Donald J. Hernandez is a Professor in the Department of Sociology at Hunter College/City University of New York and a member of the Technical Work Group of the National Research Center on Hispanic Children & Families. This blog is based on his remarks at a recent Center webinar.
Overall, nearly two-thirds of Hispanic children live in low-income families falling below 200 percent of the federal poverty level: 55 percent of those with U.S.-born parents and 71 percent of those with immigrant parents. However, looking more closely at the subgroups within these numbers, we see enormous differences across country of origin and parental education and English fluency. Collecting the right data on Hispanic children and families in the U.S. is crucial to their health and academic and economic success, as these nuances have implications for how we serve their educational and health needs.
A brief from the National Research Center on Hispanic Children & Families (Center), Improving Data Infrastructure to Recognize Hispanic Diversity in the United States, looked at 34 large-scale data sets commonly used to examine topics critical to the well-being of children and their families. The data sets were analyzed to identify which include 10 data elements recommended by the Administration for Children and Families’ Hispanic Research Work Group—including Hispanic ancestry/heritage subgroup, country of birth, English language proficiency, among others. A recent webinar discussed the implications of including these measures for our understanding of Latinos in the U.S.
To understand and address the needs of Hispanic children and families in the U.S., major national surveys should include as many of the 10 data elements as possible. Although 33 of the 34 surveys examined in the brief include some of these questions, none collects all 10. Let’s look deeper at the surveys’ inclusion (or not) of what’s recommended.
Ten of the 12 education surveys examined in the brief ask about Hispanic ancestry or heritage, but the National Household Education Survey and the brand new National Survey of Early Education and Care do not collect even this most basic information. Most, but not all of the surveys, ask about country of birth and parents’ country of birth, which makes it possible to determine whether the respondent is first, second, or third and later generation from a specific country. However, most do not ask about U.S. citizenship, and none ask questions that distinguish undocumented immigrants from documented immigrants.
In 1982, the U.S. Supreme Court decision Plyler v. Doe, determined that all children in the U.S. have the right to an education, including those who are undocumented. Despite this decision, and the fact that most of the 5.5 million children with undocumented parents are U.S. citizens (having been born in the U.S.), these children often have poorly-developed cognitive skills as early as age two that impedes their ability to take full advantage of their right to education. These cognitive challenges, frequently due to factors such as economic hardship and avoidance of center-based care, can negatively influence their future school performance and, eventually, their job prospects.
Yet, none of the major education surveys provide information to estimate the size, characteristics, or geographic dispersal of the children of undocumented parents. Without this, many are destined to fall behind in school and in life, because experts do not have the data to identify their needs or to provide them with the services they need to thrive.
Perhaps even more surprising is that four of these 12 education surveys collect only partial information on English-speaking proficiency or the ability to read and write in any language. Furthermore, four of the education surveys collect no information on reading or writing literacy. Without the most basic information about literacy, public and private policies that rely on these data sets are flying blind.
As was true for education, most of the health surveys reviewed in the brief collect data on Hispanic ancestry or heritage, country of birth, and parents’ country of birth. Noteworthy exceptions are the Behavioral Risk Factor Surveillance System, National Immunization Survey, National Survey of Children’s Health, and Youth Risk Behavior Survey.
These four surveys also collect little or no information on country of birth, generational status, home language, English-speaking proficiency, or literacy in any language. Surely policy makers and program planners would benefit from the inclusion of this basic information in key national health surveys. Finally, none of the health surveys reviewed in the brief, except the California Health Interview Survey, asks questions needed to distinguish between undocumented and documented immigrants. This is particularly troubling because the undocumented population often lives in the shadows of the healthcare system.
Family Economic and Well-being Surveys
Of the 14 family economic and well-being surveys reviewed, most obtain partial or full information on Hispanic ancestry or heritage, country of birth, parents’ country of birth, and U.S. citizenship.
However, yet again, English-speaking proficiency, and especially literacy in any language, are seldom included in surveys focused on family economics or well-being. Additionally, these surveys, which may be used to develop economic development, employment, or training policies or programs, fail to ask about educational attainment outside the U.S., except in the Panel Study of Income Dynamics. This is a missed opportunity, given that monitoring the needs of Hispanics related to education and training may be especially useful in developing policies and programs.
Two surveys stand out. The National Assessment of Adult Literacy and the National Agricultural Workers Survey were the only surveys to obtain information for nine of the 10 data elements recommended in the brief. These surveys merit special recognition for going the extra mile because they collect nearly all the data elements needed to unpack the diversity of Hispanic populations. With small additions to data collection, they would have perfect scores.
Room for Improvement
At the other extreme, four surveys collect only one or none of the 10 data elements. These are the Behavioral Risk Factor Surveillance System, the National Immunization Survey, the Youth Risk Behavior Survey, and the National Assessment of Educational Progress.
These include three of the nation’s national health system resources, and the national report card on the educational progress of our children. There are, no doubt, technical difficulties in collecting all 10 recommended data elements, but most data collection efforts do better than collecting none or one. I hope that the agencies responsible for these surveys will consider assessing the possible value and technical opportunities for including additional data elements needed to unpack the diversity of Hispanic populations with regard to their health and their educational accomplishments.