The data set that I am looking at from the City of Los Angeles is a chart containing payroll information for different departments and their positions. Each position has different pay values by the year from 2011 up to 2015. The pay is described in detail through columns that contain data for hourly wage, type of employment (i.e. full-time or part-time), and quarterly disbursements. In addition, overtime, bonuses, and healthcare costs are documented in the data set to show the total budget for the Los Angeles City Departments.
The way this data is organized is useful to people who manage or analyzes finances in different Los Angeles departments. They would be able to see how each department disperses their funds to which position. This data would help financial analysts figure out how much should go into the department and perhaps understand future plans for the company.
Because the data has columns to show quarterly payments and a total projected amount, the data can show how it may update on four times per year. This phenomenon is expected as it is written in the dataset’s description. I think that by understanding how the ontology works, it may be possible to easily automate the data entry and update. This can happen by a scheduled update on a quarterly basis and analyzing the current year.
The dataset does not display any specific names. I think this is important for the sake of personal financial privacy, and this may look bad for the department if there may be a trend as to why some earn more than others. In other words, by having a name as part of the data, analysts may want to understand more. What is this person’s age, sex, and race? How do those factor into payroll?
If there is a data analyst or a financial analyst who is looking at company budget for payroll, this list actually may not do so well since it does not collect total values per department. I would rearrange the data by the department. Instead of showing payroll per position, I would show a total value based on all of the positions related to that department. Then other data could be listed such as averages per year, or total values over the course of several years. This would be easier to analyze trends and perhaps graph increases or decreases in payroll growth for business related discussions.
This was a really informative and thoughtful blog post. There were two main points that stood out to me the most. First, I think its really interesting that you noted an understanding of how the data is currently being reported and observed there was room for this data to be automatically updated. It is tedious to think of of the physical manpower that it would take to import all this data into the dataset each quarter but our modern machines have more than enough capability to learn and configure this data for us. Second, adding factors like gender, age, and sex into the dataset would provide really awesome insight onto gendered pay gaps, racial biases, and even age discrimination against the young and the elderly. Great job!