Immigrants Raising Citizens: Undocumented Parents and Their Children

Appendix: Overview of Study Design and Methods

Appendix Overview of Study Design and Methods All the data reported in this book are drawn from the early childhood studies of the Center for Research on Culture, Development, and Education. Data from two phases of the study were included: a first pilot qualitative study and the main birth cohort study. The center’s mission is to examine how the contexts of home, peers, child care and school, parental work, and public policy affect the development of young children in the first years of life and of youth during adolescence. As a researcher in public policy, parental work, and child development, I was interested primarily in these topics as one of the center’s principal investigators. The other principal investigators were Catherine Tamis-LeMonda, Diane Hughes, and Niobe Way, all professors of applied psychology at New York University. Co-investigators who directed the birth cohort and embedded qualitative study, which together form one of the center’s two cohort studies, were Ronit Kahana-Kalman, Ajay Chaudry, and Diane Ruble. As the data collection progressed in an initial qualitative study and in the subsequent first waves of a larger birth cohort study, my initial interest in economic hardship and survival strategies among the immigrant families in our sample gradually was overtaken by the realization that documentation status appeared to be a significant and substantial theme in the qualitative analysis. Two empirical puzzles led to the emergence of this theme. First, the Mexican mothers in the birth cohort, despite reporting lower incomes than the Dominicans and African Americans, also reported lower economic hardship (as measured by events such as not being able to make the rent or pay other bills). Something else must have been going on in these families, and as we collected the data on their immigration experiences, the very high rates of undocumented status among the Mexican population struck me and the qualitative team as important. The second puzzle occurred early in the birth cohort study, when the Chinese mothers told us that they were not planning on breast-feeding their infants. This was a puzzle given that knowledge of the benefits of breastfeeding is widespread. In the ensuing months, as we followed up our sample with six-month phone calls and interviews, we realized, to our appendix 152 surprise, that most of the Chinese infants had been sent back to China. The story of Ling from the pilot qualitative study, it turned out, was a much more common story than we had expected. Pilot Qualitative Study Methods As principal investigators of the Center for Research on Culture, Development , and Education, Catherine Tamis-LeMonda, Diane Hughes, Niobe Way, and I did not feel comfortable immediately engaging in hypothesisconfirming social science without initial exploratory work. Ajay Chaudry (at that time a professor of public policy at the New School) and I therefore led an initial, intensive qualitative study. In addition, two professors of applied psychology at New York University, Gigliana Melzi and Niobe Way, helped supervise subgroups of field-workers. Twenty-four mothers of Chinese , African American, Puerto Rican, Dominican, Mexican, and European American origin were recruited into this qualitative study with the assistance of New York City community agencies that served low-income families . These groups were chosen because together they represented over 80 percent of the New York City population and encompassed some of the largest immigrant groups in the city. We recruited a larger range of ethnic groups than is presented in this book because this pilot qualitative study informed not only the birth cohort study that is the focus of this book but also the adolescent study that included a much wider range of ethnic groups. In addition to ethnicity, we used the following eligibility criteria: resident of New York City; recruited from an agency that serves neighborhoods with high concentrations of the ethnic and immigrant groups included in our study; age eighteen or older; and household income below 250 percent of the federal poverty threshold for the prior year. Children in the families were either between three months and three years old (an early childhood subsample) or between ten and thirteen years of age (an early adolescent subsample). From the subsample of families with young children, four Mexican families, two Dominican families, and one Chinese family were drawn for analyses in the book. (For the names of the mothers and focal children used in this book, which are all pseudonyms, see in the notes to table A.1.) A combination of semistructured interview and participant...