Fight or Flight? Immigration, Competition, and Language Assistance Resources in Metropolitan Atlanta

Abstract

As the Latino/a immigrant population increases, racial conflict historically understood in terms of Black and White in the U.S. South has expanded to include new contestants in metro-Atlanta public schools. By examining market and sociological competition theoretical perspectives, this study investigates how language assistance resource implementation for English language learners is influenced by supply–demand, school, neighborhood, and political factors. The findings suggest that programmatic decisions about language assistance are driven predominantly by supply–demand factors. However, neighborhood and school racial competition measures also influence the availability of language assistance for English language learners, particularly in terms of onsite language assistance instruction.

Keywords:

• English language learners
• Latinos
• resource competition
• language assistance
• competition

ACKNOWLEDGMENTS

I gratefully acknowledge support from the Spencer Foundation. I would also like to thank Karlyn Crowley, Tynisha Meidl, Stan Thangaraj, and Regina Werum for their helpful comments.

Notes

¹The terms English language learner and limited English proficient are used interchangeably in this study. Although English language learner is the term more commonly used in higher education research, limited English proficient is the classification adopted by the Georgia Department of Education. Both terms categorize students whose first language is not English.

²LAPs were initially coded into one of four categories—no assistance offered, pullout, floating translators, and ESOL courses—to perform an ordered logistic regression (Ferris & Hedgcock, 1998). However, there was not enough differentiation in the variable. LAP was thus recoded as a dichotomous variable.

³Previous research suggests diminishing returns to school expenditures (Burtless, 1996). I ran the functional form fit to determine which function best captured the expenditure variable. The quadratic form had the best fit (highest r ²); thus, expenditures were squared for the regression analyses.

⁴The grade-level organization of a typical high school does not neatly correspond with the Census Bureau's block-level age categories. For instance, some children may enter high school early or stay after 18 years of age. Thus, the age category 15–17 does not match the grade structure perfectly but serves as best available approximation.

⁵In the 2000/01 school year, 110 public high schools were in operation in the 20-county metro-Atlanta region.

⁶Other non-White rates were not included because of a small size in 147 of the 152 schools.

⁷Exposure indices refer to exposure rates of Whites to each non-White subgroup and of each subgroup to each other: Whites to Blacks, Whites to Latinos, and Blacks to Latinos.

⁸Racial isolation is a district-level measure of segregation and is calculated as the number of students in a particular racial group experiencing isolation in schools with more than 50% minority enrollments across school districts (Orfield & Lee, 2006). Schools with more than 50% minority enrollments are coded as predominantly minority.