Correlation Explained, with Ancient Egypt example

One of the most basic statistical analyses is also one of the most revealing: correlation.  Simply put, correlation means when you tend to see (or not see) one variable, the other variable tends to appear (or not appear) in likewise fashion.

Rain and umbrellas are highly correlated, for example, while rain and sunglasses, which rarely occur together, have a strong negative correlation. Tetrachoric is the type of correlation coefficient that specifically applies to binary data, and it can illuminate many connections simultaneously.  One of the most valuable aspects of this type of analysis is that it reveals the full spectrum of both positive and negative connections. Correlation is not about causation, however—these studies only reveal the patterns apparent in the data, not the reasons behind those patterns.

For example, sunshades have the highest correlation to scenes where the king is seated or standing in a balustrade, those where he appears among other humans, and those where he wears an enveloping pleated cloak.

Example of a sunshade:


Pharaoh seated with sunshades behind him:


Share Button

The Art of Counting is dedicated to the memory of Margery Meilleur, who first taught me to view history through the eyes of the images we create.