An Introduction to the Use of Ecological and Robust Regression in Historical Research

On the Radical Views of Coal Miners in the Early Twentieth Century in Southern Illinois

“…We follow Goodman (1959) and Booth and Booth (1988). Let us consider the case where we have some marginal values for a series of contingency tables of the form of Table 1, with each table from a different county.…”

“…Since this is a straight line regression all of the usual least squares assumptions must be met if the least squares method is used to estimate p and r with the above equations. Those assumptions are given in Booth and Booth (1988) and many other places as well.…”

“…In this paper we consider a problem discussed previously by Booth and Booth (1988). In the early 1930s, a civil war erupted in the coalfields of southern Illinois.…”

“…To be specific, we wish to consider a problem discussed previously by Booth and Booth (1988). We wish to consider whether or not there was a difference in the proportion of support for socialist candidates between coal miners and non-coal miners in a ten-county region of Southern Illinois in the early twentieth century.…”

“…The major quantities of interest are p, the proportion of coal miners that are also socialist voters, and r, the proportion of non-miners that are also socialist voters. The marginal totals and frequencies are available for Illinois counties, but the cell frequencies are not (Booth and Booth 1988). We will attempt to recover these frequencies using ecological inference techniques.…”