|
|
|
|
L406.1 Goldenberg, C. ,Gallimore, R., & Reese, L. (2001). Using Mixed Methods to Explore Latino Children's Literacy Development. Paper prepared for Conference on "Discovering Successful Pathways in Children's Development: Mixed Methods in the Study of Childhood and Family Life." MacArthur Foundation Research Network on Successful Pathways through Middle Childhood, Santa Monica, CA , January 26, 2001 
complete
version of paper in PDF format
Bronfenbrenner (1979) chided the developmental research community more than 20 years ago for neglecting the everyday contexts in which children develop and learn. As a result of this neglect, "our ability to address public policy concerns regarding contexts of child rearing is correspondingly limited" (p. 844). Much has changed since 1979, and attention to context is no longer a novelty. In addition to Bronfenbrenner's "circle of influences" as a metaphor of contextual influence, other perspectives have also flourished, for example a "woven fabric" of ecological/cultural features (Cole, 1996; Weisner, 1984). From the work of many research communities, over the past four decades much has been learned about context and development, including effects of cultural, community, neighborhood, family, and school factors. Taking account of context effects demanded re-thinking of the methodologies that dominated the psychology and related disciplines for a half century (Cronbach, 1975). Since Bronfenbrenner's admonitions, a wider range of methodological and conceptual tools have come into use, although much debate accompanied his challenge and continues to this day. Many remain uncomfortable because so much of the data generated by contextual studies does not live up to the traditional methodological requirements of their disciplines. Others argue for turning away from conventional methods toward interpretive branches of the social sciences and to humanities for methodological foundations (Cole, 1996, p. 4). As epistemological debates continued, and dualistic rhetoric escalated, some concluded that the methodology wars were not likely to be ended in our lifetimes, and a purist approach would not get much research done (Miles & Huberman, 1985; Greene & Caracelli, 1997). A more ecumenical or "multiplist" approach suggested mixing methods from intense local observation to random-assignment experiments (e.g. Campbell, 1974: Cook & Reichardt, 1979; Cronbach, 1975; Houts, Cook, & Shadish, 1986; Sechrest, Babcock, & Smith, 1993; Webb, et al, 1966). The multiplist approach rejects single methods in favor of juxtaposing multiple probes using heterogeneous methods to seek stable and convergent results and interpretations across contexts, times, populations, datasets, analytic strategies, and perspectives. It assumes all research is affected to varying degrees by values and preferences, and that "individual passion and intellectual commitments provide the life force of science," which are best minimized by "trying to represent multiple preferences and values in a research program" (Houts, et al., 1986, p. 62-63). Included in those individual preferences and values are methods choices. Multiple, competing approaches provide one way to estimate the degree convergence of findings and interpretations, as well as force out conflicting assumptions. Multiplist approaches are claimed to be especially helpful for problems where little is yet known or understood (Cook & Reichardt, 1979): They can reveal unsuspected relationships; suggest unanticipated variables and effects; provide a basis for more cost-effective prelude to more ambitious and expensive undertakings; and ground "defensible interpretations of what may be true about the world" (Houts, et al, 1986, p. 61). Our research team has been influenced by Campbell, Sechrest, Cook, and others, and we have been purposefully ecumenical, mixing methods to study an under-studied population: Spanish-speaking children of immigrant Latino parents. We chose this approach also because the existing research literature seemed markedly inconsistent with our own professional and personal experiences in Latino communities. These inconsistencies were especially marked regarding Latino children's literacy development and academic attainment. Some inconsistencies directly related to questions of how to help Latino children succeed in American schools in general and how to assist their literacy development in particular. The importance of the questions we confronted could hardly matter more to the children and their parents and to the school personnel who try to serve them. Latino children comprise a large and growing portion of the U.S. school-age population. Despite progress and the narrowing of gaps between majority- and minority-group students, Latinos continue to be at risk for poor school attainment (Goldenberg, 2001). In this paper we will focus on the family part of the equation and review some of what we have learned about factors that influence Latino children's educational outcomes, most especially in literacy. Over the past 15 years we have tried many combinations of research methods to uncover family and school factors influencing the academic development of Latino children of immigrant parents. What literature there was on this population often asserted that Latino parents de-value formal education either because of economic circumstances or because experiences with discrimination against Latinos have led them to conclude education will not help their children get ahead (e.g., Ogbu & Matuti-Bianchi, 1986; Suárez-Orozco & Suárez-Orozco, 1996). Another perspective, no longer so common in the research literature (although still heard informally), attributed Latino children's school failure to traditional values--family ties, honor, masculinity, and immediate gratification (Heller, 1966, pp. 33-34). These perspectives, different as they are, have at least one thing in common: They attribute the difficulties Latino youngsters have in U.S. schools to discrepancies, or discontinuities, between family values and beliefs about schooling and the values and beliefs assumed to be important for school success in this country. However, based on personal and professional experience, we were skeptical about the validity and comprehensiveness of these characterizations of Latino students and family. For example, we had observed that despite differences in cultures and outlook between Latino immigrant parents and educators in the schools their children attended, there were also considerable commonalities in their values and beliefs. Both parents and teachers want children to do well and succeed in school; both parents and teachers see formal schooling as important for economic and social mobility. Moreover, and despite attempts to maintain links with their native cultures, some immigrant parents made self-conscious attempts to move away from the educational values espoused by their parents and provide greater educational opportunities for their children than they felt they themselves had. At the same time, most of the children did not come from homes that afforded a wealth of literacy experiences. Although literacy (and other academic learning) opportunities were not non-existent, neither were they as plentiful as they tend to be in middle- or upper-middle class homes. The reality lay somewhere in between. In short, a complex portrait of commonalities and differences, continuities and discontinuities, seemed more plausible than the widely held stereotypes about this population (Goldenberg & Gallimore, 1995). Using several samples and mixed methods, our findings eventually converged on several broad ecocultural categories of contextual influence on children's literacy experiences and literacy development. These influences included family history and community demography, job-related constraints and enablers, domestic routines and roles, institutional influences, natal cultural schemas, and exposure to alternative cultural schemas. In the following, we summarize what we learned in our longitudinal studies about literacy learning opportunities in low income Spanish-speaking households, including some of the ecocultural factors that either constrain or enable those opportunities. We provide illustrations of how we have used quantitative and qualitative methods reflexively and interactively to pursue questions about home influences on Latino children's literacy development. Finally, we give a brief indication of how we are currently attempting to gain comparable understanding of home influences on these children's mathematics attainment.
|
