practice were taken into consideration for constituting the different GP panels. The first group comprised teachers and GPs trained in the use of HBPM (some GPs had a university degree in hypertension and cardiovascular risk). The second group comprised GPs occasionaly working with one of the authors and known to them as non-user.
Consent: All participating GPs gave their agreement to the recording of the interviews. All of the comments expressed during each focus group were meticulously transcribed, preserving the anonymity of those involved. All participants were informed regarding the goal of this study, the modalities of focus group attainment and the processing of collected data.</p
Data for: Exposure to trade and postcolonial divergence on the island of Hispaniola
No full textData Overview
These documents focus on political, economic and social development processes in Haiti and the Dominican Republic during the post-colonial period. In particular, they provide a window into the diverging trajectories of trade integration, industrialization, and long-run economic development. They were collected in public and non-profit archives located in Santo Domingo, DR and Port-au-Prince, Haiti. They were captured primarily using a digital camera placed on a tripod over a period beginning in the summer of 2009 through the summer of 2012
Data for: From Pews to Politics: Religious Sermons and Political Participation in Africa
No full text Project Summary
Does religion influence political participation? This book takes up this pressing debate using Christianity in sub-Saharan Africa as its empirical base to demonstrate that religious teachings communicated in sermons can influence both the degree and the form of citizens' political participation. McClendon and Riedl document some of the current diversity of sermon content in contemporary Christian houses of worship and then use a combination of laboratory experiments, observational survey data, focus groups, and case comparisons in Zambia, Uganda, and Kenya to interrogate the impact of sermon exposure on political participation and the longevity of that impact. From Pews to Politics leverages the pluralism of sermons in sub-Saharan Africa to gain insight into the content of cultural influences and their consequences for how ordinary citizens participate in politics.
The sermon data were collected in order to describe and understand religious messages in our research sites, with the goal of then investigating the effects of exposure to such religious teachings on political behavior. We wanted to have a deep and broad understanding of what clergy-congregant communication across a range of contemporary churches actually looks like, without prejudgment. Therefore, we sampled churches of different denominations and conducted church-level surveys and participant observation in sub-Saharan African cities. In Nairobi, along with a team of research assistants, we collected texts of Sunday sermons, observed worship services and other church activities, and collected information on congregation characteristics. We compare churches’ social organizations, weekday programming, and social welfare activities. In order to approximate a random sample of churches, we generated two comprehensive lists of (1) Pentecostal churches and (2) Catholic and Mainline Protestant churches in Nairobi based on Google Maps and then, using a random number generator, created a sample of one hundred Pentecostal churches and fifty Mainline/Catholic churches, expecting that there might be more variation among Pentecostal churches (see also McClendon and Riedl 2016). Because some churches are new, poor, or not officially registered, we did not work from government registration lists, nor did we rely on capturing information only from church websites. This method allowed the highest degree of church variation and self-categorization. We collected sermons online, where available, and in person through attending a service and transcribing the message. In total we collected sermon content from sixty of the one hundred selected Pentecostal churches and from twenty four of the fifty selected Mainline and Catholic churches, with random replacement for churches that could not be located at all. Sermons span from 2008 – 2017; data collection took place from 2003 – 2017.
In other contexts across sub-Saharan Africa (Accra, Lagos, and Kampala), we conducted a cursory examination of sermon content by searching for Pentecostal, Mainline Protestant, and Catholic churches that appeared in Google Maps. We then collected any online sermons, recording, or texts from the churches that had Websites, YouTube channels, or social media pages with sermon content. This method allowed us to examine the sermons of seven Pentecostal and seven Mainline/Catholic churches in Accra, eight Pentecostal and three Mainline/Catholic churches in Lagos, and three Pentecostal and two Mainline/Catholic churches in Kampala (for a total of thirty churches). Most churches had their most recent sermon, if any, available online, so the coverages of these searches is limited to 2017. We also conducted an exercise in Johannesburg that paralleled the one done in Nairobi but we don't include those sermons in this depository because they are owned also by Dr. Maria Frahm-Arp at University of Johannesburg and discussed only briefly in the book.
Data Overview
Nairobi sermons: This file folder contains the .txt files of the sermons collected, transcribed and recorded from Nairobi churches. Sermons have been de-identified by removing any specific references to the church name, location and pastoral teams. Each sermon file (.txt) is labeled with a unique identifier that contains a church identifier (rather than name) and date on which the sermon was delivered, where available.
Sermons outside of Nairobi: This file folder contains the .txt files of the sermons collected, transcribed and recorded from churches in other cities (not Nairobi). Sermons have been de-identified by removing any specific references to the church name, location and pastoral teams. Each sermon file (.txt) is labeled with a unique identifier that contains the country name from which it was collected, a church identifier (rather than name) and date on which the sermon was delivered, where available.
Meta data: McClendonRiedl_SermonData.xlsx contains information on churches that delivered the sermons. Not all churches were visited or all priests/pastors interviewed, so meta data are available only for a subset of churches.
Codebook: McClendonRiedl_Codebook.xlsx contains explanations and values for each variable listed in McClendonRiedl_SermonData.xlsx
</ol
Data for: “Ninguém me chama” - “Nobody asked me”: Gendered patterns of participation in sea turtle conservation in Northeast Brazil
No full textThis is an Annotation for Transparent Inquiry (ATI) data project. The annotated article can be viewed on the publisher's website.
You will need to use the Chrome browser with the Hypothesis extension installed to view the ATI annotations
Data Overview
This manuscript employs a mixed methods approach to identifying gendered characteristics related to access, attitudes and agency to explain gendered patterns of participation in conservation at the community level. The data presented in this manuscript was part of a larger study conducted over a nine month period in four small coastal communities on the northeastern coast of Bahia, Brazil.
Data Generation
Household questionnaires and focus groups were the primary methods of data generation for this study. Determining a sampling frame and procedures for the questionnaires involved creating hand drawn maps for each community, physically counting all existing houses, and comparing this information with the number of households registered with the local health clinic. The number of vacation homes in the area complicated initial attempts to randomly sample houses by neighborhood in the larger communities, with the decision being made to approach all homes in the smaller communities. Time and transportation were also a constraint as the roads to some of these communities were not paved and often full of large potholes. No public transportation was available to three of the four communities (the author lived in the largest of the four communities – Sítio do Conde).
Focus groups were conducted only with members of one community, the largest of the four included in this study. The original plan to host 2-3 focus groups in each community did not seem necessary or even feasible after data collection using the questionnaires. The intention of the focus groups was to meet with groups of community members already working together (as part of the church, fishers association, etc.), instead of creating groups of people that may or may not already know each other well or have had a shared experience. Because of the lack of functioning community groups in the other communities (mainly owing to their size and the centralized nature of groups to Sítio do Conde and Conde, the municipal center), it would have been difficult to gather enough people that belonged to the same community organization for a focus group. Due to time and transportation constraints, I began with the community groups in Sítio do Conde and re-evaluated the need for additional focus groups in the other communities later. Also, the sea turtle conservation project is physically located in Sítio do Conde and community members in this area have more direct contact with them. Alternatively, it would have been interesting to conduct focus groups in the other communities for comparison of perceptions and experiences by community as well as gender. Because the household questionnaires were conducted in all four communities I was confident that I had captured some of this already and saw little difference during preliminary data analysis.
The focus groups were also segregated by gender, partially by design and partially as a result of working with existing community groups, which are already fairly segregated. The fishers association was supposed to have two focus groups, one for men and one for women, however both men and women showed up to the first focus group. Only two women attended this focus group among 12 men, and only one of them spoke. I decided to separate out her comments for analysis. Another focus group of just women fishers was not conducted due to lack of support by the fishers association and a concern on my part of over taxing the same women that already attended a focus group and participated in other aspects of the larger study.
Data Analysis
I will focus my discussion on the focus groups here, with only a little treatment of the household questionnaires as more sophisticated statistical analysis was conducted using that data and published in another manuscript. Only descriptive statistics were used in this manuscript to provide a broader overview of the communities highlighting gendered patterns of participation and show the basis for further exploration through the use of focus groups.
The focus group dialogues were digitally recorded with the coding being done to the audio file rather than being transcribed and coding the text. This decision was made primarily because of the time and cost constraints involved in transcribing multiple speakers in Portuguese. Coding the audio file also allowed for segments of the conversation to be coded, capturing more of the exchange between speakers, rather than simply the individual quotes. This coding procedure also highlighted how much time was spent discussing a theme or sub-category within a theme. A number of factors were considered in determining what aspects were most relevant, including the number of times a code appeared in a focus group, overlap of codes between themes, the amount of time or length of comments related to the topic, as well as the number of groups mentioning these issues. However, this was not a purely quantitative assessment of frequency or absence/presence, with greater emphasis on what and how people were discussing an issue in addition. The quantitative aspects of the codes were left out of this manuscript because I feel it detracts from the overall findings, especially with a small sample size for the focus groups. In a small community with participation focused in a small number of groups This was appropriate. Furthermore, the alignment of the focus group findings with the questionnaire reinforce my conclusion that they are representative of the larger community.
Logic of Annotation
I chose to focus my annotations on providing analytic notes and source excerpts to offer additional context and interpretation, while also illustrating links between data, analysis, and conclusions. Because this manuscript was part of a larger project there was data collected previously that informed the instruments used in this study, as well as the conclusions that were made. The intent of the annotations was to make this more transparent and help paint a more complete picture. My decision to link (or not) to original sources was primarily based on ethical constraints related to the protection of human subjects. As a result, I did not link directly to any audio recordings from the focus groups, but did provide extended source excerpts and translations. Furthermore, I only included photographs that I had taken myself, and ones that did not include research participants. </p
Data for: Socioeconomic determinants of nutritional status among 'Baiga' tribal children in Balaghat district of Madhya Pradesh: A qualitative study
No full textProject Summary
The design of the study is qualitative, it includes data from twenty in-depth interviews with the mothers of Severe acute malnutrition (SAM) and Moderate acute malnutrition (MAM) children from seven different villages to explore the social and cultural factors affecting the nutritional status among the children of Baiga tribes. Seven formal interviews with the public servants - one Integrated Child Development Services (ICDS) supervisor, one PDS (Public Distribution System) supplier, five Anganwadi worker (AWW), were conducted to gain insight into stakeholders’ perspectives and challenges faced by them.
Data Overview
Both random and purposive sampling techniques have been used to select the villages and participants. In all, seven villages were selected from those pockets where the Baiga tribal population is high, but sampling of the mothers of SAM and MAM affected children, public servants and other stakeholders viz Local Registered Medical Practioners (RMP ), medical officer and district official were purposive. Detailed demographic and social profiling of the villages was carried out with specific focus on villages under study. Interviews were conducted in Hindi and transcribed and translated to English by the author
Data for: Why efforts to promote the rule of law in Afghanistan failed
No full textThis is an Annotation for Transparent Inquiry (ATI) data project. The annotated article can be viewed on the publisher's website.
You will need to use the Chrome browser with the Hypothesis extension installed to view the ATI annotations
Promoting the rule of law in Afghanistan has been a major U.S. foreign policy objective since the collapse of the Taliban regime in late 2001. Policymakers invested heavily in building a modern democratic state bound by the rule of law as a means to consolidate a liberal post-conflict order. Eventually, justice-sector support also became a cornerstone of counterinsurgency efforts against the reconstituted Taliban. Yet a systematic analysis of the major U.S.-backed initiatives from 2004 to 2014 finds that assistance was consistently based on dubious assumptions and questionable strategic choices. These programs failed to advance the rule of law even as spending increased dramatically during President Barack Obama's administration. Aid helped enable rent seeking and a culture of impunity among Afghan state officials. Despite widespread claims to the contrary, rule-of-law initiatives did not bolster counterinsurgency efforts. The U.S. experience in Afghanistan highlights that effective rule-of-law aid cannot be merely technocratic. To have a reasonable prospect of success, rule-of-law promotion efforts must engage with the local foundations of legitimate legal order, which are often rooted in nonstate authority, and enjoy the support of credible domestic partners, including high-level state officials.Case Selection
The core of the empirical work of this article is a case study of U.S. engagement aimed at promoting the rule of law from the intervention in late 2001 through to the election of President Karzai’s successor in 2014.
Afghanistan forms a compelling case study because it has had a major role in US foreign policy since 2001. Afghanistan’s connection with the 11 September 2001 Al Qaeda terrorist attacks and major international involvement in state-building after the fall of Taliban government in 2002 and the subsequent emphasis placed on rebuilding the country in a way that will prevent it serving as a haven for launching terrorist attacks have ensured that it will be of significant interest to policymakers in the US and beyond for the foreseeable future. Afghanistan constitutes a crucial test case for international state-building efforts as well as a rich source of “lessons learned” for policymakers.
This article looks at U.S. rule-of-law initiatives systematically over the arch of largescale U.S. engagement. It also surveys the Afghan justice sector holistically through a focused, structured comparison of major U.S. rule-of-law programming over time.
Methods
Causal inference is based on process tracing.
Here it involves a close examination of the major US initiatives to promote the rule of law in Afghanistan from 2002 to 2014.
Process tracing demands extensive data. For my research, the availability of robust contemporary documentation and the ability to interview key actors makes process tracing feasible. It allows for additional methodological rigor, assists in both the generation of new insights and in testing their plausibility as well as increasing the explanatory power of potential explanations, while helping to eliminate competing accounts.
The main sources of data used for identifying the relevant causal process observations are primary sources produced by relevant states, donors, NGOs, and other actors (for example, program reports and evaluations, strategy documents, and legislation), anonymous interviews with key informants, and secondary literature.
Materials produced by implementers, donors, and government agencies can illuminate their strategies, activities and goals. They reflect assumptions regarding causal mechanisms as well as the link between their activities and desired change. These materials articulate clear views on the motivations of key actors and institutions, which can then be subject to investigation. International judicial state-builders are quite forthright about their objectives (though not necessarily their motivations). This makes it possible to trace the evolution of rule of law efforts over time and how actors responded, or failed to respond, to unfolding events, as well as to critically examine the ends pursued and means selected.
The bulk of interviews were conducted were extensive background interviews in Kabul, Afghanistan during January and February 2014 with additional interviews over Skype. It also drew on insights from a previous research visit to Kabul in May 2009.
In situations that were unclear or contested, the strategy for making causal inferences was triangulation.
In no instance has a claim been accepted based on an anonymous interview that has not been collaborated by another functionally independent source.
Overview of Materials Included in the QDR
The material included in the repository focuses on the primary source written material cited in the published article