Methodological considerations in using complex survey data: an applied example with the Head Start Family and Child Experiences Survey

Complex survey data are collected by means other than simple random samples. This creates two analytical issues: nonindependence and unequal selection probability. Failing to address these issues results in underestimated standard errors and biased parameter estimates. Using data from the nationally representative Head Start Family and Child Experiences Survey (FACES; 1997 and 2000 cohorts), three diverse multilevel models are presented that illustrate differences in results depending on addressing or ignoring the complex sampling issues. Limitations of using complex survey data are reported, along with recommendations for reporting complex sample results.     

Complex sample data recommendations and troubleshooting

Complex survey data, as highlighted in this issue of Evaluation Review, provide a wealth of opportunities for answering methodological and/or applied research questions. However, the analytic issues of nonindependence and unequal selection probability must be addressed when analyzing this type of data. Thus, to ensure that research questions are accurately answered when using complex survey data, researchers must take extra precautions to ensure complex survey data are correctly analyzed. The purpose of this article is to provide software recommendations for analysis of and tips on troubleshooting when analyzing complex sample data.     

Playing the scales: Regional transformations and the differentiation of rural space in the Chilean wine industry

Globalization and industrial restructuring transform rural places in complex and often contradictory ways. These involve both quantitative changes, increasing the size and scope of operation to achieve economies of scale, and qualitative shifts, sometimes leading to a shift up the quality/price scale, towards finer spatial resolution and identification with place of origin. This paper examines the transformation of the Chilean wine industry noting its expansion and orientation towards export production. As the industry has changed, it has become apparent that firms have adopted different scalar strategies, sometimes downscaling by seeking low-cost production, homogenisation of product and a weak identification with place, and sometimes upscaling by improving quality, claiming exclusiveness and fixing products to ever finer definitions of place. Places have been defined, reconstructed, promoted and significantly differentiated as a result.     

Noisy zones of proximal development: Conversation in noisy classrooms

Despite the importance of context in studies of language use, sociolinguists have ignored the impact of noise on conversational interaction. This inattention is of particular concern in classrooms where language is a learning tool. Our research on interaction in noisy settings took place in English language elementary school classrooms with students in grades 3, 5, and 7, whose first language was English. Students were observed during regular classroom activities. Employing a novel method, in which students wore ear‐level microphones, we obtained stereophonic recordings of the noise and conversation that reached each listener's ears. A dosimeter measured the noise levels in each classroom. Analyses of students’ patterns of conversation suggest that noise levels impeded the intended development of complex conversational interaction and collaborative learning. This study also questions the place of acoustics in understanding context, and the significance of the hearer's perspective in sociolinguistic studies of conversational interactions.     

Fertility and development in Brazil

In the past, one of the concomitants of development has been a sustained reduction in fertility. As a result of this experience, demographers hypothesize that in a society in which fertility is lower in urban areas, among the upper socioeconomic status groups and the better-educated, fertility will decline to a moderate level as the country changes from a rural, agricultural socioeconomic structure, with low levels of living and education, to an urban, industrial structure, with rising levels of living and education.The data analyzed in this study indicate, however, that though substantial social and economic development (as measured by changes in industrial structure, per capita income, urbanization, and education) occurred in Brazil from at least 1920-40 to 1960, during which time fertility differentials of the kind indicated above existed, fertility has shown little or no tendency to decline. Between 1940 and 1960, in fact, the birth rate appears to have remained fairly constant around 43. With the death rate steadily dropping, the rate of natural increase and population growth (given a small net in-migration) has been accelerating. p ]From a theoretical point of view, these facts reinforce a growing realization, based on similar findings in some other developing countries, that the prevailing theoretical ideas concerning the relationship between development and fertility require modification, particularly in the direction of greater specificity. On the practical side, the question is raised whether Brazil's rate of economic development during the postwar period up to 1960 can be maintained, let alone increased, in the face of a population growth rate which will probably average 3.2-3.5 percent for the period 1960-70 and which, in the absence of a decline in fertility, is likely to accelerate further.     

The relationship between maternal depression and child outcomes in a child welfare sample: implications for treatment and policy

Maternal depression is an extremely important parenting variable in relation to child general health. This investigation addressed the issue of maternal depression as it relates to the lives of children seen by child welfare authorities. Maternal depression was investigated in the context of the increasing rate of children coming to the attention of the London and Middlesex Children's Aid Society. A variety of child outcomes including those of particular relevance to child welfare, specifically disorders of attachment and neglect and physical abuse, school‐related variables, conduct disorder and psychological distress, were investigated. The relative contribution of maternal depression along with measures of socioeconomic status and social isolation to childhood risk was examined. Results indicated that the rate of maternal depression doubled during the years 1995–2001. Maternal depression was related to children entering care through wardship. Additional poor child outcomes such as attention deficit disorder, conduct disorder and emotional adjustment were also related to maternal depression. The findings are discussed in terms of treatment and policies that can lower the risk to children through effective intervention with depressed mothers.     

Language Policy and Provision in Social Service Organizations

This paper reports the findings of a survey into minority languagepolicy and provision in social service departments and largevoluntary social service agencies in England and Wales. It foundthat many organizations lacked specific policies on the provisionof services to minority language communities and that most werepoorly prepared to meet their needs.     

Volunteers, families and children in need: an evaluation of Family Friends

This paper explores the findings from a small‐scale research project that analysed the impact of Family Friends – a voluntary agency that provides support to families under stress who have children aged between 5 and 11 years. The study, funded by Family Friends, evaluated service users’ perceptions of the support they received, specifically in relation to the significance and impact of the volunteer befrienders. The research identified that there are a proportion of families in need who fail to attract services from mainstream statutory agencies. It also identified that the Family Friends voluntary agency makes a particular contribution to service provision by offering a friendly, non‐stigmatizing, caring and responsive service.     

THE ASSOCIATION OF POVERTY WITH CHILD WELFARE SERVICE AND CHILD AND FAMILY CLINICAL OUTCOMES

The influences of poverty, family status and violence in the lives of children and families seen at a large urban children's aid society are investigated. Rates of social assistance dependency of child welfare families in 1995 and 2001 are compared. While the rate of poverty of child welfare families is stable across the two time periods, the ability of families to cope with poverty has dramatically changed, with the percentage of children from impoverished families being admitted into care nearly doubling across the six-year period. Violence as a mitigating factor in the relationship between poverty, the extent of child welfare service and child and family outcomes is also reported. Suggestions for enabling women who are caught in the complexity of choosing between physical safety and poverty are reported.

Les influences de la pauvreté, de la position sociale de la famille et de la violence sur la vie des enfants et des familles vues à une grande société urbaine d'aide à l'enfance sont examinées. Les taux de dépendance des familles vis-à-vis des programmes de protection de l'enfance en 1995 et 2001 sont comparés. Tandis que le taux de pauvreté des familles qui se servent du système de protection de l'enfance est stable à travers les deux périodes de temps, la capacité des familles de faire face à la pauvreté a radicalement changé, le pourcentage d'enfants de familles appauvries qui sont placés dans des familles d'accueil ayant presque doublé à travers la période de six ans. On fait aussi compte-rendu de la violence comme facteur atténuant dans le rapport entre la pauvreté, l’étendue du service de protection des enfants et les résultats pour les enfants et les familles. On résume aussi des suggestions favorisant l'habilitation des femmes qui se trouvent coincées dans la complexité de choisir entre la sûreté physique et la pauvreté.