By Reynolds Farley, Ph.D.
I worked as a crew leader for address canvassing from March 23 to May 7. Reflecting the economy, five of the 18 members of my crew had post bachelor’s degrees. Several had given up lower-paying jobs to canvass for the Census Bureau. All had been told that they would have six to eight weeks of work. Canvassing took place from April 17 through May 6.
Our area was a rural one with numerous lakes and isolated homesteads not visible from unpaved roads. In many places, dirt roads lacked names, homes lacked numbers and residents claimed that the post office did not deliver their mail. The vehicles of three canvassers became stuck in mud. Maps on the hand-held computers bore no more than a remote relationship to what we found. Quite often we came upon an array of a dozen or two mailboxes sitting side-by-side at the end of a dirt lane. Some had numbers, some did not. Then there would be a dozen or two homes scattered about a lake, an estuary or a river front. Matching numbers with residences was extremely time consuming, if possible.
Address canvassing went well from April 17 through May 1. Our local census office was located in a suburban area adjoining a major metropolis. Officials there appeared to be unfamiliar with canvassing in a remote rural area. We were told that our district was the only one in the local census office not completed by the week-end of May 2.
Rather than letting us work for another week to finish the job competently, canvassers from urban areas were sent to our area in great numbers and at considerable cost. There appeared to be no interest at all in quality control. The emphasis was solely upon completing the canvassing before an arbitrary deadline.
The canvassers who started with this crew believed they would be employed for six to eight weeks worked three weeks at most. I suspect that the very many new canvassers who were sent in to complete the area had little, if any, familiarity with the rural area where we worked.For my entire career, I have used U.S. Census data in my teaching and research. The area we canvassed is one in which no address list could be complete and accurate. The canvassers working with me were serious and cautious. Two-thirds of the area was competently canvassed. One-third of the 29,000 address lines were canvassed in extreme haste implying that several hundred housing units may not receive a questionnaire when they are mailed next March 17. I hope that this emphasis upon speed rather than quality was a rare happening.
I had the good fortune of working with many excellent an dedicated canvassers in this brief period and a very competent Field Operations Supervision. I am, however, less sure about the dedication of some higher level local census office administrators to the important issues of minimizing undercount in the 2010 Census by getting an excellent address list.
Dr. Reynolds Farley is Professor Emeritus at the Population Studies Center and Department of Sociology at the University of Michigan, where he served as Chairman of the Sociology Department. Dr. Farley’s research interests concern population trends in the United States, focusing on racial differences, ethnicity, and urban structure. A recognized leader among social scientists who study race relations in the United States, Reynolds Farley is among the top echelon of social demographers, a leading authority on the demography of African Americans, and a penetrating and creative analyst of racial and ethnic relations over the past 40 years. His pioneering studies of the causes and implications of massive and continuing racial segregation have enlightened the national discourse on social policies concerning families, welfare, health and education. His current work includes an investigation of the residential consequences of revitalization in the Rust Belt. He earned his Ph.D. from the University of Chicago and first worked for the Census Bureau in 1962.