Appinio Research “What is Data Analysis? Definition, Tools, Examples” 11.04.2025 Source: Appinio
For the purpose of this project, the dataset we are analyzing is Kaggle’s “Executions in the United States, 1976-2016”. As the title suggests, the dataset documents executions carried out in the United States since the death penalty was reinstated in 1976. Each row in the dataset represents an individual who was executed and includes the following demographics and case-related information such as name, execution date, age, sex, race, victim count, victim sex, victim race, and county. In an accompanying dataset, we also have additional variables including state, region, method, juvenile status, volunteer status, Federal status, and Foreign National status variables.
Our datasets allows researchers to identify broad patterns and trends in capital punishment across time, location, and demographic categories. For instance, one can use it to analyze change in the different execution methods over time, highlight which states exercise the death penalty and executions most often, and perform quantitative analysis of the demographics such as race and gender. Naturally, from this perspective the dataset is extremely useful in identifying institutional patterns in the administration of the death penalty.
Data Sets:
Our dataset was generated through manually compiling execution records from several independent sources. The state Department of Corrections are responsible for carrying out executions and by extension generating official documents that are related to each execution, a vital source in this data set. Execution records are then cross-checked with news reports, as executions are typically covered by the national and local media, providing extra affirmation. Some legal advocacy organizations also monitor capital punishment and track executions independently. After we collected the information we cleaned it and organized it into the standardized spreadsheet form. This process involved standardizing the qualitative categories, such as race and execution method, and resolving inconsistencies across states and time periods.
In terms of where our sources originated, the primary data comes from official records maintained by State Departments of Corrections, which contain the legal documentation of execution records and information about individuals executed in the United States. These records provide key variables such as execution date, method, demographic information, and jurisdiction. News reports are also used to cross-reference these records and confirm details about execution events, including the procedures and circumstances surrounding each case. Additionally, supplementary information is drawn from legal research publications such as the NAACP Legal Defense Fund’s Death Row USA reports, which provide statistics on death row populations, executions by state, and demographic patterns within the capital punishment system. Together, these sources allow researchers to verify execution records while also analyzing broader institutional and demographic trends related to the death penalty in the United States (NAACP Legal Defense Fund; Bureau of Justice Statistics).
The organizations that funded, compiled and published the dataset were the Death Penalty Information Centre. Victim details, including quantity, sex, and race, were acquired from the Criminal Justice Project’s Death Row USA report. The Data Penalty Information Centre is a non-profit intended to provide analysis and further report information about U.S. capital punishment; they are subsequently funded through foundations and grants.
As for information left out of the database, we do not have information as to the legal reasoning for sentencing, if there are any appeals, or overturns on the cases, meaning new criteria for those cases that may have resulted in an alternative outcome. If there are any mental health diagnoses or disability statuses that could have been a contributing factor. We also do not have the narrative of the individual or their corresponding families. We also do not have information about their private counsel, whether they are court-appointed or privately retained? Typically, court-appointed representation are too occupied with other cases to be able to properly show their defense. To be blunt, we do not have the information about why they received the death penalty. It could be because of the quality of defense, prosecutorial discretion, or judicial bias. The left-out information will prevent us from deeply analyzing the society’s political punishments.
The dataset’s structure inherently positions people’s genuine experiences and legal issues into absolute numerical categories like age, race, state, etc. In placing these executions in this table structure, the dataset frames capital punishment as simple, countable and comparable data outcomes, completely eradicating its context as a human and legal process. This type of structure indirectly makes the viewers shift focus to patterns and trends, naturally distancing analysis from the moral and emotional weight of each case within the dataset. As a result, the dataset’s ontology can lead viewers to interpret the death penalty as a broader term for an execution method. If we were to view this data set as the sole source of information, we would omit several personal identifiers that provide color to these cases. The personal narrative of the individuals executed, personal experiences of their families, legal reasoning. All these omissions risk a detrimental simplification of the death penalty into numbers alone, which potentially discredits the entirety.
Dr. Maisha Tabassum “Data Doesn’t Speak — Until You Make It Tell a Story” Source: Medium