EDA
(Exploratory Data Analysis)

Topics

In this module, we look at methods of analyzing and visualizing correlation matrices, including multidimensional scaling, treating correlations as networks, working with generalized proximity matrices, using principal components analysis, and cluster analysis.

Software

You will need to install the latest version of borgworld. Instructions are on the Software page. 

• • devtools::install_github("StephenBorgatti/borgworld")

  • library(borgworld)

Readings

• DeJordy, R., Borgatti, S. P., Roussin, C., & Halgin, D. S. (2007). Visualizing proximity data. Field Methods, 19(3), 239-263. 

• Hierarchical cluster analysis

• Eigenstructures and factor analysis 

• Multidimensional scaling

  • Classical multidimensional scaling

• Correspondence analysis

  • Correspondence analysis

Optional readings

• Kruskal and Wish. Multidimensional Scaling

Datasets

• 1-mode

  • Holidays.xls

  • cities

  • animal pilesorts

  • emotions

  • illnesses pilesorts

• 2-mode

  • occup.xlsx

  • animal traits 

  • bcancer

  • nuts 

  • Greenacre Doctorates.xlsx

  • Marks.xlsx

• Cases & variables

  • music

  • hdi dataset

  • self-monitoring

Class notes

• Agenda | EDA