Correlation

This week we discuss levels of measurement, and use this to discuss measures of association, similarity and dissimilarity. Of particular interest will be the Pearson correlation coefficient, which will then form a basis for discussing regression. An important part of the discussion will be appropriate normalizations of data.

Please bring a dataset to class that would be interesting to compute correlations or distances on.

Primary Readings

• • Osherson & Lane. Levels of measurement

  • Borgatti. Measurement

• Trochim. Correlation

• Borgatti. Distance and correlation

  • Borgatti. Notes on similarity measures

• Borgatti. Normalization

• Zegers, F. E., & ten Berge, J. M. F. (1985). A family of association coefficients for metric scales. Psychometrika, 50(1), 17-24 [PDF])

Worksheets

• measurement.xls

• binary.xls

Useful website: http://www.tylervigen.com/spurious-correlations