Exploring the Correlations of Economics, Education and Employment
Click the Poster to View Full Screen, Right click to save image
Max Kaiser
CoPIs:
Richard Angueira
College:
The Dorothy and George Hennings College of Science, Mathematics, and Technology
Major:
Computer Science
Faculty Research Advisor(s):
Ching-yu Huang
Abstract:
The economy has long been a substantial part of people's lives and has widely been regarded as among the utmost of importance. There exists issues that span across all cultures and can continue across generations that occur from this. Class issues have become much more apparent over years with the gradual emergence of egalitarianism. Gone are the days where people are born and die within the same economic class like how their lives were for most of history. That being said, it continues to be difficult to make determinations about the methods of which changes to individual parties can be made when it comes to this. There have surfaced many ways of which people will express are advantageous for moving up in social class and there have, alternatively, been ways that have been expressed that can do the opposite, ways of which are hoped to be avoided.
Data about such subjects exist across the United States, among other nations and can be observed to make potential determinations about how they affect each other. Measures of wealth, and related themes, are reflected in household income, poverty rates, unemployment rates and, less directly, education rates. A worthwhile resolution to come about from information like this would be to determine how exactly these things affect each other tangibly and mathematically. Questions about how education can affect poverty rates and how unemployment rates can affect median household income can lead to potentially helpful conclusions. Correlations between these is the start to coming to some idea of a resolution on this, but other methods will also be of significant use. Consider outlier detection in attempting to figure out ideas on why some places are wildly better or worse off economically when compared to contemporaries or the entirety of the nation. Despite how obvious some of these things may seem, real data and data mining is paramount in coming to real conclusions, because some things that may seem obvious are not quite what they seem when data is involved.
The datasets being used contain county level data with attributes on household income, education, poverty, population and unemployment. With data for each county, there are over 3000 records for information to perform data mining on. All of the attributes (aside from population) are relevant to this question. The datasets are sourced from United States government sources including the Bureau of Economic Analysis and the Department of Agriculture.