Table of Contents

 

 

 

Case Study Files

 

 

Flint Case Study (Folder)

 

• pdf (this document)
• pdf (the case to be analyzed)

 

Gephi network files (Sub-Folder)

 

• combined-example.gephi
• combined-original.gephi
• individuals-original.gephi
• organizations-original.gephi

 

Raw network files (Sub-Folder)

 

• edge-combined.csv
• edge-organizations.csv
• edge-individuals.csv
• node-combined.csv
• node-organizations.csv
• node-individuals.csv

 

Gephi resources (Sub-Folder)

 

• CMN212_network_analysis_overview
• pdf
• html
• gephi_introduction

 

 

 

 

The Flint Michigan Water Crisis was selected as a case study due to the widespread inter-organizational communication that contributed to flawed decision-making and unprecedented man-made contamination of a public water supply (See Case Materials for additional background). This case study focuses on the events that led up to the crisis. During this stage of pre-crisis, organizations played a critical role in contributing to systemic failure, and as organizational communication scholars, it is our opportunity to analyze the organizational contribution to this crisis to ensure that it does not happen again.

 

This project requires you to read case materials, perform a network analysis of the organizations and individuals involved, relate results to the relevant organizational communication theories discussed in this course, and interpret your findings in a comprehensive paper. Your paper is expected to reference relevant materials, use visual outputs from your analysis as supporting figures, and elaborate on the results of your network analysis to reach new theoretical and practical solutions.

 

 

This document provides step-by-step guidelines for conducting this project. The steps of this project include:

 

1. Reading the case materials outlined in the following section (Section 3)
2. Conducting a network analysis using Gephi (Section 4)
3. Interpreting analysis results (Section 5)
4. Authoring a paper (Section 6)

 

In addition, this project is broken into four assignments (Noted in RED below):

 

1. Preparatory part 1: “A conceptual mental model of the organizational communication patterns in the Flint Water case”
2. Preparatory part 2: Use your mental model and answers from assignment 1 to form research questions supported with course readings
3. Preparatory part 3: “Network data analysis”
4. Final report

 

See the course syllabus for due dates of these elements.

 

 

You will read the following sections of the March 2016 Flint Water Advisory Task Force Final Report (FWATF, 2016):

 

• Executive Summary (Pages 1-3)
• Background and Summary Timeline of Key Events (Pages 15-21)
• Roles of Government Entities in the Flint Water Crisis (Pages 26-52)

 

You are encouraged to take notes while developing a conceptual mental model of the organizational communication patterns (Midterm preparatory part 1) in the Flint Water Advisory Task Force Final Report (FWATF, 2016). After reading the case materials, students should be able to answer the following questions:

 

1. What were the key organizations involved in the case? (About how many total?)
2. Who were the key individuals involved in the case? (About how many total?)
3. Did these organizations and individuals communicate concerns about water quality and safety effectively?
4. What constraints contributed to a reduced flow (no pun intended) of information between organizations?
5. Which organization do you believe were primarily responsible for the Flint Water Crisis?
6. Sketch a picture of the relationships between the organizations and individuals in the case.

 

In this project, you will take the role of an organizational communication expert. Based on the information you received from the Flint Water Advisory Task Force Final Report and course materials you will develop research questions to guide your exploration and report on this case (Midterm preparatory part 2).

 

In developing your research questions from your notes in midterm preparatory part 1, you are encouraged to think broadly about inter-organizational relations and approaches we have discussed, but consider the following in reference to the Flint Water Crisis case:

 

• institutional or technical environments
• propositions of contingency theory
• life cycle analysis
• elements of institutional theory
• elements of structuration theory (based on readings assigned February 21)
• elements of the communicative constitution of organizations (requires readings assigned March 7)

 

 

 

 

Before beginning this section of the project, you should have read the Flint Water Advisory Task Force Final Report located in the Case Materials folder.

 

In this step, you will perform a network analysis using a data set created for this case study. This section provides detailed instructions for using the network analysis software Gephi.

 

 

The network data in this project was extracted from the Background and Summary Timeline of Key Events (Pages 15-21) of the Flint Water Advisory Task Force Final Report. To develop a data set with limited scope and manageable features, the following assumptions were made when creating the files provided in this case study:

 

• Individuals and organizations are combined in one data set.
• Each “edge” (or “link,” or “tie”) in the network corresponds to an instance of communication between individuals and organizations.
• Nodes were created when an individual or organization introduced new information in the case, or performed some type of communication with one or more individuals or organizations.
• Nodes are linked (by edges, links, or ties) through directed connections, meaning, if “Individual A” is linked to “Organization B” with an arrow originating at Individual A and terminating at Organization B (Individual A à Organization B), this relationship should be interpreted as “Individual A communicates with Organization B.”

 

 

 

Gephi is an open source graph (network) analysis software used by data analysts to discover patterns in network data (Bastian et al., 2009). Gephi was selected because it can be installed on a personal PC or Mac, and has the statistical tools you will need to explore the patterns of communication among the actors and organizations in this case study. Furthermore, it allows you to explore the data set and develop new and interesting ways to visualize the organizational communication network.

 

We will be using pre-installed Gephi in the computer lab in Rm 4051 Lincoln Hall. However, you can also download it on your own machine. To learn how to install Gephi and how to open graph files, follow these directions.

 

 

1. Launch Gephi

 

Navigate to a graph file contained in the ‘Gephi network files’ folder of the case study: Flint_Case_Study > Gephi network files. For example, open ‘combined-original.gephi’

 

At the top of the screen, click ‘Window’ and make sure the following windows are selected: Appearance, Layout, Graph, Context, Filters, and Statistics.

 

If the Gephi file does not open or if you would like to learn how to load raw network data into the software please consult the following resource:

In the Statistics window, Run Average Degree for Network Interpretation: Directed. Once this has run, go to the ‘Data Laboratory’ window, and in the Node Table, new columns for the score of ‘Degree, In-Degree, and Out-Degree’ have been calculated. You can click the top row to order them from lowest to highest or vice versa. Now, in the ‘Overview’ window, under ‘Appearance’, ‘Node’ ‘Attribute’ includes Degree, In-Degree, and Out-Degree, visualize the graph using these parameters. Interpret the results.

 

“Highest average degree shows the most active organization, which is the one that has many connections or communications with more organizations regardless of direction of the connection.” (Abbasi and Kapucu, 2012)

 

“Out-degree computes the number of links sent to another actor, while In-degree refers to the number of links received by each actor.” (Valente, 1995)

 

 

In the Statistics window, Run Network Diameter for Network Interpretation: Directed. Once this has run, go to the ‘Data Laboratory’ window, and in the Node Table, four new columns have been calculated: Eccentricity, Closeness Centrality, Harmonic Closeness Centrality, and Betweenness Centrality. For the purpose of this case study, you will only be responsible for interpreting Betweenness Centrality. Now, in the ‘Overview’ window, under ‘Appearance’, ‘Node’ ‘Attribute’ includes Betweenness Centrality, visualize the graph using this parameter. Interpret the results.

 

 

In the Statistics window, Run Graph Density for Network Interpretation: Directed. Interpret the results.

 

“Density describes the general level of linkage among the actors in a network (graph) (Scott, 1991). The more actors connected to one another, the denser the network is. It is the proportion of existing links, to all possible links. The densest network is the one in which all points are connected with each other but such networks are very rare. The precise definition of density can be shown as below, where L is the total number of links (ties) among actors in the network and n is the number of actors and thus the denominator is the maximum possible number of links in the network.” (Abbasi and Kapucu, 2016)

 

 

Gephi is a tool for exploring your network data. Try some of the following techniques to visually investigate the interesting patterns in this dataset.

 

 

It is also possible to change the size of the nodes in your network to represent values you have calculated using the Statistics window. For example, you can use the Appearance window to resize nodes by their degree using the following sequence of steps:

 

 

 

 

 

 

Include the network maps/visuals that you created using Gephi for each of the three network modes (individuals, organizations, and combined).

 

 

Address three of the five following problems (you may address more, but no extra credit will be given). In your paper use the headings provided to refer to which problems you choose to discuss, and support your observations or analysis with citations to the readings.

 

1. Life Cycle analysis: Consider the lecture notes on life cycles of organization and apply the concept to one or more organizations in this case. Is an organization in a period of growth, maturity, or decline? Explain your reasoning.

 

Institutions: What are the institutional issues at work on organizations and departments of organizations in the case? Can you see evidence of isomorphism, rational myths, beliefs, practices, institutional messages, or other established arrangements of long duration?

 

Communicative Constitution: Discuss evidence in the case for the communicative constitution of Saint Pearl. In particular, cite any evidence you can identify of the four flows described by McPhee and Zaug.

 

 

 

You should write a well-argued paper with full supporting evidence, network graph(s), citations from the course readings, and citations of any additional and relevant literature. The paper should follow a standard format: abstract, introduction, methodology, results, conclusion, and references.

 

You should attach your completed exam to an email. Please use the following format in naming your exam: “JonesN 412Midterm” where N is your first initial and Jones is your last name. You are encouraged to study with colleagues, but grading of the exam (which is worth 25% of your course grade), will be based upon 1) application of the concepts given in the questions, 2) attention to detail, 3) originality, and 4) expository style (a rubric will be available on the Moodle page). This step is due by 8:00 a.m. March 14, 2016.

 

 

 

 

Use any citation style you prefer, as long as it is consistent throughout your paper.

 

e.g., using APA-6^th :

 

FWATF. (2016). Flint Water Advisory Task Force Retrieved from the Office of Governor Rick Snyder State of Michigan.

 

 

 

To cite this case study, please use the following reference:

 

Pence, J. & Proulx, J. D. (2017). Network Analysis of Inter-Organizational Communication Leading to the Flint Water Crisis. Department of Communication. University of Illinois Urbana-Champaign.

 

 

 

References

 

Abbasi, A., Kapucu, N., 2012. Structural dynamics of organizations during the evolution of interorganizational networks in disaster response. Journal of Homeland Security and Emergency Management 9.

Abbasi, A., Kapucu, N., 2016. A longitudinal study of evolving networks in response to natural disaster. Computational and Mathematical Organization Theory, 1-24.

Bastian, M., Heymann, S., Jacomy, M., 2009. Gephi: an open source software for exploring and manipulating networks. ICWSM 8, 361-362.

Blau, P.M., 1977. A macrosociological theory of social structure. American Journal of Sociology, 26-54.

Borgatti, S.P., Halgin, D.S., 2011. On network theory. Organization Science 22, 1168-1181.

Di