The places immigrants settle and how they disperse across the U.S. has more to do with family relationships than it does with economics, say Dartmouth researchers. In a study that compared the geography of immigration based on households rather than on individuals, professors Mark Ellis and Richard Wright found that, among other things, generational status, not income, has the most effect on immigrant dispersion. The team's findings, reported in this week's Proceedings of the National Academy of Sciences not only challenge current demographic analyses that track heads of households but reveal some of the social, political, and economic implications of doing so.

Unlike previous researchers who have generated settlement patterns of individual immigrants using such markers as time and age, Ellis and Wright are part of a growing group of demographers who consider social and economic variables along with physical location. According to the team, ignoring the complexities of families containing both U.S.- and foreign-born members exaggerates subtle differences between them and creates a more pessimistic view of immigrant assimilation. For example, if home-ownership rates were compared between U.S.-born heads of households to immigrant-headed households, the results would show a higher rate of ownership among U.S.-born breadwinners. But if the rates were calculated again taking into consideration homes that may have an immigrant presence, such as a spouse who is not necessarily the head of the household, the rate of home ownership among immigrants would go up.

In order to tease out these differences, Ellis and Wright identified seven different kinds of households based on the immigrant presence within--for instance, a household in which the head of the family represents the second generation of immigration. All but one category of household is connected to an immigrant directly by the presence of the person or indirectly through the presence of a parent who was foreign-born.

The numbers of people living in these distinct circumstances vary widely: In the U.S., 12 million people live in immigrant-only households, but 23 million live in an immigrant/second-generation situation. Understanding this mix could provide more useful data to states containing a larger number of immigrant-only households, whose members may have greater cultural and language barriers to overcome as they try to assimilate. Conversely, in states where immigrants live with U.S.-born families, the financial and emotional costs are shouldered by the family. "The family eases adjustment to American life for immigrants. And when these immigrant families span generations, especially when they include third-generation-plus folks, that adjustment will speed up," Ellis remarks.

When mapped, each of the seven household categories revealed a different pattern of settlement across the U.S. compared with maps that track individual immigrants. For example, a map that focuses only on individual immigrants shows that a large portion of the U.S. has less than half the immigrants they would have if these groups were distributed evenly across the country in proportion to each state's population. But a map that tracks households containing second- and third-generation immigrants (who have one or more foreign-born parents) reveals a wide distribution of states with large populations of households that have ties to first-generation relatives. The researchers found that in Los Angeles, for one, mixed-nativity households are less likely to live in ethnically distinct neighborhoods than households comprising first-generation immigrants.

Such detailed information is not available for every metropolitan area in the country and as a result, the researchers still have questions about what accounts for the variations in immigrant living arrangements across the country. However, they write in their report, "the judicious use of the tools of social geography can show some of the ways in which immigrants constitute 'us,' not 'them.'"