top of page

Quickcode Use Case: Power, Conflict, & Cultural Differences

Research performed by Rachel Pacheco, Wharton Ph.D. Candidate

My academic research at Wharton explores how power, conflict, and cultural differences impact large, multi-organization projects. Recent work looks at projects from over 120 countries that are funded by the World Bank. To better understand why certain projects experience conflict and what drives that conflict, I began an exploration of 7,000, 70-page reports that are written about each project.

Using Quickcode, I was able to quickly analyze these long text reports. Quickcode’s inclusion and exclusion feature helped me to build a query around conflict (an unbiased "bucket of words"), quickly showing me how conflict and disagreement is talked about in these reports, including how it might show up in a way that does not measure what I am looking for. Using this, I was able to understand from these 7,000 reports, where conflict was present, and more importantly, what level of conflict was present. The Quickcode analysis also included a query score that showed how frequently the query showed up in a report. Using word count of each report to normalize the query score, I was able to understand the projects that had more or less conflict. For my research, this was particularly compelling: what characteristics of these projects might drive more or less conflict amongst the organizations?

Perhaps even more interesting was what else I learned about my data from the Quickcode

interface. As I was building my conflict query using Quickcode’s inclusions, I noticed that another phenomenon was coming up through the machine learning algorithm. Along with companies, contractors, and firms experiencing conflict, there was a subset of projects where collusion and bid-rigging were also present. I have now begun a separate exploration around anti-competitive practices in these projects to begin to understand how and why collusion and other similar activities are occurring. I am now able to link the output data built through Quickcode’s analysis with quantitative World Bank data to explore this important dynamic in these development projects. Quickcode helped to identify the unknown unknowns in my open text data in a way that I wouldn't have been able to do previously.

bottom of page