Breaking the Unequal Cycle: Parental Involvement in Hispanic Non-Native English Speakers’ Postsecondary Enrollment

ABSTRACT

This study examines the extent to which parental involvement factors predict the likelihood of Hispanic non-native English speakers’ (NNESs) postsecondary education (PES) enrollment after controlling for socioeconomic status and linguistic factors. A nationally representative Education Longitudinal Study dataset and a binary multilevel logistic regression model were used to examine the differences in the effects of parental involvement factors on Hispanic NNESs PSE enrollment. The results provide evidence in support of a conceptual model for Hispanic parental involvement, differentiates models from other NNES family groups, focusing on Hispanic NNESs’ unique familial characteristics and support system that could affect students’ PSE enrollment.

KEYWORDS:

• enrollment
• Hispanic
• non-native English speaker
• parental involvement
• post-secondary education
• Latino/a children and families

Notes

¹ F2BYWT is a label item in the ELS:2002 database. F2BYWT is the panel weight and it produced for all ELS:2002 sample members who responded in the base year and in the second follow-up, or who responded in the second follow-up and had key base-year data that were collected in the first follow-up.

² F2EVRATT stands whether participants ever attended a postsecondary institution.

³ MCAR refers to the missing data on a variable are not related to other measured variables. This is the only type of missing data that can be verified (Little, 1988).

⁴ MAR is identified as a systematic relationship between the propensity of missing values and the observed data, but not the missing data. MAR is not possible to be verified statistically, so researchers usually reply on its substantive reasonableness (Little, 1988). For instance, males are less likely to fill in an anxiety survey but this is irrelevant to their level of anxiety.

⁵ MNAR means when the missing data on a variable are related to the values of that variable itself, even after controlling for other variables (Little, 1988). For instance, when data are missing on standardized test scores and only the students with low scores have missing observations for this variable.