Methodology

Data Collection and Analysis Tools

Interviews were conducted predominantly over the Zoom platform, which allowed for virtual interactions and ensured a safe and convenient environment for participants. They were an hour long on average and recorded within the Zoom platform. For interviews that took place in person or over the phone, recording was handled by the Apple Voice Memo app. The subsequent transcription of interviews was assisted with Otter.ai, aiding in the accurate preservation of participants' narratives. Thematic analysis was employed to identify recurring patterns and themes within the interview data, assisted by NVivo software, which will be further discussed in the Analysis section.

Ethnographic Observation

Prior to this research, I was unfamiliar with the city of Chicago. Desiring to better grasp the city’s dynamics, cultural fabric, and disparities between neighborhoods, as well as the lived experiences of planners working to mend the fabric of the city, I decided to gather firsthand experiences and insights through ethnographic observations. For a month and a half (from May-June 2023), I lived in an apartment on Chicago’s North side. During that time, I worked at a design firm in the urban planning division, and I hoped to observe the preparation for and participation of planners in community-engagement meetings, the process of engaging with clients across the city, and gaining a deeper understanding of the methods through which planners connected with their projects and the residents who they wanted to get feedback from. 

Through this work I was able to develop relationships with urban planners at this firm, and others included in my study, including some that worked for the city or county. My work for the firm was somewhat limited: contributing research and writing, participation at a community-engagement meeting, notes during discussions with clients, etc. Through this process, however, I was able to better understand the complexities involved with the practical implications of residents’ engagement in shaping their surroundings. At a meeting on the South side of Chicago, which I attended as a part of my job shadowing experience, I helped display data maps and engage people walking by. Detailed field notes were taken to document observations, interactions, and reflections on the experiences encountered. These notes provide important context to inform the analysis and interpretation of interviewee narratives, as well as a few stories to complement their observations.

I distinguish my ethnographic observations from the participant observation (which I will explain below) based on my own level of engagement or visibility within the setting. While working at the urban design firm, it was known that I was there as a researcher, and I was making observations on the events I was directly engaging in and my own experiences. The data I gathered through participant observation, on the other hand, was done in a purely observational sense, meaning that I did not do anything to disturb the course of the meeting or engage in any way as a participant or an organizer. They are intertwined in their necessity: to provide the necessary context I needed to understand the responses of interviewees. However, due to the differences in the way I conducted myself when collecting the data, I see it fitting to distinguish these as two separate methods.

Participant Observation

During my time in Chicago, I had the opportunity to seek out community-engagement meetings in which residents, local government officials, and planners could be observed. I incorporated this participant observation in my study in order to enhance my understanding of residents’ roles within these settings and their interactions with those running the meeting. Selection of these events was based on what was taking place while I was in Chicago, and my desire to witness a variety of formats and locations. I observed a total of 4 meetings. Two of these meetings were in-person, and two were online. Two of these meetings were general ward meetings, and two were based on specific infrastructural proposals.

During these conversations I took on an observatory role, as I did not want my presence to interrupt the flow of the meeting or influence the way anyone behaved. At each of these opportunities, I was able to take note of the structure of the meeting, the setting, the subjects discussed, which voices found expression, and the manner in which comments and questions were received. These observations were recorded with detailed field notes, as well as audio recordings and photos to ensure every part of the meeting could be captured.

Separate from these conversations was also the opportunity to observe recordings of research team meetings from the Housing & Neighborhoods Pillar of the We Will Chicago Plan. As a part of We Will Chicago, there were 7 research teams, each composed of a team of residents selected from applications to the city to participate. The research team meetings that I observed were from the Housing & Neighborhoods Pillar, which focused on how the city could improve its policy related to the availability of housing, the ways people could shape their neighborhoods, and how engagement with the city in these regards had impacted them. Each of these recordings was approximately 1.5-2 hours in length, and I observed 8 recordings total. I watched each of these recordings from start to finish, taking note of relevant themes and topics of discussion similarly to how I recorded observations at the community-engagement meetings I observed in-person. The insights from all of these participant observations further enrich the interview data by backing up the stories and feelings of the interviewees with examples witnessed and recorded by me in real-time and contribute their own narrative findings on my research questions related to the influence of ownership and trust on people’s experiences with city-engagement.

City-Wide Data Sets

While my interviews and other qualitative observations form a rich tapestry of stories, in order to better understand these patterns across the city I felt it beneficial to include quantitative data on relevant themes. The quantitative data served as a valuable tool to corroborate and contextualize the qualitative findings, in addition to providing their own answers. The city of Chicago offers many publicly available data sets, collected through surveys of residents that I have included to enhance my qualitative findings.

Data Sets and Relevant Variables

For my study I mainly relied on the data made available through the Chicago Health Atlas. This site compiles data about Chicago from many sources, including the American Community Survey (ACS), the City of Chicago Data Portal, and others. I also use data from the Chicago Parks Department. Throughout my findings chapters, I specify the source of data I am pulling from as well as whether I accessed it from the Chicago Health Atlas or another data source.

Relevant Variables

For my research, the relevant variables I examined included: trust in local government, owner occupied building rates, neighborhood cleanliness, the Hardship Index, the racial composition of each community area, the amount of vacant land owned by the city, and more. The selection of these variables was guided by their relevance to the themes explored through qualitative methods and their potential to contribute meaningful context to the analysis, such as neighborhood cleanliness as a metric for care and sense of ownership. Within each chapter, I introduce the variables mentioned and explain their relevance to the question I am exploring.

Analytical Approaches

Qualitative Approaches

In the analysis of my interviews and field notes, I employed thematic analysis to identify common themes across my interviewees’ responses. I coded for mentions of challenges or victories related to the resident’s alderman with regard to ideas residents proposed or residents perceptions of interactions with them, perceptions surrounding graffiti or murals, the impact of the city’s history on their trust or involvement, contributions to sense of ownership, relationship with neighbors, instances where neighbor relationships were helpful in advocacy, feelings surrounding vacant land, the impact of understanding the city on involvement, perceptions of difference between owners and renters, residents’ trust in the city, planners’ trust in residents, and the impact of ward boundaries on residents. To classify modes of involvement, I also coded for mentions of using the city’s 311 service, attending community meetings, participating in DIY action, and involvement in We Will Chicago.

Beyond coding for obvious themes, I also centered the principles of Interpretive Phenomenological Analysis.[1] The open nature of my research questions lent themselves well to this approach, along with the narrative style of much of the data I collected. Through this approach, I strove to understand the underlying thoughts and feelings of my interviewees and synthesize them, while also remaining true to what the interviewees said directly. This process allowed me to pursue an exploratory mindset and employ an inductive approach to identify key themes, building my conclusions around observations I found repeating themselves across multiple interviews.

Quantitative Approaches

In the analysis of city-wide data sets, I employ a combination of descriptive and inferential statistics. I use descriptive statistical methods throughout the findings chapters to describe the occurrence of certain phenomena across the city such as racial concentration, presence of public parks, trust in local government, etc. Inferential statistics, on the other hand, I employ to attempt to make connections between different phenomena across the city, and what those connections suggest with regard to my research questions. The primary method of inferential statistics I employ is correlation analysis. While correlation cannot determine the order or cause of variables, it can identify the degree of association between variables. Using this analysis, I compare two variables (such as rates of litter and rates of homeownership) to identify whether or not there appears to be any statistically significant relationship between them. I display these in my findings with tables and graphs that I generated using the R programming language in RStudio.

To run my correlation analysis, I used R, specifically the ggplot2 package, to construct figures that highlight the correlations between selected variables throughout my work. The code snippet:

geom_smooth(method = "lm") + stat_cor(p.accuracy = 0.001, r.accuracy = 0.001)

incorporates the methods used within the ggplot2 framework for linear regression analysis and correlation coefficient computations. I used the geom_smooth function with the method = "lm" option to fit a linear model to the scatter plots. The least squares method was used in this approach to minimize the sum of squared differences between observed data points and model-predicted values, yielding a linear equation that captures the general trend in the data.

In addition to the graphical representation of the linear model, I incorporated the stat_cor function to display the correlation coefficient (denoted as R) and its associated p-value directly on the figures. The p-value indicates the statistical significance of the observed correlation, and a value less than 0.001 suggests a highly significant relationship unlikely to occur by random chance. To enhance the precision and clarity of the presented statistical results, I set specific accuracy levels (p.accuracy = 0.001, r.accuracy = 0.001) for the displayed p-value and correlation coefficient. This precision control ensures a clear representation of results while maintaining a reasonable level of detail.

Ethical Considerations

Anonymity and Confidentiality

Throughout the research process, the protection of anonymity has been a paramount concern. All participants' identities, whether interviewees, ethnographic subjects, or engagement meeting attendees, have been meticulously guarded through the use of pseudonyms and adjusted descriptors. This approach mitigates any risk of unintended identification, fostering a climate of trust and safeguarding privacy.

In order to secure that privacy, steps were also taken to ensure confidentiality. Throughout my study, I was the only person with access to my field notes, recordings, and pictures. My field notes and photos were stored locally on an iPad notebook that only I had access to unlock. Interview recordings were also stored locally on my personal computer and uploaded only to Otter.ai for transcription. Interview transcripts, which were void of personally identifiable information, were stored securely within Google Drive. Their only other location was NVivo for the purposes of analysis.

Informed Consent

The principle of informed consent was upheld with utmost transparency. For interviewees, comprehensive consent forms were administered, detailing the study's objectives, the voluntary nature of participation, and the participants' authority to withdraw at any juncture. To ensure the confidentiality of interviewees, their identities have been covered by pseudonyms, preserving their privacy and fostering open dialogue. For city-wide data sets, these data sets are anonymized and free to be used by the public, so consent was not a concern.

For ethnographic and participant observations, it was not announced that I was making observations in order to preserve the nature of the events that I was observing. However, each of the meetings that I attended and have included in this research were open to the public and could have been attended by anyone, thus this consent was not necessary. During these meetings I limited my participation to observation of the meeting, taking written notes as the meeting progressed of content pertinent to themes of trust, ownership, and processes related to shaping the physical aspects of the residents’ community. The same level of protections on identity that I granted to my interviewees were granted to all those I observed during these data collection methods.

Reflexivity

I am aware of the ways in which my background and life experiences may influence the ways in which I interact with and interpret the data within my study. As a biracial woman of both Black and White descent, I am more cognizant of the racial implications between these two races than I may be toward other racial interactions. Despite the time that I spent living in Chicago during my data collection, having grown up within the suburbs of East Kansas, I am more familiar with social interactions in that setting than I am of those within a huge city like Chicago. My role as an outsider to this city created both new opportunities for me to see trends that others may have overlooked as well as challenges to connecting with residents within the area, especially those on the South and West sides of Chicago. Sociology as a field, and particularly at Harvard where I have been trained in that field, seems to me often oriented toward the idea of “justice,” defined by each researcher within their own moral framework. Within my own experience in this discipline, I have come to see most phenomena as a convergence of historical impact, individual agency, and communal culture. It is important to acknowledge how all of these elements form the lens through which I see and move through the world and may influence my analysis of the data.

[1] “Interpretative phenomenological analysis (IPA) is concerned with the detailed examination of personal lived experience” (Eatough and Smith 2017).