NCRS - The Changing Midwest Assessment

The maps depicting change in the density of seasonal housing units must be interpreted with care. The definition of a seasonal housing unit used by the Census Bureau changed significantly in 1990. Furthermore, the tabular data published by the Census Bureau and the definitions provided with those data exhibit inconsistencies such that it is very difficult to know exactly how the data for seasonal housing in the 1980 census should be compared to the data in the 1990 and 2000 censuses. Because of the incongruities between the definitions in 1980 and 2000, the magnitude of the changes shown in the change map on slides 3 and 6 likely exceeds the actual change. However, we believe that the relative rates of change are valid and demonstrate the variations in relative rates of change across the Midwest. For example, we are confident that counties shown as having >500% change in the density of seasonal housing (red counties) did experience the greatest growth in seasonal housing density, though the magnitude of the change was likely much less than 500%.

The map on this slide show the change in seasonal housing density between 1990 and 2000. The change map looks markedly different than slides 3 and 6 for several reasons. First, the same definition of seasonal housing units is used in 1990 and in 2000, reducing the likelihood that the same housing unit receiving the same use would be classified as seasonal in one census and permanent in the next. Changes in social trends also give rise to differences; seasonal housing density approached a saturation point in the upper Midwest in the 1980s, resulting in much slower growth during the 1990s. Furthermore, many housing units which were seasonal homes in 1990 had become permanent homes by 2000, a trend that is reflected in the declines in seasonal housing density seen in many counties that had high seasonal housing densities in 1990.