This continues our series of student reflections and analysis authored by our research team.
A Focus on Organization and Uniformity
As the project grows and matures, the processes and content are changing, which creates non-uniformity in the dataset between our past cases and the cases we are adding and coding currently. In our class, our weekly assignment consists of submitting five new and completely coded cases to add to the dataset. These cases are then checked every two weeks by a steering team member where mistakes are pinpointed and corrected. This has been successful thus far, but as we continue in the process, I believe it is not the most effective way to move the project forward.
The first reason for this is miscommunication and confusion. Since we update the codebook often, some students are left behind on what was changed and are still coding in the ‘old way,’ completely unaware that anything has changed. For example, we recently changed how charges are coded in the spreadsheet. Though before we did not have a uniform way of coding charges, we now add formal charge numbers and note how many counts there are for each charge by placing the number between two brackets ( [ ] ).
Such as in the case of Ardit Ferizi, his charges are coded (correctly) as: 18 U.S.C, § 2339B Conspiracy to Provide Material Support and Resources to a Designated Foreign Terrorist Organization [2 counts]; 18 U.S.C. § 1030(a)(2)(C) Unauthorized Computer Access and Obtaining Information; 18 U.S.C. § 1028A(a)(2) Aggravated Identity Theft. However, when looking at the “U//FOUO” (our finished cases) doc, I noticed that the ‘charges variable’ is coded incorrectly for about 95% of the cases, and this is not the only variable that stands un-updated on multiple cases.
I think a part of the class should be spent focusing on the uniformity of specific variables in the “U//FOUO” doc. For example, we would spend a class, or more if needed, splitting up years between each other and checking if all the charges are written according to the updated codebook.
We also have a problem of students coding cases they think are new, but have actually already been included in our dataset. My partner and I have run into many problems where we thought we were adding new cases, but actually had done unnecessary work because the cases were already fully coded and included in our dataset. Focusing on the cases we do have will help students become familiar with the cases and reduce the likelihood of duplicate cases. However, for new cases, when someone finds a new case, they can add it to the appropriate “New Cases” folder. Then, students can create case starters and code those cases for extra credit, but it would not be required each week. The focus regarding new cases will be on renaming, sorting, and filing all of the files in the “New Cases” folder while identifying what files contain actual new cases and what files contain updated information for pending cases.
Class could also be spent coming up with a better way to monitor and manage pending cases. Right now, we have a calendar noting when specific defendants are due to declare a plea, receive a verdict, or be sentenced. I think it would be helpful to have some kind of notification system so we are alerted when one of these cases moves further in the prosecution process. We currently have cases in the “Pending Cases” doc that date back all the way to 1993 (meaning they should not still be pending). Turning more of our focus to updating pending cases will help us organize the dataset as well as allow us to focus on staying up to date with cases that are actually pending and not fall behind.
Though there are project members outside of the class working on various things, including auditing, the main focus of the project class is adding and coding cases. I think this is helpful because it helps us expand and grow our dataset; however, I think a focus on organization and uniformity would be more productive and efficient for the project. As Miles, Huberman, and Saldana say in their book, Qualitative Data Analysis: A Methods Sourcebook, “Codes are primarily … used to retrieve and categorize similar data chunks so the researcher can quickly find, pull out, and cluster the segments relating to a particular research question, hypothesis, construct, or theme” (Miles, Huberman, Saldana, 2014, p. 72).
In order to use our codes, they need to be uniform and organized so we can categorize similar data chunks. We already have an auditing process in place, but I believe that organizing our dataset will help us come up with more uniform procedures and pinpoint reoccurring problems that we find. This will also help prevent us from adding more cases that are coded incorrectly and help clear up confusion on how certain variables should be coded.
ISIL-Linked Kosovo Hacker Sentenced to 20 Years in Prison. (2016, September 23). Retrieved March 29, 2020, from https://www.justice.gov/opa/pr/isil-linked-kosovo-hacker-sentenced-20-years-prison
Miles, M. B., Huberman, A. M., & Saldan?a Johnny. (2020). Qualitative data analysis: a methods sourcebook. Los Angeles: SAGE.