The issue of student debt is one of the most pressing for the millennial generation. Millions of young people every year leave college with an ever increasing burden of debt, threatening their purchasing power and financial security. The data bears these facts out, as the tableau story below will demonstrate. The purpose of this story is to demonstrate that student debt continues to be an unaddressed issue and is affecting the viability and ultimate value of the current higher education model.
The story type I used was change over time. Using the college scorecard data set, I was able to pull a decade of tuition, debt and income data for Non-Profit and Public institutions. The story looks at change in tuition, total debt and debt-to-income from 2003 to 2013. I provided three story points in order to tell the story. Each story point should logically build on the previous one – setting the stage with the increase in tuition leads to the consequential increase in debt burden. Because tuition exceeds the increase in income over the period – student debt-to-income naturally increases as well.
When it comes to transforming data into information, the tool used is often as important as the data itself. The College Scorecard data-set will require data cleansing and any tool used to create visualizations will need to be able to handle large data volumes. The first step to analyzing data is to understand the structure of your data-set.
To begin, I will use Microsoft Excel to open the .CSV files in the raw data dump and better understand how the data is structured. This will give me an idea of what is relevant and usable in the data-set. It will also give me the beginning of an understanding of what data needs to be cleansed and what columns or rows will be irrelevant to the analysis. Excel can provide quick and easy formulas, sums and averages for simple data calculations.
Second, I will look to use a tool such as Microsoft PowerBI to attempt to analyze and visualize the data. PowerBI allows powerful visualization and the adding of fields and filters to present information in the most relevant manner. Finally, it allows the establishment of relationships between fields, tables and even data-sets. Understanding a data-set, its components and the relationships between fields and tables is important to utilizing it as effective information.