This study examines the international flow of ideas and goods in eighteenth and nineteenth century New England port towns through the experience of a Boston-based commercial network. It traces the evolution of the commercial network established by the intertwined Perkins, Forbes, and Sturgis families of Boston from its foundations in the Atlantic fur trade in the 1740s to the crises of succession in the early 1840s. The allied Perkins firms and families established one of the most successful American trading networks of the late eighteenth and early nineteenth centuries and as such it provides fertile ground for investigating mercantile strategies in early America. An analysis of the Perkins family's commercial network yields three core insights. First, the Perkinses illuminate the ways in which American mercantile strategies shaped global capitalism. The strategies and practices of American merchants and mariners contributed to a growing international critique of mercantilist principles and chartered trading monopolies. While the Perkinses did not consider themselves "free traders," British observers did. Their penchant for smuggling and seeking out niches of trade created by competing mercantilist trading companies meant that to critics of British mercantilist policies, American merchants had an unfair advantage that only the liberalization of trade policy could rectify. Following the Perkinses allows for a reconsideration of the Anglo-American relationship in the East Indies, especially China. For example, the special relationships the Perkinses established with the Wu family of Canton as well as the London-based Baring Brothers & Co. proved critical to their success in business. Yet these relationships developed out of the Perkinses' geopolitical position as Americans. Further, the project shows that family life, gendered ideals, and particular visions of the life cycle were central to how Americans came to terms with expanding trade and evolving markets. In the late eighteenth century, Americans began to exalt family as a sentimental unit whose central aims were personal fulfillment and the raising of future citizens. But this new ideology of family masked the institution's continued political and economic utility. Family has never been the promised "haven from the heartless world" of market perils; in fact, well into the nineteenth century it was the opposite: family was a core market institution used for protection from risk and speculation. Even as the Perkinses embraced the speculative potential of commerce and investment, familial and gendered ideals shaped how they understood profit, risk, and even what it meant to be a merchant. Finally, in recent years, scholars have integrated New England into the Atlantic World; I demonstrate the importance of New Englanders in shaping American involvement in Asia and the Pacific as well. The Pacific continues to be a central space of American empire and influence, from former colonies to trust territories. Its history merits a more robust place in American historical consciousness.

Employment trade-offs are defined as the sacrifices that employees make in their job because of their family/dependent care responsibilities (Mennino & Brayfield, 2002). They represent an employee's decision to restrict their work responsibilities and devote their time and attention to their family when time and attention cannot be given to both their work and family responsibilities. Research on employment trade-offs has been primarily theoretical and qualitative, with the few empirical studies primarily examining demographic and attitudinal correlates to an employee's decision to participate in employment trade-offs without considering the mediating mechanisms between these variables and employment trade-offs. This dissertation extended the literature on employment trade-offs by examining family-work conflict as a mediator of the relationship between family and work domain variables and an employee's decision to engage in employment trade-offs among Federal government employees. This study also investigated the relationship of participating in employment trade-offs to workplace withdrawal behaviors, family-friendly benefit utilization, and turnover intentions. The family domain variables include dependent care responsibilities and childcare characteristics, while the work domain variables include organizational supports. The 2006 Federal Employee Dependent Care Survey (U. S. Office of Personnel Management, 2006) was analyzed and hypotheses were tested using both multiple regression and logistic regression analyses. The results showed that family-work conflict partially mediated the relationship between the variables of type of care responsibilities, perceived job schedule flexibility, childcare arrangement satisfaction, childcare quality, and childcare costs and employment trade-offs. Multigenerational caregiving responsibilities, number of dependents, and supervisor support were not related to family-work conflict and no mediation effect for family-work conflict was established between these variables and employment trade-offs. Analyses also found that family-work conflict was positively related to employment trade-offs, workplace withdrawal behaviors, and family-friendly benefit utilization. In addition, employment trade-offs were positively related to workplace withdrawal behaviors and family-friendly benefit utilization, suggesting that individuals do implement a variety of family adaptive strategies to manage competing work and family demands. Finally, employment trade-offs were found to be positively related to turnover intentions. Contributions to the work-family literature, implications for practice, future research directions, and limitations of the study are discussed.

The skills acquired in the first few years of life are critical in preparing children for school and for long term development. Families play a primary role in the development of cognitive and social skills as well as physical health. Changes in family structure that have occurred over the last several decades have resulted in fewer children growing up in a two parent married household; however, few children are raised by just one parent. Many children spend time in a three-generation family household, in which a grandparent, parent and child coreside. To date, little research has described the prevalence or correlates of three-generation family households or looked at the association between three-generation family coresidence and child wellbeing during early childhood. To fill this gap in the literature this dissertation was structured around three empirical chapters (papers) and the findings from those studies are described below. Using data from the Fragile Families and Child Wellbeing Study (N = 4,898), Chapter 2 investigates how the share, correlates, transition patterns, and duration of three-generation households vary by mother's relationship status at birth. Nine percent of married mothers, 17% of cohabiting mothers, and 45% of single mothers live in a three-generation family household at the time of the child's birth. Incidence over time is much higher and most common among single-mother households: Sixty percent live in a three-generation family household at least 1 wave. Economic need, culture, and generational needs are associated with living in a three-generation household; correlates vary by mother's relationship status. Three-generation family households are short lived, and transitions are frequent. Kin support through coresidence is an important source of support for families with young children and in particular families in which the parents are unwed at the time of their child's birth. Chapter 3 investigates to what extent stable and unstable three-generation family households (grandparent, parent, child) are associated with child health, socioemotional and academic wellbeing over the first three years of a child's life. Using longitudinal data from the Fragile Families and Child Wellbeing Study (N=4,009) differences in the association by mother's relationship status and race/ethnicity are investigated. Results suggest stable three-generation family households are associated with child wellbeing whereas unstable or transitory three-generation households are not. Living in a stable three-generation family household is protective against child behavior problems for married families but detrimental for single or Black mothers. Stable three-generation coresidence is associated with higher PPVT scores but also higher odds of being overweight for some groups. Using nationally representative data from the Early Childhood Longitudinal Study - Birth Cohort (N~10,700), Chapter 4 investigates the associations between stable and unstable (or transitory) three-generation coresidence over the first five years of life and school readiness, and how those associations vary by race/ethnicity. With a few exceptions, the findings suggest that three-generation family coresidence is not associated with cognitive development, psychomotor development, or physical health. However, coresidence with a grandparent is associated with a higher likelihood of obesity across all race/ethnicities, as well as more externalizing behavior for Whites and less externalizing behavior for Hispanics. Although differences between stable and unstable coresidence are mostly insignificant, stability appears to matter for behavior, but in different ways for Black and Asian children. Black children who unstably coreside and Asian children who stably coreside with a grandparent experience more internalizing and less prosocial and positive learning behaviors.

This dissertation attempts to learn the amount of satisfaction and comfort single custodial fathers feel in different parenting areas and which variables are most highly associated with the father's satisfaction and comfort. A questionnaire was placed in the May-June, 1982 issue of The Single Parent, the membership magazine for Parents Without Partners. This approach yielded a non-representative sample of 1136 fathers with children 18 years old or younger, approximately 15 to 20 percent of the population that was believed to have qualified for the study. Five different parenting areas were examined. The father must (1) maintain the house and provide child care; (2) feel satisfaction with how the children living with him are progressing; (3) work and balance the demands of child rearing; (4) adjust to being single again; and (5) establish or resolve his relationship with his ex-wife vis-a-vis her involvement with the children. It was hypothesized that the following variables would be associated with either satisfaction or comfort in the five areas: the age, sex, and number of children being raised; the number of years the father had sole custody; the father's income; whether the father sought custody; and the level of involvement of the ex-wife. By using multiple regression, it was found that only one of the seven variables were useful predictors of the father's satisfaction and comfort in the areas. Fathers were more likely to be experiencing satisfaction and comfort if they earned a higher income or if the income earned was felt to be adequate to meet their needs. This variable was also highly associated with the ex-wife's involvement with the children. The father's satisfaction with the child care arrangements made emerged as a significant predictor of their comfort. It was found that fathers feel satisfaction and comfort with running the household and with their children's progress. The areas involving working while raising the children and adjusting to being single proved more problematical. Fathers were mixed as to their satisfaction with their ex-wives' involvement with the children, though fathers who described their ex-wives as being very involved experienced greater satisfaction in some of the areas explored.

The purpose of this field study was to determine the association between clients' perceptions of the level of four relationship factors and scores on 10 outcome criteria measures. The outcome criteria were the adjusted difference in seven before-after marital and psychological behavior measures, a change index, clients' assessment of improvement in family relationships and four outcome groups related to aspects of continuance and termination. The four relationship factors measured between pretest and post-test were clients' subjective experience of the level of accurate empathy, congruence or genuineness, regard, and unconditional positive regard communicated by their social workers. A fifth relationship variable was the sum score of the four individual factors. This highly intercorrelated system of five variables was also compared to twenty-two other independent variables for their relative ability to account for as great or greater a proportion of the variation in scores on the ten outcome criteria. These twenty-two other variables were classified into the domains of aspects of the client, aspects of the worker, aspects of the treatment system, and post-test case status. A further attempt was made to find combinations from all twenty-seven independent variables together that consistently predicted the greatest proportion of variation in outcome scores. Prior research in other than primary social work settings and theoretical literature supported the expectation that a positive association would be found between the relationship factors and outcome in this sample of clients receiving social case-work treatment. The total sample was composed of eighty-three clients reporting family related problems who were seen by twenty-three social workers with MSW degrees in three family service agencies. Clients were not randomly assigned but rather entered the study on a first come first served basis within the data collection time limit. Measurement instruments, all uniformly self administered, were the Locke-Wallace Marital Adjustment Test, the Semantic Differential, a measure designed by the FSAA, several questionnaires, and the Barrett-Lennard Relationship Inventory. The statistical methodology was the use of multiple regression techniques including step-wise analyses. It was found that clients' subjective experience of the level of relationship factors predicted the scores on all of the ten outcome measures. Levels of significance ranged from .05 to .001. Relationship factors predicted as great or greater a proportion of outcome variance as any other single variable 9 out of 10 times, and in 8 out of the 10 outcome measures as any other set of independent variables that were grouped together. However, various combinations of all twenty-seven variables together were always able to predict a greater proportion of outcome variance than the five relationship factors by themselves. These factors did not account for even half of the total variation in outcome scores despite the fact that they were the most consistent significant outcome predictors. Therefore, the experienced level of the relationship factors might be necessary but not sufficient conditions to account for all change. It was concluded that further research is needed to see if this positive association applies to a wide cross section of social casework samples employing different interventive procedures and different outcome criteria. If the association is universal in social casework practice, it is relevant to discriminate specific worker behaviors, as well as interaction processes, that in general facilitate clients' experience of higher levels of the relationship factors. Also the relationship factors need to be accounted for as intervening variables in future studies of the relative effectiveness of different models of social casework intervention.

This research was undertaken to analyze and to describe the social support networks of a non-random sample of 55 legally blind and visually impaired young adults, 20 to 50 years, inclusive. Modified versions of the Arizona Social Support Inventory Scale (Barrera, 1981, 1983), and the Network Analysis Profile (Cohen and Sokolovsky, 1978) were used to examine key aspects of network structure and to evaluate the attributes of network links. Results from this study indicate that with the exception of network size, the level of visual impairment may have less impact on network structure than such factors as age of onset of blindness, type of school attended, acceptance of blindness, marital status, gender, and mastery. Study findings also contribute some support for optimism with regard to the level of social integration achieved by study subjects. The majority of men and women in the sample showed evidence of access to all essential varieties of social support including: companionship, advice, material assistance, physical assistance, affirmation, and emotional support. A relatively small percentage of the total sample lacked access to all six of the above listed dimensions of social support. Only two of fifty-five subjects had networks that contained fewer than five persons. The average network contained ten persons. Subjects with the smallest networks were prone to be less educated, unmarried/formerly married, and unemployed. Stepwise multiple regression procedures identified employment status, mastery, level of functional vision, and gender as significant predictors of expanded networks. Young adult subjects clearly considered kin as their first line of social support. Kin supporters outnumbered nonkin supporters by close to two to one, however, nonkin proved to contribute a larger proportion of total support than did kin. Degree of visual impairment did not influence the observed pattern of support provision, nor did age. Subjects also demonstrated heavy reliance upon friends, spouses and siblings in the form of a high percentage of multi-dimensional ties. Subjects identified their immediate social surroundings as the context most frequently associated with the origination of new friendships. However, organized programs sponsored and operated by agencies that serve blind and visually impaired persons were frequently associated with the origination of new nonkin ties. Significant relationships also linked residential school attendance to a number of psychosocial measures that indicate successful adjustment to blindness.

In recent years the child welfare system in New York City has come under criticism for placing most of its resources in providing services to the child away from home. Due to the reimbursement agreement between the public sector responsible for these children and the voluntary agencies which provide care, few "in home" or preventive services have been available. In response to this criticism the Preventive Service Demonstration Project was established, in which intensive family services were provided to families of children in jeopardy of being placed. Evaluation of this project showed that such services did, in fact, reduce the number of children entering foster care and the time spent in care of those who entered. This research investigates the impact of preventive service units on the workers' attitudes and perceptions of the need for placement. Five groups of child welfare workers were participants in the study. Two of the groups were primarily concerned with providing preventive services (n=55), while three provided traditional under care services (n=109). The subjects in the research were administered an instrument which collected social/demographic information and measures of six attitudes. In addition, the subjects were presented five case analogues and asked to judge six case elements and whether the child should be placed in an appropriate foster care setting. Results of the analysis showed that workers in preventive units were different in their attitudes than workers in traditional settings -- they were more likely to feel preventive services were useful, to see the continuing importance of biological parents and to feel that foster care was a damaging experience for children. In addition. while all workers saw the elements of the five cases in about the same way, workers in preventive units placed fewer of the five children in the case analogues (a Guttman scale of Placement Proneness) than other workers. The greatest variation in the placement decision occurred in the "mid-range" case, confirming the results in a number of other studies. Several of the social/demographic variables were also related to the workers' attitudes. and these variables as well as the workers' attitudes were related to the judgment of case elements and the decision to place a child. In order to determine the importance of the variables in explaining a workers' placement proneness score a number of regression analyses were performed. The worker's setting was shown to be a strong predictor of the placement proneness score. In addition, the worker's attitude toward preventive services, judgments of a number of case elements, attendance at courses, ethnicity and the client group with which he/she had contact were also found to be predictive of this score. A total of 34% of the variance in the placement proneness score and 48% of the variance in the placement decision on the mid-range case was explained by these variables. The research gave rise to the following recommendations: (1) the creation of educational preventive units which, at least initially, are administratively separate from the under care units of the agency and the establishment of new funding patterns in the foster care system in order to facilitate their creation; (2) the redefinition of jobs within the foster care system so as to encourage contact between all workers and all parties in· the foster care triangle; (3) the education and training of workers in the area of preventive services in order to increase the workers perceptions of their effectiveness; and (4) an increased emphasis in the training of workers on the skills needed to discern strengths in clients.

This study was a comparative, cross-sectional analysis of maternal reactions to separation from their young children at school entrance. The first observation was of mothers who were entering their children in school for the first time. The second observation, made concurrently, was of mothers who entered their children in school at least six months before the observation. These were the "experienced" mothers. The sample population consisted of 177 mothers drawn from nine schools. Data was collected by means of a written questionnaire, distributed to the mothers at the school. The study sought to discover factors which might influence the ease or difficulty which a mother experienced during separation. Six major and seven minor hypotheses were tested. The first hypothesis stated that reactions to separation change over time and will be most intense at the initial separation. It was confirmed. The separation reactions of the "inexperienced" mothers were significantly different from those of the "experienced" mothers, and in the predicted direction. The second hypothesis stated that separation reactions change according to the mothers' view of her relationship with her child and her perception of the degree of independence between her child and herself. It was hypothesized that a mother who was able to recognize her child as a separate individual, with needs different from her own, would have less difficulty at parting than a mother who saw her child as an extension of herself. In the operationalization of the hypothesis, four different content areas were created. In the findings, two of these areas proved to have significant relationships to the mothers' difficulty separating: her early mothering experience and her assessment of the child's current level of independence. The mothers who indicated great satisfaction with the early mothering experience were the ones who had the most difficulty separating. Those who only moderately enjoyed the early mothering experience had a much easier time separating. Second, mothers who saw their children as able to function fairly independently were able to undergo the separation with much less anxiety themselves. The third hypothesis examined the effect of a mother's social network on her difficulty separating. It was predicted that the more isolated a mother was, the more difficulty she would have separating at school entrance. The findings confirmed the hypothesis for the "inexperienced" mothers. The ones receiving the lowest amount of support from their husbands, parents, and relatives were very clearly the ones who had more difficulty separating. The fourth hypothesis stated that mothers who had difficulty separating were less likely to find new activities and also less likely to have an improvement in their relationship with their children. The findings were not significant. The fifth hypothesis stated that differences in the mothers' responses to separation related to their views of the school. Did they see the school as helpful and supportive, indifferent, or even intrusive? It was hypothesized that a mother who feels that the school cares about her child and herself will have an easier time separating. The findings were that there was no relationship between a mother's difficulty separating and her view ·of the school. Hypothesis 6 stated that separation reactions related to the amount of separation experienced prior to school entrance. It was predicted that a mother who has frequently been apart from her child will experience less separation anxiety. Conversely, a mother who has never left her child will experience greater separation anxiety. The hypothesis was confirmed. The mothers who spent more time away from their children before school entrance had an easier time separating from them at school entrance. Hypotheses 7 through 13 explored the relationship of the main dependent variable, "difficulty separating" and seven antecedent variables: "general morale", "age", "social class", "ethnic group", "religion", "length of employment", and "number of children". None were found to relate significantly to "difficulty separating".

Previous research has shown that adult children are the predominant service and health care providers to the impaired elderly. However, relatively little is known about the conditions under which caregiving is either enhanced or hindered. Therefore, the purpose of this study was to systematically examine the caregiving relationship in order to better understand its causes and consequences. Data were collected via in-depth structured interviews with a sample of adult children (n = 131) identified as the primary caregiving relative to an older parent currently receiving home care or day care services. Bivariate and multivariate analytic techniques were utilized to identify the most salient variables associated with the two dependent variables of interest: caregiving involvement (the task and time commitment) and caregiving consequences (the perceived impact of providing care). Study findings indicate that the typical caregiving child is a late middle-aged daughter who holds concurrent responsibilities to other family members as well as working outside the home. Emotional support was the most universal caregiving activity although substantial proportions also assisted with linkage tasks, instrumental services as well as financial assistance. The primary strains of caregiving were found to be the result of the emotional aspects of providing care and the restrictions on time and freedom necessitated by caregiving responsibilities. The most salient independent predictors of caregiving involvement were: the parent's level of impairment; the quality of the parent-child affective relationship; the child's sex and marital status; and the degree of anticipatory planning for caregiving. Contrary to expectations, the child's employment status did not impinge upon fulfilling caregiving responsibilities. The significant variables predicting perceived negative consequences included: the extent of caregiving involvement; the parent's level of impairment; the perception of unmet service needs; the quality of the parent-child affective relationship; and the child's social class, sex, and health status. The extent of formal service utilization did not emerge as a significant predictor of caregiving consequences although the qualitative data gave support to the hypothesis that service input reduced caregiving strains. Implications for policy, practice, and service delivery in support of families caring for frail older relatives are discussed.

Little research has been undertaken to show how women in South Africa are integrating their work and family lives. The purpose of this study was to generate data on one area which could facilitate such integration, namely, maternity benefits. Although employers in South Africa are prohibited from employing pregnant women for one month prior to and two months after confinement, they are not compelled by law to transfer women to less strenuous work during pregnancy or to reinstate these women after the enforced period of leave. Whilst the state has not legislated such transfers or reinstatement, they have urged employers to be "sympathetic" to requests for transfers and reinstatement. The study was designed to provide an exploratory and descriptive perspective on maternity-related rights and benefits in the textile industrial sector in selected areas of Natal. Such research data is essential for the purposes of policy advocacy and policy development. Data for this study were collected through the use of the social survey method. The main part of the study included structured interviews with all textile employers in the geographical areas of the study who employed more than ten women. The subsidiary part of the study included structured interviews with selected female employees who had become pregnant whilst working in the textile industry in the area of the study. The purposive sampling method was used. Forty-six textile employers were interviewed and 301 employee interviews were conducted at 31 of these establishments. It emerged from the study that the governmental policy of merely requesting employers to provide benefits has not been effective. In nearly all cases where a maternity benefits package was provided this was as a direct result of pressures to which employers found themselves subject. Unions were the major factor in supplying such pressure. However, even where maternity benefit packages did eventuate, there were two important factors that inhibited their availability. Firstly, women lacked sufficient education about the benefits and the ability to assert their rights sufficiently. Secondly, even where these difficulties could be overcome, no adequate legal machinery exists for the enforcement of the women's maternity-related rights.