QGIS offers 5 methods of classifying vector map data (quantiles, equal interval, etc). What is the theoretical base of these methods? If there are certain methods suitable for certain types of data or so. Or if you know some webpages dealing with this. QGIS user guide just names them, doesn't explain.
"the range is then divided by the number of classes"
" for visualizing continuous data that is not distributed normally"
"method designed to optimize the arrangement of a set of values into 'natural' classes"
"A normalized measure of the magnitude of deviation from the mean within a set of measurements"
see related question Standard deviation classification in QGIS
Besides the good answer of Mapperz, you may also find the explanations in the ArcGIS Help of some use. It refers to ArcGIS, but the underlying principles are the same in QGIS:
Classifying numerical fields for graduated symbology