Past and recent studies have consistently found that ethnicity significantly affects varied outcomes of social mobility among different immigrant groups and that such divergent outcomes in turn lead to further changes in the character and salience of ethnicity. Much of the intellectual debate on ethnic differences is between the cultural perspective – emphasizing the role of internal agency and the extent to which ethnic cultures fit the requirements of the mainstream society – and the structural perspective – emphasizing the role of social structure and the extent to which ethnic groups are constrained by the broader stratification system and networks of social relations within that system.

Social scientists from both perspectives have attempted to develop statistical models to measure quantitatively the effects of “culture” and “structure” for the upward social mobility of immigrant groups. Under ideal circumstances, these models would include indicators illuminating pre-migration situations. But because of data limitations, many social scientists typically attempt to control for “structure” by documenting specific contexts of exit, identifying aspects of post-migration social structures, and operationalizing those components for which they have data. This is not only a conventional practice but also a reasonable approach, since many post-migration social structural differences (in the socioeconomic status of persons who came to the United States as adults) are likely to either reflect, or be carryovers from, pre-migration differences. However, even the most sophisticated statistical model accounts for only a proportion of the variance, leaving a large residual unexplained.