Racially biased high-school support? The heterogeneous effects of support on bachelor-degree enrollment for first-generation immigrants

ABSTRACT

First-generation immigrants (FGIs) are eager to achieve the American dream and acquire the advantages a bachelor’s degree affords. FGIs compared to their second-generation, third-generation and native peers fare better educationally. But, this comparison obscures across racial variation within a single-nativity status. As such, it is important to understand what mechanisms might produce inequality in college access among FGIs. One mechanism may be differences in how schools support students for college-going. The content of support may vary if racial stereotypes affect how support is given. Such variations could be decisive for FGIs’ educational trajectories. We find that Latinx and black FGIs are less likely than other FGIs to enroll in bachelor-degree programs. The content of school support is a mechanism that can help partially explain this differential for black FGIs. Our findings indicate that how schools support black FGIs for college-going is different and is likely at least implicitly racially biased.

KEYWORDS:

• First-generation immigrants
• racial bias
• school support
• college-going

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. We recognize that achieving the American dream is not only associated with obtaining a four-year degree but also with the prestige of the college from where the degree is earned. Our data is limited to whether someone is enrolled in a four-year degree program, so we focus on this aspect of achieving the American Dream.

2. Roughly 16% of those born outside the United States have two parents who were born in the United States. We ran models where we included and excluded this group and results were robust, so we include them in our final models.

3. Per NCES restricted-use data requirements, we round all sample size numbers to the nearest 10 to protect the identities of the respondents.

4. We created scales with different combinations of these variables and the pattern of results was consistent.

5. We control both for parental SES (which includes educational attainment) and whether a parent has a BA. Our measure of whether a parent has a BA is designed to capture parents experience with four-year programs specifically while SES is designed to measure resource support (economic or cultural) for college-going more broadly. While parental SES and having a BA are highly correlated, including both did not lead to issues with heteroskedasticity.