Classmate and close friend Jeff Hudson and I were the first two student employees of the nascent Rose Institute of State and Local Government (then the Claremont Institute of State and Local Government) beginning in 1972. Organized by Alan Heslop, the Institute was guided by Tom Hofeller, a CMC and CGI graduate. Tom was the political/technical brain behind the main work, which was to marry demographic data and voting history with specific geo/political boundaries to allow a better understanding of voting behavior and to forecast the consequences of changes in political boundaries. At a basic level, the method was to aggregate voting data statewide by voting precinct (the smallest available units of voting behaviors), demographic data by census tract and census block (the smallest available units of demographic information) then painstakingly mapping the boundaries of voting precincts, census tracts and existing geo/political units in as granular a way as was then possible.

These were days of the rudiments of box-upon-box IBM punch cards, dot-matrix printers, and late night/early morning visits to the Colleges’ Seaver Center, where the only adequate computing power for the work of the Institute was located. Voting data was only available on machine readable form in a few counties and, even then, without the benefit of the Internet, expensive and difficult to gather. The work was tedious and inexact because the boundaries of units of measurement did not always coincide. In some ways like eating the elephant. How things have changed.