PLO 6

For this assignment, I described my data wrangling, cleaning, and analysis journey with the labor employment data I used. I explained why I chose the topic, supporting dataset, origins of the dataset, and my experience with wrangling the data. The data gathering process showed me how to decipher when a dataset will be useful for analytical purposes. Additionally, I used Python3 and JupyterLab to clean and access the data I collected. Within Python, I utilized the pandas library to help filter results as well.

Throughout this assignment, I was using a prescribed framework designed to approach data analysis problems from an organized standpoint. This framework simplified the data analysis research to a few steps: have a goal, find appropriate data sets, and use technology as an assistant to augment my effort. It was evident to me that technology is extremely beneficial for munging and filtering large datasets, but I should not center my research around what technology is available to me. Technology that I use for analytical purposes will be for supporting my workflow rather than directing my attention.

I reviewed a presentation of Stamford Hospital’s transition from their original data analytics framework to their adoption of Tableau. The original setup was very fragmented and did not offer nearly the same amount of valuable insight that was garnered through the use of Tableau. When incorporating Tableau, the hospital was able to save money, identify areas of improvement, and catalog resource intensive areas. I found that the presentation highlighted the vast improvements offered by Tableau’s analytical capabilities. The presentation I reviewed showed me how beneficial analytics can be in a healthcare setting. What was particularly interesting was the fact that the insight derived from the analytical software influenced multiple departments, from patient care to financial reporting.

In this briefing, I reviewed data sources in the realm of healthcare. I explain how new technology can obfuscate the value of healthcare data if standardization processes are not in place. Having too many data sources with just as many file types will not produce insights and will only complicate matters further. Once the mountain of data has been analyzed, the insights may already be outdated because by the time the analysis took place, more data was accumulated. Also, faulty collection devices could lead to incorrect insights, therefore ongoing maintenance of data collection devices is necessary.

The sheer amount of unstructured data in healthcare settings is enough to warrant the implementation of software and devices that analyze such data. Machine learning can assist data analysis; however, this novel tool requires a learning period. The training sessions are crucial for machine learning because it sets the foundational knowledge for the algorithms and this period is when researchers must discuss and prevent biases from being passed on. I found that there are many ways to collect data and that this data collected also comes in different formats. This assignment emphasizes my understanding of the different types of data and their limiting factors. Additionally, I consider the significant improvements and benefits that technology can bring about to solve the problem of too much data, especially in the healthcare setting.