Technologies Used: Python, Pandas, Sklearn, Tensorflow, wordcloud, nltk, PyDavis, Plotly, Seaborn
Related Fields: Language and Topic Modeling, Textual Analysis
Concerns regarding the effectiveness of elementary and middle school’s science education have elevated significantly over the past few years. With a polarizing political landscape and an impasse between federal and state government over the increasing education achievement gap among students with drastically different family and financial backgrounds, an investigation of some of the most commonly asked science exam questions at elementary & middle schools across the U.S. can shed more light on how the current administration (both local and national) navigates the difficult task of providing a impartial, accurate, and modern science education while aligning with the socioeconomic and political interests of the regional voters. The goal of this project is to visualize and explore the following questions surrounding elementary STEM education in the U.S.:
- What scientific topic(s) do K-12 schools in the U.S. most commonly focus on?
- How does the emphasis on the questions vary through the descriptions of the questions across each scientific subject category?
To view the analysis, please check the link https://paulshaoyuqiao.github.io/k12-stem-exam-topic-modeling/index.html