Data Analysis

Following data compilation, we analyze the data as follows.

• We inductively categorize qualitative answers to open-ended interview questions and textual information extracted from secondary sources. Each distinct idea is first assigned a unique numeric code. The codes are then categorized into two levels of higher-level categories. For example, local residents’ demands for more jobs and an industry association’s disappointment with a project’s local hiring practices are both grouped into the broader “jobs” category. Along with government officials’ demands for higher tax revenues and a local mayor’s request for greater financial aid to support migrants, the “jobs” issues are then further aggregated into the top-level “financial” category.
• We classify stakeholders strategically by assigning them to categories based on their quantitative responses to agree/disagree questions, from which we infer such attributes as level of SLO granted, influence power, and issue sensitivity.
• We summarize and evaluate data on “social ties” among stakeholder organizations using network graphs, which reveal relational characteristics such as cohesion, isolation, and fractionalization.
• We combine the results of the network analysis with those of the strategic stakeholder classification analysis to identify common perceptions and concerns among different stakeholder organizations, and ultimately, opportunities for influence and collaboration toward shared goals that reduce the level of sociopolitical risk or increase the level of sociopolitical opportunity confronting the client.
• We may use the results of the analyses described above to develop a set of hypothetical alternative engagement initiatives. Using our proprietary GIST software, we model each initiative’s potential future effect on stakeholder attitudes toward the client or its local operations and company, as well as the effect of maintaining the status quo (i.e., undertaking no new initiatives).