Abstract
My main concern in this article is with statistics relating to such basic matters as the sizes of farm-holdings, the output and yield of crops, household income and expenditure, occupational statistics, cattle ownership, the sizes of villages, etc.—though I shall also range more widely. While the distinct and professional field of demographic statistics is necessarily outside my scope, I shall criticize some features of the Karnataka population census.Although since 1953 most of my fieldwork has been undertaken in the West African countryside, I am obliged to take most of my examples of bad statistics from south India, since West African statistics, which were never abundant, are now scantier than ever. Throughout my discussion I take it for granted that the lack of reliable statistics gravely impairs our understanding of the working of tropical rural economies.

This publication has 3 references indexed in Scilit: