This dissertation studies a case of Enlightenment art created in feudal conditions of servitude. The Sheremetevs, one of the richest and most powerful families in eighteenth-century Russia, had some of their hundreds of thousands of serfs trained as painters, architects, opera singers, and musicians. Two of these serfs, Ivan and Nikolai Argunov, became successful portraitists who painted a range of sitters from Empresses to fellow serfs. Tensions between social rank and individuality, already a preoccupation for eighteenth-century portrait painters, became particularly pronounced in this situation. While recent scholarship has focused on the Argunovs' cosmopolitan influences, their paintings of fellow serfs and others of low rank are sometimes visually and iconographically distinct from their usual output. This category of portrait, this dissertation argues, should be considered within the context of the other artistic projects of the Sheremetev household. Despite strong Western European influences on the Argunovs, the painters were also exposed to extremely personal and local precedents. These include earlier portraits, garden prints, an atlas project, the Sheremetevs' many collections, and operas staged by the family's renowned serf theater. Working within this visual environment, Ivan and Nikolai Argunov painted their subjects in intricately allusive ways. Their portraits represented and negotiated the complications of serfdom in a setting where unusual social change was possible.

No study, to date, has systematically examined the interplay of social contact, depression, functional disability, and cardiovascular health when examining the relation between religious activity and all -cause mortality. This study used Cox regression models as well as a series of structural equation models to elucidate these relations and resultant mortality over a 5-year period. This sample included 3,607 persons, age 65 and older, who participated in the Cache County Study on Memory in Aging, and who were not demented. Results indicate that when using Cox regression modeling, after controlling for other variables related to mortality, both religious activity and social contact remained statistically significant predictors of survival time. Based on hazard ratios obtained from the Cox regression models, it was found that subjects who attend church activities at least once a week or more are 41.6% less likely to die than subjects who attend church less frequently. Subjects who increase their social contact by each additional level gain 3% protection against mortality. Surprisingly, depression was not related to mortality in any analyses. Therefore, the best-fitting structural equation model did not include depression. Possibly, the most interesting findings from this study were the mediating effects found between functional disability, religious activity, social contact, and all-cause mortality. Using a nested series of structural equation models, we found that social contact mediates the relation between functional disability and mortality and that religious activity mediates the relation between functional disability and social contact These results indicate that social contact may be a crucial underlying mechanism, which is triggered by religious activity, and therefore acts as a mediator between functional disability and mortality. Limitations of this study include narrow or unidimensional measures, as well as problems w1th reliability. Due to the homogeneity of this sample, it may be very difficult to justify generalizing these results to a different population. Despite these limitations, this study finds that both religious activity and social contact converge in their effects on mortality and their interconnectedness is evident from these results. Both religious activity and social contact have important implications for the health of our elderly. Nevertheless, many multilayered aspects of religious behavior and social networks have not been addressed in this study. Future work investigating the consequences of the longitudinal aspects of religious belief, social networking, and depression is needed.

The American Association of Marriage and Family Therapy (AAMFT) recently sponsored Head Start-Marriage and Family Therapy (MFT) program partnerships. MFT programs can begin building similarly successful collaborative partnerships with Early Head Start (EHS) programs through using all or portions of this research study. This study has been dedicated to describing the occurrence, co-occurrence, and predictive characteristics of marital discord and depression in families served by EHS programs. This identification of at-risk families can then be used to bolster existing treatment efforts, develop new maritally based interventions, and facilitate increased referrals. Marital discord and depression are two often interrelated problems EHS mothers are at increased risk to face because they have low incomes and very young children (up to age three). Previous research has demonstrated the negative effects of marital discord and maternal depression on child, adult, and family development. Research with various married samples has further identified variables predictive of marital discord and depression. It was the aim of this study to reexamine these predictors and test couple measures to find the most effective identifying variables. Cross-sectional and prospective longitudinal research analyses were conducted from surveys with 148 EHS married mothers and their spouses to answer specific research questions. In general, results revealed that EHS married mothers were (a) slightly less depressed and maritally discordant than what might be expected of lower income parents, (b) more prone to experiencing these problems the more children they had, and (c) more accurately identified by considering couple data, which included similarity in earlier marital discord, earlier depression, religious activity, attachment attitudes or demographic variables. The limitations of this study included weaknesses in measurement and analytic procedures largely resulting from the use of data originally organized at a national level with Jess complementary purposes in mind. In the future research should address the limitations and incorporate the findings of this study into development and testing of theoretically driven marital interventions in EHS samples. Systemic implications and managing ethical concerns of using the proposed marital interventions in EHS- MFT collaborative effort are also discussed.

The purpose of this study was to examine the quality of family life presently experienced by women who were premaritally pregnant as teenagers. Self-esteem and general life satisfaction were measured in order to evaluate the results of pregnancy decisions made five to 10 years earlier. A survey design was used to (1) obtain demographic data, (2) analyze the relationships between these women and their parents, (3) determine the level and type of education completed by the women, (4) compare mothers who married before the birth of the child, single mothers, and mothers releasing their children for adoption on many variables constituting well-being. Data were gathered during personal interviews with 46 women. In this small unrepresentative sample, the results show no significant differences in the present well-being of women based on their previous decisions. Each of the test groups exhibit interesting characteristics with regard to present marital status, amount of education completed, and relationships with children. The mothers who show higher levels of life satisfaction are also those who are employed either part-time or full- time and have higher family income levels. In general, women in the study were found to be satisfied with their lives and seemed to be functioning well. Some trends were noted, indicating a need for further study, particularly in the area of mothers relinquishing children for adoption.

Data from the National Survey of Children were used to study the relationships between children's perceptions of parental support and control and measures of self-esteem and behavior problems over time. Data were collected in 1976 , when the children were aged 7-11; 1981, when the children were in their early to mid teens (age 12 to 16); and 1987, when the children were in their late teens and early 20s (age 17 to 22). Parenting measures , based on children's reports, were developed for each wave from items included in the data; constructed variables measuring self -esteem and internal and external expressions of behavior problems were also comprised of individual items drawn from each wave of data. Preliminary analyses showed that parental support was positively correlated to children's self-esteem and negatively related to behavior problems. Although the parental control measures had little effect on the outcome variables , the effect that was present showed that parental coerciveness, rejection, and permissiveness were negatively correlated with children's self-esteem, while being positively associated with both internal and external behavior problems. The preliminary results also showed that the parental effects of mothers and fathers differed for daughters and sons. LISREL analyses were done in an effort to more fully investigate the interactive effects between the constructed variables of interest. Generally, the measures were related as expected , although the relationships were not as strong as anticipated. Of the parenting measures, parental support showed the strongest effects on child outcomes; parental control measures had very little effect on children's self-esteem or behavior problems . In 1987, parenting style had virtually no effect on youths' self-esteem or on behavior problems. For this wave, self-esteem was the strongest predictor of behavior problems in young adults.

Research for this thesis investigated factors predictive of how committed married couples are to make positive efforts for their marriage. Respondents consisted of 72 husbands and wives from a semi-urban area. For each gender, correlations were conducted between marital commitment to invest and egalitarian ism, decision-making power, and conflict communication style. Finally, regress ions were conducted with these measures and nine demographic variables. As projected, commitment to invest in marriage correlated negatively with husband demand-withdrawal communication and positively with husband and wife mutually constructive communication. The stepwise regression predicting husband commitment to invest included demand-withdrawal communication and total months knowing one's spouse. For wives, the regress ion consisted of mutually constructive communication. Finally, the couple regress ion included mutually constructive communication and total months knowing one's spouse. The main implication of this thesis is that conflict communication styles may be assessed for and incorporated into marital therapy because of their possible saliency with commitment to invest.

The stress parents feel affects how they fulfill their roles as parents and their own psychological well-being. Social support has been shown to help parents deal with the demands of their parenting roles while maintaining psychological health. Compared to married parents, divorced mothers are most at risk for parenting stress and negative feelings of well-being. Low-income can add further to the levels of stress in parenting and increase the need for sources of social support. This study compared low-income divorced mothers to low-income married mothers. It explored mothers' perceptions of the stresses of parenting and feelings of wellbeing in relation to their use of social support resources. Results indicated that divorced mothers who used informal sources of social support (i.e., relatives or fiiends) were more likely to feel in control of their lives and have a more positive perception of their preschool child. On the other hand, married mothers who used formal sources of social support (i.e., agencies or professionals) felt more positive about interactions with their children and felt less distress in their parenting roles. This study also looked at social support as a moderating variable interacting with marital status to affect feelings of well-being and parenting stress. Social support as a moderating variable was not supported for this sample. It was concluded that researchers must be careful in selecting an instrument used to measure the concept of social support. To assist in clarification of the effects of social support on stress and psychological well-being. measures should encompass not only how often social support is used and the sources of support but also the types and quality of the support received.

This study primarily examined the relationship between role model choices and body image of female adolescents. More specifically this study sought to examine the relationship between the reasons that adolescent females give for choosing a role model and body image. Because body image dissatisfaction has been found to be associated with self-esteem and eating disorders such as anorexia and bulimia nervosa, these variables also were investigated. Body image was examined by using two measures of the body image construct· the Figure Rating Scale and the Multidimensional Body-Self Relations Questionnaire (MBSRQ). Both were used in order to capture a greater portion of the body image construct, as well as to examine possible differences between the measures. Eating disordered behaviors were measured using the Eating Attitudes Test (EAT-26) and self-esteem was measured with the Rosenberg Self-Esteem Scale (RSES). Participants were asked to list one role model within their family and one role model outside of their family, and to list why they selected those specific people as their role models. Role models were coded based upon the reasons provided by participants. Those who listed physical reasons such as "she has a good body," "she is beautiful ," or "she is skinny" were coded as "Group 1 - Physical," and those who chose all other reasons were labeled "Group 2- Non-physical." Females, ages 13-15 and 18-23, voluntarily participated in this study. The majority participated in exchange for extra credit in specific courses at their freshman center or university A prepared packet (including consent forms, instructions, and questionnaire) was sent home with willing participants to fill out and return. In total, 159 packets were returned. Statistical analyses indicated that body image is highly correlated with eating disordered behaviors and low self-esteem. Those scoring high on the measure of eating disordered behaviors (EAT-26) had, on average, even lower self-esteem as indicated by scores on the Rosenberg Self-Esteem Scale (RSES). However, statistically significant differences between body image and reasons for choosing a role model {physical vs. non-physical) appeared only on weight and/or shape-related aspects of the body image measures. It is not known whether differences between type of role model and body image occurred due to actual weight I Body Mass Index (BMI), or if those with greater weight and/or shape-related concerns chose more physical role models. Age differences among participants appeared on type of role model chosen. with the younger participants, ages 13-15, choosing more role models based upon physical characteristics than those ages 18-23. However, no differences were found between age of participant and level of body 1mage.

Fear of crime and the public's satisfaction with police has been a focus of criminologists for a number of years. Most studies, however, have focused on the general population as a whole. What is not known is how fearful American youth are of the crime in their neighborhoods and how they perceive those in law enforcement. The purpose of this study was to explore this subgroup of the population and to begin the investigation of youths' fear of crime and their perceptions of law enforcement. Using a national sample of 1,897 youth ages 16-25 collected from 12 cities throughout the United States, data were analyzed to explore the strength of demographic and community variables as they related to youths' fear of crime and perceptions of law enforcement. Results for fear of crime indicate that in general, American youth are not very fearful of crime in their neighborhoods. Older youth (18-25) tended to be more fearful than younger youth (16-17), females were more fearful than males, married youth were more fearful than non-married youth, youth living in the Northeast census region were the most fearful of the four census regions, and all non-White youth were more fearful of crime when compared to Whites. Youths' overall satisfaction with police indicated that in general, American youth were satisfied with police in their neighborhoods, with White youth and those youth with some college education being most satisfied with police. When addressing the variance explained by demographic and community variables, youths' perceptions regarding their quality of life accounted for the most variance in both fear of crime and satisfaction with police. The strength of this predictor variable was so strong in fact, that other demographic and community variables were negligible by comparison.

Low interpersonal skill level in younger youth has shown to lead to higher rates of delinquent behavior in older adolescents. Utah State University Extension established a program titled Youth and Families with Promise (YFP). This is a research-based mentoring program where mentors were screened, matched to individual youth, and activities planned that focus on 8 of the Search Institute's 40 assets. The program provided services for 156 at-risk youth ages 7 to 16 in six Utah counties, 66 of whom completed a pretest and posttest assessment. This research-based, intervention/prevention program has shown that the interventions used were effective in building skills in youth, assisting them in planning for the future, making better decisions, and resolving conflict nonviolently. This research evaluates three of those eight assets: problem solving, decision making, and peaceful conflict resolution. Participating college-age mentors from the same six counties were matched with one or two at-risk youth. Group family activities based on the three assets were held monthly. By using a youth self-assessment pre- and posttest survey and a posttest survey for parents, teachers, and mentors, the program was evaluated. This research project is a study of the YFP at-risk youth mentoring program during an 8-month period. Although results were not statistically significant, youth gained more assets and had a higher score on the posttest, with the exception of four items: avoidance of making decisions, hanging out with troubled friends, doing what friends want even if it will get them into trouble, and pushing back when they themselves are pushed.