I chose to analyze the Gender Breakdown of City Workers by Department. The data type for this dataset is presented in a table which seems to mostly display categorical department titles with the percentage male and percentage female. Employee Payroll and male/female salary are also included within the table. A record in this dataset is the various department titles.
According to Wallack and Srinivasan, a dataset’s ontology “act as objects” and “negotiate boundaries between groups.” Bassically they are tags that manage and organize information so that you can use it to compare to even more information. The ontology in this dataset is department title.
This ontology of breaking down city jobs will make the most sense to a person who is looking at the bigger picture of all jobs in the city and their gender payroll differences. The dataset takes a lot of information and presents it in a digestible way. It is specifically designed to possibly answer the question that is presented in its title: “Gender Breakdown.” The ontology may be too strict to answer questions on specific jobs and the different positions within that one job. For example if I wanted to just analyze the role of a police officer, there is no ontology available for me to look at the various different types and rankings of police officers. While this makes the bigger picture clearer, some details that someone may try to find gets left out.
The gender breakdown dataset seems to tell the viewer that although not true in all cases, males typically have a higher salary under the same department titles than females. The dataset can show the wage gap phenomenon by specific jobs, and possibly by taking the average of all job salaries.
As i was saying earlier specific ranking and positions within a broader job department gets left out in this dataset, along with how long an employee has been working there and how many hours spent on the clock. These factors are just additional details that could result in less conclusive correlations being made about the gender gap in salary and profession.
If I were to start over with a different ontology I could go with years employed, looking from the perspective of someone saying that the gender wage gap exists regardless of the amount of years spent at a job. I could break down the categories as less than a year at the job, one to five years at a job, five to ten years, and over ten years at the job.
Another ontology could come out of someone who didn’t want to focus on gender at all and instead wanted to take a closer look at race. It would be a similar ontology, except male/female could be replaced with black/white/asian/hispanic.