~ '— “3-..": rt .Lllaii c; r‘?« tl « mmm WWW 27/3é3/7'" IlllflllllwflljllfllflljwllfllfllllHUI LIBRARY Michigans ate Unim‘béy This is to certify that the dissertation entitled Perceived Economic Nell—being of Three Cohorts of Rural Female Household Financial Managers presented by Constance Young Kratzer has been accepted towards fulfillment of the requirements for Ph. D. degree in FCE %Wbflfl¢/ Major professor Date M MYIIi-n- Afl‘ -‘ A ‘ " 0-12771 ~fi-_*,A——£J “" ' PLACE IN RETURN BOX to remove thls checkout from your record. TO AVOID FINES return 02 or before date duo. DATE DUE .31.; , V .33 M UE DATE DUE *- r- ‘ . ‘ B?»26:994 Ii «5 w v. 1W1“, " 3 ELL-Lass. "'5 MSU In An Arlirmdlvo AotlonEqual Opportunity lndltutlon em PERCEIVED ECONOMIC WELL-BEING OF THREE COHORTS OF RURAL FEMALE HOUSEHOLD FINANCIAL MANAGERS BY Constance Young Kratzer A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Family and Child Ecology 1991 ABSTRACT PERCEIVED ECONOMIC WELL-BEING OF THREE COHORTS OF RURAL FEMALE HOUSEHOLD FINANCIAL MANAGERS BY Constance Young Kratzer The purpose of this research was to study the relationship of locus of control orientation, financial management practices, and managerial and household characteristics to the perceived economic well-being (PEWB) of three cohorts of rural women. PEWB was measured using an index that included statements about\satisfaction with the household financial situation and perceptions of income adequacy and change in financial conditions. The study assesses which variables were best able to predict PEWB for each of the cohorts, and the adequacy of the index used to predict PEWB. This study utilized data collected for the NC-182 regional research project, ”Family resource utilization as a factor in determining economic well-being of rural families”, which was supported by Agricultural Experiment Stations from the eight participating states. Because people are shaped by the historical times in which they mature, the cohort to which a woman belongs may influence the relative contribution of selected independent variables to perception of economic well-being. Based on this premise, three cohorts of women were selected from the larger sample. The women in Cohort 1 reached the age of 18 during the depression years 1929 to 1934, and were 72 to 77 at the time of the study. The women in Cohort 2 reached the age of 18 in the years between 1950 and 1955 when the economy was booming and were 51 to 56 at the time of the study. The women in Cohort 3 were among the first of the Baby Boom cohort to reach 18 in the years between 1964 and 1968 and were 36 to 42 at the time of the study. The sample was predominately white, married and middle income. The results of stepwise regression indicated that the best predictors were income and locus of control for Cohort 1 (B? c .64); income, locus of control, net worth, education, and financial management practices for Cohort 2 (B? a .63); and income, health status, locus of control, employment status and number of insurance sources for Cohort 3 (3’ - .35). There was not a significant difference in PEWB among the cohorts, but the predictors and the amount of variance explained were different. Cohort differences described in this research have implications for future research and public policy formulation. ACKNOWLEDGMENTS This dissertation project could not have been accomplished without the contributions of several individuals whom I wish to acknowledge and thank. Among them are the following: Dr. Barbara Ames was of immense assistance throughout the entire project--from proposal ideas to final draft revisions. Her standards of excellence in organization and writing were challenging and rewarding. She was a source of inspiration and endless support which was greatly appreciated. Dr. Rosemary Walker served both on the NC-182 project and the dissertation committee. Her commitment to high standards of research and her practical suggestions were very helpful. Dr. Sharon King and Dr. Larry Schiamberg also provided helpful input not only in terms of the production of this dissertation, but throughout my graduate program. I also want to thank Dr. Dennis Keefe, Principal Investigator on the NC-182 research project, for giving me an opportunity to be a part of this research project. He also provided assistance in the selection of a dissertation topic. ii Kathy Lau provided valuable assistance in preparation of the manuscript as well as immeasurable moral support. My children, Don and Deborah, and my parents, Donald and Mary Young, were most supportive and encouraging during the process. I especially want to thank Dan, who kept my computer running; John, who kept my house from falling apart! Veronica and Mary, who kept me from falling apart; and my soccer teams, who cheered me on. iii LIST OF LIST OF CHAPTER I. CHAPTER II. CHAPTER III. TABLE OF CONTENTS TABLES O O O O O O O O O O O O I O O O O O C FIGURES O O O O O O O O O O O O O O O O O O 0 INTRODUCTION . . . . . . . . . . . . Purpose of the Study . . . . Relationship of this Study to NC-182. Conceptual Framework . . . . . . Human Ecological Framework . . . . . Life Course Perspective . . . . . . . Assumptions . . . . . . . Assumptions of the Ecological Perspective Assumptions of the Life Course Perspective Assumptions of the Study . . . . . . . . . Limitations . . . . . . . . . . . . . . . . . REVIEW OF LITERATURE . . Economic Well-Being . . . Conceptual Framework . . Life Course Perspective Human Ecological Approach Cohort Studies . . . . . . . Locus of Control . . . . . Financial Management Practices Rural Women and Their Families Summary . . . . . . . . . . . . METHODOLOGY . . . . . . . . . Core Sample Selection . . . . Description of Core.Sample . All Women in the Core Sample Description of Study Sample . Research Variables . . . . . iv Page vii 10 10 11 13 13 14 14 14 16 16 31 31 35 37 45 52 59 62 65 66 67 68 69 75 CHAPTER CHAPTER IV. Dependent Variable . . . . . . Perceived economic well-being . Satisfaction with financial situation . Perception of income adequacy . Perception of changing financial conditi Independent Variables . . . . . Manager characteristics-demographi Cohort . . . . . . . . . . . . Educational level . . . . . . . Marital status . . . . . . . . Employment status . . . . . . Race or ethnic background . . . Manager characteristics-behavioral Locus of control orientation Financial management practices Household Characteristics . . Household income . . . . Per capita income . Sources of income . Insurance . . . . . Net worth . . . . . Debt/income ratio . Health status . . . Household size . . Research Questions and Hypothesis Procedure Used . . . . . . . . . Research Instrumentation-Core Study Research Instrumentation-Study . . Indices . . . . . . . . . . . . . . Data Analysis Procedures . . . . . RESULTS . . . . . Research Question Research Question Research Question Research Question Research Question Research Question Research Question Summary . . . . . ONIOtU'l-bUNH' ......... ... ...... ... . ..... C O O O O O O O O O O O O O O O O O O O O O O O O O o O O O O n Page \l U! 75 76 76 77 77 77 77 78 78 78 78 79 79 80 82 82 82 82 83 83 83 84 84 85 89 9O 92 93 95 106 106 111 114 115 116 120 124 130 CHAPTER V. APPENDIC APPENDIX A. LIST OF Page DISCUSSION, CONCLUSION, IMPLICATIONS . . . . . 134 Discussion . . . . . . . . . . . . . . . . . . 135 Household Characteristics . . . . . . . . . 135 Summary of Household Characteristics . . . . 143 Manager Characteristics - Demographic . . . 143 Summary of Manager Characteristics - Demographic . . . . . . . . . . . . . . . . 146 Manager Characteristics - Behavioral . . . . 147 Summary of Manager Characteristics - Behavioral . . . . . . . . . . . . . . . . . 149 Perceived Economic Well-Being . . . . . . . 149 Conclusion . . . . . . . . . . . . . . . . . . 157 Implications for Research . . . . . . . . . . . 160 Professional Practice . . . . . . . . . . . . . 164 PUblic palicy O O O O O O O O O O O O O O O O O 165 ES Cohort Time Line . . . . . . . . . . . . . . . 169 Human Subjects Approval . . . . . . . . . . . . 170 Instrument O O O O O O O O O O O O O O O O O O 172 Correlation Tables . . . . . . . . . . . . . . 194 REFERENCES 0 O C O O O O O O O O O O O O O O O 19 8 vi Table 1. 10. LIST OF TABLES Age of Cohorts - 1988 . . . . . . . . . . Marital Status by Cohort . . . . . . . . Educational Attainment by Cohort . . . . Employment Status by Cohort . . . . . . . Response Ranges of Independent Variables by cahort O O O O O O O O O O O O O O O 0 Means of Independent Variables by Cohort Means of Components of Perceived Economic Well-Being by Cohorts . . . . . Regression Results of Perceived Economic Well-Being (PEWB) with Income, Net Worth and Debt/Income Ratio Entered Separately Regression Results of Perceived Economic Well-Being (PEWB) with Income, Net Worth and Debt/Income Ratio . . . . . . . . . . Regression Results of Perceived Economic Well-Being (PEWB) by Locus of Control (Locus) and Financial Management Practices Entered Separately . . . . . . . . . . . vii Page 107 108 112 117 118 121 TABLE 11. 12. 13. 14. 15. 16. 17. 18. 19. Regression Results of Perceived Economic Well-Being (PEWB) by Locus of Control (Locus) and Financial Management Practices All: Results of Stepwise Regression of Perceived Economic Well-Being (PEWB) with Households and Manager Characteristics Cohort 1: Results of Stepwise Regression of Perceived Economic Well-Being (PEWB) with Households and Manager Characteristics Cohort 2: Results of Stepwise Regression of Perceived Economic Well-Being (PEWB) with Households and Manager Characteristics Cohort 3: Results of Stepwise Regression of Perceived Economic Well-Being (PEWB) with Households and Manager Characteristics All: Results of Stepwise Regression of Perceived Economic Well-Being (PEWB) with Households and Manager Characteristics, Using Per Capita Income (Percap) . . . . . Cohort 1: Results of Stepwise Regression of Perceived Economic Well-Being (PEWB) with Households and Manager Characteristics, Using Per Capita Income (Percap) . . . . . Cohort 2: Results of Stepwise Regression of Perceived Economic Well-Being (PEWB) with Households and Manager Characteristics, Using Per Capita Income (Percap) . . . . . Cohort 3: Results of Stepwise Regression of Perceived Economic Well-Being (PEWB) with Households and Manager Characteristics, Using Per Capita Income (Percap) . . . . . viii 123 125 125 126 126 127 128 128 129 TABLE 20. Cohort 1: Results of Backwards Regression of Perceived Economic Well-Being (PEWB) with Households and Manager Characteristics . . 130 ix Figure 1. LIST OF FIGURES Page Indicators of Perceived Economic well-Being O O C I O O O I O O O O O O O O O O O 5 Household Size by Cohort . . . . . . . . . . . . 71 Household Income by Cohort . . . . . . . . . . . 72 Indicators of Perceived Economic well-Being O O C O O O O O O O O O O O O O O O 153 Indicators of Perceived Economic Well-Being: Cohort 1 . . . . . . . . . . . . . 154 Indicators of Perceived Economic Well-Being: Cohort 2 . . . . . . . . . . . . . 155 Indicators of Perceived Economic Well-Being: Cohort 3 . . . . . . . . . . . . . 156 CHAPTER I INTRODUCTION Economic well-being, or the degree to which families or individuals have economic adequacy, is assessed in both subjective and objective terms. At the policy level there is concern with objective income adequacy in order to provide at least basic necessities for all citizens. Objective measures that are used include family or household income, per capita income, net worth, or debt/income ratio. The household income also may be compared to standards or averages to determine adequacy. Use of objective measures reveal whether a family is above or below poverty level or how an individual's income compares to others of the same age or occupation. But this type of measure leaves other questions unanswered. For example, why do some persons perceive their income as adequate while others at the same income level feel that their income is insufficient? Studies of economic well-being have indicated that while the traditional objective economic factors are important, (Mammen, Helmick 8 Metzen, 1983), they are not sufficient in explaining perceptions of present or 2 anticipated well-being (Williams, 1983). A more complete picture of economic well-being can be obtained if subjective measures of economic well-being are included. Perceptions of income adequacy, perception of change in financial condition and the anticipation of financial improvement have been found to be indicators of financial satisfaction. Other indicators of financial satisfaction are satisfaction with wealth, satisfaction with consumption and satisfaction with financial security (Davis and Helmick, 1983). Many of the perceptions individuals have concerning income adequacy are formed in relation to the economic history of the time and the particular family or community in which they grew up. One explanation of why a given income is perceived as adequate or inadequate may be how skilled the manager is in utilizing the resources available. The individuals' perceived control over financial situations and/or life in general may be another explanation. The family as well as the social context influences _,,,»~ management practices and locus of control orientation. These may be moderated by changes in economic climate and/or changes in family context as the individual grows older. Individuals' current financial management practices and locus of control orientation are the result of cumulative life experiences, and these may be modified by further life experience. 3 As the population of the United States grows older and the numbers of elderly persons increase, much concern has been voiced as to the economic well-being of future retirees. Early economic experiences, family status and the economic conditions through the life course are cumulative. Therefore, in order to assess the economic well-being of future retirees, it is useful to be able to identify aspects of their economic well-being over the life—course (Hareven, 1981). Cohorts who will become the elderly of the let century have been on the leading edge of the family revolution (Goldscheider, 1990). The rapid growth of the labor force participation of women, the tremendous rise in divorce and child-rearing outside of marriage, and the overall decline in marriage and remarriage will be a part of the life-course experience for these individuals. Women in these cohorts will enter old age with a different economic history than many of the women who make up the retirees of today. The diverse economic histories of the cohorts may result in different perceptions of economic well-being. In order to explore this premise, the sample for this study was three cohorts of women. The cohorts were selected based on the historical events occurring at the time these women reached 18 years of age. This is the time period that most young adults are making decisions about further schooling, marriage and first jobs. Because people are shaped by the 4 historical times in which they mature, the cohort to which a woman belongs may influence both whether she is the financial manager of the household and the relative contribution of selected independent variables to perception of economic well-being. The independent variables and the components of the dependent variable, perceived economic well-being, that were used in this study are presented in Figure 1. These include characteristics of the household and of the manager. Household characteristics are: income, number of sources of income, number of types of insurance, net worth, debt/income ratio, health and household size. Manager characteristics are divided into two parts: demographics include the cohort to which one belongs, education attained, marital and employment status and race, while behavioral characteristics include locus of control orientation and financial management practices. The economic well-being of tomorrow's older women is shaped by their educational attainment, employment history and family status. This study examined three cohorts of women who have had different life experiences to explore differences in the factors that determined their perceived economic well—being. The opportunity for women to study for a degree, to take a job or pursue a career has been increasing since the 1950's (Ogden, 1986). Even though more women are employed, Household Manager Characteristics Characteristics Income Demographic Sources of Income C0110". Education “sum“ Marital status Net Worth Employment Debt/Income ratio Race Health W. Household size Locus of control orientation Financial management practices Perceived Economic Well-Being Satisfaction with Financial Situation Perception of Income Adequacy Perception of Changing Financial Condition Indicators of Perceived Economic Well-Being 6 the employment experience of women and men is not yet equal. Men are much more likely to have a continuous work history, are employed in a wider variety of occupations, have higher work earnings and better retirement benefits (Kahne, 1985). However, womens' economic well-being is tied increasingly to their own work history. Married couple families have higher per capita incomes than unmarried individuals (Coulson, 1990; Family Economics Review, 1988). Unmarried men, on the average, have higher incomes than unmarried women. Current lower marriage and remarriage rates could affect economic well-being of future cohorts. Women are usually poorer after a divorce than they were before. Women more often have custody of minor children, and child support payments are frequently inadequate or nonexistent. One effect of increased demands on/less income is an inability to save for future financial security. ”brewing up in a single parent household may affect the future economic well-being of the children as well. They may be unable to attend college or have to borrow money for hschool and start their young adult lives with heavy debt repayments (Wallerstein, 1989). Female life expectancy has continued to increase relative to that for males (Hess, 1985). Currently, by age 65 more women than men are living alone and are poorer than men. Poverty in old age is overwhelmingly a problem for 7 older widows and single women. The income of older unmarried women is about 40 percent of the income available to older married women (Uhlenberg and Salmon, 1986). The unmarried woman may find it more difficult to save as a result of lower economic resources over the life course. Public policy with respect to the elderly has not been gender neutral (Hess, 1985). Despite the fact that the majority of elderly persons are female, most policies reflect traditional gender roles and family structures and have been framed in terms of the male life course. For example, Social Security benefits for widows are tied to the husband's earnings. Social Security benefits for unmarried women are frequently lower, reflecting the lower earnings in more traditional women's occupations. The occupations held by men are more likely to provide private pensions, and most of these pension plans do not provide automatic widow benefits even though more recent legislation has sought to correct these inequities. This study was conducted in order to learn more about _ ...—- the economic well-being of future cohorts of older women. , “The information gained can be used by educators, both formal Hand informal, to increase their understanding of family differences in economic satisfaction whenmobjectiyewincome is the same. Programs can be developed to help families manage financial resources in order to increase the satisfaction with their financial condition. The results t' 8 also can be used to evaluate public policies so that they can better meet the needs of future cohorts, particularly of women.77 / Pur ose of t Stud The purpose of this research was to study the relationship of locus of control orientation, financial management practices, managerial characteristics, and household characteristics t9_the perceived economic well- being of three cohorts of rural women. This study builds on the work of Davis & Helmick (1985) who measured financial satisfaction using a combination of objective and subjective indicators. Perceived income adequacy and perception of changes in financial conditions were added to the satisfaction measures to create an index of perceived economic well-being. ~The study assessed which variables were best able to predict perceived economic well-being, whether the strength of the predictors varied among the cohorts, and the adequacy of the index used to measure the outcome variable of perceived economic well-being. The three dimensions of perception of economic well- being that were explored were 1) satisfaction with current financial situation 2) perceived income adequacy and 3) ‘perception of changes in financial condition over time. Objective income was measured by 1) income, 2) debt/income ratio and 3) net worth and was used to determine the 9 potential of these objective measures to predict perceived economic well-being. The relationship of managerial and household characteristics such as education, marital status, health, and employment to perceived economic well-being also was studied. e a 'ons - This study utilized data collected for the NC-182 regional research project, "Family resource utilization as a factor in determining economic well-being of rural families”, subsequently referred to as the core study. Support for the core study was provided by Agricultural Experiment stations and the California Cooperative Extension Service. Eight states are involved in the project: Arizona, California, Illinois, Indiana, Iowa, Kansas, Michigan and Minnesota. This study used data collected from three cohorts of female financial managers from all eight states. The core research project (NC-182) focused on the effects of family resource utilization in determining the economic well-being of rural families. The study has two primary foci: model development, instrumentation and data collection; and dissemination of research findings to the public and professional sectors (Bauer, 1990). Data were collected from over 2500 families in economically growing/sustaining and declining rural areas. In addition 10 to assessing demographic and external factors, measures of functioning styles, interpersonal relations and family resource use were employed. The scope and magnitude of the core data facilitated this study of differences among cohorts in perceived economic well-being. o ce tua m wo um co 0 'ca ramewor This research utilized a human ecological framework. The human ecological approach focuses on human beings interacting with their environment. The environment can be defined as the sum total of the physical, biological, social, economic, political, aesthetic, and structural surrounding of the individual or household (Bubolz, Eicher, 8 Sontag, 1979). The individual or unit in interaction with its environment is called an ecosystem. An ecosystem approach to aging over time is especially appropriate as each cohort represents a unique set of ecological actors. Current research points to several distinctive features: (1) older persons appear to be more sensitive to environment variation (2) the local environment for action and choice constricts with old age (Ward, La Gory, S Sherman, 1988) and (3) while individual response to environmental change may be patterned and normative in early life it becomes increasingly less so in 11 middle age and late adulthood (Campbell, Abolafia, & Maddox, 1985). A Personal resources/characteristics are influenced by the external environment and the individual's interaction“ with this environment.— Locus of control orientation is a personal characteristic that influences how one perceives the environment. In this study the personal characteristics such as financial management practices and locus of control orientation influence how the individual manager interacts with the environment.Previeuskresearch has shown that(satisfaction with income is related to a sense of wellfbeing. \The extent to . which income level affects the individual's sense of well-/ 18 being depends upon past experiences as well as upon the social environment, values and ideas about fairness and, equity (Strumpel, 1976).” Individuals measure their income against the societal average and are more or less satisfied depending on how they perceive their income to match the average (Ackerman 8 Paolucci, 1983). A measure of reference has been poverty thresholds, i.e. whether one is above or below thehihcome defined as the poverty level. The dollar value assigned to poverty level is most often determined by multiplying the cost of the economy level food plan by three. This is adjusted for the number, age, and gender of persons in the household as well as regions of the country. These figures are published by the United States Department of Agriculture (USDA). Income as related to age of household head has been another reference point. Several reports have compared income of older and younger households (Institute of Rural Poverty, 1983; Radner, 1987; Ryscavage, 1987). In 1966, the average income of families, head 65 and over, was 49% of the average income for all families. In the decade of the 70's Social Security payments were increased and an automatic cost of living adjustment, tying benefits to the Consumer Price Index, was instituted. By 1983 this average income for older families was 63% of the average income for all families (Institute of Rural Poverty, 1983). It has been noted that factors other than income need to be considered 19 in comparing households of older and younger adults. Older households are more likely to own their homes and have fewer household members. Another factor is marital status. The median income in 1985 for families, head 65 and over, was $19,162 contrasted to $7,568 for unrelated adults over 65. For households, head 25 to 34, the median income for families was $26,023 and for unrelated individuals $17,211. The gap between old and young was greater for unrelated adults than for families (Ryscavage, 1987). These studies reported only objective measures of income. In order to know more about the economic well-being of younger and older households, studies have added subjective measures related to satisfaction with income and other aspects of their financial situation. Metzen and Helmick (1975) developed the income _improvement index to measure how the addition of a secondary worker's income affected family income adequacy. {In order to assess the relative economic well-being of families, it was necessary that family income be considered in the context of family size and composition; A family's essential income was considered to be three times the economy level food plan of the USDA. Base income was defined as the income contributed by the primary worker (in this study, the husband) and income from sources other than earnings. An ineome adequacy_index was derived by dividing base income by essential family income. They found that 20 contribution of the secondary income was more important to family income adequacy where the base income was below or close to the essential family income. As base income increased in relation to essential family income, the contribution of the secondary income to family income adequacy declined. Metzen and Helmick's (1975) finding of margieelaregegesw on increased income after a certain level is reached is supported by the following studies. Campbell (1981) found that individuals who are materially well off have stronger feelings of well-being than those at the bottom of the ladder. However income and material goods were not sufficient to guarantee well-being or the health of the family system. Foa and Foa (1973) stated that a shortage of non-economic resources can result in economic and ecological costs. The less one is satisfied spiritually and emotionally, the greater the demand for material goods. But economic goods, even in increasing amounts, are not conducive to ever increasing satisfaction. Foa and Foa also suggested that provision of psychological non-economic resources may reduce the demand for economic ones and increase satisfaction. Various measures of economic well-being have been used in studies of the concept. Fletcher and Lorenz (1985) used total family income and satisfaction with family income to measure economic well-being for different age, sex and 21 racial groups. They found that there was a less strong relationship between subjective income and objective income for older respondents than for younger ones. Satisfaction with income increased with age without a corresponding increase in income. There were no differences in females and males in the strength of relationship nor between the races. For both groups, increases in family income were directly related to increases in satisfaction with family income. The nonwhites were less satisfied over all but the relationship was the same. Ackerman 8 Paolucci (1983) studied subjective and objective income adequacy as related to over—all life quality. Objective income was measured by comparing the reported income to the moderate income of the Bureau of Labor Statistics budgets adjusted for area of the country and family size. Perceived income adequaey was measured by asking if the family income was enough for the family to live as comfortably as they liked. They found that life quality measures of satisfaction increased with an increase in either subjective or objective income adequacy. However, subjective adequacy accounted for more variation of scores on satisfaction with income and over-all life quality, while objective adequacy was more related to satisfaction with consumption. Subjective adequacy is more personal and unique to the family involved, as is consumption. 22 The findings of Ackerman and Paolucci (1983) agree with findings of other studies relating objective income adequacy to satisfaction with family income. Satisfaction has been found to increase most at the extremes of income distribution. Increases in objective income for those groups having lower incomes and for the group with the highest objective income resulted in significant increases in subjective adequacy for these groups. For those income groups in the middle of the objective income distribution, subjective adequacy did not increase significantly with an increase in actual income (Campbell, 1981; Douthitt, MacDonald, 8 Mullis, 1990; Strumpel, 1976). In a study of factors contributing to quality of life, middle-aged, married women were asked to rate the importance of life concerns (Sontag, Bubolz, 8 Slocum, 1979). The respondents were asked to indicate the degree of satisfaction or dissatisfaction with the domains listed as important. Respondents also were asked to indicate how satisfied they were with the degree to which important values were being fulfilled. Family life, children, and love and affection were rated the most important aspects of life. Financial security and total family income were rated sixth and ninth respectively, although at least 80 percent considered both factors of very high or high importance. Most of the women were satisfied with their family life and children. Though 35% were very satisfied with their 23 total family income, 20% had mixed feelings or were very dissatisfied with family income. Feelings about life-as-a- whole were most influenced by satisfaction with family life, self, family income and housing (Sontag, Bubolz, 8 Slocum, 1979). Their life concerns reflect both material and human resources as contributing to perception of quality of life. Satisfaction with family income was a contributing factor but family life was a more significant predictor. The quality of life of middle aged women was studied further to discover the relationship of various stressors and resources to quality of life. The variables studied included family, economic stressors and health stressors, and resources represented by level of income, employment status and education. Income adequacy, marital happiness and quality of family life were used as both dependent and independent variables. Level of income, number of dependents, employment status and education were directly related to income adequacy but not to other domains of life. Income adequacy was related to quality of family life. Health stressors appear to affect all aspects of life as they were related to income adequacy, health symptoms marital happiness, quality of family and quality of life (Walker, Lee, Bubolz and Keefe, 1990). A study of factors affecting quality of family life for mid-life men and women found marital happiness to be the strongest predictor of quality of family life. Income 24 adequacy, health symptoms and education were among other variables that had a direct influence. Income adequacy had a positive effect, with health stressors and educational level relating negatively to quality of family life. More health symptoms were negatively related to marital happiness, while the higher the income adequacy and the more money management strategies used, the greater the marital happiness rating. Income adequacy was positively related to higher levels of income and education, and negatively related to a higher discrepancy between money management strategies desired and those employed. For women, perceived income adequacy was more important to quality of family life than per capita income, while for men perceived income adequacy had no effect on quality of family life and per capita income had a direct effect (Walker, Bubolz 8 Lee, 1991). Three objective measures of economic well-being were used in a study to examine the relationship between an individual's perceived over all life satisfaction and economic well-being (Douthitt, MacDonald, 8 Mullis, 1990). The three measures included (1) Modigliani and Brumberg's ”Life Cycle Income Hypothesis" which states that consumers' sense of well-being is contingent on their being able to maintain a relatively level pattern of consumption over the life course, (2) Duesenberry's "Relative Income Hypothesis" which argues that the consumer's reference groups have a 25 direct impact on the extent to which economic resources are regarded to be adequate or inadequate and (3) Kyrk's "Resource Deficit Hypothesis" which suggests that the greater the discrepancy between the consumers' actual economic resources and the level that one aspires to the lower the reported psychological well-being. {/Results from each of the three measures supported the relationship between current income and overall life satisfaction, with satisfaction increasing with higher income levels.l Life cycle stage also was found to be a factor in explaining life satisfaction. The study did not examine whether this was due to demands on income or other family factors. All three models of economic well-being were useful in explaining psychological well-being. But the ”Relative Income Hypothesis” model explained the largest percentage of variance in life satisfaction. Results indicate that how far the consumer is from average in terms of expenditure patterns is important. If income is held constant and expenditures increase above those of the average household, satisfaction is diminished. Satisfaction is increased when expenditures are less than average, holding income constant (Douthitt, MacDonald, 8 Mullis, 1990). Several recent studies have attempted to further define the concepts of economic well-being and perception of economic well-being and to determine which variables should 26 be included in the measure (Hayhoe, 1990; Porter 8 German, 1990; Bauer, Danes, 8 Rettig, 1990). Although a systems approach to family well-being assumes the interdependence of economic, social-psychological and physical attributes and conditions, the concept of economic well-being is not clearly defined. It is often assumed that economic well- being is the same for everyone and that it does not influence other aspects of family well-being. Bauer, Rettig 8 Danes (1990) outlined a comprehensive measure of economic well being including subjective and objectives indicators at the global and micro-unit level over time. They proposed that the purpose of a given study would guide which measures to be stressed. An example used was when health concerns are the focus of the study a more comprehensive measure of economic well-being is appropriate as the flow of money income can be influenced be a decline in health. Income, level of insurance and family wealth may be variables of interest when focusing on health concerns. Health status can affect resources available by limiting ability to work and creating additional demands on resources. This inability to work and increased demand on resources may be reflected in perception of economic well- being (Hayhoe, 1990). The cost of poor health is of great concern to older persons. About 65% of elderly persons are covered by some type of private health insurance. About 27% of health care costs are out of pocket (Ebersole 8 Hess, 27 1985). About 30% of the total personal health care expenditures are made by elderly persons even though they make up about 11% of the population. Measures of health or health status have been included in studies of satisfaction with quality of life and the financial domain, perceived economic well-being, and financial satisfaction (Davis 8 Helmick, 1983; Helmick 8 Metzen, 1983: Slusher, Helmick 8 Metzen, 1983). Health was related to life satisfaction for men but not for women and was not found to be a significant factor in the other studies. The form of measurement used and the lack of variance in measured health status for women was suggested as reasons for this finding (Helmick 8 Metzen, 1983). It was suggested that future studies include a measure of health and of costs related to health. Hayhoe (1990) tested the extent to which active participation in financial management in the household contributed to perceived economic well-being. Perceived economic well-being was measured by responses to questions of satisfaction with level of income, money available for necessities, and financial emergencies, amount of money owed, level of savings for future needs and feelings about the families' economic and financial security. The input variables included total income, number of years married, number of people in the family, age, health status, locus of control, respondent's view of future and past financial 28 condition, reference to the outcome of others, and involvement in the family's financial decision making process. Income, improvement of financial situation eyer the past 5 years? and reference to outcomes of others were significantly related to higher perceived economic wellf Beggg‘while more external locus of control, low expectations for financial futures, being employed and involvement in financial decision making were negatively related to perceived economic well-being for women. Health was not significantly related to perceived economic well-being for women, though it was for men. The difficulties of comparing well-being of different size households was addressed by Blaylock and Blisard (1990). The use of per capita income assumes no economies of scale; whereas the use of household income fails to take into account the demands of that income. Per capita income measures put black households at the lowest income levels, as they tend to have larger households. However, the use of \1 household income tends to over report single persons at the lower income levels. Other income-ddjustment scales in use also introduce bias, so the method used needs to be selected carefully, and taken into account in interpreting results.)r Family economic well-being is not only dependent on the income and the internal conditions in the family but also on conditions outside or external to the family. The economic conditions of the country and the culture influence how the 29 individual or family perceives income in relation to their cohorts and affects their feelings of well-being (Moen, Kain, 8 Elder, 1983). Moen, Kain, 8 Elder presented a model of economic loss to study the effect of change in economic status (1983). Families tend to judge their current economic status on the basis of past economic conditions as well as their future expectations. This model is supported by the research findings that perceived change and expectations for the future were predictors of satisfaction with family income (Davis 8 Helmick, 1985, Hafstrom and Dunsing, 1973; Year, 1976). Changes in conditions external to the family such as inflation, recession or wars affect the family's sense of security. Fear of job loss or concern that current income or savings will be inadequate frequently results in diminished perception of economic well-being even though income is unchanged. Changes internal to the family such as illness of a member or change in family composition may have the same effect. Changes in family composition may be additions to the family such as the birth of a child or an aging parent joining the household, or decreases through divorce or widowhood. These may be reflected in increased demands or decreased resources resulting in a decrease in feelings of economic well-being. The degree to which any of the above listed changes affect economic well-being is shaped by the individual's 30 definition of the situation. This definition is determined by the feelings individuals have in regards to control of their lives, resources available and past history (Moen, Kain, 8 Elder, 1985). For example, change in family composition through divorce frequently results in fewer economic resources for families as they must establish separate residences. The change in economic status and the feeling of lack of control may have as much of an impact on perceived economic well-being as the actual income available. As current status is compared to the family's former economic condition, feelings of economic well-being may be diminished. Previous experience of overcoming adversity and/or high expectations for the future can have a positive effect on perceived economic well-being (Moen, Kain, 8 Elder, 1983; Wallerstein 8 Blakeslee, 1989). Multivariate regression analysis was used to identify the factors having the greatest impact on financial satisfaction in a study by Davis and Helmick (1985). Financial satisfaction was measured using an additive index including satisfaction with consumption level, satisfaction with wealth, and satisfaction with financial security. The measure of financial security included satisfaction with proportion of income saved, satisfaction with emergency reserves and satisfaction with future financial security. Inputs included family income, family net worth, debt/income ratio, remaining number of child-rearing years, and number 31 of earners. Also included were desire for financial improvement and perceived change in financial condition as reference point variables. Husbands and wives from each of three states were analyzed, creating six groups. In each of the six groups, desire for financial improvement and perceived change in financial improvement emerged as strong predictors of financial satisfaction. Net worth was a strong predictor in four of the groups, while family income was not a strong predictor in any of the groups (Davis 8 Helmick, 1985). The findings are consistent with previous research results. Objective indicators such as family income or net worth have a significant influence on the measure of financial satisfaction. The best predictors, however, appear to be perceived change in financial condition and desire for financial improvement. Previous studies have found both to be significant predictors of financial satisfaction (Hafstrom and Dunsing, 1973; Year, 1976). o ce Life Course Perspective The life course perspective focuses on the interaction between individuals and collective timing of family transitions as they are shaped by different historical conditions. The life-course approach links the individual's 32 life story with collective behavior as part of an ongoing continuum of historical change (Hareven, 1981). The most obvious dimension in the study of the life course is aging or life time. Aging is composed of biological, psychological and social processes. Lives are played out in particular social and historical circumstances that must be recognized if the life course of the individual is to be understood. Social time may be defined as the set of norms that specify when particular life transitions or accomplishments are expected to occur in a particular culture or segment of society. Historical time refers to macro-developments, often represented by one major event such as the Great Depression, or a world war. Historical events may directly affect social time (Campbell, Abolafia, 8 Maddox, 1985; Hareven, 1981; Clausen, 1986). As individuals develop and change, social institutions both limit alternatives and are changed by new demands placed on them. The relationship between individuals and society are interactive and dialectic (Campbell, Abolafia, 8 Maddox, 1985; Elder, 1978). Development and change occur throughout the life course. According to Campbell, Abolafia, and Maddox (1985), change may be patterned and normative in early life, but becomes increasingly less so in middle age and late adulthood. Schaie and Willis (1986) state that age graded influences may be more important in childhood and old age, 33 with history graded and non-normative changes more important in early and middle adulthood. "Cohort effects" on adolescents may be greater than the effect of age alone (Schiamberg, 1988). Individuals in different birth cohorts are likely to have experienced different or unequal historical circumstances that result in differing paths of development. Thus a full understanding of the historical background of an individual is an important dimension of the life-span perspective. How an individual moves through the life course is dependent on (1) the person's own attributes such as intelligence and health, (2) the sources of socialization, support and guidance that provide the individual's initial orientation to the world (e.g.family and peers), (3) the opportunities available to the person as influenced by class, age, gender,and ethnic group as well as the effects of war, depression and other major social changes that affect particular birth cohorts differently and (4) individuals' investments of effort on their own behalf (Clausen, 1986). Latten (1989) explored life-course patterns to determine if there was a general life course of life satisfaction for everyone irrespective of cohort. Using data collected from 3 year age groups from 18 to 75+ at four different time periods, he found that the similarity in life satisfaction curves was striking. At around age 30 a 34 decline starts which reaches its lowest point in midlife. Thereafter the curve begins to turn upward. From this, Latten concluded that there is a general life course for all persons regarding satisfaction with life, but this general level of satisfaction is not dominating. Fluctuations in the individual's life-course are more important than the general course. Hareven (1981) used the life course perspective to discuss historical changes in the timing of family transitions. Life course is concerned with the timing of transitions over an individual's career, particularly the balancing of entry and exit from roles as they relate to collective family behavior. A second important feature is the impact of historical processes on the individual and family transitions. A third feature of the life course is the cumulative effect of earlier transitions on later ones. The adaptation of individuals to social and economic conditions when they reach old age is contingent on the paths by which they reach old age. The cumulative effect of life course experiences is documented in Elder's Qhi1dIen_gfi_tng_§;eg;_geng§§ign‘ Within the same cohort of unemployed adults, coping abilities differed not only in terms of availability of resources and family backgrounds but also in terms of earlier transitions-such as how long the individual had been working (Elder, 1974). 35 The life course perspective combines the approaches of family change and social change and recognizes the interplay between external economic change and internal family experience. Moen, Kain, and Elder (1983) use this approach in developing a construct of family economy, which views the family as a flexible unit whose economic status depends upon its family members as well as the workings of the outside economy. Economic adversity affects conditions in the home. These conditions prompt adaptations, which in turn prompt economic change. For example, more family members may enter the work force, which will have a cumulative effect on~the economy at large. WWI: «a The human ecological approach includes human beings interacting with the total environment. It assumes that the development of a human being is affected by and affects not only the immediate situation or system of which it is a part but all systems that directly or indirectly relate to it. The delimitation of human ecology to the immediate or near home and living environment does not exclude the consideration of broad-based complex societal problems such as population growth, use of energy resources, pollution and resource distribution (Kilsdonk, 1983). The nearer the environment, the more influence it has on the individual, but all environments influence and are influenced by the individual. 36 ‘.The environment can be divided into three interacting environments: the natural physical-biological environment, which includes physical and biological components; the human built environment which includes alterations and transformations made by humans to the natural environment (e.g. roads, urban settlements, oil spills); and the social- cultural environment. The social-cultural environment includes 1) other human beings, 2) abstract cultural constructions such as language, laws, and cultural patterns, and 3) social and economic institutions (Bubolz 8 Sontag, 1921). The ecological approach involves the human being interacting with these three environments as well as the interactions of the environments with each otherIM/ The individual or family selects, shapes and modifies its neer environment to be consistent with its goals and yalues. The adequacy of the near environment for the achievement of goals will depend in part on the family's management plan. Adequacy also will be determined by the resources available for use and the individual's perceived needs and expectations. ._....._.-.—-. ___——-- organisms in relation to their environment. He continues to say that all species live in characteristic habitats and engage in a form of behavior called habitat selection. This selection of habitat is an active process in which certain cues play an important role, so that 'psychological' factors 37 may be as important as physiological in making choices. For humans, cultural characteristics play a large part in determining which environments are seen to be supportive. Whenever they can people leave undesirable or unsuitable environments and seek out more suitable ones. The choice of environment is based on culture, and more specifically, lifestyle and activity systems. Lack of choice may in itself be a major environmental problem. a Aging takes place in and is affected by the environment of the individual. Although all of the environments interact with one another, in the human ecosystem the interaction between people and their environment is the most important to individual well-being. Thus it would appear that the interaction of the individual with the economic institutions in the environment would have an effect on the economic well-being of the individual.“ W A cohort is a group that moves along together through the life course and experiences historical events at the same time. A cohort may be a group who were born the same year and would be the same age, or it could be a group that experienced a given phenomenon at the same time. For example, a work cohort would be all those persons who started work at the same time. The cohorts' placement in historical time provides information about the opportunities available and the constraints that are placed on its 38 members. An understanding of an individual's current position and perspective of the future depends on that individual's cumulative life history and the location in historical time (Clausen, 1986; Kain, 1990; Hareven, 1983). Each cohort has a distinctive composition and is predisposed to characteristic modes of thought reflecting the circumstances of its unique place in history. Cohorts experience somewhat different cultures or subcultures (Clausen, 1986; Ryder, 1965). A cohort may differ in size relative to the one before and after it, in educational attainment and income levels. Cohorts differ in the way they experience historical factors such as the economic conditions at a particular transition time. Cohorts are not homogeneous but have differences within. These differences may influence how individuals within the cohort are affected by the social and historical experiences of the cohort (Campbell, Abolafia, 8 Maddox, 1985; Ryder, 1965). Cohort analysis may be used to study the cohort flow and cohort differences (Ryder, 1965). Elder (1974) used the cohort concept to look at differences within the cohort of the effect of the same historical experience as well as differences in the way cohorts experience the same historical phenomenon. The influence of cohort size on personal welfare was described by Easterlin (1987). Easterlin's thesis is that cohort size shapes the economic climate felt by young adults 39 and this in turn affects their attitudes and behaviors in relation to work, marriage, childbearing and aspired standard of living. Smaller cohorts have an advantage as they have fewer competitors for jobs. Farraro (1990) studied retirement preparation of different cohorts in 1974 and 1981 to explore Easterlin's thesis. Retirement preparation was measured by response to eight questions about activities related to retirement planning. These included such items as increasing savings, buying a home, preparing a will and development of leisure activities. It was expected that earlier cohorts would prepare more as they were closer to retirement. Also according to Easterlin's theory, as these earlier cohorts were smaller in size, they would have had the advantage of less competition for employment and advancement. Thus they would be in a better financial position to save. He found that retirement preparation varied considerably over the life course and between cohorts. As expected, in 1974 the earlier cohorts (birth years 1929-1935 and 1936-1942) had made more preparation. These cohorts were older and smaller and had grown up in conservative financial times. But in 1981, the Baby Boomer cohort (birth years 1950-1956) had increased their savings more than these two cohorts. One explanation for this was that even though this was a large cohort they entered the work force in a unique social 40 climate. Their preparation reflected a major change in values and looking forward to early retirement. The decrease in retirement preparation from 1974 to 1981 in the older cohort reflected the period effect of a depressed economy, high energy costs, inflation, government cutbacks and unemployment. Cohort effects were still evident but were modified by historical and social context. Burns (190) studied retirement saving of the Baby Boom cohort and an earlier cohort using data from the 1983 Survey of Consumer Finances. Because the Baby Boom cohort is so much larger than previous groups, it is believed that they exhibit different economic behavior than earlier population groups. The sample consisted of households which contained a head of household between the ages of 27 and 37 in 1960 and 1983 and focused on privately provided retirement assets and net worth. Measures for comparison that were created were private retirement assets/income ratio and net worth/income ratio. The variation in demographic characteristics reflected changes in the society over that time. The 1983 cohort was more educated, had more female head of households, and was more likely to be a divorced or separated head of household. The 1963 sample had larger households and fewer earners in the spending unit. There were no significant differences in the net worth/income ratio and the retirement assets/income ratio for the two groups. This would indicate that the Baby Boom cohort is 41 not different from earlier cohorts in their retirement preparation. Ryff (1985) has suggested the use of cohort studies to separate out cohort effects from aging effects. Rossi (1985) criticized previous literature on adult personality development of 'cohort particularity' i.e. it is based on findings of studies of persons born between 1920 and 1930 and has not been updated. The personality traits of that cohort may have resulted from the social and economic conditions of that period. Later cohorts of the same chronological age may develop different personality traits as the social and economic conditions vary. Studies using only men as subjects also have been criticized (Ryff, 1985). Cohort is particularly important in comparing women of the same age at different points in time. The women of 50 in 1960 may have a very different set of values and expectations in comparison to the woman who will be 50 in 1990 (Block, Davidson, 8 Grambs, 1981). Hareven (1983) suggests that there has been an historical shift in the timing of passage from one role to another in families with timing becoming characterized by greater uniformity. She found that cohorts varied in how they segmented the life course and what they viewed as key transitions. Older cohorts were found to be less conscious of stages. Younger cohorts were more conscious of timing and transitions and were found to prefer earlier ages for 42 educational/occupational events and later ages for family events. An historical consideration of family places some changes in proper context (Fallo-Mitchell 8 Ryff, 1982; Hareven, 1983). The women in Cohort 1 were 72 to 77 years of age at the time of the study. These women reached the age of 18 during the depression years of 1929 to 1934. The economic conditions at the time made it difficult to find work and educational opportunities were limited. The 1930 census counted 23 million housewives and 10 million women in gainful employment. Unemployment figures climbed from 8.7% in 1930 to 23.6% in 1932 for all persons 16 to 64 (Ogden, 1986). Women who worked in the 1930's were predominantly in domestic employment, elementary teaching or replaced men in menial jobs for which women were paid less. Marriage to a good provider was considered the most lucrative career. Marriage was often postponed due to financial circumstances and fewer children were born to those couples who married, as children were considered an economic liability (Ebersole 8 Hess, 1985; Wetzel,1990). For many women it became a challenge to make the most of limited resources as their husbands lost their jobs and a paycheck was no longer forthcoming. For these women, homemaking was very labor intensive (Elder, 1974). The women in Cohort 2 were 51 to 56 years of age at the time of the study. These women reached young adulthood in 43 the early 1950's and were members of the cohort that first moved to the suburbs. Women were expected to get enough education to meet a good provider but not spend so long in school as to waste their youth. The prevailing sentiment was that every woman should marry, that she should have children and her family roles would be the only roles important to her. Marriage per se was the focus of interest (Elder, 1974; Ogden, 1986). Few of the women in Cohort 2 worked outside the home as young adults. Most women went from school to marriage, and if they worked it was only until marriage or the birth of the first child. These women married earlier than the women in Cohort 3; the economy was booming and conditions were ideal for starting and raising a family. There were 10.4 million wives working but these were mostly middle-aged women who had held a job before marriage. The goal of realizing personal ambitions in the outside world had receded into oblivion for most housewives. During this time the focus was on family, and the nuclear family resurfaced as the norm (Ebersole 8 Hess, 1985). These women were the super mothers of the 1950's. The American home became a 'hot house' of permanent family happiness. The housewife was considered the mainstay of society and was indispensable in maintaining the stable happy family which she was assured was the central core of American life. Politically it was the cold war era of 44 nationalism and patriotism. Mrs. America's role was to shore up the family against liberalism, socialism and communism. The economy was booming and it was an era of consumerism (Ogden, 1986). The women in Cohort 3 were 37 to 42 years of age at the time of the study. These women reached the age of 18 between 1964-1969 as part of the first wave of the Baby Boom cohort to become young adults. The 1960's were an historical time of change and questioning of values. The student movement began in Berkeley in 1964. In 1965-66 the anti-war movement was at its height. The 1960's were also the time of the civil rights movement, the women's movement and the counter culture movement (Bengston 8 Schaie, 1989). It was a time of rebellion for young adults, the days of the hippies and yippies (Ebersole 8 Hess, 1985). Many of these women went on to college and entered the work force. Many lived away from the family, either alone or with roommates before marrying, instead of moving straight from the parental home to the marital home. These women married later than previous cohorts, had children later and had fewer children. Families also began to question the indulgence of children and to recognize that children needed limits (Ebersole 8 Hess, 1985). For all of these cohorts the 1970's and 80's have brought changes in the number of women entering the work force, an increase in single parent households, and earlier 45 retirement. In the 1970's there was a boom in the farm economy followed by the farm crisis in the mid 1980's. Each cohort weathered these changes from a different perspective depending on their position in the life course (Wetzel, 1990). (See Appendix A for a time line chart). Campbell, Abolafia 8 Maddox (1985) have stressed the need for continued cohort studies so that there is a base of data on the characteristics of cohorts over time. This data can be used to begin to separate cohort effects from the effect of aging. Looking at developments over time enables us to better assess the uniqueness of present conditions and distinguish between long-term trends and temporary developments. Mattel Locus of control is not a characteristic to be discovered within individuals but is a construct-a working tool in social learning theory which allows for interpretation of an individual's response to causality (Lefcourt, 1976). Personal control is a self-report individual difference variable which indicates to what extent an individual believes that outcomes or performances are due to his or her own doing as opposed to forces outside the self (Lachman, 1990). The effects of an individual's perception of control, whether one believes that one can determine his or her own fate within limits, is of critical 46 importance to the way in which an individual copes with stress and engages in life's challenges. Control can be ff’ defined as the active belief that one has a choice among responses that are differentially effective in achieving a desired outcome (Langer, 1983). People are not totally internals or externals. The terms are used to depict an individual's more common tendencies to expect events to be contingent or noncontingent upon personal actions (Lefcourt, 1976). A person's sense of control consists of beliefs and expectancies about the self and about the environment. Beliefs about the causal nature of environment focus on whether environment is seen as lawful or orderly as opposed to random forces (Abela, 1990). Perceived control is positively associated with access to opportunity. Internal control expectancies are more likely to be held by persons who through position and social environment can more easily attain those things they value. An environment that encourages control is one that encourages a sense of mastery in individuals (Lefcourt, 1976). Sense of control may be modified by personal experience and the experience of others through reference group processes. One's general cultural beliefs also influence the beliefs and expectations constituting one's sense of control (Abela, 1990). The desire for control may vary by individual. Varied cultural, social-structural and institutional processes brought about 47 by different historical circumstances lead to differences in emphasis placed on importance of self-direction and individual personal control (Rodin, 1990; Schooler, 1990). The largest body of empirical data about perceived control derives from Julian Rotter's social learning theory (Rotter, 1976). The premise of this work is that actions are predicted on the basis of values, expectations and the situations in which persons find themselves. Rotter used general scales to measure views about locus of causality which compared multiple factors involving different domains or types of control. A list of paired items reflecting internal or external control beliefs was used to determine the degree to which individuals perceived themselves to be in control of their own lives. Without an expectation of internal control, the postponement of immediate pleasures and the organizing of one's time and efforts would be unlikely. Individuals must entertain some hope that their efforts can be effective ‘ before they can be expected to make sacrifices that are the prerequisite for achievement. The idea that locus of control orientation is related to cognitive activity appeals to common sense. Persons holding internal control expectancies are usually more cautious and calculating about their choices and more attentive to information relevant to decision making than are persons holding external control orientations (Lefcourt, 1976; Rodin, 1990). 48 Perceptions of control begin in infancy and have a life course trajectory. Across the life span every significant developmental transaction provides new challenges for perceived and actual control. In young adulthood the individual learns about commitments and responsibilities of being an adult (Rodin, 1990). Youths have not yet developed a variety of roles to meet their needs; their ability to control everyday lives may be constrained (Heise, 1990). In middle adulthood and old age new demands and issues impact on one's sense of control (Rodin, 1990). Changes in the middle years are frequently in response to successes and provide the middle aged with a multitude of roles from which to choose. One view is that the loss of roles, norms and reference groups through retirement and bereavement may lower perception of control in old age. Although it is uncertain that even if actual control may be diminished in these circumstances, whether the perception of control diminishes with age (Rodin, 1990). Another view is that older people hold multiple roles which limit their ability to chart their own courses as they please (Foner, 1990). Locus of control orientation at different ages and whether or not there are changes over the life course have been the subject of many studies. The results are not consistent. Skinner 8 Connell (1986) see individuals becoming more internal from childhood to adolescence and not much change between older groups. In a comparison of 49 several studies, Lachman (1986) found some reported a decrease in internal control, others reported no change and still others reported an increase in internal control with age. Lachman's approach to explaining these inconsistencies has been to look at multidimensionality and domain specificity. Multidimensionality recognizes multiple sources of control such as the self, chance and other people which may operate in a general way across all behavior domains or within specific domains or spheres of behavior. She suggests that a multidimensional approach may be more useful in examining changes in adulthood. It is proposed that growing older sensitizes one to the role of luck or chance particularly as it affects health and causes a loss of roles due to retirement, prompting adults to become more external with regard to environmental responses but making no difference in feelings of personal efficacy. Internal control is less likely to decline because older adults have a vast experience, and the longer one lives the more likely one is to have a feeling of competence and mastery. The domains in which one feels that one has control may differ by age. Studies suggest stronger external control beliefs of the elderly than the young in health and intellectual functioning but not for generalized control beliefs. Those older adults with internal control have higher levels of intellectual function in later life. In 50 another study, older persons were more likely to indicate a feeling of less control over health and economics while younger ages said they felt less control in the domains of work and family (Gatz, Seigler, George, 8 Tyler, 1986). Walford-Kraemer and Light (1984) studied midwestern women between the ages of 21 and 63 to determine the relationships between depression scores, internal locus of control scores and several demographic variables. They found that depression was significantly related to internal locus of control. Those women whose locus of control was more internal had lower depression scores. They also found that married women had the lowest mean depression scores. Income was related to depression in that the highest depression scores were held by women in the lowest income groups, while women with annual incomes over $30,000 had the lowest depression scores. Education was related to perceived control in that the higher te level of education the more internal the perceived control. Perceived control also differed by residence with rural women indicating the most perceived internal control, and women who lived in small towns the least perceived internal control. (\A correlation between high income level and internal locus of control has been found in studies by Hoff and Hohner (1986) and Strumpel (1976). )Iytton and German (1988) studied locus of control as related to financial satisfaction. Individuals with an internal locus of control 51 approach to life reported more financial satisfaction. Twice as many of the respondents who were financially distressed felt they had little control over the events affecting their lives.§LE:sense of control over one's life also was found to be espe 1ally important to the amount of stress experienced by individuals with high financial and work demandeL] (Lefcourt, 1983; Pearlin 8 Schooler, 1978). Keating (1987) found the personal resource of mastery (internal control) was the most important predictor of stress for both farm men and women. High stress farmers felt that their fate and their livelihood were out of their control. They felt controlled by the fluctuations of the market and the weather. All farmers face the same demands, yet high mastery men and women were confident that they could manage the problems. i / In a study of rural women and economically stressed men and women, Danes, Rettig 8 Bauer (1991) found that locus of control was negatively affected by larger household size. They also found that more internal locus of control was negatively related to the gap between the standard and level of living. The more internal control reported, the smaller the gap and the greater the satisfaction with the gap. Locus of control can be viewed as a mediator of involved commitment in life pursuits. The use of perception of control as a powerful predictor will be more profitable if an assessment device is designed in relation to the 52 specific domain of interest. Control orientation of specific domains in conjunction with an understanding of individuals' generalized perceptions of control will provide more utility in the study of locus of control orientation (Lachman, 1986; Lefcourt, 1983). a c a me Financial management practices as described in the literature primarily deals with what to manage and how to manage it, rather that why to manage. In a review of resource management texts, financial management practices were found to include goal setting, planning or budgeting, and implementing financial plans, which includes record keeping and evaluating the success of the plan (Gross, Crandall, 8 Knoll, 1973). Nickell, Rice and Tucker (1976) list ten steps to more successful financial management. Despite the importance of family financial management for families, little is known about what families actually do. Even less research has focused on whether the use of these procedures are effective in producing positive results for families (Godwin, 1990). A budget process variable was developed by Beutler and Mason (1987) to address questions concerning family cash flow management: 1) how frequently was a formalized budget used?, 2) what were the characteristics of families who used it?, and 3) what were the potential benefits associated with 53 the use of a formalized budget? They found that less than 10% of the families always kept written plans of up to a year, kept written records of expenditures, and usually or always reviewed their expenditures. About 25% of the families reported using plans that were usually written with a three to four month planning horizon. Over a third of the sample reported no advanced planning and no records or evaluation of spending. Other studies have found that a small percentage of families engage in formal money management practices (Godwin 8 Carroll, 1986; Mullis 8 Schnittgrund, 1982; Titus, Fanslow, 8 Hira, 1989). Godwin (1990) suggests a need for a theory of cash flow management to be studied over time. She sees family cash flow management as a dynamic process that evolves over the life course of the family. Those who used plans were more highly educated, two spouse families (Beutler 8 Mason, 1987). Households headed by a younger person and families in stages of the family life cycle characterized by high demands used more formalized plans. Total household income had no effect on the use of plans. The tendency to budget was more closely related to an increase in demands than to a change in income. The use of plans had a positive impact on the four output variables of net worth, preparation for financial emergencies, reported satisfaction with level of living and perceived adequacy of income. Satisfaction with level of 54 living and perceived income adequacy were much less influenced by the budgeting process than satisfaction with preparation for emergencies, the objective measure of net worth. Titus, Fanslow 8 Hira (1989) found that financial planning behavior was positively related to families' net worth. Families reported a higher net worth when they estimated their household income and expenses, calculated their net worth regularly, reviewed and evaluated their spending and set financial goals. The financial situation of households with similar resources can differ depending on their credit practices. Mueller 8 Hess (1984) studied use of credit as an indicator of financial management or mismanagement. They found that the greater the number of credit sources and the more credit used the lower the level of solvency. A study of credit use by Danes and Hira (1986) investigated relationships between credit knowledge, credit attitudes and credit practices (1986). A positive relationship between knowledge of credit and attitudes was found. Education of the money manager, household size and income were significantly related to knowledge of credit. Income had a significant path to credit attitudes, but none of the variables studied were directly related to credit practices. Jensen and Reynolds (1986) found education and income to be a significant determinate of credit. They found credit users were generally better educated, younger, and more 55 often had children. Installment purchases debts were less likely among older persons. Use of credit by older persons was more likely to be for convenience rather than credit per se. Access to credit has played a role in economic well- being by adding to the level of money income at a given point in time while committing future income for repayment. It also can affect wealth accumulation if so much credit is used that debts are greater than assets (Bauer, Rettig 8 Danes, 1990). Kinsey 8 Lane (1978) investigated the effect of debts on perceived financial well-being of families. They found that debt-asset ratios were generally non-significant, indicating that family units accept consumer credit as a normal part of doing business within the household. Increased debt-asset ratio was considered an indication of increased ability to incur and carry debt. Consumer credit was considered as one of several money management tools available to the family. Smaller family size was related to the probability of feeling better off, as was rurality, marital status and income. Being a single household head increased the probability of feeling worse off with a higher debt-asset ratio. Another aspect of financial management in families is who is responsible for the management. There has been a trend over the past few decades for the wife to assume growing responsibility for money management in the family. 56 Ferber 8 Lee (1974) studied young married couples over a period of three years. Over those three years, the number of wives serving as family financial officer increased and the number of couples acting jointly as the financial officer decreased. The couple was most likely to act jointly when they had a set goal for total savings. The wife was most likely to be the financial manager if she gave more priority to saving, was more concerned with high quality, more economy minded and more bargain minded. Neither education nor employment levels of the wife or husband made a difference in who was the financial manager. It was found that in those households where there was a substantial difference in savings priorities the wife was more likely to take charge. When wives were the financial managers, savings were more apt to be in more conservative forms such as insurance and savings accounts. When husbands served as the financial manager a higher proportion of income was saved and was more often saved in the form of real estate and negotiable securities. Crossman and Edmonson (1985) studied displaced homemakers to explore the personal and family resources that were most useful in helping these women assume the provider role and make a financial adaptation. The women were middle-aged and were either divorced or widowed. The measure to determine 'success' was the number of weeks the women took to reorganize their families' lives after the 57 event that required them to become the family provider. The personal resources that were most helpful to both groups of women were continuous employment (currently employed and employed both before and during marriage), financial support from at least two sources, higher educational level, and very good health. Women who had a high level of these personal characteristics were able to reorganize their financial lives in a shorter time than those women who had fewer personal resources. The more rapid organizers expressed satisfaction with their adaptation to their new role as financial provider in an average of 13 weeks as compared to an average of 24 weeks required by the other women in the study. Wilhelm and Ridley (1988) examined the influence of changes in financial management practices on individual level of stress reported by husbands and wives immediately following the layoff of the family's primary breadwinner. Six financial management factors were identified as having increased. These were increased borrowing or credit use, planning for major expenditures, keeping closer tabs on check writing and bill paying, increasing saving behavior and checking the ability to pay one's bills, keeping more detailed records of income and expenses and budgeting for daily expenses. The net worth of the participants was considered a coping resource of the couple. Only two of the financial management changes were associated with stress. 58 The attempt to increase saving behavior and higher net worth were associated with less stress for both husbands and wives. Planning for major expenditures was associated with more stress for the wives. This may be because it is the wives who tend to make the day-to-day financial decisions of the household (Ferber 8 Lee, 1974; Wilhelm 8 Ridley, 1988). The effectiveness of families' financial management practices has had limited study. Fewer than 20 % of families use written plans but over half of the families estimate future household income and expenses (Beutler 8 Mason, 1987; Titus, Fanslow 8 Hira, 1989). Formalized planners have been found to be in young, married, and well- educated households which are experiencing relatively high demands on their resources (Beutler 8 Mason, 1987). Education and employment experience were found to aid women household heads in organizing their financial lives, but had no effect on who managed the family finances in the married couple household (Crossman 8 Edmonson, 1985; Ferber 8 Lee, 1974). Satisfaction with savings was found to be dependent on a specific form of management behavior i.e. saving as opposed to spending, which in turn was dependent on household resources, income constraints and motivation to save. Level of saving was related to household size and education (Davis 8 Schumm, 1987). Objective outcomes such as net worth and preparation for financial emergencies were more often influenced positively by planning for income use 59 while the measured effect on satisfaction with income and level of living was limited (Beutler 8 Mason, 1987; Davis 8 Helmick, 1985; Titus, Fanslow 8 Hira, 1989). ural Wome nd F The effect of economic and political policies of the late 1970's and early 1980's had an impact on the financial and work demands of rural families. For many farm families the second half of the 1980's was a time of little promise. Many farm families were in financial difficulty because they had borrowed heavily, paid high interest rates, suffered severe droughts and/or floods, received lower farm prices, faced a decreased market for exports and been caught in a political battle to lower farm supports (Little, Prouix, Marlowe, 8 Knaub, 1987). Predictions for the future of farm families included a continued decrease in the number of total family farms, a gradual decline in economic well- being, increased off-farm employment (Scholl, 1986) and the reduction of local and governmental services as the result of lower community tax bases (Little et al., 1987). Farm family life has certain unique characteristics (Kain,1990). In farming the family is more closely associated with the productive processes than in most occupations. There is less separation of work and home. Income is unpredictable and personal skills and decisions may have less impact on income than the weather or shifts in 60 price supports. There is also a wide seasonal variation in work requirements and on farms with livestock, especially dairy cattle, there is very little flexibility in time demands of chores. Kain (1990) used National Opinion Research Center General Social Survey data to examine differences between farm and non-farm families in the United States. He found that farmers were more likely to be married than non-farm respondents. Farm women were less likely to have ever been divorced. The mean age of the farm respondents was older, and the family size was larger. Farmers were much less likely to have completed higher education degrees, and their mean number of years of education was lower. Farm women had more education than farm men. Family income was lower for farm than non-farm respondents. Farm women were less likely to work for pay than non-farm women, but this difference is decreasing. Women who live on farms are increasing their participation in off-farm employment. The labor force participation rates for farm women have increased twice as fast as the employment levels of non-farm women (Scholl, 1983). Walker and Walker (1987) found that the stress levels of women working off the farm in addition to their farm duties were higher than women not working off-farm. It has been suggested that because of more traditional values, employment of rural wives and mothers is perhaps more 61 conflictive and problematic than it is for urban women. Mertensmeyer and Coleman (1987) found that financial insecurity may decrease self esteem in rural parents as it is often essential that rural mothers of young children be employed even though it is incongruent with their values. Meiners and Olson (1887) studied the time use of rural and urban women. They found that the typical farm woman in the sample worked an average of 48.5 hours per week in the household, while rural nonfarm and urban women worked about 46 hours per week. Farm women spent more time in unpaid work (either as a volunteer or unpaid worker in the family business) than urban or rural nonfarm women. Farm women worked an average of 8.4 hours per week for pay. Ten percent of the farm women were engaged in paid work as farm laborers. For each hour of paid work, farm women reduced household work by one-half hour, but no reduction in household work was indicated for time spent in unpaid work. More rural elderly people live with a spouse than do urban elderly people. Even though more rural elderly persons are married than widowed, the widowed make up a large proportion of the rural elderly population (Mercier, Paulson, 8 Morris, 1988). Keith and Nauta (1988) compared rural and urban unmarried older persons regarding use of leisure time and its relationship to well-being. Health contributed most to well-being regardless of residence or gender. Both education and employment contributed to 62 greater happiness of rural women. Lower income did not diminish the happiness of rural persons, but it did for urban residents. Lower income did not act as a restraint for rural women in participation of leisure activities. Scott 8 Roberto (1987) found that rural residents reporting good health, higher perceived income and living in the country as opposed to a small town, had higher morale scores . £11m Economic well-being has been measured using objective and subjective indicators. Objective economic measures include family income, per capita income, net worth, per capita net worth, and debt/income ratios among others. Subjective measures include perception of income adequacy, satisfaction with family income, satisfaction with level of consumption, satisfaction with amount of savings available for emergencies and satisfaction with net worth, perception of change and expectations for the future. Most studies agree that there is a direct relationship between satisfaction with income and the amount of income. Fundamental values and behavior necessary for human cooperation and economic and social welfare are learned in the family. The health of the family system depends on adequacy of both economic and psychological resources. It is generally recognized that basic needs for food, clothing, 63 shelter and physical health must be met before consideration can be given to meeting higher level needs. Economic conditions over the life course vary by cohort. The effects of life events are cumulative and may influence the economic well-being of the cohort both objectively and subjectively. Locus of control orientation has been related to decision making and stress in times of increased financial and work demands. A relationship of locus of control orientation to money management practices may exist as those individuals who believe their actions produce results would be more likely to practice money management. The effect of increased use of money management practices to economic well-being is ambiguous. Increased use of money management practices has been related to increased demands on income, loss of income and to increased stress of wives. A positive relation between increased net worth and satisfaction with preparation for emergencies and the use of financial management practices has been found. Younger, more educated, two spouse households in the early stages of the family life-cycle used more money management practices than older, less educated, single parent households. The financial demands created by the economic and political climate in the late 1970's and early 1980's have had an influence on farm families. One result has been the exit from farming of many long term farmers. Another has 64 been an increase in off-farm employment by farm women and men and more part-time farming. The economic well-being of farm families will be challenged as a result of these changes, with some families making these transitions more easily than others. Many factors have been identified as relating to perceived economic well-being. Perception is important as it is the affective component of the relationship of objective conditions to satisfactions with economic conditions and their contribution to well-being. CHAPTER III METHODOLOGY This study explored the relationship of perceived economic well-being to objective measures of household and individual characteristics of rural female household financial managers.g The data utilized were collected for the NC-182 regional research project, "Family Resource Utilization as a Factor in Determining Economic Well-being of Rural Families”, subsequently referred to as the core stedy. Support for the core study was provided by ~ Agricultural Experiment stations and the California Cooperative Extension Service. Eight states are involved in the project: Arizona, California, Illinois, Indiana, Iowa, Kansas, Michigan and Minnesota. This chapter includes the selection and description of the core sample, the description of the study sample, the research questions and hypotheses, the research variables, procedures used, the instrumentation and the data analysis procedures. 65 66 Co am Se ec The sample was selected from rural counties in each state defined as economically growing or declining. The ceenty was defined as rural if at least twenty percent or more of the employed persons were engaged in the occupations of agriculture, livestock, forestry, mining and/or fishing. Counties were defined as economically growing or declining based on the per capita income change from 1979 to 1985. The percent of change was computed and the counties were ranked in order from high to low based on the per capita income change from 1979 to 1985. The list of ranked counties was divided into quartiles, and one county was randomly selected from the top quartile and one from the bottom quartile. These counties were labeled growing for the top quartile and declining for the bottom quartile. The sample in each state was randomly selected from a commercial directory service lise_which was updated in November before the sample was drawn in December, 1987. The listings were based on telephone directories that were supplemented with auto registration information and checked for double entry. The overall response rate for the eight states was about thirty-three percent. Useable questionnaires were returned by 2,510 persons identified as ”financial managers" and 1,348 persons identified as "other adult" in the A household. The questionnaires returned by the financial 67 managers were almost evenly divided between the two counties; 51.9% from the declining county and 48.1% from the growing county. Minnesota had 460 questionnaires returned by financial managers while the number of returned questionnaires from each of the other states ranged from 275 to 312. esc o C Sam Of the core sample, 49.2 percent of the financial managers were men and 50.8 percent were women, a total of 2,510. The age range for the financial managers was 18 to 97 years, with a mean age of 51.5 years and a median of 51 years. The financial managers were primarily white (91.6 percent). Native Americans and persons of Spanish descent were the largest groups of non-white respondents. The majority of the non-white financial managers were from Arizona and California. The average years of education attained by the financial managers was 12.8 years. The median income range was $20,000-$24,999 for all financial managers. Fifty-eight percent of the financial managers were employed or self-employed, 3.7 percent were unemployed, 9.4 percent were full-time homemakers and 28.6 percent were retired. Almost 72 percent of the financial managers were married, and 60 percent were in first marriages. The average length of marriage was 24.8 years. Unmarried 68 financial managers were widowed (12.5 percent), divorced (8.1 percent) or never married (6.9 percent). Preliminary analysis of each state's data has not shown significant differences in demographic characteristics between the growing and non-growing counties. Descriptive statistics on selected socioeconomic characteristics and attitudes regarding income adequacy do not differ greatly between the two counties within or among states. Thus it may be concluded that rural residents in the regional sample are a homogeneous group (Hira,1990). W me There were 1242 women from the eight states in the core sample. They ranged in age from 18 to 96 with a mean age of 56 years. About 51% were married, 20.5 % were widowed, 11.4% remarried, 10.7% were divorced or separated, and 6.6% had never married. The average years of marriage to the present spouse was 24 years. Household size ranged from 1 to 8 persons, with 26% having one person, 34% two persons, 33% 3 to 4 persons and 7% having 5 or more members. The mean years of education was 12.8 years. Fifty-five percent had finished high school or more. The average income for all women was in the $20,000 to $24,999 bracket. About 52% of the women were employed, 27% were retired, 17% were full time homemakers and 4% were unemployed. The 69 sample was 91% white, with Native Americans and persons of Spanish descent being the predominant ethnic groups represented. Almost 27% of the women reported that someone in the family had a health problem that was of concern. Descpiptiop pf Study Sample The women in the sample represent three specific cohorts that were selected from the 1242 female financial managers who responded to the questionnaire. The women in Cohort 1 reached the age of 18 during the depression years 1929 to 1934. They were 72 to 77 years of age in 1988, at the time of the study. The women in Cohort 2 reached the age of 18 in the years between 1950 and 1955 when the economy was booming. They were 51 to 56 years of age at the time of the study. The women in Cohort 3 were among the first of the Baby Boom cohort to reach 18 in the years between 1964 and 1969. This was a time of many societal changes and a questioning of previously held values. They were 37 to 42 years of age in 1988. For all three cohorts, the racial mix was 90.2% white and 9.8% non-white. . Tables 1-4 and Figures 2-3 present demographic data by Cohort. Table 1 reports the number of women in each cohort, the mean ages and percentage of total sample, and Table 2 presents marital status. 70 In Cohort 1, 69.1% of the women were not married, most of these as the result of being widowed. Over 70% of Cohort 2 and Cohort 3 were married. Age of Cohorts - 1988 [ N=350 ] Cohort l Cohort 2 Cohort 3 Age Range 72-7 7 5 1 -56 37—42 Median Age 74.2 53.5 39.2 # in Study 96 109 145 'FabfleJZ Marital Status by Cohort tumflunsauu. Cbmm1l Cbhm12 commas Than [ [1:94] [ ll=lo9 ] [ ll=l44 ] [ [1:347 ] a. %> a» ‘% Married 21.3 66.1 56.9 50.1 Remsrried 9.6 11.9 17.4 13.5 Widowed 62.8 9.2 .7 20.2 Separated/Divorced 3.2 12.8 18.8 12.7 anrhhufkd :12 00 63 as 71 [8 10’. [8 It: IS 18. [8 [8 IS III In. IM 13 13.1 .° o 0:6' I. I ... on 'o 6 llwtlxt‘llnltl \l/t' (H ('l liltin ltlllJl-l COHORT l [n = 96] — TOTAL [N = 350] Household Size by Cohort - “0.10.1“ “No.0 0.3 72 'llHItl. a... > .... a 0 u c :.:.. 's o c 34': O O o c s ..... a I I .- _4 g. gr -- -— -_- ’. v1 c t o s o ''''' ‘lll.l‘.'t‘ c c e 0:0:0 00000 c o n o 0 s ..... tlt‘t ) . I 0:0:0 ..... o o ... n:o:o o c o o c 0:0:0 ... o a l o o ... ttttt n n I a 0 0.5.. o':‘: 0:. ::: ... o . . ':':' ooooooooo ........... ... ooooooo 00000000 oooooooooooooo 0:. '''''''''' ''''''''''''''''' eeeeeee .................... 00000000000000 ..................... under 10.(XX) 2011!) 30. 000 40 .000 50 .000 60 .000 10.” - 19.999 -29 .999 - 39 .999 -49 .999 -59 .999 + over lllilIHK‘lllllkl lmtmlc In l)ul|.nu COHORT l [n = 83] COHORTZ [ n = 108] _ COHORT3[n=l4l] _ TOTAL [n=332] Household Income by Cohort 73 Figure 2 displays household size by cohort. Household size ranged from 1 to 8 members. About one-third of the women lived in two person households, with fewer than 9% living in households with 5 or more members. Figure 3 portrays the household income for each cohort. In Cohort 1 about 46% had income of 0 to $9,999, while another 32.5% had income between $10,000 and $19,9999. Less than 10% had incomes above $30,000. Almost 19% of Cohort 2 had incomes below $9,999, 50% had incomes between $10,000 and $29,999, 22% had incomes from $30,000 to $49,999 and almost 10% had incomes above $50,000. Only 10.6% of Cohort 3 had incomes below $9,999. About 47% were between $10,000 and $29,999, 24% had incomes between $30,000 and $49,000 with 8.5% having incomes above $50,000. Tables 3 and 4 present the figures for educational attainment and employment status of the cohorts. The mean years of education were 11.9 for Cohort 1, 12.4 years for Cohort 2 and 13.4 years for Cohort 3. The majority of the women in Cohorts 2 and 3 were employed while most of the women in Cohort 1 were retired. While most of the women in Cohort 1 reported that they were retired, it is not known if they had retired from employment or indicated "retired" because they were of retirement age. 74 Table 3 Educational Attainment by Cohort Education Cohort 1 Cohort 2 Cohort 3 Total [n=91 ] [n=109 ] [ [1:144] [n=344] % % % % 1-8 years 17.6 6.4 0.7 7.0 9-11 years 7.7 10.1 6.2 7.8 HS. Grad 37.4 47.7 39.3 41.4 Some College 26.4 23.9 33.1 28.4 College Grad 9.9 3.7 1 1.0 8.4 Post College 1. 1 8.3 9.7 7.0 Table 4 Employment Status by Cohort Emploment Status Cohort 1 Cohort 2 Cohort 3 Total [ ll=85 ] [ “=1“ l [ ll=l4l ] [ [1:332 ] % % % 96 Employed 4.7 67.0 78.0 55.7 Unemployed 1 .2 3. 8 5.0 3.6 Homemaker 4.7 22.6 16.3 15.4 Retired 89.4 6.6 .7 25.3 75 Resea V ’ab es This section cgntains the conceptual and operational definitions of the dependent and independent variables. Dependent Variables e ce ved conomi Well-Bein Conpeptual definition: The individual's feelings with regard to over-all satisfaction with the financial situation and expectations for the future. Qpeznpionai definitipn: Respondents were asked to respond to seven subjective items related to satisfaction with aspects of their financial situation, perception of change in financial status and perception of current income adequacy. Each question had five possible responses. The responses were summed to create an index of perceived economic well-being. The higher the score the greater the perceived economic well-being. The composite measure of perceived economic well-being (PEWB) was created using the procedure outlined below. The definitions of the subjective indicators used in the composite measure of PEWB follow the outline of the method used. Perceived Economic Well-being = PEWB PEWB - S + P + CH Range 4 to 20 S — Satisfaction with financial situation s-s,+sz+ss+s,/4 S1 == satisfaction with income (D1) S == satisfaction with material things (D2) a satisfaction with ability to meet emergencies(D3) 8, == satisfaction with net worth (D4) 76 S will range from 1 to 5 very dissatisfied dissatisfied mixed feelings satisfied very satisfied U'I-bUNH ll ll II I ll P= Perception of income adequacy (E3) Range 1 to 5 not at all adequate can meet necessities only can afford some of the things we want can afford about everything we want can afford about everything we want and still save money UIbUNH CH - Perception of changing financial situation CH = E1 + E2 Range 2 to 10 E1 = financial situation compared to 5 years ago E2 = expectations for financial situation in 5 years 1 = much worse 3 = about the same 5 = much better 2 - worse 4 = better w 3C Cpnpgp;nn1_dgfiinipipn: The level of gratification received from one's income and its uses. ngznpipnn1_definitipn: Respondents were asked to respond to four statements related to satisfaction with total income, satisfaction with the material things they have, the resources available for emergencies, and the amount of net worth. Each question had five responses that ranged from very satisfied to very dissatisfied. Responses were summed and divided by four. The higher the score the greater the satisfaction with the financial situation. W ggnpeppnn1_gefiini;ipn: The individual's feelings about how well current income meets the household's needs. 77 ona ef nitio : Perceived income adequacy was measured on a five point scale: 1) Not at all adequate, 2) Can meet necessities only 3) Can afford some of the things we want 4) Can afford about everything we want 5) Can afford about everything we want and still save money. an ' a 'a o ua e i : The individual's feelings as to whether the family's present financial condition is better than it was in the past and if there is an expectation of future improvement. e a e t' : Respondents were asked to respond to two items: 1) a comparison of their current financial condition to five years ago and 2) the expectation for their future financial condition. Each question had five possible responses which ranged from much better to much worse. A higher score reflected better current conditions and expected improvement in the future. This variable has twice the weight of the others. QQDQIS gpnpgppnnl_ggfinipipn: A cohort is a group that moves along together through the life course and thus experiences historical events at the same age (Clausen, 1986). ngzg;ionn1_definitipn: Three cohorts are defined. Cohort 1 reached the age of 18 in the years 1929 to 1934 and 78 were 72-77 years of age in 1988 (the time of the study); Cohort 2 reached the age of 18 in the years 1950 to 1955 and were 51-56 years of age in 1988; Cohort 3 reached the age 18 in the years 1964 to 1969 and were 37-42 years of age in 1988. Educational_lexel e u de n ' n: The number of years of school attended. t' a e ' ’tio : The respondent was asked to indicate the highest number of years in school completed. ”Marital_§tatu§ £2nceataal.definition= The respondent's current marital situation. Qpeznpipnnl_definipipn: Respondents were asked to indicate marital status: first marriage or a remarriage, separated, widowed, divorced or never married. n I e tus gpnggppnnl_dgfinipipn: The level of participation in paid work. ngznpipnn1_ggfiinipipn: The respondent was asked to indicate if she was employed, unemployed, a homemaker or retired. Base_or_ethnio_baokgroand gpngepgnn1_gefiinitipn: The group of people with whom one identifies based on genetic, nationality or geographical characteristics. 79 Qpennpional definition: Racial or ethnic background was indicated by self-selecting from the following responses: Black, White, Native American, Asian or Pacific Islander, Spanish descent (Mexican, Puerto Rican, Chicano, other Spanish) or other as specified. Manager Characteristics-Behavioral Lpgns pf ppntnpi pzientatipn Conceptual definition: The effect of an individual's perception of control. Internal locus of control is defined as the belief that one can determine one's own fate within limits. External locus of control is defined as the belief that external forces such as luck or chance determine fate. o a e ’ ' ' : The respondents were asked to indicate the extent to which they agreed or disagreed with the following statements relating to locus of control orientation: F1 When I make plans, I am almost certain that I can make them work F2 It is not always wise to plan too far ahead because many things turn out to be a matter of good or bad fortune anyhow F3 Many times I feel that I have little influence over the things that happen to me F4 What happens to me is my own doing F5 My financial situation depends on my control of the situation F6 It is impossible for me to believe that chance or luck plays an important role in my life F7 Sometimes I feel that I don't have enough control over the direction my life is taking 80 F8 Sometimes I feel that I don't have enough control over the family income Five responses ranging from 1 = strongly disagree to 5 = strongly agree were offered. Responses were recoded for questions F2, F3, F7 and F8 so that the higher number reflects internal locus of control beliefs. When the responses were summed, the higher the score the stronger the internal locus of control orientation. The instrument consisting of the statements F1, F2, F3, F4, F6, and F7 developed by Bugaighis 8 Schumm (1983) was abbreviated from Rotter's (1966) 29 paired-item scale used to assess internal-external locus of control beliefs. Two additional statements, F5 and F8, were added to obtain information about the specific domain of family financial situation. W gpnpeptnn1_definipipn: The planning, controlling and evaluating of activities related to money management, consumer buying and credit use. Money management practices consist of planning and record keeping relating to use of money. Consumer buying is the use of money related to the buying of goods. Credit use is the number of sources of credit used and debt repayment practices. ngpngipnn1_gg1inipinn: Respondents were asked to respond to the questions below about how they handle their finances. Five responses that ranged from 1= 'never' to 5 = 'most of the time' could be selected. The responses to the 81 statements were summed so that a higher score indicated more participation in tasks related to management. Financial Management Practices = FMP FMP= MMP + CB + CU Range = 13 to 65 Money Management Practices = MMP MMP 8 Ala + Alb + Alc + Ald + Alg Range = 5 to 25 How often do you Ala Make plans on how to use money? Alb Save on a regular basis for goals? Alc Write down where money is spent? A1d Use a written budget? Alg Keep bills and receipts where you can find them? Consumer buying = C8 CB= Ale + Alf + Ali + A1j + Alk Range=5 to 25 How often do you Ale Evaluate spending on a regular basis? Alf Evaluate your needs before you buy? Ali Buy on impulse? A1j Feel sorry you bought something? Alk Make a list before you shop? The codes were reversed for Ali, Alj so that 1 = most of the time and 5 = never, thus the higher score indicated good management practices. Credit use - CU CU- Alh + Alo + A2 Range-3 to 15 How often do you Alh Pay interest on charge accounts? A10 Make only minimum payments on charge accounts? A2 Types of credit -- can select up to 7 sources which range from banks to pawn shops. 82 The codes were reversed for Alh, Alo so that 1 = most of the time and 5 = never, thus the higher score is an indication of good management practices. A2 was recoded in the following manner: 1 = do not use credit 1-2 sources 3-4 sources 5-6 sources 7 sources UltDUN HouseholdJncome Qpnceptuai definitipn: The dollar amount of income before taxes available to the household in 1987 from all sources. Qpeznpipnn1_definipipn: The respondent was asked to indicate the income categories that represented the household's income. ’W e a ' t : The dollar amount of the household income in 1987 available per person. ngpnpipnn1_gefiinipipn: The income category the respondent indicated as representative of the household's income was divided by the number of persons reported to be in the household. SourceLoLincome e ' it : The activities and agencies that supply income to the household. Qpe:atignnl_defini§ipn: A list of income sources was given and the individual was to check all that provided 83 income to the family. The responses were counted to indicate number of sources of income. IDSBIQDQQ £9n28232a1_gefinitipn: Protection purchased by the family against the risk of loss. a e ' o : The respondent was asked to indicate the types of insurance that the family had. The responses were counted to provide a total number of insurance types held by the household. "mm gpnpeptnn1_defini§ipn: The difference between the ._-—..—_.... I. Assets include any material possessions the respondent owns or is buying, while liabilities are the obligation or debts of the respondent. ngznpipnnl_ggfiinipipn: Respondents were asked to check one category representing household assets and one category representing household debts. The mid point of the household debt response was subtracted from the midpoint of the asset response to get a dollar amount for net worth. Wig gpnggppnnl_ggfiinipipn: The ratio of reported debt to reported household income. Qpe:atignal_definitipn: The midpoint of the category of total debt in dollars indicated was divided by the 84 midpoint of the category of total household income reported in dollars. W e tua e ' 't'o : The presence of a chronic health problem for any member of the household. ng;npipnnl_gefiinipipn: The respondent was asked if anyone in the household had a chronic health condition and if this created a financial hardship for the family. Health will be coded as 1 = no health problem, 2 = presence of a health problems and 3 = health problem which is a financial hardship. Health was recoded as a dichotomous variable where 0 = no health problems and 1 = the presence of health problems whether or not the problem caused a financial hardship, for use in the regression analysis. 893M113? ce tua e ' ' n: The number of persons living in the residence. ng;npipnni_§gfiinipipn: The respondent was asked to list the age of each person living in the household. These responses were counted to obtain household size. 85 sea ti 5 The research questions and hypotheses directed by the review of the literature follow: 1. Do cohorts differ on the following household and manager characteristics? a. income b. number of sources of income c. number of types of insurance d. net worth e. debt/income ratio f. health status 9. household size h. education 1. marital status j. employment status k. race 1. locus of control orientation m. financial management practices Hla: Cohort 2 will have higher incomes than either Cohort 1 or Cohort 3. Hlb: There will be no difference among the three cohorts on the number of sources of income. ch: There will be no difference among the cohorts on the number of insurance types. Hld: The net worth of Cohort 2 will be higher than the net worth of either Cohort 1 or Cohort 3. Hle: Cohort 1 will have a lower debt/income ratio than Cohort 2 or 3. Hlf: Cohort 1 will have more health problems than Cohort 2 or Cohort 3. H19: The household size of Cohort 3 will be larger than Cohort 1 or Cohort 2. th: Hli: Hlj: Hlk: H11: Hlm: 86 Cohort 3 will have more years of education than Cohort 1 or Cohort 2. Fewer women in Cohort 1 will be married than in Cohort 2 and Cohort 3. More women in Cohort 2 and Cohort 3 will be employed than in Cohort 1. There will be no differences in the racial composition of the three cohorts. There will be no differences among the cohorts in internal locus of control orientation. Cohort 3 will be involved in more financial management practices than Cohort l or Cohort 2. Is there a difference in responses among the cohorts to the indicators used to create the composite measure of perceived economic well-being as listed below: a. satisfaction with income b. satisfaction with material things c. satisfaction with ability to meet emergencies d. satisfaction with net worth e. perceived income adequacy f. financial condition compared to five years ago 9. expectation of change in financial condition in the next five years h. perceived economic well-being H2a: H2b: There will be no differences in satisfaction with income among the three cohorts. There will be no differences in satisfaction with material things among the three cohorts. 87 H2c: There will be no differences in satisfaction with the ability to meet emergencies among the three cohorts. H2d: Cohort 1 and Cohort 2 will be more satisfied with their net worth than Cohort 3. H2e: Cohort 2 and Cohort 3 will report more improvement in financial conditions compared to five years ago than Cohort 1. H2f: Cohort 1 will have less expectation of improvement in financial conditions in the next five years than Cohort 2 or Cohort 3. H29: Cohort 1 will perceive their income to be more adequate than Cohort 2 and Cohort 3. H2h: There will be no differences in perceived economic well-being among the three cohorts. What is the relationship between locus of control orientation and financial management practices? H3: Financial managers who are more internal in locus of control orientation will be involved in more financial management practices. Do financial management practices differ by income levels? H4: Financial managers in the middle income levels will be involved in more financial management practices. 88 How much variation in perceived economic well-being is explained by income, net worth, and debt/income ratio, for the sample as a whole and for each cohort? H5a: Of the objective economic measures, income will H5b: explain the most variation in perceived economic well-being, followed by debt/income ratio and net worth. The variation in perceived economic well-being explained by the objective economic variables will not differ among cohorts. How much variation in perceived economic well-being is explained by locus of control orientation and financial management practices separately and combined for the sample as a whole and for each cohort? H63: H6b: H60: The portion of variance of perceived economic well-being explained by locus of control orientation is expected to be small but significant. Financial management practices will explain more variance in perceived economic well-being than locus of control orientation. Locus of control orientation and financial management practices will explain more variance in perceived economic well-being for Cohort 3 than for Cohort 1 or Cohort 2. 89 7. Which manager and household characteristic variables are the best predictors of perceived economic well- being for the sample as a whole and for each cohort? H7a: Household characteristics of income, net worth, health status and household size, and manager characteristics of education, marital status and financial management practices will contribute the most to the explanation of variance in perceived economic well-being. H7b: Income is expected to explain less variance for Cohort 1 than for Cohort 2 and Cohort 3. H7c: Health status is expected explain more variance for Cohort 1 than for Cohort 2 and Cohort 3. H7d: Financial management practices and household size are expected to explain more variance for Cohort 3 than for Cohort 1 and 2. ce u s The core study, Project NC-182, used a survey research .FEFEEQL— Data were collected from rural households by means of mailed questionnaires. A modified Dillman method of data collection was used. _The_prgcedure consisted of four steps. A pgstcerd was sent to inform potential respondents that a questionnaire was to be mailed. The second mailing consisted of a COYEFPLEEESI explaining the project and inviting participation, an 90 informed consent form, a form to enter a drawing for those /--F\\_ “-“fl- -— ‘1. _ whose questionnaires were returned by a certain date (in Michigan), two questionnaires (one for the financial manager and one for another adult, if any, in the household) and two postage paid self-addressed envelopes. The third_neiling was a reminder postcard which was sent one week after the survey mailing. Those households from which there had been no response received 3,5395Fh and final mailing which contained a new set of questionnaires. Prior to conducting the core study the procedures and instruments were approved by the University Committee on Research Involving Human Subjects (UCRIHS). This research was reviewed and received approval from UCRIHS. Additionally, the NC-182 Research Committee approved a request to use the regional data for this study. (See Appendix B). Beseapgh Instrnmentation-Qppe Stngy The research instrument utilized for this study was developed by the committee members of the Agricultural Experiment Station regional research project NC-182, North Central Region.) The questionnaire represented the interests and expertise of the various researchers as they related to the over-all project. A section also was added to each state's version of the questionnaire that was of particular interest to the researchers from that state. 91 The research instrument consisted of tge self- .“ "' "'\“.-._....._ edninisterednqeeetionnaires nailed to each household. Onefiquestionnaire was to be completed by the self-selected financial manager of the household.p The second questionnaire was to be completed by one other adult in the household, if any, and did not contain questions regarding household members, household income, assets and debts or children who were no longer living at home. Otherwise, the questionnaires were identical. ’The research instrument for the financial manager took approximately thirty to forty-five minutes to complete. The instrument was pre-tested in each state and the results compiled to identify problems. .Aflpiletfletudy was then conducted in Minnesota to further test the usefulness of the instrument. One of the objectives of the core study was to develop indexes and measures of economic well-being and resource utilization. The analysis of this data is in its preliminary stages. One outcome that is being pursued by the research team is the development of a medel for assessing economic well-being. Appendix C contains a copy of the full questionnaire and the selected questions used for this research. 92 sea 5 - Questionnaires completed by female financial managers in the selected three cohorts were used. Items pertinent to this study are presented below. The measure for locus of control orientation (questions F1 through F8, page ) consisted of two parts. Items F1, F2, F3, F4, F6, and F7 are from Rotter's (1966) paired item scale to identify internal-external locus of control. Items F5 and F8 were devised to identify locus of control orientation in the domain of family income. Three paired items from Rotter's scale were used to create a separate six-item Likert-type scale (Bugaighis and Schumm, 1982; Edwards and Booth, 1976). The three items for which a response of 5 indicated the most external belief were recoded prior to analysis. The higher score reflected internal locus of control orientation. Reliepilityaas measured by Cronbach's alpha was .66 and had a correlation of .62 with Rotter's 23 paired-item scale. The reliability of the single item measure of locus of control had an estimated Cronbach alpha between .30 and .40. Evidence is ‘offered that the single-item measure provides a valid alternative brief measure of locus of control (Bugaighis and Schumm, 1982). Davis and Helmick (1985) used a simple additixc. index to measure financial satisfaction. They asked fi/liii respondents in six samples to use a 7 point scale to 93 indicate satisfaction or dissatisfaction with their consumption level, wealth and financial security. The scores were summed to create a measure of financial satisfaction. A coefficient of reliability (Cronbach's alpha) was computed to assess the reliability of the summed variables as an estimate of the case's true score. The coefficient of reliability for the index was over 0.74 for each of the six samples in their study, indicating a high level of common variance. This supports the assumption that all the items included in the measure of financial satisfaction are measuring different aspects of the same underlying dimension. The questions D1 through D4 on page 1 of the survey instrument were used to measure the dimension of financial satisfaction. That these questions measure financial satisfaction is supported by the Davis and Helmick study (1985). ces Indices were built for the dependent variable, perceived economic well-being (PEWB), and the two independent variables, locus of control orientation (locus) and financial management practices (FMP). A reliability procedure was used on the set of statements believed to measure each variable. (See operational definitions for the items included in each index.) Reliability was measured by Cronbach's alpha, a correlation coefficient of how well the 94 scale would repeatedly measure the variables in question. A coefficient of .60 or above is considered an indication of the usefulness of the scale in measuring the variable (Nunnally, 1978). PEWB included seven statements that were believed to measure that variable. Reliability for the scale using all seven statements was .85. The locus of control variable consisted of eight statements reflecting internal or external control orientation. All statements were coded so that higher scores reflected a more internal locus of control orientation. Cronbach's alpha for the scale was .66. No statements were dropped. The financial management practices index was built in three parts, credit use, consumer buying, and money management. Reliability was increased by eliminating 'the number of credit sources used' and 'sorry you bought something' from the index. The resulting correlation coefficient was .65. Estimates of reliability are based on the average correlation among items within a test and concern 'internal consistency'. Coefficient alpha is the basic formula for determining the reliability based on internal consistency. It sets an upper limit to the reliability of tests constructed in terms of the domain-sampling model. If the alpha is very low, either the test is too short or the items 95 have very little in common. Coefficient alpha provides a good estimate of reliability in most situations, since the major source of measurement error is due to the sampling of content. The reliability coefficient is used to summarize the amount of measurement error expected from using the instrument, and a satisfactory level of reliability depends on how the test is being used (Nunnally, 1978). Data na sis ocedu es Each state coded its own original data and entered it into the micro-computer using SPSSPC data entry. Missing “ha—......“ _H___~ “5......- data were coded as 9 and an 8 was recorded if the question did not apply to that respondent. .Frequencies were computed on all variables as a check for errors in data entry. A disk containing the Michigan data was sent to Iowa State University, and a tape containing data from each of the eight states was returned. Further data cleaning was done at the time the regional tape was made. A data‘file was constructed which combined the data from the eight states.~\It_included the variables used in this study for all female financial managers from each of the eight states.” Only data for the women from the three cohorts previously defined were used in this study. Cohort 1 had 96 cases, Cohort 2 had 109 cases, and Cohort 3 had 145 cases for a total N=350 cases. The actual number of cases differed on some items due to respondent omission. 96 The core study was designed to compare resource use by families in growing and declining counties. Preliminary analysis has not revealed significant differences between the counties on the variables used in this study. The data from the eight states will be treated as one data set as preliminary analysis has not found state differences that would bias the results. "The variables used were primarily nominal, ordinal and interval. For purposes of statistical analysis some of the ordinal data were treated as interval data. _Likert-type‘ response scales were used in several items of the research If, instrument. This is not considered to be true interval data because the differences between the response points may not always be equal. However, Likert response scales are frequently treated as interval data for analysis. _Althoegh some small error may result when ordinal variables are treated as if they are interval, this is offset by the use “—5 -__ of more powerful, more sensitive, better developed, and more clearly interpretable statistics (Babbie, 1986; Blalock, 1980; Nunnally, 1978);("Descriptive statistics, frequencies and means, will be used to summarize the data and determine whether normal distribution occurredIC Data analysis procedures for each research question follow. Question 1: Do cohorts differ on the manager and household characteristic variables? 97 The hypothesis related to Question 1 is concerned with differences among the cohorts and the household and personal characteristics used as independent variables. Indices were constructed for the financial management and locus of control variables in question one. The items included are listed in the operational definitions. The scale procedure in SPSSPC was used to identify which items would be included in each of the indices. The procedure provided a)‘i§EEE:» item correlations--including the mean, minimum, and maximum, b) item-total correlations--which is a Pearson r between each item and the overall index score and c) an overall internal consistency measure, Cronbach's alpha (Babbie, 1986; Norusis, 1988; Nunnally, 1978). (ANOVAQwas used to test for differences among the cohorts 6; the independent variables. The analysis of variance formula is: Y 8 + n + e where Y a the independent variable being tested, = the population mean, 9 s the effect of the cohort to which one belongs, and e = the residual error. //Analysis of variance examines the variability in the sample to determine whether there is a reason to believe that the group means are unequal.5?The statistical test for the hypothesis that the means of all the groups are equal is the E statistic. xThe z statistic is used to determine whether a significant difference exists between the means. 98 A significant 2 indicates that the means are unequal but does not specify where the differences are. To test for which means were different in the event of a significant F statistic, p tests were used for the hypotheses where a difference was expected. For those hypotheses relating to variables where no difference was expected, a Tukey Range Test of multiple means was used:-*The Tukey Range Test begins by rank-ordering the means in order of their size and then comparing the largest pairwise difference. When a nonsignificant range is encountered with the largest mean, the next largest mean is tested against the smallest mean. The procedure continues until there are no more significant differences. <€The use of parametric statistics assumes that the observations are from a normally distributed population and that the observations are random samples from that population (Babbie, 1986). The data were tested for normality of distribution using procedures available in SPSSPC. ’The assumption of a random sample was met through the sampling procedure used. KgThe assumptions of the analysis of variance model are 1) continuous dependent variables with equal appearing intervals, 2) a sample drawn from a normally distributed population 3) independence of observations, and 4) homoscedasticity- the homogeneity of variance (Glass 8 Hopkins, 1984). The first two assumptions have been 99 discussed earlier and were assumed to be satisfied. '"Independence of observation was assumed to be met as the questionnaires were to be completed by one individual without interaction with any other person in the study. Further, each respondent could belong to only one cohort. The homogeneity of variance was tested to determine if the .- fourth assumption was met. The Oneway procedure in SPSSPC was used to produce a summary description of the y.§I§;EEE\ which was used to test for similarity of variance. One measure that can be used is Cochran's C. If the significance level associated with the value of Ceehran's C_ is small, ANOVA should not be used. In practice if the number of observations in each of the groups is fairly similar, analysis of variance gives good results even if the / normality assumption is not met (Norusis, 1988). Question 2: Is there a difference in responses among the cohorts to the indicators used to measure personal economic well-being? related to it for differences among the cohorts on each of the subjective indicators that comprise the dependent variable of perceived economic well-being. The formula used was: Y = + g + e where Y = a subjective indicator, = the population mean, n a the effect of the cohort to which one belongs and g = residual error. 100 Question 3: What is the relationship between locus of control orientation and financial management practices? a CorrelationFmeasured by Pearson product moment/was used to test the relationship between locus of control orientation and financial management measures. The Pearson correlation coefficient 1 is a measure of association and indicates the strength and the direction of the relationship. Values of 1 range from -1, a strong negative linear relationship to a +1, representing a strong positive relationship. Tnemeeeeaniens related to the use of this measure of correlation are linearity, a random sample, normal distribution and interval level data. The data were plotted to test for linearity. The other assumptions were discussed previously. ' Crosstabe in SPSSPC were used to construct a contingency table to further study the relationship of the two variables, locus of control orientation and financial management practices, using financial management practices as the dependent variable. The,Chi-square statistid?enables the researcher to test whether the observed differences in the sample are sufficiently large to refute that the two variables are independent. The higher the chi-square value, the less probable it is that value can be attributed to sampling error alone. Question 4: Do financial management practices differ by income levels? 101 For Question 4, ANOVA will be utilized to test the / ... ..____... hypotheses related to it for differences in financial management practices and income levels. Question 5: How much variation in perceived economic well-being is explained by income, net worth and debt/income ratio, for the sample as a whole and for each cohort? The hypotheses related to Question 5 were tested using linear regressiOn to determine the amount of variation in the dependent variable, perceived economic well-being (PEWB), explained by income, net worth and debt/income ratio. The regression was run for each variable,gfpr the sample as a whole and for each cohort. The regression was, rerun adding the variables all at once for the sample as a whole and for each cohort. The regression formula used was: Yig+nx1+nzxz+n3x3+e where Y - the dependent variable PEWB, n = the intercept between PEWB and the economic variables, p s the increase or decrease in Y for a one-unit change in X, X1={/income, X2== net worth, x, = debt/ income ratio and e is that-Residual error. The use of regression analysis assumes a 1) linear relationship between the predictor(s) and the dependent variable, 2) normal distribution of errors, 3) homoscedasticity-the variance of the error term is constant, and 4) errors are independent (Lewis-Beck, 1980):\’ 102 The SPSSPC regression procedure was used to test that the basic assumptions were not violated. The output included seatterplots and histograms of residuals to test for normal distribution of errors and consistency of variance./’ Question six: How much variation in perceived economic well-being is explained by locus of control orientation and financial management practices separately and combined for the sample as a whale and for each“cchort? 'Linear regression analysis/was used to describe variation in PEWB related to locus of control orientation and financial management practices. The analysis was run_ separately for each variable, then for the two variables together, for the sample as a whole and for each cohort. Question 7: Which manager and household characteristic variables are the best predictors of perceived economic well-being for the sample as a whole and for each cohort? Analysis of Question 7 usedimhltiple_regression, ypreeeduresp7 The independent variables marital status, employment, health, and race were recoded to dichotomous variables for the regression analysis associated with Question 7. The correlation matrix of all independent variables was examined to determine which indicator to enter ..4'.‘ first into the equation. Si:forced entry procedure which /“ included all of the independent variables was used to \ ..-- ——-~*.. - -— —-—--— a 7—— a-“ _ ___,,. kfi—n‘ measure the contribution of each independent variable to 103 perceived economic well-being. The regression analysis was - _ __ .. _-—_.——.———_.—-—~ run fpur times, once for each cohort and once for the sample as a whole.’ The regression equation and independent variables used for Questions 7 were as follows: Y=g+n1x1+nzxz ....+pnxn+g where: Y = PEWB perceived economic well-being of the particular sample, n = value of Y if X is equal to 0, p = change in Y for each unit of X when all other independent variables in the model are held constant, X .....X = the values of the independent variables, and e = th9e error term. The independent variables used in the analysis are: X1 = income 15 categories X2 = net worth (continuous assets-debts) X3 = debt/income ratio continuous X, = health dichotomous variable 0 = no 1 = yes X5 = marital status 0 = non-married 1 = married X6 = financial management practices 11 to 55 X7 2 household size 1 through 8 Xe = locus of control orientation 8 to 40 X9 = education number of years Xw:= insurance sources 1 through 9 X11 = sources of income 1 through 15 Xn‘= employment 0 = nonemployed l = employed X13 = race 0 = nonwhite 1 = white Analysis of Question 7 used the stepwise multiple regreseion_procedure to determine whiCh variables were-the best predictors. The stepwise procedure enters independent variables one at a time. The independent variable that explains the greatest amount of variance in the dependent variable will be the first to be entered into the equation. 104 The rest of the independent variables will be entered into the equation such that the variable which explains the greatest amount of variance unexplained by the variables already in the equation enters the equation at each step (Babbie, 1986; Norusis, 1988). The analysis was completed for each cohort and for the sample as a whole. \ Multiple regression analysis results in a regression equation which estimates the values of a dependent variable for the values of several independent variables. BE measures the proportionate reduction in total variation of the dependent variable associated with the set of independent variables. The closer EE is to the value 1, the greater is the association between the set of independent variables and the dependent variables. 33 increases with the addition of each independent variable and tends to be an optimistic estimate of how well the model fits the population. The statistic adjusted 3: tries to correct 33 to more closely reflect the goodness of fit of the model in the population. ‘ The standardized regression coefficients (beEee) which indicaterfhéigelegige importance of the independent: veriables were used to identify the independent variables that most affected perceived economic well-being.1,The significance of the coefficients were calculated using the p statistic for the standardized regression coefficients and E-ratio for the coefficient of determination. These 105 procedures were used to accept or reject the hypotheses for questions 5, 6, and 7. SPSSPC, a statistical software package was used for analysis of relationships between variables. Hypotheses were supported at p < .05 level. CHAPTER IV RESULTS The findings are presented in seven sections corresponding with the research questions. Hypotheses were described as supported when p <.05 level of significance. The first and second questions examined the independent variables and the components of the dependent variable, perceived economic well-being, for differences among the cohorts. Question three examined the relationship between locus of control orientation and financial management practices, and question four explored the relationship of objective financial measures to financial management practices. Questions five through seven dealt with the relationship of the independent variables to perceived economic well-being for each of the cohorts and for the entire sample. Research Question 1 Do the cohorts differ on the managerial and household characteristics included in the study? Table 5 provides the variance and Table 6 gives the means for all of the independent variables, except marital status, health status, employment status and race, for each 106 cohort. differences among the cohorts. each hypothesis. Table 5 107 Analysis of variance was used to test for The results are given for Response Ranges of Independent Variables by Cohort Income Sources of Income(#) Insurance Typed!) Net Worth Debt/Income Ratio Household Size Education (yrs.) Locus of Control Financial Mgt. Practices Cohort l [n=96] $2500 to $55000 l to 5 1 to 8 -$5250 to $200000 0 to 3.5 l to 4 1 to 17 15 to 33 28 to 55 Cohort 2 [ n=l69 ] $2500 to $ 90000 1 to 7 l to 7 -$75000 to $200000 0 to 8.3 l to 8 1 to 20 14 to 37 25 to 55 Cohort 3 [ n=145 ] $2500 to $90000 1 to 8 l to 8 -$81250 to 3100000 0 to 8.3 l :07 8 to 21 13 to 37 20 to 53 108 'Fabdeifi Means of Independent Variables by Cohort Cohort l Cohort 2 Cohort 3 All Response [ n=96 ] [ n=169 ] [ n=l45 ] [ n=350 ] Range Income $1 4,307 $25,694 $28.91 8 $24,217 2500 to 90.000 Sources of Income(#) 2.5 2.5 2.3 2.4 0 to 15 Insurance Types“) 4.3 4.9 4.8 4.7 o m 9 Net Worth $23,936 $23,670 $9,342 51 6.681 42.500 to 200.000 Debt/Income Ratio .28 .72 .86 .70 0 to 8.3 Household Size 1.4 2.3 3.5 2.5 1 to 8 Education (yrs.) 11.9 12.4 13.4 12.7 0w22 Locus of Control 25.8 25.8 26.1 25.9 8 w 40 Financial Mgt. Practices 43.1 41 .3 40.1 41.0 11 to 55 file Cohort 2 will have higher incomes than’either Cohort 1 or Cohort 3. This hypothesis was not supported by the data. Both Cohort 2 and Cohort 3 had incomes significantly higher than Cohort 1, 2(2, 329) = 23.64, p < .0000, but there was no statistically significant difference between the incomes of Cohort 2 and Cohort 3. More women in Cohort 1 (14%) than in the other cohorts did not provide income information. nib There will be no difference among the cohorts on the number of sources of income. The data supported this hypothesis as no significant differences were found among the cohorts in the number of sources of income, {(2, 347) a 1.3881, n = .25. However, the types of income varied among the cohorts. 109 81c There will be no difference among the cohorts on the number of insurance types. The data did not support this hypothesis. Cohort 2 and Cohort 3 reported more types of insurance than did Cohort 1, {(2, 339) = 4.8468, g < .01. 81d The net worth of Cohort 2 will be higher than the net worth of either Cohort 1 or Cohort 3. Analysis of differences in net worth indicated that Cohort 1 and Cohort 2 were very similar in net worth, with Cohort 1 being slightly more than Cohort 2. The hypothesis, therefore, was not supported by the data. However, the net worth of both Cohort 1 and Cohort 2 was significantly higher than the net worth of Cohort 3, {(2, 274) = 4.9205, p < .01. H1e Cohort 1 will have a lower debt/income ratio than Cohort 2 or 3. The debt/income ratio of Cohort 1 was significantly lower, {(2, 283) = 6.4932, 2 < .002, than Cohort 2 or Cohort 3. The hypothesis was supported. H1f Cohort 1 will have more health problems than Cohort 2 or Cohort 3. The data did not support the hypothesis. A health problem that caused concern was reported more often by Cohort 2 than Cohort 1 and Cohort 3. The number of health problems reported by Cohort 2 was significantly greater than the number reported by Cohort 3, {(2, 297) = 3.9272, 9 < .03. 110 819 The household size of Cohort 3 will be larger than Cohort 1 or Cohort 2. The household size of Cohort 3 was significantly larger than the household size of Cohort 1 or Cohort 2, which supported the hypothesis. The household size of Cohort 2 also was significantly larger than that of Cohort 1, {(2, 347) = 123.5580, 2 < .0000. 31h Cohort 3 will have more years of education than Cohort 1 or Cohort 2. The number of years of education attained by Cohort 3 was significantly greater than the number of years of education for Cohort 1 and Cohort 2, {(2, 342) = 10.7350, 9 < .0000. The hypothesis was supported by the data. H11 Fewer women in Cohort 1 will be married than in Cohort 2 and Cohort 3. This hypothesis was supported by the data. More women in Cohort 2 and Cohort 3 were married than were the women in Cohort 1, {(2, 344) = 36.3043, 9 < .0000. H1j More women in Cohort 2 and Cohort 3 will be employed than in Cohort 1. More women in Cohort 2 and 3 were employed than in Cohort 1, which supported the hypothesis, {(2, 329) = 97.4508, 9 < .0000. 111 H1: There will be no differences in the racial composition of the three cohorts. No significant differences in racial composition were found among the cohorts. The hypothesis was supported by the data, {(2, 345) = .1807, p = .8348. H11 There will be no differences among the cohorts in internal locus of control orientation. This hypothesis was supported by the data. No significant differences in locus of control orientation were found among the cohorts, {(2, 313) = .2474, p = .7810. 81a Cohort 3 will be involved in more financial management practices than Cohort 1 or Cohort 2. Cohort 1 used significantly more financial management practices than either Cohort 2 or Cohort 3, {(2, 280) - 12.8175, p < .0000. Cohort 2 used more financial management practices than Cohort 3, but the difference was not significant. The hypothesis was not supported by the data. Research Question 2 Is there a difference in responses among the cohorts to the indicators used to create the composite measure of perceived economic well— being? Table 7 provides the means of each of the items used in the index for the dependent variable, perceived economic well-being, by cohort and for the total sample. Analysis of variance was used to test the following hypotheses: 82a There will be no differences in satisfaction with income among the three cohorts. 112 The data did not support this hypothesis. The women in Cohort 1 were more satisfied with their total family income than the women in Cohort 2 and Cohort 3, {(2. 331) - 8.5648, g < .001. Means of Components of Perceived Economic Well-being by Cohort Variable Cohort l Cohort 2 Cohort 3 All (Range 1 toS) [ “=96 ] [ [1:169 ] [ [1:145 ] [ N=350 ] Satisfaction with: Income 3.59 3.14 2.98 3.18 Material Goods 3.93 3.72 3.37 3.62 Emergency Savings 3.56 2.93 2.38 2.85 Net Worth 3.54 3.17 2.88 3.13 Financial Condition: Compared to 5 yrs ago 3.08 3.14 3.41 3.24 Expectation in 5 yrs 2.81 3.16 3.76 3.32 Perceived Income Adequacy 3.35 2.94 2.75 2.97 32b There will be no differences in satisfaction with material things among the three cohorts. This hypothesis was not supported by the data. The women in Cohort 1 and Cohort 2 were significantly more satisfied with the material things they had and used than the women in Cohort 3, {(2, 335) 8 11.5164, 9 < .0000. 82c There will be no differences in satisfaction with the ability to meet emergencies among the three cohorts. 113 The women in Cohort 1 were significantly more satisfied with the resources available to meet financial emergencies than those in either Cohort 2 or Cohort 3. The women in Cohort 2 were significantly more satisfied with resources available to meet emergencies than those in Cohort 3, {(2, 334) = 30.3961, 2 < .0000. The hypothesis of no differences among the cohorts was not supported. H2d Cohort 1 and Cohort 2 will be more satisfied with their net worth than Cohort 3. Cohort 1 and Cohort 2 were significantly more satisfied with their net worth than Cohort 3. Cohort 1 also was significantly more satisfied with their net worth than Cohort 2. The hypothesis was supported by the data, {(2, 332) = 10.7905, p < .0000. Hze Cohort 2 and Cohort 3 will report more improvement in financial conditions compared to five years ago than Cohort 1. Analysis of variance to determine differences among the cohort in 'financial situation now compared to five years ago' indicated no differences among the cohorts. However, as {(2, 334) = 3.0889, 9 < .05, t-tests were performed between the means of the cohorts to test for differences. The t-test between Cohort 1 and Cohort 3 indicated that Cohort 1 reported less improvement than Cohort 3, §(227) = -2.21, p < .05. The observed 3(191) -.38, p =.705, was not significant when Cohort 2 was compared to Cohort 3. The hypothesis was partially supported by the data. 114 32: Cohort 1 will have less expectation of improvement in financial conditions in the next five years than Cohort 2 or Cohort 3. The hypothesis that Cohort 1 would have less expectation for financial improvement in the future than Cohort 2 and Cohort 3 was supported by the data, {(2, 333) = 367.9647, 9 < .0000. The expectations of Cohort 3 also were found to be significantly higher than the expectations of Cohort 2. 829 Cohort 1 will perceive their income to be more adequate than Cohort 2 and Cohort 3. As expected Cohort 1 did perceive their income to be significantly more adequate than either Cohort 2 or Cohort 3. The hypothesis was supported, {(2, 333) = 8.5583, 9 < .001. 82h There will be no differences among the cohorts in perceived economic well-being. This hypothesis was supported by the data as no significant differences in perceived economic well-being were found among the cohorts, {(2, 288) = 2.6276, p_< .074. Research Question 3 What is the relationship between locus of control orientation and financial management practices? 83 Financial managers who are more internal in locus of control orientation will be involved in more financial management practices. A Pearson correlation was completed for the locus of control variable and financial management practices. The 115 correlation coefficient was .110, p = .073, which was not significant. However, the correlation coefficient for financial management practices associated with credit and locus of control was .164, p < .01, indicating more internal locus of control was associated with a greater use of financial management practices associated with credit. Contingency tables did not produce any significant differences between locus of control orientation and financial management practices. The hypothesis was not supported by the data. Research Question 4 Do financial management practices differ by income levels? H4 Financial managers in the middle income levels will be involved in more financial management practices. Analysis of variance was used to test for differences in use of financial management practices by income levels. No differences were found using household income as the independent variable. Household size was added as an independent variable, but did not affect the finding of no differences. Per capita income was computed by dividing income by household size and used as the independent variable to test for differences in financial management practices, but no significant differences were found. The hypothesis was not supported by the data. 116 Research Question 5 How much variation in perceived economic well-being is explained by income, net worth, and debt/income ratio, for the sample as a whole and for each cohort? The independent variables income, net worth, and debt/income ratio were entered into a regression equation using perceived economic well-being as the outcome variable. The procedure was run for each of the economic variables separately, then the three together for each cohort and for the entire sample. Table 8 shows the results of this procedure when each variable was entered separately. Table 9 presents the results when the three economic variables are entered together. 85a Income will explain the most variation in perceived economic well-being, followed by debt/income ratio and net worth. Although income was the largest contributor, net worth contributed more to variation than did debt/income ratio. When the variables were entered separately, {3 for income was .22, for net worth, {3 s .09 and for debt/income ratio, {3 = .06, for the total sample. Therefore, this hypothesis was not supported by the data. The amount of variation in perceived economic well-being explained by income, net worth and debt/income ratio for the total sample was {3 - .31, p < .0002. Table 8 Regression Results of Perceived Economic Well-being 117 (PEWB) with Income, Net worth and Debt/Income Ratio entered separately Sample Cohort l [n=39] Cohort 2 [ n=99 ] Cohort 3 [ n=90 ] All [n=282] I"p<.05 Variable Income Net worth D/I Ratio Income Net worth D/I Ratio Income Net worth D/I Ratio Income Net worth D/I Ratio Beta .48591 .22479 -.49018 .58183 .44001 -.38026 .48606 . 13736 .06236 .473 14 .30699 -.25287 T-value 4.448 1 .441 -3.601 7.046 4.518 -3.857 5.964 1 .461 .661 8.987 4.987 -4.091 88:1 .0000* .1517 .0008* .0000* .0002* .0000* .1468 .5098 .0000* .0000* .0001* R .224 .026 .222 .332 .184 .135 .230 .010 -.005 Sig. E: .0002* .0000* .1468 .5058 .0000* .0000* .0001* Table 9 Regression Results of Perceived Economic Well-being 118 (PEWB) with Income, Net worth and Debt/Income Ratio . Sample Cohort 1 [ n=39 ] Cohort 2 [ n=99 ] Cohort 3 [ n=90 ] All [ [1:282 ] *p<.05 Variable Income Net worth D/I Ratio Income Net worth D/I Ratio Income Net worth D/I Ratio Income Net worth D/I Ratio Beta .45786 .0449] -.34698 .45542 .21722 -.l7263 .47674 .17402 .11795 .46297 . 17089 -.13433 T-value 3.289 .326 -2.168 5.016 2.353 -1.918 5.749 1 .844 1 .249 8.425 2.930 -2.337 84;; .0023* .7467 .0130* .0000* .0210* .0585 .0000* .0679 .2145 .0000* .0037* .0203* .375 .406 .241 .312 Sig. F. .0001* .0000'“ .0002 * 119 85b The variation in perceived economic well-being explained by the objective economic indicators will not differ among the cohorts. This hypothesis was not supported by the data. For Cohort 1 each of the economic indicators used to predict perceived economic well-being were significantly related and explained a portion of the variance of perceived economic well-being when entered alone. When all three were used together, only income, ; = 3.29, p < .002, and debt/income ratio, 9 = .013, were found to be significant; net worth was not significant, p = .5158. The variation in PEWB explained by income and debt/income ratio was {E = .38, p_< .0001. Each of the economic variables was significantly related to perceived economic well-being when tested separately for Cohort 2. When all of the variables were entered into the equation, income and net worth were significantly related to PEWB and contributed to the explanation of variance 33== .41, p_< .0000. For Cohort 3 income was the only one of the three economic variables that was a significant predictor of perceived economic well-being, t = 5.964, p < .0000. In combination, only income was significant although the presence of the other two variables added 1% to variance explained, 33 = .24, p < .0000. The economic variables that were significantly related to PEWB for Cohort 1 were income and debt/income ratio. For Cohort 2, the variables in order were income, net worth, and 120 debt/income ratio. For Cohort 3 only income was significant in explaining variation in PEWB. The amount of variation in perceived economic well-being explained by the three economic predictors was &2 = .31 for the entire sample, g2 = .40 for Cohort 1, BE== .41 for Cohort 2 and B = .24, for Cohort 3. Research Question 6 How much variation in perceived economic well-being is explained by locus of control orientation and financial management practices separately and combined for the sample as a whole and for each cohort? Regression procedures were run for locus of control, for financial management practices and for the two together for each of the three cohorts and for the sample as a whole. Table 10 gives the results for each cohort when the variables are entered separately. 86a The portion of variance of perceived economic well-being explained by locus of control orientation is expected to be small but significant. The portion of variance explained by locus of control was significant, which supported the hypothesis. However, it was larger than expected. Twenty-four percent of the variance in PEWB was explained by locus of control for the sample as a whole, 9 < .0000. Table 10 Regression Results of Perceived Economic Well-being (PEWB) by Locus of Control (Locus) and Financial 121 Management Practices(FMP) entered separately Sample Cohort l [ n=61 ] Cohort 2 [n=95] Cohort 3 [ n=121 ] All [n=277] Cohort l [n=46] Cohort 2 [ “=84 l Cohort 3 [ n=112 ] All [ n=242 ] " p < .05 Variable Locus FMP Beta .50590 .61689 .37149 .48936 .29672 .06333 -.03007 .09333 T-value 4.505 7.559 4.365 9.306 2.061 .575 —.316 1 .452 S811 .0000* .0000* .0000* .0000'“ .0452* .5671 .7529 .1470 R2 .243 .373 .131 .237 -.008 -.008 .005 Sig. I; .0000* .0000* .0000* .0000" .0452* .5671 .7529 .1478 122 86b Financial management practices will explain more variance in perceived economic well-being than locus of control orientation. This hypothesis was not supported by the data. Financial management practices explained only .5 percent of the variation in PEWB and was not significant, 3 = .508, p = .1478 when entered alone. When both locus of control and financial management practices were entered into the regression equation, 19.1% of variance was explained. Table 11 gives a summary of the results when financial management practices and locus of control are entered together. 86c Locus of control orientation and financial management practices will explain more variance in perceived economic well-being for Cohort 3 than for Cohort 1 or Cohort 2. The hypothesis was not supported. Locus of control provided more explanation of variance for Cohort 2 (33 == .37) than for either Cohort 1 (33 = .24) or Cohort 3 (33 = .13). Financial management practices increased the amount of variance explained in perceived economic well-being only for Cohort 1. Table 1 1 Regression Results of Perceived Economic Well-being (PEWB) with Locus of Control (Locus) and Financial 123 Management Practices(FNIP) Sample ' Cohort l [ [1:41 ] Cohort 2 [ n=81 ] Cohort 3 [ n=112 ] All [ [1:234 ] *p<.05 Variable Locus FMP Locus Locus Locus Beta .36068 .33396 .60803 -.07850 .34035 .06492 .44029 .03 203 T-value 2.577 2.386 6.578 -.849 3.758 -.717 7.412 .508 S81; .0140* .0221* .0000* .3938 .0003* .4749 .6117 R2 .222 .341 .191 Sig. E; .0033* .0000* .0012* .0000* 124 Research Question 7 Which manager and household characteristic variables are the best predictors of perceived economic well-being for the sample as a whole and for each cohort? A correlation matrix of all the independent variables and the dependent variable, perceived economic well-being, for the entire sample and for each cohort was made to explore the relationships between the variables. See Appendix D. Regression procedures relating all of the independent variables were then run entering all variables to see how perceived economic well-being was affected by .each variable. A Stepwise method of regression was completed to determine which of the independent variables were the best predictors of PEWB. These procedures were completed for each Cohort and for the sample as a whole. See Tables 12 through 15. 87a Household characteristics of income, net worth, health status and household size, and manager characteristics of education, marital status and financial management practices will contribute the most to the explanation of variance in perceived economic well-being. This hypothesis was not supported by the data. The household characteristics that were the best predictors of PEWB were income, net worth, and health status. The only manager characteristic that was significant was locus of control orientation. These explained 45% of the variance in perceived economic well-being ,p < .0000, for the sample as a whole. 125 Table 12 All: Results of Stepwise Regression of Perceived Economic Well-being (PEWB) with Household and Manager Characteristics [ n=175 ] Variable Beta T-value Sig t_ R2 Chaagze Sig. If: [1 Income .41715 7.037 0000* .298 .0000“ Locus .24040 3.951 0000* .389 .091 .0000" Net worth .22572 3.860 .0002“ .427 .038 .0000" Health -.16839 -2905 0042* .452 .025 .0000" " p < .05 Table 13 Cohort 1: Results of Stepwise Regression of Perceived Economic Well-being (PEWB) with Household and Manager Characteristics [ n=l8 ] Variable Beta T-value Sig ; R2 crating: Sig. I; [I Locus .56673 3.702 .0021 .475 .0009" Income .44720 2.921 .0105 .643 .168 .0002" ‘p<.05 126 Cohort 2: Results of Stepwise Regression of Perceived Economic Well-being (PEWB) with Household and Manager Characteristics [ n=61 ] Variable Beta T-value Sig t R Change in R 1 Income .52260 5.552 .0000“ .393 Locus .40266 4.273 .0001" .515 .122 Net worth .33476 3.8 14 .0003“ .576 .061 Education -.26225 -2.644 .0106" .605 .029 FMP -.l6859 -2.048 .0453“ .626 .021 " p < .05 Table 15 Cohort 3: Results of Stepwise Regression of Perceived Economic Well-being (PEWB) with Household and Manager Characteristics [ n=96 ] 2 Variable Beta T-value Sig t R Change in R 1 Income .29663 2.946 .0041“ .222 Health -. 19644 -2.3 1 1 .023 1* .270 .048 Locus .26020 2.939 .0042‘“ .300 .030 Employ -.20734 -2.380 .0194“ .324 .024 Insure (it) .21726 2.1 86 .03 14* .35 1 .027 *p<.05 127 87b Income is expected to be less important in explaining variance in perceived economic well- being for Cohort 1 than for Cohort 2 and Cohort 3. This hypothesis was partially supported by the data. Household income was the most important predictor for Cohort 2 and Cohort 3 and second for Cohort 1. Although when per capita income was used, it was the most important predictor for Cohort 1 and Cohort 3 and second for Cohort 2 (See Tables 16 to 19). Tatflee1fi All: Results of Stepwise Regression of Perceived Economic Well-being (PEWB) with Household and Manager Characteristics, Using Per Capita Income (Percap) [ n=l75 ] Variable Beta T—value Sig t R cmllllzg: Sig. E Percap .36760 5.261 .0000 .262 .0000* Locus .25316 4.353 .0000 .372 .1 10 .0000* HHS .16130 2.584 .0106 .410 .038 .0000* Health -.18361 -3.226 .0015 .436 .026 .0000" Net worth .18398 3.133 .0020 .460 .026 .0000" Insure .14194 2.281 .0238 .47 1 .01 1 .0000" ‘p<.05 128 Cohort 1: Results of Stepwise Regression of Perceived Economic Well-being (PEWB) with Household and Manager Characteristics, Using Per Capita Income (Percap) [ n=1 8 ] Variable Beta T-value Sig ; R Chalfigg Sig. E ['I Percap .53199 3.562 .0028 .526 .0004" Locus .47332 3.1 70 .0063 .697 .17 1 .0001 * " p < .05 Table 18 Cohort 2: Results of Stepwise Regression of Perceived Economic Well-being (PEWB) with Household and Manager Characteristics, Using Per Capita Income (Percap) [ n=61 ] Variable Beta T-value Sig t R2 Change Sig. I; in R 2 Locus .39117 4.166 .0001 .315 .0000"' Percap .38096 4.148 .0001 .494 .179 .0000'“ Net worth .26506 2.771 .0075 .546 .052 .0000" ‘p<.05 129 'Tatflee1Sl Cohort 3: Results of Stepwise Regression of Perceived Economic Well-being (PEWB) with Household and Manager Characteristics, Using Per Capita Income (Percap) [ n=96 ] Variable Percap Insure Health Locus Employ " p < .05 870 This hypothesis was not supported by the data. Beta T-value Sig ; .25 849 2.779 .0066 .28524 3.150 .0022 -. l 95 1 5 —2.283 .0248 .26203 2.944 .0041 -.23334 -2.649 .0095 R . 164 .228 .270 .301 .345 Change in R 3 858- E. .0000* .0000" .0000* .0000" .0000“ Health status is expected to be more important in explaining variance in perceived economic well- being for Cohort 1 than for Cohort 2 and Cohort 3. Health status emerged as a significant predictor only for Cohort 3. The presence of a health problem was negatively related to perceived economic well-being, t - 2.946, p < .001. 376 This hypothesis was not supported by the data. Financial management practices were significant as a Financial management practices and household size are expected to explain more variance in perceived economic well-being for Cohort 3 than for Cohort 1 and Cohort 2. 130 predictor only for Cohort 2. The use of more management practices was negatively related to perceived economic well- being. Household size was not one of the better predictors for any of the cohorts. Stepwise regression procedure enters the most powerful predictor first and adds variables until no additional variables are available using p = < .05 as the cut off criterion. The best predictors of the outcome variable are selected in this procedure. When the regression procedure Backward is used, all variables are entered into the regression equation and removed one at a time (p = <.10) until those left provide the most explanation of variance in the dependent variable. The results of this procedure gives a set of variables that explain the most variation in the outcome variable. With the Backward procedure, no additional variables were added except for Cohort 1 (See Table 20). For Cohort 1, the set of variables that was used to explain variation . in PEWB was income, locus of control, health status, marital status, household size, number of income sources and education. The variation in perceived economic well-being explained increased from 63% using the stepwise procedure with two variables to 84% using the additional variables. 131 Table 20 Cohort 1: Results of Backwards Regression of Perceived Economic Well-being (PEWB) with Household and Manager Characteristics [ n=1 8 ] Variable Beta T-value Sig ; Married 1 .88759 3.758 .0037 HHS -1 .56930 -3. 129 .0107 Income 1 . 1 3960 4.5 1 6 .001 1 Health -.78249 -4. 1 14 .0021 Insource -.3458 1 -1 .838 .0959 Locus .24792 1 .82 l .0986 Education -.218 1 0 - 1 .898 .0869 R2 = .843 Sig. _E = .0002 fiflflflfll! Question 1 looked at differences among the cohorts in manager and household characteristics. Cohort 1 was found to have significantly lower income, less net worth, fewer insurance sources and a lower debt/income ratio than Cohort 2 and Cohort 3. Significantly fewer women in Cohort 1 were married or employed than in Cohort 2 and Cohort 3. Financial management practices were used significantly more often by Cohort 1. Cohort 2 had significantly more health problems that Cohort 1 or Cohort 3. Cohort 3 had significantly larger households and more education than Cohort 1 and Cohort 2. No significant differences were 132 found among the cohorts in number of sources of income, race or locus of control orientation. Question 2 examined the items used to create the dependent variable, perceived economic well-being, for differences among the cohorts. Cohort 1 was significantly more satisfied than Cohort 2 and Cohort 3 with both the total income available and their ability to meet financial emergencies. Perception of income adequacy for Cohort 1 was significantly higher than for Cohort 2 and Cohort 3. Cohorts 1 and 2 were more satisfied than Cohort 3 with both the material things they had and their net worth. Cohort 3 had significantly higher expectations for their financial situation to improve in the next five years than did Cohort 1. For Question 3, no relationship was found between financial management practices and locus of control orientation for any of the cohorts. Analysis related to Question 4 found no differences in financial management practices by income level. Question 5 addressed the variation in perceived economic well-being explained by income, net worth and debt/income ratio. Income explained significantly more variance than either net worth or debt/income ratio for all the cohorts and the total sample. In addressing Question 6, locus of control was found to be a significant predictor of perceived economic well-being, while financial management practices did not explain 133 additional variance. This finding was significant for all three cohorts. Question 7 asked which manager and household characteristics were the best predictors of PEWB. Regression procedures were used, and the best predictors were locus of control orientation and income which explained 64% of the variance for Cohort 1; income, locus of control orientation, net worth, education, and financial management practices which explained 63% of the variance for Cohort 2; and income, health status, locus of control orientation, employment status, and number of insurance sources which explained 35% of the variance for Cohort 3. For the sample as a whole, income, locus of control, net worth and health status were the best predictors of PEWB explaining 45% of the variance. CHAPTER V DISCUSSION, CONCLUSIONS AND IMPLICATIONS The perceived economic well-being of three cohorts of rural women was studied within a conceptual_framework that blended the human ecological approach with a life course perspective. The cohorts were selected based on the historical events occurring at the time these women reached age 18. This definition of cohort is compatible with the life course perspective that one's perception of and adaptation to life is affected by events occurring throughout the life span. The women in Cohort 1 reached age 18 during the depression (between 1929 and 1934) and were 72 to 77 at the time of the study. The women in Cohort 2 reached 18 between 1950 and 1955, coinciding with an increased emphasis on the nuclear family and the move to the suburbs. They were 51 to 56 at the time of the study. The women in Cohort 3 were the first wave of the Baby Boom cohort. They reached 18 between 1964 and 1969 and were 37 to 42 at the time of the study. 4' This chapter includes a discussion of#thg_!igbilitylof\ the model (Figure 1) introduced in Chapter 1 for the sample as a whole and for each cohort. It also includes the ...— 134 135 F-«r—w— -~.__ I - //conclusions reached and the implications for research, professional practice and public policy. Figure 4 is a revised version of the model based on the findings. Figures 5 through 7 present models specific to each cohort. Discussion The contributions of the household characteristics as presented in the model are addressed first. All correlations reported in this section of the discussion are for the sample as a whole and were significant at p < .05, unless otherwise stated. (Correlation tables are in Appendix D). Correlations for the individual cohorts are reported in the discussion of results for each cohort when appropriate. IDQQEE Household income explained 17% of the variance for Cohort 1, 40% for Cohort 2, 22% for Cohort 3 and 30% for the total sample. Income was either the first or second best __.__.._-— upredictor of perceived economic welljbeing for all of the "'7" an“... _ cohorts. When per capita income was substituted for \.____..__*__ ‘_/ household income in the regression equation, it explained ...-Mn“... 52.6% of the variance for Cohort 1, 17.9% for CohortMSTShd 16.4% for Cohort 3. The differences were a_rg§glt of the ...---—a order of 995rY19§,th° variablefi<:The importance of income as e--l' _q" ...-.. --_ ___,_._.-F"'_ no." a predictor of perceived economic well-being is consistent “...—a/h —. 136 with previous research which has found income to be a significant predictor of financial satisfaction gutcomes (Ackerman & Paolucci, 1983; Davis & Helmick, 1985; Lorenz & Fletcher, 1985)(/ The amount of variance_explained~by income “~— for each cohort differed depending on whether per capita income or household income was used. Household-ihcome explained less variance for Cohort 1 as it was the second predictor entered, so any amount of variance in income related to locus of control was included in the variance explained by locus of control. The use of per capita income reversed the order of entry for these variables. The effect of different measures of income may have been more pronounced in this cohort because of the large number of one person households. \. 0ne explanation for the importance of income in explaining perceived economic_welljbeinggfgrngghgrtfii‘may be related to their age and location in the life course. ‘K— “...—..-... Persons in this cohort are at the height of their earning power and most likely to be either in the launching stage or early empty nest stage with heavy demands on income. i/Agpexpecggd} Cohort 2 and Cohort 3 had higher household k incomes than Cohort 1. For the total sample, income was ___.._..-—-—'" significantly correlated with net worth, locus of control orientation, education, household size, marital status, race, the number of income sources, the number of insurance sources and employment. Not unexpectedly, those women who 137 reported higher incomes had more education, were white, were married and lived in larger households and reported more internal locus of control orientation, more income sources and more insurance. The presence of a health problem in the household was negatively correlated with income at p = .071. The association of less income and the presence of more health problems may be attributed to the interference of health problems with the ability to earn income. It also has been noted that health problems appear more frequently in lower income households which may be a result of insufficient resources to obtain preventive health care. e ' om A list of sources of income was provided and respondents were asked to check all that applied. It was thought that more sources of income would provide a greater sense of security (Institute for Poverty, 1983). The cohorts did not differ significantly in number of sources. The average number of sources was 2.4, but the types of sources were different. The most frequently indicated sources of income for Cohort 1 were Social Security and interest on savings, while income from employment was listed most frequently by Cohort 2 and Cohort 3. Although Social Security benefits have a cost of living feature, the incomes of Cohort 1 are relatively fixed (Institute of Poverty, 1983; Radner, 1987). 138 The costs of some goods and services needed by this cohort, such as medical care, have risen faster than costs in general. Income perceived as adequate now may not continue to be adequate if costs rise faster than income. However, the income from employment received by Cohort 2 and Cohort 3 is more likely to increase as they are in the prime years of their work life. IDEEIQEEE The number of insurance types was a significant predictor of perceived economic well-being for Cohort 3 only. The number of types of insurance was included as a measure of preparation for financial emergencies (Davis & Helmick, 1985). Health, household, and car insurance were the most frequent insurance types listed. Cohort 1 had significantly fewer types of insurance than did Cohorts 2 and 3, as these two cohorts more often had life and disability insurance. One explanation for the importance of insurance to Cohort 3 may be that having more insurance instills confidence and is therefore reflected in satisfaction with ability to meet financial emergencies. Another may be that sufficient income to purchase insurance reflects higher income, which is associated with perceived economic well- being. Also, as insurance is frequently a benefit of employment, the significance of insurance may reflect the significance of being employed to this cohort. 139 N§&_EQIED Net worth was a significant predictor of perceived economic well-being for the total sample and for Cohort 2. This is consistent with the findings of Davis and Helmick, (1985). For the sample as a whole, net worth explained an additional 4% of variance and for Cohort 2 an additional 6% of variance. This variable reflected the amount of estate building the household had accomplished. Net worth was positively correlated with income, locus of control orientation, financial management practices, race, and the number of income sources (p < 05). It was negatively correlated with debt/income ratio. The finding of a positive relationship between net worth and perceived economic well-being is consistent with findings in previous studies (Buetler & Mason, 1987; Titus, Fanslow, & Hira, 1989). It was expected that Cohort 2 would have higher net worth than Cohort 1, as they would be saving for retirement. Even though Cohort 1 and Cohort 2 reported similar net worth amounts, the significance of net worth to the perceived economic well-being of Cohort 2 reflects the importance of building net worth for this cohort. The similar current levels of net worth of Cohort 1 and Cohort 2 suggest that Cohort 1 had a higher net worth to begin with, they have not begun to spend their savings, or most of their assets are in a house or other non-liquid assets. 140 Cohort 1 and Cohort 2 had significantly higher net worth than Cohort 3. The younger cohort still has children at home and has more demands on income than the older cohorts. It is conjectured that their net worth will be more like the older cohorts as they move through the life course, have fewer demands on resources and can begin to save, and as they accumulate more equity in their houses. W Kinsey 8 Lane (1978) suggest that families accept consumer credit as a normal part of doing business as a household so that debt, except in excess, does not decrease feelings of well-being. Consistent with this, debt/income ratio was not found to be a significant predictor of perceived economic well-being. A higher debt/income ratio was negatively correlated with perceived economic well- being. A probable explanation for the difference between the correlation and the regression results would be that the correlation tables used all cases, but the cases with missing data on any variable were omitted in the regression analysis. It also may be that the portion of variance that was explained by debt/income ratio when entered alone was absorbed by income or net worth in the stepwise regression analysis. A higher debt/income ratio was negatively correlated with financial management practices and positively correlated with employment. This may indicate that those 141 managers that used more financial management practices utilized resources better, lowering debt/income ratios. Employment as associated with debt/income ratio may reflect that women in younger cohorts are more often employed, and younger households have higher debt/income ratios. Increased employment also may increase the potential to obtain more credit (Kinsey 8 Lane, 1978). 1321mm Health was a significant predictor of perceived economic well-being for the sample as whole, for Cohort 1 and Cohort 3. For the sample as a whole, the presence of a health problem was negatively correlated with perceived economic well-being, locus of control orientation, marital status and household size (the more persons in the family, the more likely someone has a health problem). Health status was expected to affect perceived economic well-being either as a demand on resources or a barrier to obtaining resources (Hayhoe, 1990). Interaction of the household with the health delivery system may also affect the individual's perception of well-being. Health problems are frequently on-going so that the duration of the problem could affect perceived economic well-being from a life course perspective. As health problems increase with age, it was expected that Cohort 1 would report the most health problems (Ebersole 8 Hess, 1985). However, Cohort 2 reported 142 significantly more health problems and health problems that caused a financial concern. This may be because Cohort 1 has become used to minor health problems and did not report them. It also may be because Cohort 1 is the only one that is eligible for Medicare, which reduces their anxiety about the financial worries of poor health. Another explanation could be that for Cohort 2 the presence of health problems is relatively new and as such very noticeable, or that these problems are more restrictive of life style for Cohort 2. Although the presence of more health problems was not a significant predictor of perceived economic well-being for Cohort 2, health was negatively correlated with perceived economic well-being. It also was negatively correlated with income, locus of control and financial management practices, all of which were significant predictors of perceived economic well-being for Cohort 2. W Household size was not a significant predictor of perceived economic well-being for any of the cohorts. This is consistent with findings in a study of perceived economic well-being by Hayhoe (1990). For all cohorts, larger household size, being married, and higher income was significantly correlated. It appears that the effect, if any, of household size on perceived economic well-being may have been reflected by one of these variables. 143 Household size was negatively correlated with net worth; the larger the household, the less the net worth. Increased demands on resources keep larger households from building up savings. mm o o se o d ha acte s 'c 0f the household characteristics in the model, income, net worth, health and number of insurance sources were significant in explaining variance in perceived economic well-being for one or more cohorts. Although debt/income ratio and number of income sources may provide a more complete picture of the financial status of the household, they were not significant in explaining perceived economic well-being. Even though household size did not explain a significant amount of variation in perceived economic well- being, its inclusion in the model may be useful in other studies. It would appear that the effect of household size on the economic well-being of the household may be determined more by the membership of the household as reflected in resources available and demands on resources than the number of persons (Bauer, Danes 8 Rettig, 1990; Blaylock 8 Blisard, 1990). - em This section examines the manager characteristics as they relate to the model used to predict perceived economic well-being. 144 Education Education was a significant predictor of perceived economic well-being only for Cohort 2 when the stepwise regression procedure was used. When the backwards regression procedure was used, it was significant, p < .10, for Cohort 1. Less education had a negative effect on perceived economic well being for both Cohorts 1 and 2. More education was positively correlated with perceived economic well-being, having more income, having more internal locus of control orientation, larger household size, the likelihood of being married, being white, having more income sources and insurance types and being employed. The relationship between more income, employment, number of income sources and number of insurance sources is an expected one. .The relationship between more education, marital status and increased household size is probably explained by the fact that Cohort 2 and Cohort 3 were more educated, more likely to be married and lived in larger households than Cohort 1, so that this is an effect of cohort not education. 8321mm Marital status was not a significant predictor of perceived economic well-being for any cohort. Marital status was positively correlated with increased income, more internal locus of control orientation, more education, larger household size, more health problems, number of 145 insurance types and employment. The relationship of marital status to education, household size, income, number of insurance types and employment is probably because more women in Cohort 2 and Cohort 3 are married and these variables were related to having membership in these two cohorts. The presence of more persons in the married household increases the probability of someone in the household having a health problem. e a s Employment status was a significant predictor of perceived economic well-being for Cohort 3, with not being employed having a negative effect on perceived economic well-being. Employment serves as both a source of income and interaction with other environments outside the household. As expected, more women in Cohort 2 and Cohort 3 were employed. Employment was positively correlated with income, debt/income ratio, locus of control orientation, education and household size, marital status and insurance types, as previously noted. Being employed was negatively related to financial management practices. This could be because the additional income from being employed means fewer management strategies are felt to be needed or that less time is available to manage. 146 BEE: Race was not a significant predictor of perceived economic well-being. Income, net worth, education, and number of income sources were positively correlated with being white. The number of non-white women in the sample was about 10% and did not differ among the cohorts. Because there were more non-white women in two of the states a separate analysis was done for those states, but no differences were found. Previous studies have found that the factors that explain variance in economic well-being and satisfaction are the same for persons of different races (Fletcher 8 Lorenz, 1985). Thus the effect of race is reflected in lower income and net worth, fewer income sources and less education of non-whites. WW Education was a significant predictor for Cohort 2, and explained additional variance in perceived economic well- being for Cohort 1. Employment was a significant predictor of perceived economic well-being for Cohort 3. Employment was not relevant for Cohort 1 as most of this cohort reported themselves retired. The inclusion of these two variables measured different aspects of perceived economic well-being for the different cohorts. Neither race nor marital status was a significant predictor of perceived economic well-being. The effect of 147 race is probably included in income, education and employment so that it is not necessary to the model. The effect of marital status was probably included in household size. Marital status does provide information about the number of potential earners in the household. Whether or not to include it in the model may be dependent on the population being studied. Locus of control orientation was a significant predictor of perceived economic well-being for the entire sample and for all three cohorts. It was significantly correlated with income, net worth, education, and employment. Bauer, Danes 8 Rettig (1991) found a significant negative relationship between larger household size and locus of control orientation using this same measure. The correlation was negative but not significant in this study. When entered into a regression equation alone, locus of control orientation explained 24% of the variance in perceived economic well-being. The more control the individual felt over financial situations, the higher the perceived economic well-being. Finding ways to increase feelings of control may be as important as more income in increasing perception of economic well-being. 148 No differences were found among the cohorts in locus of control orientation. An individual's locus of control orientation develops over the life course but may be highly influenced by things that occur in adolescence (Schiamberg, 1988). There also have been some research findings that indicate that locus of control orientation may change with age (Clausen, 1986). Based on this, it was suspected that the different historical backgrounds of the cohorts would result in differences in locus of control orientation among the cohorts, but none were found. This may be because this measure was specific to the financial domain. There may be differences in locus of control orientation related to other domains (Lefcourt, 1976). W Financial management practices were significant predictors of perceived economic well-being only for Cohort 2. When more financial management practices were used, there was a negative effect on perceived economic well- being. This concurs with another study where the increased need for financial management practices was associated with limited resources and perceived as stressful (Wilhelm 8 Ridley, 1988). Cohort 1 used a significantly higher number of financial management practices than did Cohort 2 or Cohort 3, but this use was not a significant predictor of perceived economic well-being. This cohort started adulthood during 149 an economic depression when most households had limited resources. It was surmised that practicing financial management was simply a way of life. Financial management practices were significantly correlated with debt/income ratio; when more financial management was practiced debt/income ratio was lower. There was no linear relationship between financial management practices and locus of control. The premise that financial management would more likely be practiced by larger households, more educated and middle income households was not supported by this data (Buetler 8 Mason, 1987). aummaIx_9f_Maneser_Qharasteristisa_:_nehaxieral The use of the locus of control measure specific to the financial domain was a significant predictor of economic well-being for all three cohorts. The use of financial management practices was significant for one cohort and with improvement in measurement as discussed in the instrumentation section, could be more useful. Both variables are kept in the model. 0 ' We - e The model was used to predict perceived economic well- being. Perceived economic well-being was defined as the individual's feelings with regard to over-all satisfaction 150 with the financial situation and expectations for the future. In this study, the dependent variable, perceived economic well-being, was an index created from statements about the respondents' satisfaction with financial situation, perception of income adequacy, and perception of changing financial conditions. Other studies have used satisfaction with some aspect of the family financial situation, either one measure or a series to measure economic well-being (Davis and Helmick, 1985; Douthitt, MacDonald 8 Mullis, 1990; Fletcher and Lorenz, 1985). In a previous study, Davis and Helmick (1985) found perception of change in the financial condition to be a good predictor of financial satisfaction. Perception of change in financial situation was included as part of the index for the outcome variable in this study. As a life course perspective was used, it seemed appropriate to include feelings about the previous financial situation and expectations for the future as part of current feelings. The statements used to make up the index were all significantly correlated. The reliability coefficient for the index was .83. This measure of perceived economic well-being appeared to be a viable research instrument. The inclusion of the perception of change component in the measure of perceived economic well-being encompassed the household's interaction with other environments and experience over time. 151 The model explained 64% of the variance for Cohort 1, 63% for Cohort 2, 35% for Cohort 3 and 45 % when the three cohorts were treated as one group. Not all of the variables included were significant predictors. The use of per capita income instead of household income increased the variance explained to 69% for Cohort 1. The use of per capita income reduced the number of variables entered and explained only 54% of variance for Cohort 2. There was no change in the amount of variance explained or the variables used for Cohort 3. The revised version of the model (Figure 4) used household income as the measure for income. Except for household size, the predictors included were significant for at least one cohort. Although household size was not a significant predictor in this study, it may prove important in further studies. Additional predictors need to be explored to increase the viability of the model. One way people evaluate their economic well-being is by comparing it to others (Douthitt, McDonald 8 Mullis, 1990). One or more variables comparing the household income to others in the neighborhood or to their household of origin might be useful in explaining perception of economic well-being. Also, because the household is affected by economic conditions external to it, asking about these changes as perceived by household members could provide additional information (Moen, Rain 8 Elder, 152 1983). Further research with different populations will be useful in testing the index used to measure perceived economic well-being and the model used to predict it. The variables that were significant predictors for each cohort are summarized in the following paragraphs. Figures 5, 6 and 7 show the model as specific to each cohort. Locus of control and income explained 64% of the variance in perceived economic well-being for Cohort 1 when the stepwise regression procedure was used. The more control these women felt they had over their financial condition and the more their income, the higher their perceived economic well-being. Using the backward regression procedure, the independent variables health, education, number of income sources, marital status and household size were added, increasing the variance explained to 84%. Being married had a positive effect on perceived economic well-being. The presence of health problems, less education, larger household size and fewer income sources had a negative effect on perceived economic well-being. The results of the stepwise regression procedure for Cohort 2 showed that income, locus of control, net worth, education and financial management practices explained 63% of the variance in perceived economic well-being. The negative effect of use of financial management practices may be attributed to the fact that they were used to meet increased demands, and therefore were perceived negatively. 153 Household Manager Characteristics Characteristics Income Demtgraphic Insurance Cohort Net Worth Education t In Health 9 oymen Household size Behavioral Locus of control orientation Financial management practices Perceived Economic Well-Being Satisfaction with Financial Situation Perception of Income Adequacy Perception of Changing Financial Condition Indicators of Perceived Economic Well-Being Revised Model Household Characteristics Income Sources of Income Health Household size 154 Manager Characteristics Demographic Education Marital status Behavioral Locus of control orientation Perceived Economic Well-Being Satisfaction with Financial Situation Perception of Income Adequacy Perception of Changing Financial Condition Indicators of Perceived Economic Well-Being Revised Model: Cohort l 155 Household Manager Characteristics Characteristics Income Net Worth Behavioral Locus of control orientation Financial management practices Perceived Economic Well-Being Satisfaction with Financial Situation Perception of Income Adequacy Perception of Changing Financial Condition Indicators of Perceived Economic Well-Being Revised Model: Cohort 2 156 Household Manager Characteristics Characteristics Demographic Employment Behavioral Locus of control orientation Perceived Economic Well-Being Satisfaction with Financial Situation Perception of Income Adequacy Perception of Changing Financial Condition Indicators of Perceived Economic Well-Being Revised Model: Cohort 3 157 The independent variables that were significant for Cohort 3 in predicting perceived economic well-being were income, health, locus of control, employment and number of insurance of types. This combination of variables explained 35% of the variance. Having more income, more internal locus of control and more insurance types increased perceived economic well-being. Not being employed and having more health problems decreased perceived economic well-being. One explanation may be that insurance is frequently a benefit of employment and also is associated with the ability to pay for costs related to more health problems. Conclusions This study provides an assessment of the usefulness of a model to predict perceived economic well-being for three cohorts of rural women. The model included locus of control and financial management practices as behavioral characteristics of the financial manager, as well as selected household characteristics and demographics. There were no differences among the cohorts in their perceived economic well-being. However, the variables that were significant in predicting perceived economic well-being were different for the cohorts. The model explained 64% of the variance for Cohort 1, 63% of the variance for Cohort 2 and 35% for Cohort 3. 158 The conceptual framework used was the examination of cohorts from a human ecological approach within the life course perspective. The human ecological approach focuses on human beings interacting with their environment, while the life course perspective focuses on individual and family transitions as they are shaped by different historical conditions. A cohort is a group of individuals that move together through the life course and experience historical events at the same time. The human ecological approach acknowledges the effect of the individual's interaction with other household members, the social environment and current economic conditions. A life perspective view utilizes the individual's past history as an explanation of current behavior. Income and locus of control were the only two variables that were significant predictors for all three cohorts. Income was the most important predictor for Cohort 2 and Cohort 3. The most important predictor for Cohort 1 was locus of control. This variable measured feelings of control over the financial domain. Its importance to Cohort 1 may be a result of their economic history. They entered adulthood during the depression when the majority of individuals felt that they had little control over financial aspects of their lives. Income was a better objective indicator of perceived economic well-being than either net worth or debt/income ratio, the other objective measures in 159 the model. The reason for this may be that these two measures differed more in importance among the cohorts. Whether this was an effect of age differences or different attitudes to different economic histories can not be determined. When data were analyzed in aggregate, cohort differences were masked. Cross-sectional data can not be used to determine if differences are a function of age, past history or current social conditions. Longitudinal and time-lag studies of these and other cohorts would aid in the understanding of which factors are related to the aging process and which reflect different life style preferences as a result of diverse historic and social perspectives. Longitudinal studies observe the same cohort at two or more points in time. The data produced reflect age changes and are not clouded by potential differences between the cohorts. Time lag-studies are used when the researcher is interested in the study of cohort effects. This type of study involves the observation of people of the same age at different times (Schaie 8 Willis, 1986). Based on the amount of variance explained by the model for Cohort 3, perhaps additional variables should be examined to reflect the diverse lifestyles of younger families. For example, about 19% of these women were divorced, and single parents may have some concerns that other women do not. Also the significance of employment to 160 this cohort may indicate the need to explore factors related to work and family as associated with perception of economic well-being. It is important not to assume that the cohorts in this study will remain the same over time or that Cohort 2 will be the same as Cohort 1 when they reach age 72. This is a picture of where the cohort is now, and it is expected that their future development will be modified by historical change. There also is a need to recognize that there is an 18 year age spread between the oldest and the youngest of the Baby Boom cohort and that some aspects of the life history of members within that cohort are not the same. 1W Continued work with perceived economic well-being measures with other populations, including other cohorts of women, men and urban populations will further test the usefulness of the model. In addition, other factors that are effective in predicting economic well-being need to be identified. Additional study of how the social, economic and cultural environment of individuals and families affects their perception of economic well-being would be worthwhile. The human ecological approach to this study of perceived economic well-being is recommended. This approach includes identification of behavioral characteristics that are unique 161 to individuals and families, such as attitudes, perceptions and coping skills and how these influence the interaction with the environment in terms of recognition of resources available and interpretation of demands to be met. The life perspective approach also is recommended as it would help researchers to identify reasons for differences in individual perceptions and attitudes based on past experience. It could be used to aid researchers in identifying factors that affect the perception of economic well-being in specific cohorts, cultures or regions of the country. Comparison of the index used to measure perceived economic well-being in this study with others currently being used is needed to standardize the meaning of the concept for researchers. This measure also needs to be tested with various populations and the same populations over time. These methodologies would strengthen the research findings and continue to refine the conceptual framework for the study of economic well-being. Exploring other statistical tools to increase the understanding of the relationship of variables used in this model is recommended. For example, path analysis might be used to look for relationships between the independent variables used in the model. The effect of marital status on perceived economic well-being needs further study. When marital status was 162 defined as married or not married, no differences were found for these particular cohorts. Additional research is needed to determine if there are differences between first married and remarried women and between widowed, divorced and single women in the factors that predict perceived economic well- being. Marital status is confounded with age, as older women are more likely to be widowed, and younger ones are more often divorced or never married. The effect of marital status and the age effect of cohort would need to be separated. There also may be cohort effects in attitudes toward marital status. 5 me Additional exploration of some of the factors included in this model would aid in understanding how these factors influence perceived economic well-being. For example, measures of effective financial management practices need to be developed. This study measured only how often a management practice was used; frequency is not necessarily related to effectiveness. Credit use was included as one of several management tools. It may be that a separate. variable on use of credit would be more functional as the use of credit increases. Better measures of economic variables are needed. Exact measurement of economic variables is difficult as people are reluctant to disclose this information. Use of 163 narrower, equal interval categories would make this information somewhat more accurate. Employment was measured as a dichotomous variable, with individuals identified as employed or unemployed. The unemployed category included persons who identified themselves as unemployed, homemaker or retired. Additional information that could make this variable more descriptive would be: whether employment was full or part-time, whether the person was retired or a homemaker, and length of time for employment] unemployment. For those persons who identify themselves as retired, information as to whether they had previously worked and for how long is needed. Given the increased number of women in the work force, further exploration of the impact of employment/unemployment could prove effective in explaining perceived economic well- being. Predictors of perceived economic well-being may be different for women who identified themselves as unemployed and therefore warrants further study. The education variable was measured in years of school. Additional information of highest degree attained would make this number more meaningful. As many persons are furthering their education at later ages, when the education was obtained might be relevant to its effect on perceived economic well-being. 164 W An implication for Human Ecology professionals is the support for a human ecological and life course perspective approach to understanding individuals and families. Cohort differences as a result of life course history and experience need to be considered when attempting to understand and work with individuals and families. Several findings have implications for program planning. The importance of internal locus of control needs to be recognized. If help is offered in a way that the individual still feels in control of the situation, it may be more acceptable. Also programs could be developed to help individuals increase their feelings of control over situations. As professionals who believe that good management increases well-being, there is a need to do something about the negative connotation of financial management practices. The word budget brings up images of constraints and doing without. Perhaps stressing spending and savings plans in relation to financial goals would be one approach. The relationship between budget and locus of control also could be pursued, i.e. it is the individual not the budget that is in control of how dollars are used. When working with families and individuals in matters of financial concern, it is critical to recognize that although income is important in predicting perceived 165 economic well-being, other factors contribute to this perception. The perceptions, attitudes and skills of the individuals and families affect the use of resources and how environmental factors are perceived. More households currently are headed by women who may be living alone or as single parents with children. These households usually have less income available, as there are fewer wage-earners. The factors that predict economic well-being may be different for these households than for two parent households. Women in different cohorts will have different information about and expectations for financial management. Also, the reasons for financial management may differ among cohorts, based on experience. Programs designed to meet the needs of specific cohorts could prove more effective. The recognition of cohort differences in all aspects of program planning can be used to make programs more effective in helping individuals and families build on past experience in meeting current needs. {ublig {Q1191 If the findings of differences among the cohorts hold over time, tomorrow's older women are going to be different and have different needs than the women who are older now. They will be better educated, more likely to have been in the labor force, more likely to be divorced or never married and have fewer if any children. As stated previously, the 166 cohorts who will be the elderly of the 21st century have been on the leading edge of the family revolution (Goldscheider, 1990). Much of current policy is based on the traditional family consisting of two parents and their children. This traditional model assumes male only employment, is inconsistent with current family structure and is often detrimental to women in nontraditional roles. Recognition of cohort differences can be used to evaluate current policy in order to provide the support and programs needed by women in nontraditional roles. Consideration must be given to current policies affecting women's education, labor force participation, and economic status. Policies that encourage young women to stay in school and invest in their future are needed. There also is a need for policies that provide education and job training for women who have been at home full time and need to reenter the work force. Adequate, inexpensive day care will be necessary for women if they are to return to school or participate in job training. At the policy level the concern has been with objective income adequacy in order to provide basic necessities. As the number of female primary earners increases, strategies for reaching this goal need to be reevaluated. The effect of discrimination on wages for all women is of particular concern for women trying to maintain 167 their families. Continued support for equal employment opportunities and pay for women of all ages is an important policy direction. It also has been noted that public policy with respect to the elderly has not been gender neutral (Hess, 1985). Our current Social Security system which is based on average earnings may provide less income for older women than for older men. The years women are out of the work force to raise families are averaged in as zero earnings. Even when women work, they are often in lower paying jobs than men. Therefore benefits based on their earnings are frequently lower than a spouse's portion (half the husband's Social Security) if they are married and collecting in that role. The most equitable change in Social Security policy for married women would be a division of the couple's earnings between two accounts. This would give credence to the value of the homemaker/mother's job even though not a paid position. As individuals develop and change, social institutions both limit alternatives and are changed by new demands placed on them. The recognition of cohort differences can provide direction for changes in institutions and policies. Each cohort has a distinctive composition and is predisposed to characteristic modes of thought reflecting the circumstances of its unique place in history. The cohort's placement in historical time provides information about both 168 opportunities and constraints. An understanding of an individual's current position and perspective of the future depends on that individual's cumulative life history and the location in historical time. Lives are played out in particular social and historical circumstances that must be recognized if the life course of the individual is to be understood. APPENDICES APPENDIX A COHORT TIME LINE $0.50 5.09 _J ......0 Es..- .l.l|.. APPENDIX A COHORT TIME LINE 559.0.— 00 50>) .u< are... .....U ... as)? ... ... 00380.8( ...-58> ....ll. .003 lee..- soc-cs. i“ .llJ] coca—e un- .... 80.. 3.9.0.8800 ... «03.000 anchov- .llll. 3 g .lllJ >.—. 80.00 ces—t.— are... 0.50.! 8.....- ... :33 ..ll. .3003 9.8g .lllu. gm “flan 86m moan Gan 30¢ Sam “had Oman “Tau 3m 9:3 62...... ...v ...-u 8.... ... ... UHF—Baa 30050.5 ... 8:329... sec ...-to .925 :35 ... ....eudesno ... ... u >93 ..lJ «Bug . 800i is. 803i 18.-.330; q .4 . .llllJ _ _t. _ _ _ _ _ _ “Had Gna— Wflau GHQ“ “dad Quad “an 86a 8.3.3 nee. 8.3.3 3... 86!.6. ...: tastes. . 833...... Beer «Inn a o a "drum ......Q A _ . . . A u n «Lon—OD 8.8.! one. 8.3.8 one. coo—3N 80.. 88.5.2 ease—.....fl ease? 3:3 . l l o and. ......a A _IIJ ] WIJ N 80:00 8.3 «no. 8.3.8 one. 08.3.3 8.: oust-2 . 39.5....— ease? tune I l l «alum ......E a. . . . _ . . 20:00 05.05:. 169 APPENDIX B HUMAN SUBJECTS APPROVAL \ll('lll(w\\ “All l\l\'l-R.\l'l'\' , APPENDIX B --_... l\|\ll’\ll\ ill\l\llllll Il\ Ni‘l \lil ll l\\lil\l\'. l\\l l-\\\l\(t 0 \ll( lll(u-\\ ... u \\ .1 um I\ .u ( mm. HUMAN SUBJECTS APPROVAL .‘~‘~ -\|)\vl|\l\ll\’\llll.\ lll lll)l\t. 0 488211040 ISl‘l 48% Ill“. M ay 26, 1988 Dennis R. Keefe Family & Child Ecology 107 Human Ecology Dear Dr. Keefe: Subject: "FAMILY RESOURCE UTILIZATION AS A FACTOR IN DETERMINING ECONOMIC WELL-BEING OF RURAL FAMILIES [RBfi 85-1 1" UCRIHS' review of the above referenced project has now been completed. I am leased to adv15e that smce rewewer comments have been satisfactorily addressed, the conditional approvai given by the Committee at its May 2, 1988 meeting has been now changed to full approva . ‘ You are reminded that UCRIHS approval is valid for one calendar year. If you plan to continue this project beyond one year, please make provisions for obtaining appropriate UCRIl-IS approval oanonth prior to May 2. 1982 Any changes in procedures involving human subjects must be reviewed by the UCRII-IS prior to initiation of the change. UCRIHS must also be notified promptly of any problems (unexpected side effects, complaints, etc.) involving human subjects during the course of the work. Thank you for bringing this project to our attention. If we can be of any future help, please do not hesxtate to let us know. Sincerely, John K. Hudzik, PhD. Chair, UCRIHS JKH/sr 170 MICHIGAN STATE UNIVERSITY 171 ()HICE Oi VIC} PRISIIIENT POI RESEARCH EAST LANSING 0 MICHIGAN 0 48814-1046 AND DIAS OI Illi. CRADL'ATE SCHOOl March 13, 1991 Ms. Connie Kratzer Unit 3 Paolucci Building RE: ECONOMIC WELL-BEING OF THREE COHORTS 0F RURAL FEMALE HOUSEHOLD FINANCIAL MANAGERS, IRB#9l-OBS Dear Dear Ms. Kratzer: The above project is exempt from full UCRIHS review. I have reviewed the proposed research protocol and find that the rights and welfare of human subjects appear to be protected. You have approval to conduct the research. You are reminded that UCRIl-is approval is valid for one calendar year. If you plan to continue this project beyond one year, please make provisions for obtaining appropriate UCRIHS approval one month prior to February 28, 1992. Any changes in procedures involving human subjects must be reviewed by the UCRIHS prior to initiation of the change. UCRIHS must also be notified promptly of any problems (unexpected side effects, complaints, etc.) involving hmn subjects during the course of the work. Thank you for bringing this project to our attention. If we can be of any future help, please do not hesitate to let us know. Sincerely, - 2. vid 3. Wright, Chair, ucarns DEW/deo cc: Dr. Dennis Keefe MSU 'u all A/fl'ruurl're Attics/Equal Opportunity Institution APPENDIX C INSTRUMENT APPENDIX C INSTRUMENT SELECTED QUESTIONS FROM THE FAMILY ECONOMIC WELL-BEING SURVEY: A01. Scale FINANCIAL MANAGER QUESTIONNAIRE Most of us have ways to handle our finances and time. Please circle how often you: 1 - Never 4 — Usually 2 - Seldom 5 - Most of the time 3 - Occasionally a Make plans on how to use your money 1 2 3 4 5 b Save on a regular basis for goal(s) 1 2 3 4 5 c Write down where money is spent 1 2 3 4 5 d Use a written budget 1 2 3 4 5 e Evaluate spending on a regular basis 1 2 3 4 5 f Evaluate your needs before you buy 1 2 3 4 5 g Keep bills and receipts where they are easy to find 1 2 3 4 5 h Pay interest on charge accounts 1 2 3 4 5 i Buy on impulse 1 2 3 4 5 j Are sorry you bought something 1 2 3 4 5 k Make a list before you shop 0 Make only minimum payments on charge accounts . 1 2 3 4 5 2. What types and sources of credit do you or your family normally use, excluding mortgage and/or business credit? (Circle all that apply.) a Credit card e Pawn broker b Bank f Friends c Credit union g Family d Finance company h Don't use credit 3. What kinds of types of insurance does your family have? (Circle gll that apply.) a Health, HMO, and or e Automobile Hospitalization f House b Health catastrophe g Flood, earthquake c Life h Liability d Disability 172 173 Do you or anyone living in your household have a chronic health condition that concerns or worries you? N0 (IF NOT, SKIP TO QUESTION 6) YES Which family members? YOU b. SPOUSE C. CHILD d. OTHER Do you or anyone living in your household have a chronic health condition that is a financial strain for you? NO 2 YES C.3. At the present time are you: 1 Employed or self-employed (SKIP TO QUESTION 5) 2 Unemployed 3 At home full time (not employed) 4 Retired (SKIP TO QUESTION 5) D. Please circle the response that best describes how satisfied or dissatisfied you are with each. SCALE: , 5 - Very satisfied 2 - Dissatisfied 4 - Satisfied 1 - Very dissatisfied 3 - Mixed feelings 1 Current total family income 5 4 3 2 1 2 The material things (food, clothing, housing, transportation) you have or use 5 4 3 2 1 The resources you have available to meet a financial emergency 5 4 3 2 1 The amount of your family's net worth (All assets minus debts) 5 4 3 2 1 174 These questions ask about your financial situation. Akin ltxh the "mil mtter gtter gm mrse worse 1. Compared to 5 years ago, would you say that your family’s present financial condition is: 5 4 3 2 1 2. Thinking of the future, 5 years from now, do you expect that your financial condition will be: S 4 3 2 l 3. Thinking about your current income, to what extent do you think your income is enough to live on? Not at all'adequate Can meet necessities only Can afford some of the things we want Can afford about everything we want Can afford about everything we want and still save money manure—- Please indicate the extent to which you AGREE or DISAGREE with the following statements, as they apply to you: Stmngl y Strongly slim Em Lixed Ages _m 1. When I make plans, I am almost certain that I can make them work 1 2 3 4 5 2. It is not always wise to plan too far ahead because many things turn out to be a matter of good or bad fortune anyhow l 2 3 4 5 3. Many times I feel that I have little influence over the things that happen to me What happens to me is my own doing 1 2 3 4 5 My financial situation depends on my control of the situation 1 2 3 4 5 6. It is impossible for me to believe that chance or luck plays an important role in my life 1 2 3 4 5 7. Sometimes I feel that I don’t have enough control over the direction my life is taking 1 2 3 4 5 8. Sometimes I feel that I don’t have enough control over the family income 1 2 3 4 5 175 We’d like to ask you a few questions about the people who usually live with you. Please start with yourself. 1. who lives in your household? Sex Age on last (Circle) birthday a You 1 M F 2 b Spouse (if any) 1 M F 2 Children and/or stepchildren living with you. Fill in the ages; leave blank if no children living with you. c Sons: 1 2 3 4 5 6 d Daughters: 1 - 2 3 4 5 6 Other adults and/or family members living with you. Please fill in the age and relationship (friend, grandfather, etc.) for each person: Age Relationship e l 2 f 1 2 g 1 ____ 2 2. What is your present marital status? (Circle only one.) First marriage number of years Separated Hidowed Divorced Remarried number of years Never married oscianuana—o 3. what is the highest number of years in school you have completed (high school graduate-12 years; 1 year college or trade school=13 years; B.S.-16 years; etc.) years 4. Please indicate your racial or ethnic background. (Circle one.) 1 Black 5 Spanish descent (Mexican, 2 White Puerto Rican, Chicano, 3 Native American other Spanish) 4 Asian or Pacific Islander 6 Other (please specify) 176 These questions are about your shared family or household income. 5. There are many sources of household income. you have for you and your family. mm 0.0 U!” 310 Wages or salary from job Your own business Savings interest Investments Pensions Alimony, spousal maintenance Child support Scholarships for education 0 3 a -x- ¢~.a- Circle all the sources Gifts from family Social security (Survivors, disability, retirement) AFDC, General Assistance, SSI Worker’s Compensation Farm support programs Rental income Other (specify) Now, think about the total 1987 income before taxes for your household from all the above sources. applies for your household) U‘J-Q’hih‘ The ‘total amount is: (Please circle the gag that Less than 55,000 6 5 5,000-5 9,999 7 510,000-514,999 8 515,000-519,999 9 520,000-524,999 10 525,000-529,999 530,000-534,999 535,000-539,999 540,000-544,999 545,000-549,999 550,000-559,999 560,000-569,999 570,000-579,999 580,000-599,999 $100,000 and over Please circle all the household and business/farm assets you (and your spouse, if married) own or are currently buying. d-SWO #10 050 0’9 Household Own home Second home, vacation home Any vehicles Checking account Savings account Certificate(s) of deposit Stocks or mutual funds IRA, KEOGH, 403b funds Other: jewelry, antiques, household possessions, 5 —J7th etc. u in F Business Vehicles if used in business farm land, rental property Livestock, farm equipment and buildings, etc. For only your household assets, what is your estimate of the total value? 1 2 3 4 S 5 5,000-5 7,499 5 7,500-5 9.999 510,000-514,999 515,000-519,999 (Circle the one that applies.) 5 0-5999 6 51.000-51.999 7 52,000-52,999 8 53,000-53,999 9 54,000-54,999 10 $20,000‘5299999 530,000-549,999 550,000-574,999 575,000-599,999 5100,000-5199,999 $200,000 or more 10 177 Please circle all the categories in which you (and your spouse, if married) have any debts. Household Mortgage on own home Mortgage on rental property or other home(s) or real estate Automobile or other vehicle loan(s) Credit card Home improvement loan(s) Educational loan(s) Doctor, dentist, hospital, nursing home bills 0“” Brink; O :1 P-L'T \QHHDQ: Other: Bus' ess Farm Your business Vehicles Farm land Rental property mortgage Livestock, farm equipment and buildings, etc. Loan(s) owed to friend, family member Think about all of your household.debts circled above» What do you estimate is the total amount of these household debts? (Please circle the one category that applies to you.) 1 $0; no debt 6 s 3,0 O-$ 3,999 2 $1-$499 7 s 4,0 O-$ 4,999 3 $500-$999 8 $ 5,000- $ 7,499 4 $1,000-$1,999 9 $ 7, 500- $ 9,999 5 $2,000-$2,999 10 $10, 000- $19, 999 11 $20,000-S29,999 12 $30,000-$49,999 l3 $50,000-$74,999 14 $75,000-$99,999 15 $100,000 or more FAMILY ECONOMIC WELL-BEIN G SURVEY 178 A. 1. Most of us have ways to handle our finances and time. often you: 413-th00'“? 3' a d r u. do 179 Make plans on how to use your money Save on a regular basis for goal(s) Write down where money is spent Use a written budget Evaluate spending on a regular basis Evaluate your needs before you buy Keep bills and receipts where they are easy to find Pay interest on charge accounts Buy on impulse Are sorry you bought something Make a list before you shop Combine shopping with job or errands feel you are doing a good job with finances Worry about where the money will come from to pay bills Make only minimum payments on charge accounts Make plans on how to use time Have financial concerns that affect your relationship Discuss your finances without getting upset Do things when they need to be done Do not have money problems gfldan NNNNNN NNNNNN (tra- $011311! wwwuww wwuuww (gually .5 &&#A& A‘bbhh Please circle how anLof thetfime mmmmmm 010101010105 2. 180 What types and sources of credit do you or your family normally use, excluding mortgage and/or business credit? (Circle all that apply.) a Credit card e Pawn broker b Bank f friends c Credit union 9 Family d Finance company h Don’t use credit What kinds or types of insurance does your family have? (Circle all that apply.) a Health, HMO, and/or Hospitalization Automobile b Health catastrophe House e f c Life 9 Flood, earthquake d Disability h Liability Do you or anyone living in your household have a chronic health condition that concerns or worries you? 1 N0 [IF NO, SKIP T0 QUESTION 6] 2 YES Which family members? a YOU b SPOUSE c CHILD d OTHER Do you or anyone living in your household have a chronic health condition that is a financial strain for you? I N0 2 YES During the past year did you have a loss of income at any time? 1 N0 2 YES How long? (weeks) Did you have any large unexpected expenses during the last year such as uninsured dental or medical expenses, car repairs, more taxes due than planned, etc.? 1 NO [IF NO, SKIP T0 QUESTION 4] 2 YES do —1 W L4. 181 Please indicate what you did to handle the income loss or any unexpected expenses. (Circle as many as apply.) Used money saved for emergencies Used money from regular savings Used money from checking or money market account Borrowed from bank, savings & loan, credit union or other sources Borrowed money from a friend or family member Put it on a charge account Put off paying other bills Did without new clothes, entertainment, etc. Earned extra money Sold something (please specify) Didn’t pay ruda'o 450 0.0 0'” . For each statement below, circle the response which best describes how satisfied or dissatisfied you are with that part of your life. very Mixed Dissatis- dissatis- eetisfied Set Wet j;_1jng§ fied fied Ability to achieve success and get ahead 5 4 3 ‘ 2 1 Way you use all your resources 5 4 3 2 1 Amount of money your family is able to save 5 4 3 2 1 Amount of your current debt 5 4 3 2 1 Time you spend on household tasks 5 4 3 2 1 Time you have to do things you want to do 5 4 3 2 1 Amount of work your family does around your house 5 4 3 2 1 Proportion of work you do compared to others in your household Your housing Your neighborhood Your community Extent to which you control your life S 4 3 2 1 Extent to which you control your financial situation 5 4 3 2 I The things you have and the way you are living now 5 4 3 2 1 / 182 ’5. All families have some problems when it comes to spending money. How often do you have the following problem? m- mmmym a Get behind on the rent or house payment 1 b Not able to buy special things 1 c Do not have enough money for dentist, doctor, or medicine 1 d Do not have enough money to pay for health insurance e Cannot afford to keep car(s) in running order f Cannot afford to pay for utilities 1 9 Cannot afford to buy adequate insurance 1 h Cannot afford to buy new shoes or clothes I i Not able to save to have something to fall back on 1 People spend their time in different ways. amount of time you spend in the different activities listed below. I. About how many hours do you usually spend in a 25-hour eey: (round to nearest quarter of hour) a sleeping and napping? b commuting to and from work? 2. About how many hours per_ueek do you usually spend: (O-none) a earning your income? b working in a family business operation (including farming/ranching) for which you are not paid c doing home-related work (such as cooking, cleaning, care of family members, repair, yard work)? d involved in religious and community activities? e involved in educational activities for self- improvement or job training? f as leisure time? 2 2 3 3 Think about the average Mist of theifime 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 hrs./day mins./day hrs./week hrs./week hrs./week _ hrs ./week ____ hrs./week ____ hrs./week I 183 43. At the present time are you: Employed or self-employed [SKIP TO QUESTION 5] Unemployed At home full time (not employed) Retired [SKIP TO QUESTION 5] tho—I . If unemployed or not employed, which of the following beet describes your current situation? A spouse and/or parent is a full-time job [SKIP T0 QUESTION 7] Laid off temporarily Laid off long term Out on strike Stopped work for personal reasons such as school, family responsibilities, health reasons, etc. Cannot find employment in area of training No jobs available that pay enough to allow me to help my family No Jobs of any kind are available Other (specify) sodium mwat-a . If employed, self-employed, unemployed or retired, what is (or was) your main paid job (for example: salesperson, farmer, rancher, high school teacher, resort worker, electrician, janitor, cook, etc.)? a (Job I) If you have other paid work, what is it? b (Job 2) c (Job 3) . If employed or self-employed, please indicate the situation for each Job you have. Job I is your main one. Job 1 (main) Job 2 Job 3 a Average been; worked per week I 2 3 b Average geeke worked per year 1 2 3 _—‘- a... '_..-- -.. .. - ‘455/ 184 ’7. Think about your pen 1987 personal income before taxes (not your family or business/farm income). Please circle the source or sources of your personal income. IF NO personal income SKIP TO D. Wages or salary from Job h Gifts from family Savings interest i Social security Investments . (Survivors, disability, retirement) Pensions J AFDC, General Assistance, SSI Alimony, spousal maintenance k Worker’s Compensation Child support I Other (Specify Scholarships for education O'NDQC'DU’U '9?{’ .8. Think about the amount of your personal 1987 income before taxes from all these sources. Please circle the te1e1_emegnt that applies to you. 1 Less than 55,000 6 525,000-529,999 ll 550,000-559,999 2 5 5,000-5 9,999 7 530,000-534,999 12 560,000-569,999 3 510,000-514,999 8 535,000-539,999 13 570,000-579,999 4 515,000-519,999 9 540,000-544,999 14 580,000-599,999 5 520,000-524,999 10 545,000-549,999 15 $100,000 and over 9. Which of these statements best describes how regularly you receive most of yee: income? Income is received at: Same time and in same amount Same time but varies in amount Different times but is generally about the same amount Different times and varies in amount Do not receive any income (flufiIa’NH Please circle the response that beet describes how satisfied or dissatisfied you are with each. ‘me ifixai Ihssflrlhsan: Yew mmmmm 1. Current total family income 5 4 3 2 l 2. The material things (food, clothing, housing, transportation) you have or use 5 4 3 2 1 3. The resources you have available to meet a financial emergency 5 4 3 2 1 4. The amount of your family’s net worth (All assets minus debts) 5 4 3 2 1 :n\ 185 These questions ask about your financial situation. Abufl: Mxh the Itch better Better $13 119128 ELSE 1. Compared to 5 years ago, would you say that your family’s present financial condition is: 5 4 3 2 1 2. Thinking of the future, 5 years from now, do you expect that your financial condition will be: 5 4 3 2 1 3. Thinking about your current income, to what extent do you think your income is enough to live on? Not at all adequate Can meet necessities only Can afford some of the things we want Can afford about everything we want Can afford about everything we want and still save money (”bum—- Please indicate the extent to which you AGREE or DISAGREE with the following statements, as they apply to you: Stnrgty Sbnmgty mmmmmmm 1. When I make plans, I am almost certain that I can make them work 1 2 3 4 5 2. It is not always wise to plan too far ahead because many things turn out to be a matter of good or bad fortune anyhow l 2 3 4 5 3. Many times I feel that I have little influence over the things that happen to me 1 2 3 4 5 4.‘ What happens to me is my own doing I 2 3 4 5 5. My financial situation depends on my control of the situation 1 2 3 4 5 6. It is impossible for me to believe that chance or luck plays an important role in my life 1 2 3 4 5 7. Sometimes I feel that I don’t have enough control over the direction my life is taking 1 2 3 4 5 8. Sometimes I feel that I don’t have enough control over the family income 1 2 3 4 5 186 Think of what has been happening in your community Please read each statement. g1re1e_ehe_!e:g_er_phre§e which best completes the sentence. Then indicate how each of these changes has affected your household by circlingethe epp:epg1e§e_ngmger under ”better off," 'not affected," or ”worse off". How has this event affected your family/household finan- cially? We are (please circle) (Circle word or phrase to Better Not Worse complete sentence) Q££___ Affgglgfi 911.. 1. Employment conditions, such as number and types of jobs available have (improved, worsened, not changed, don’t know). 3 2 1 2. There has been an (increase, decrease, no change, don't know) in the number of businesses in town. 3 2 l 3. The number of persons seeking welfare has (increased, decreased, not changed, don't know). 3 2 1 4. Large number of persons are (loving into, moving out of, no change, don’t know) the community 3 2 1 S. The cost of health care has (increased, decreased, not changed, don't know). 3 2 I 6. The availability of community services, such as public schools, police/fire protection, garbage collection, etc. has (increased, decreased, not changed, don’t know). 3 2 1 7. The cost of borrowing money has made it (easier, more difficult, no change, don't know) to obtain money for business and/or personal use. 3 2 1 8. In the past year, farm support programs such as the federal commodity programs have benefited (lore, fewer, no change, don't know) persons in the community. 3 2 l 9. In the past year, education resources such as Cooperative Extension, have benefited (wore, fewer, no change, don't know) persons in the community. on N p—a in. 187 We’d like to ask you a few questions about the people who usually live with you. Please start with yourself. / f1. Who lives in your household? Sex Age on last (Circle) birthday a You ‘ 1 M F 2 b Spouse (if any) 1 M F 2 Children and/or stepchildren living with you. Fill in the ages; leave blank if no children living with you. c Sons: 1 Z 3 4 5 6 d Daughters: 1 2 3 4 5 6 Other adults and/or family members living with you. Please fill in the age and relationship (friend, grandfather, etc.) for each person: Age Relationship e l 2 f l 2 g l 2 ~2. What is your preeent marital status? (Circle only one.) First marriage number of years Separated Widowed Divorced Remarried number of years Never married aim-wao-a '3. What is the highest number of years in school you have completed (high school graduate-12 years; 1 year college or trade school-I3 years; B.S.-16 years; etc.) years /4. Please indicate your racial or ethnic background. (Circle one.) 1 Black 5 Spanish descent (Mexican, 2 White Puerto Rican, Chicano, 3 Native American other Spanish) 4 Asian or Pacific Islander 6 Other (please specify) 188 'These questions are about your shared family or household income. /5. There are many sources of household income. Circle all the sources you have for you and your family. a Wages or salary from job i Gifts from family b Your own business j Social security c Savings interest (Survivors, disability, retirement) d Investments k AFDC, General Assistance, 551 e Pensions l Worker’s Compensation f Alimony, spousal m Farm support programs maintenance n Rental income 9 Child support 0 Other (specify) h Scholarships for education ,q9< Now, think about the totel 1982 ineome befere texe; for your household from all the above sources. The total amount is: (Please circle the pee that applies for your household) 1 Less than 55,000 6 525,000-529,999 ll 550,000-559,999 2 5 5,000-5 9,999 7 530,000-534,999 12 560,000-569,999 3 510,000-514,999 8 535,000-539,999 l3 570,000-579,999 4 515,000-519,999 9 540,000-544,999 l4 580,000-599,999 5 520,000-524,999 IO 545,000-549,999 15 $100,000 and over Please circle all the household and business/farm assets you (and your spouse, if married) own or are currently buying. fleeeehelg Businesszfiarm Own home j Business Second home, vacation home k Vehicles if used in business Any vehicles 1 Farm land, rental property Checking account m Livestock, farm equipment Savings account and buildings, etc. Certificate(s) of deposit Stocks or mutual funds IRA, KEOGH, 403b funds Other: jewelry, antiques, household possessions, etc. d-S'CD 'h@ an 5'0! For 9311 your hegeehelg assets, what is your estimate of the total value? (Circle the gee that applies.) 1 5 0-5999 6 5 5,000-5 7,499 11 530,000-549,999 2 51,000-5l,999 7 5 7,500-5 9,999 12 550,000-574,999 3 52,000-52,999 8 $10,000-514,999 13 575,000-599,999 4 53,000-53,999 9 515,000-519,999 l4 $100,000-$199,999 5 54,000-54,999 10 520,000-529,999 15 5200,000 or more 189 9. Please circle all the categories in which you (and your spouse, if married) have any debts. Household a Mortgage on own home b Mortgage on rental property or other home(s) or real estate n loan(s) Credit card Educational loan(s) nursing home bills dofl' to 450 Oh Other: Automobile or other vehicle Home improvement loan(s) Doctor, dentist, hospital, 3 a -aarcm Loan(s) owed to friend, family member 10. Think about all of your hogeehele debts circled above. Busineeszferm Your business Vehicles Farm land Rental property mortgage Livestock, farm equipment and buildings, etc. What do you estimate is the total amount of these household debts? (Please circle the gee category that applies to you.) 50; no debt 51-5499 5500-5999 51,000-51,999 52,000-52,999 U13b9’hohd 6 7 8 9 10 5 3,000-5 3,999 5 4,000-5 49999 5 5,000-5 7,499 5 7,500-5 9,999 $10,000-519,999 II 12 13 14 15 11. Do you have grown children, MOT LIVING WITH YOU? I 12. Please tell us: 520,000-529,999 530,000-549,999 550,000-574,999 575,000-599,999 $100,000 or more 13. Do any of your grown children live in your community? 14. IF NO, why? If you live alone, TURN TO PAGE 16 for a few'more questions. If you have no spouse or partner, but live with your child(ren), SKIP TO PAGE 15, QUESTION 20. All others continue with the next page. 2 YES Education level ___ 2 ___ 2 ___ 2 1 N0 2 YES 190 These items describe various characteristics within couple relationships. Please describe your relationship with your spouse or partner. (Ice Almst in a Sare- me- Altmst mahflemesmflrm 1. We ask each other for help. I 2 3 4 5 2. When problems arise, we compromise. 1 2 3 4 5 3. We approve of each other’s friends. 1 2 3 4 5 4. We are flexible in how we handle our differences. 1 2 3 4 5 5. We like to do things with each other. 1 2 3 4 5 6. Different persons act as leaders in our marriage. 1 2 3 4 5 7. We feel closer to each other than to people outside our family. 1 2 3 4 5 B. We change our way of handling tasks. 1 2 3 4 5 9. We like to spend free time with each other. 1 2 3 4 5 10. We try new ways of dealing with problems. 1 3 4 5 11. We feel very close to each other. 1 3 4 5 12. We jointly make the decisions in our marriage. 1 2 3 4 5 13. We share hobbies and interests together. 1 2 3 4 5 14. Rules change in our marriage. 1 2 3 4 5 15. We can easily think of things to l 2 3 4 5 do together as a couple. 1 2 3 4 5 16. We shift household responsibil- ities from person to person. 1 2 3 4 5 17. We consult each other on our decisions. 1 2 3 4 5 18. It is hard to identify who the leader is in our marriage. 1 3 4 19. Togetherness is a top priority. 1 3 4 20. It is hard to tell who does which household chores. 1 2 3 4 5 191 Here are some questions about your relationship with your spouse/partner. Sore- Alnost Never Seldom $311.12.; Alum Do you and your spouse/partner talk about things which are of interest to both of you? I 2 3 4 Do you discuss intimate matters with him/her? l 2 3 4 Do you and your spouse discuss personal problems with each other? 1 2 3 4 Do you and your spouse talk over pleasant things that happen during the day? I 2 3 4 Do the two of you ever sit down just ' to talk things over? 1 2 3 4 Does he/she have a tendency to say things that would be better left unsaid? 1 2 3 Does your spouse nag you? 1 Do you and your spouse remain silent for long periods when you are angry with one another? 1 2 3 4 Do you and your spouse find it hard to disagree with one another without losing your tempers? l 2 3 4 Does your spouse insult you when ' he/she is angry with you? 1 2 3 4 Do you have a tendency to keep your feelings to yourself? 1 2 3 4 Do you fail to express disagreement with him/her because you are afraid he/she will get angry? I 2 3 4 Do you find it difficult to express your true feelings to him/her? 1 2 3 4 Do you hesitate to discuss certain things with your spouse because you are afraid he/she might hurt your feelings? I 2 3 4 192 Following are several items dealing with your satisfaction with your marriage and child(ren). Please respond only to the items that pertain to you. .2, How satisfied are you with: Bdrmmflylkmy Saunhat Sounhat \kmy deemfly satis- satis- satis- dissatis- dissatis- dissatis- ified ified ified Mixed mg _i_fi_g_ mg a your marriage? 7 6 5 4 3 2 1 your relationship with your spouse? 7 6 5 4 3 2 l c your husband/wife as a spouse? 7 6 5 4 3 2 1 d yourself as a parent? 7 6 4 1 se your children (or children’s behavior)? 7 6 5 4 3 2 1 =f your relationships with your children? 7 6 5 4 3 2 l «5 your family life? 7 6 5 4 3 2 1 Eh your children’s relation- ship with each other? 7 6 S 4 3 2 l 193 THANK YOU for taking the time to complete this questionnaire. Your willingness to share this information allows research to be completed which is useful in understanding the economic well-being of families. The information from those who participate could affect public policy and further the understanding of household economic behavior. If you have comments or questions regarding this survey, please use the space below for this purpose. We would very much like to hear your comments and will reply to questions if you submit your name and address. APPENDIX D CORRELATION TABLES APPENDIXD (DRREIATION TABLES .o. m a: 8. ma. 83 8am .8. 83 8:5 .882 .8. :8m. 83 8.3m :4... .....w. :o.~. 83 83 .o.. :48. .8.. ...mm. 83 . 828m 28:. ...... c8. :48. .....N. ...8. 83 828m 8.8... .8.- :8... 8..- 8..- .8.- «8.. 83 5.8: m8. :8... :«8. :8m. .8.. 95.. 8o: 83 m1: 8... m8: 8... «3.. 48.- 3.. n8. «.8.. 83 .3“. .8. .o... .8.. ...8. :8... ...... ....~.. 8... o... 83 38.. 8... 8.. .3... 08.. 03.. 2.... 8.. 8o. .8... 8... 83 83:5 .8.. to: ..o... 8. «B. :5... 8o. .3... :3. :8m. .88.. 83 €02.32 a... :34. ...8. :8... :9... ...8. 8... :8... 2o; :8»... 8... .8.. 83 2:8... N8. .8. 08. :8... .....o. :8... :8... . .8.. m8. :8... :08. . :8... :02. 83 9.6.. 3am 3.2m SEEM. oauw moosom moccaom 5.3: 9*. AS“. 260.. ozmm £33 3.8:. m>>mm .852 Sam... 2.50:. _5 .oz 8.6588sz Sacco... can 20:339. 53. 0531.63 QEocooo oozoocod .332 3:22.00 ”2958 .20... . .N w4m995 oauw «859w moosoo .....on or: n5“. goo. 0.3.. 3.8.2 23:. 068:. .5 5.02. 2.8:. Ema. .02 3.6.5.0220 Emacs... new 39.339. 5.3 9.3.3: o.E0cooo 328.3. .332 5.3.2.00 ”m cocoo .vw o.o.; LI ST OF REFERENCES LIST OF REFERENCES Schemas, sense of control and ag1ng. ' - Self. Abelar, R. (1990). In J. Rodin, C. Schooler, & K. W. Scha1e (Eds.) digectedness; Cause and egfigcts throughout the lite course. (pp. 85-94). Hillsdale, NJ: Lawrence Erlbaum Associates. & Paolucci, B. (1983). Objective and subjective income adequacy: Their relationship to perceived life quality measures. figgig1_lngiga§g:§ Research, 3, 25- 48. Bubolz, M. 8 Paolucci, B. (1980). An Marriage Andrews, N., ecological approach to study of the family. and Family Reviews, 1 Spring/Summer (1/2) es a . Babbie, E. (1986). he ract'ce soc Belmont, CA: Wadsworth Publishing Co. Baltes, M. N., & Baltes, P. B. (Eds.). (1986). In; c o o 0 an ' . Hillsdale, NJ: Lawrence Erlbaum Associates. Rural families and their economic well- Bauer, J. (1990). being; procedures, methods and conceptual model of In M. Carsky North Central regional project NC-182. (Ed. ) Proceedings of Age niggn 9933911 on Consume ; Interests. (p.329). Columbia, MO: ACCI. Bauer, J., Danes, S. 8 Rettig, K. (1990, November). neasurement of economic well-being in family studies, Paper presented at the Theory Construction and Research Workshop, National Council on Family Relations, Seattle, WA. Bengston, V. L. & Schaie, K. W. (Eds.). (1989). Ihg_ggg;§§ ect'o . New York: a e f esea Springer Publishing Company. Ackerman N., Family cash-flow Beutler, I.F., & Mason, J. W. (1987). budgeting. WM]. 1.6. 3-12- 198 199 Blaylock, J. R. & Blisard, W. N. (1990). Eppnpm1§_flgll; e'n ouse o e: e n v wa s a1 zi emo a ' o a ' u . (Report No. 640). United States Department of Agriculture, Economic Research Bureau. Block, M. R., Davidson, J. & Grambs, J. (1981). upp§n_pyg; forty: Vision and Realties. New York: Springer Publishing. Brofenbrenner, U. (1986). Ecology of the family as a context for human development: Research perspectives. Developmental Psychology, z;(6) Bubolz, M., Eicher, J., & Sontag, S. (1979). The human ecosystem: A model. qurpal pf flome Egpppmigg, 11(1) 28-31. Bubolz, M. & Sontag, M. S. (in press). Human ecology theory. In P. Boss, W. Doherty, R. LaRossa, W. Schuman, S. Steinmetz (Eds.) 0 ebook am pheories and methods: A contextual appzoagn, New York: Plenum Press. Bugaighis, M. & Schumm, W. (1983). Alternative measures of perceived locus of control. £§Y§h212919a1_8e22rt§ 5;. 819-823. Burns, S. (1990). Retirement savings of baby boomers and an earlier cohort: 1960 and 1983. In M. Carsky. (Ed.) oceed' s o e ' a oun ' o o sume e s 5 (pp. 322-328). Columbia, MO: ACCI. Campbell, A. (1981). he sens we - e' ' e 'c : ece at e s an en . New York: McGraw-Hill. Campbell, R., Abolafia, J. & Maddox, G. (1985). Life-Course analysis in social gerontology: Using replicated social surveys to study cohort differences. In A. Rossi (Ed. ) Gender and the life pppzs e. (pp. 301-318). New York, NY: Aldine Publishing Company. Clausen, J. (1986). rue life gpuzse. Englewood Cliffs, NJ: Prentice-Hall. Coulson, L. (1990). Female baby boomers--savings over time. In M. Carsky. (Ed.) ceed' gppncil op Consume; lntgpests. (p. 316-321). Columbia, MO: ACCI. 200 Crossman, S. M & Edmonson, J. E. (1985). Personal and family resources supportive of displaced homemakers' financial adjustment. In M. J. Sporakowski (Ed.). Family Belatipns, 14(4). (pp. 455-465). Danes, S. M. & Hira, T. K. (1986). Knowledge, attitudes, and practices in the use of credit. In V. Hempler (Ed-) Ez22eeding§_2f_Ameri§an_ggungil_2n_§9nsnmer Intgrggtg. 254-260. Danes, S. M., Rettig, K. D., 8 Bauer, J. W., (1991 in press). Locus of control, gap between the standard and level of living and satisfaction with the gap: A path model replication- Er22eeding§_2f_tne_llth_Annual Conference pf the Anerican Qounpil pn gpngnmer Interests. Davis, E.P., & Helmick, S. (1983). Composite measures of financial satisfaction: An analysis of two indices. In M. M. Dunsing (Ed.) 0 ' ' C ° C ' W - , 27-39. Davis, E.P., & Helmick, S. (1985). Family financial satisfaction: The impact of reference points. fipng Economigg Beseargn Journnl, 15, 123-131. Davis, E. P. & Schumm, W. R. (1987). Savings behavior and satisfaction with savings: a comparison of low- and high- income groups. ome c m ese , 15(4), 247-256. Douthitt, R. MacDonald, M., Mullis, R. (1990). The relationship between perceived life satisfaction and measures of economic well-being. In M. Carsky. (Ed.) Erpceedings pf Anerican Qpnnpil nn Qpnsnmgr Interggts. Columbia, MO: ACCI. Easterlin. R. (1987). Birth_and_fgrtune1_the_imnast_ef numbers pn personal welfare. New York: Basic Books. Ebersole. P-. & Hess, P- (1985)- Toward_neal§nx_agingi_ uma eds n u s o s . St. Louis: The C. V. Mosby Company. Edwards, J. 8 Booth, A. (1976). Sexual behavior in and out of marriage: An assessment of correlates. Jpnrnnl_nfi Marriage and Family. February, pp.73-81. Elder. 6. Jr. (1974). Qhildren_2f_the_great_denre§§ign- Chicago: University of Chicago. 201 Farraro, K. (1990). Cohort analysis of retirement preparation, 1974-1981. Journal pf Gergntology A; 1, $21-$31. Ferber, R. 8 Lee, L. (1974). Husband-wife influence in family purchasing behavior. Journal_9f_§2n§umer Beseargn, 1, June, 42-50. Fitzsimmons, V. S. (1987). Family life cycle changes in money management: Newlywed families and full-nest families. In M. Carsky (Ed. ). 23rd Annu nl Qpnfigrence of American Council on Consumer Interestg, 55-59. Fletcher, C. & Lorenz, F. (1985). Structural influences on the relationship between objective and subjective indicators of economic well-being. Spgigl_1ngigntpr§ Beseargn.1_, 333- 345. Foa, U. & Foa, E. (1980). Resource theory: interpersonal behavior as exchange. In Gergen, Greenberg and Willis (Eds.) Soc a xchan e' dv ces researcn. New York: Plenum Press. Foner, A. (1990). Social constraints on self-directedness over the life course. In J. Rodin, C. Schooler, & K. W. Schaie (Eds.). e f d' ectedness: use e e s u ou h ' u s . (pp. 95-102). Nillsdale, NJ: Lawrence Erlbaum Associates. Gatz, M., Siegler, I. C., George, L. K., 8 Tyler, F. B. (1986). Attributional components of locus of control: Longitudinal, retrospective, and contemporaneous analyses. In M. M. Baltes 8 P. B. Baltes (Eds.). Inn Esx2h212g1_gf_ggntrgl_and_aging- (pp. 237-265)- Millsdale, NJ: Lawrence Erlbaum Associates. Gergen, M. M. & Gergen, K. J. (1986). The discourse of control and the maintenance of well-being. In M. M. Baltes 8 P. B. Baltes (Eds.). s c o gontrgl and aging. (pp. 119-139). Hillsdale, NJ: Lawrence Erlbaum Associates. Glass. G- 8 Hopkins, K. (1986). Statistieal_metn2de_1n edugatign_and_2§x§hglosx- Englewood Cliffs. NJ: Prentice-Hall, Inc. Godwin, D. (1990). Toward a theory of family cash flow management. In R. 0. Weagley (Ed. ). 1221;__Annnnl ' -,--z' .- . 1e Assoc' . ,-1 ,o 1:1_°e °_!—‘ Tn- and_£lanning_fidugati_n- 96- 116- Godwin, D. (1990). Family financial management. Enmily WI 3.2: 221-226- 202 Godwin, D. D. 8 Carroll, D. D. (1986). Financial management attitudes and behavior of husbands and wives. Jonrnal pf gpngumer Studies nnd Home Eggngnigs, 19, 47-55. Goldscheider, F. (1990). The aging of the gender revolution. Researcno on Aging 12: 4, 531-545. Gross, I., Crandall, E. 8 Knoll, M. (1973). ppggrn_rnnili§§. New York: Appleton-Century-Crofts. Hafstrom, J. 8 Dunsing, M. (1973). Level of living: Factors influencing homemakers' satisfaction. fipne Egonomics Beseargn Journal, 2, 119-132. Hareven, T. (1981). Historical changes in the timing of family transitions: Their impact on generational relations. In R. Fogel, E. Hatfield, S. Kiesler, 8 E. Shanan, (Eds.) Aging stability and ghange in the family. New York: Academic Press. Hareven, T. (1983L American families in transition: Historical perspectives on change. In A. Skolnick 8 J. Skolnick, (Eds. ). m t 5' ° a ia e sexua t hil ea 'n an i grganizations. 4th Edition. (pp. 73-91). Boston/Toronto: Little Brown 8 Company. Hayhoe, C. R. (1990L Perceived economic well-being and active participation in financial decision making. In R- 0- .Weagley (Ed. ) 12211__Annual.£r2geeding§_ef_the $5 'at o a c u s Edugatign- 116-142- Hayhoe, C. R. (1990). Perceived economic well-being and active participation in financial decision making. In R. O. Weagley (Ed. ). ' e e 35 o n ' Educatipn. 188-204. Heise, D. (1990). Careers, career trajectories and the self. In J. Rodin, C. Schooler, 8 K. W. Schaie (Eds. L ' on e --n-ss' . se 10 ' ‘ - ! -_-!-- 1 lirg_gpnr§g. (pp. 59-84L Hillsdale, NJ: Lawrence Erlbaum Associates. Helmick, S. A. 8 Metzen, E. J. (1983L Composite measures of financial satisfaction. In M. M. Dunsing (Ed. ), o o o 9‘ ‘19. _y o g 010:: ‘ - aging, 27-40. Hess, B. (1985). Aging policies and old women: The hidden agenGA- In A- Rossi (Ed-) Eender_and_the_life_egur§e. (p.319-332). New York, NY: Aldine Publishing Company. 203 Hira, T. (1990). Residents in economically gaining and declining rural communities and their perception of changes in community and impact on household's financial situation. In M. Carsky. (Ed.) Erpggggingg gr Anerican Council pn Qonsumer Interests, Columbia, MO: ACCI. Hoff, E. 8 Hohner, H. (1986). Occupational careers, work and control. In M. M. Baltes 8 P. B. Baltes (Eds.). The Esycnglogy pf cgntrgl and nging. (pp. 345-372). Hillsdale, NJ: Lawrence Erlbaum Associates. Jensen, H. H. 8 Reynolds, 8. W. (1986) Consumer credit use and age: An analysis of the effects of preretirement. In V. Hempler (Ed. ) ee gpnsumer Interests. (pp. 261-264) Columbia, MO: ACCI. Jungen, B. (1985). Integration of knowledge in human eeolegy- Humanekglggiska_§krifter. 5 Februari Gotesburg Universitet. Kahne, H. (1981). Women and social security: Social policy adjusts to social change, Aging_nng_finnnn WI .11: 195-208 . Kain, E. (1990). e am e ' . Lexington, MA: Lexington Books. Keating, N. (1987). Reducing stress of farm men and women. , 3§,(4) 358-363. Keith, P. 8 Nauta, A. (1988). Old and single in the city and in the country: Activities of the unmarried. £1milx.Belatigns. 21. 1 79-84- Kilsdonk. A- G. (1983). Human.e2919921__ueaning_and_u§age. Human Ecology Monograph Series, No. 102, pp. 36-37. College of Human Ecology, Michigan State University, East Lansing, MI. Kinsey, J. 8 Lane, 8. (1978). The effect of debt on perceived household welfare. Jpnrnnl_pr_gpn§nmgr Affairs. 12(1). 48-61- Lachman, M. (1986). Personal control in later life: Stability, change and cognitive correlates. In M. M. Baltes 8 P. B. Baltes (Eds.). Ing_£§ygnplpgy_gfi . (pp. 207-236). Hillsdale, NJ: Lawrence Erlbaum Associates. Langer. E- (1983). The_n§xgnolggx_ef_sgntrel. Beverly Hill, CA: Sage Publications, Inc. 204 Latten, J. (1989). Life-course and satisfaction, equal for every-one? §opial Indicators Beseargn, 21 599-610. Lawrence, F., Carter, C. 8 Verman, S. Einnnginl 's a ’on 1 e1 'on to ' a ' a m Louisiana State University, Agricultural Center. Lefcourt, H. (1976). ocus 0 ° tneory and researcn, Hillsdale, NJ: Erlbaum. Lefcourt. H. (Ed.) (1983). Besear§h_xith_tne_19§n§_gf ppntrpl_ gpnstrnct Vpl,2. New York: Academic Press. Little, L., Prouix, F., Marlowe, J. 8 Knaub, P. (1987). The history of recent farm legislation: implications for farm families. Family Relatipns, 1§,(4) 402-406. Mammen, S. Helmick, S. A. 8 Metzen, E. J. (1983). Financial well-being and life satisfaction. In M. M. Dunsing (Ed.), oceed' s u e e' d Eggnomip Well-Being, 8-13. Mason, W. 8 Fienberg, S. (Eds.) (1985). Qpngrt_nnnly§i§_in sgpinl researpn. New York: Springer-Verlag. Meiners, J. 8 Olson, G. (1987). Household, paid, and unpaid work time of farm women. £nnily_3g1ntipn§, ;§,(4) 407-411. Mercier, J., Paulson, L. 8 Morris E. (1988). Rural and urban elderly: Differences in the quality of the parent-child relationship. £nmily_3g1ntipn§, 11(1), 73-79. Mertensmeyer, C. 8 Coleman, M. (1987). Correlates of inter-role conflict in young rural and urban parents. Family Relations, 35(4), 425-429. Metzen, E. 8 Helmick, S. (1975). Secondary workers' earnings and their impact on family income adequacy. B2me_E22nem12s_8e§earsh_leurn111 2. 249-259. Moen, P., Kain, E. 8 Elder, G. (1983). Economic conditions and family life: Contemporary and historical perspectives. In Nelson, R. 8 Skidmore, F. (Eds.) a am' 'es t c . (pp.213-259). National Academy Press. Washington, D.C. Money income and poverty status of households and families. (1988) £_milx_zsengmiss_3exiew 1(2). 13-14- 205 Mullis, R. J. 8 Schnittgrund. (1982). Budget behaviors: Variance over the life cycle of low income families. 12urnal_2f_Q2nsumer_Studies_and_n2me_ficengm12§. é. 113- 120. Nickell, P., Rice, A., 8 Tucker, 8. (1975). nanaganant_in fianily_1iying. New York: John Wiley 8 Sons, Inc. Norusis, M. J. (1986). SPSS/PC for the IBM PC/XT/AT. Chicago: SPSSinc. Ogden, A.S. (1986). The great American housewife: From helpmate to wage earner, 1776-1986. Westport, CN: Greenwood Press. Pearlin L. 8 Schooler, C. (1978). The structures of coping. Journal of Health and Sagial Banayigr, 12, 2- 21. Porter, N. M. 8 Garman, E. T. (1990). A conceptual framework for measuring financial well-being. In R. O. Weagley (Ed.). 1991; Annual Eroceaginga pf tna ss at' an ' o a Education. 142-152. Radner, D. B. (1987). Money incomes of aged and nonaged family unite. 1967-84. figsial_§eguritx_flulletin. 59(8), 9-28. Riley, M. (1985). Women, men, and the lengthening life course- In A- Rossi (Ed.) Gender_and_the_life ppnrpa. (pp.333-347). New York, NY: Aldine Publishing Company. Rodin, J. (1990). Control by any other name-definitions, concepts and processes. In J. Rodin, C. Schooler, 8 K. W- Schaie (Ede-)- Self_direstedne§s1__sause_and e1fe21s_threugnggt_the_life_sour§e- (pp- 1'18)- Hillsdale, NJ: Lawrence Erlbaum Associates. Rossi, A. (1985). Gender and tna lifia ggnraa. New York: Aldine Publishing Co. Rotter, J. (1976). Some problems and misconceptions related to the construct of internal versus external control of reinforcement. I2urnal_9f_ansulting_and_911nisal 2512821291- 11. 56-67- Ryder, N. (1965). The cohort as a concept in the study of social change. Amerisan_figsiglggieal_3exiex. 29.843- 861. 206 Ryff, C. (1985). The subjective experience of life-span transitions. In A. Rossi (Ed.) Qpngar_ang_tng_lira ppnrga. (pp.97-114). New York, NY: Aldine Publishing Company. Ryscavage, P. (1987). Income trends of the young and elderly. Family Econgmics Faviey, 2. 1-8. Schaie. K- 8 Willie. 8- (1986). Adult_dexeleement_and agingt Boston, MA: Little, Brown and Company. Schiamberg. L- B- (1988). thld_and_adglessent devalopment. New York: Macmillan. Scholl, K. (1983). Farm women's triad of roles. Family Fponomips Review, 1 10-15. Schooler, C. (1990). Individualism and the historical and social structures of peoples concern over self- directedness. In, C. Schooler, 8 K. W. Schaie (Eds.). self directedness: Qausa and effects tnrgngngnt tna lira_ppnr§a. (pp. 19-49). Hillsdale, NJ: Lawrence Erlbaum Associates. Scott, J. P. 8 Roberto, K. A. (1987). Informal supports of older adults: A rural-urban comparison. Family Balations, ;§(4), 444-449. Skinner, E. 8 Connell, J. P. (1986). Control understanding: Suggestions for a developmental framework. In M. M. Baltes 8 P. B. Baltes (Eds.). Ing_F§ypnplpgy_pf ppntrgl and aging. (pp. 35-71). Hillsdale, NJ: Lawrence Erlbaum Associates. Slusher, B. J., Helmick, S. A. 8 Metzen, E. J. (1983). Perceived economic well-being: The relative impact of value concordance and resource adequacy. In M. M. Dunsing (Ed.), e um 'ved ' W - ' , 14-27. Sontag, M, Bubolz, M., 8 Slocum, A. (1979). Perceived quality of life of Oakland County families: A preliminary report. Strumpel. 8- (ed-) (1976). E22n2m12_means_fgr_hnman_need§- Institute for Social Research, The University of Michigan, Ann Arbor MI. Titus, P. M., Fanslow, A. M. 8 Hira, T. K. (1989). Net worth and financial satisfaction as a function of households money mangers' competencies. fipna_annnnig§ Resear§h_lournal. 11(4). 309-318- 207 Uhlenberg, P. 8 Salmon, M. (1986L Change in relative income of older women 1960- 1980. Ina_§arpntplpgi§t. 2§(2), 164- 170. Walford-Kraemer, P. 8 Light, H. (1984). Depression and mastery in women: Differences according to personal characteristics. 2§22821221211_Be2211§. 11. 710- Walker, R., Bubolz, M. 8 Lee, M. (1991, June). 'butio am so mana e at family life at mid-lire women and pen: Paper submitted to 1991 Pre-Conference Workshop, Family Economics-Home Management Section, American Home Economics Association Meeting, Minneapolis, MN. Walker, R., Lee, M., Bubolz, M. 8 Keefe, D. (1990) Quality of life of middle-aged women: Effects of family resources and demands. In M. Carsky (Ed. ) Frpgaaginga ° !‘ ' ! ill-- 'I ° 1! pt gpnsnnar Interesta. COlumbia, MO: ACCI. Walker, L. 8 Walker, J. (1987). Stressors and symptoms predictive of distress in farmers. Fanily_3a1atign§, a§,(4) 374-378. Wallerstein, J. 8 Blakeslee, S. (1989). §agpnfi_gnanga§. New York: Ticknor 8 Fields. Ward, R., La Gory, M. 8 Sherman S. (1988). Ina_anyirpnmant E . . ! J i l l !I J ! l s . Tuscaloosa: The University of Alabama Press. Wetzel, J. R. (1990, March). American families: 75 years of change- Monthlx_Labgr_Bexi_1- pp- 4-13- Wilhelm, M., Ridley, C. (1988). Stress and unemployment in rural nonfarm couples: A study of hardships and coping resources. Eamilx_8_lat1en§ 11.(1) 56- 62- Williams, F. L. (1983). Money income, nonmoney income, and satisfaction as determinants of perceived adequacy of income. In M. M. Dunsing (Ed. ), Frpgagding_gfi_£h§ Exm22s12m_2n_2ereeixed_fisgnom19_flellzne1ng. 106-126- Uhr, E. 8 Evanson, E. (1983, Spring). The relative economic status of the aged. Fggns. University of Wisconsin-Madison, Institute for Research on Poverty. Vol 6, 2. Yaar, E. (1976). Effects of socio-psychological variables on subjective economic welfare. In Strumpel, B. (Ed.), Es9n2m12.neans_fgr_human_needs- Ann Arbor. MI: Institute for Social Research, pp.107-130.