=a I rHESl‘ r; 7' LIBRA R Y ' Michigan Stem: University This is to certify that the thesis entitled Interpersonal Resource Exchanges Preditors of Quality of Marriage and Family Life presented‘by Kathryn Dalbey Rettig has been accepted towards fulfillment of the requirements for Ph.D. degree in Family Ecology 74% W Major professor Datejfiiarc/‘S; / 75¢ 0-7 639 0 OVERDUE FINES: OCT 252003 f as 25¢ per day per its RETURNING LIBRARY MTERIALS: Place in book return to remove charge from circulation records INTERPERSONAL RESOURCE EXCHANGES AS PREDICTORS OF QUALITY OF MARRIAGE AND FAMILY LIFE By Kathryn Dalbey Rettig A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Family Ecology 1980 / (f ./ 1’! K/ -. I ABSTRACT INTERPERSONAL RESOURCE EXCHANGES AS PREDICTORS OF QUALITY OF MARRIAGE AND FAMILY LIFE By Kathryn Dalbey Rettig Purposes of the study were to: (1) Explore validity of Foa and Foa's resource exchange theory and measured indicators of the model. (2) Select the best set of indicators to predict marriage evaluation for women and men. (3) Investigate credibility of the Foa theory in predicting marriage evaluation fbr women and men. (4) Investigate the contribution of each interpersonal resource (love, status, ser- vices, information) and shared time to the prediction of marriage evalu- ation for women and men. (5) Describe differences in evaluations of marriage, family life, and life-as-a-whole for women and men. Data were collected by self-administered questionnaires dis- tributed in Oakland County, Michigan during Winter 1977-78 as part of the Quality of Life Research Project of the Departments of Family and Child Sciences and Human Environment and Design at Michigan State University. Michigan Agricultural Experiment Station funded the pro- ject (numbers 3151 and 1249) with additional support from the Univer- sity of Minnesota. The study sample consisted of 224 husband-wife couples living in the same household with at least one school-age child. Kathryn Dalbey Rettig Respondents evaluated their overall quality of life, quality of family life, and marriage. Information was obtained concerning evaluations of love, status, services and information resources received in the family, and shared time; perceived frequency of resource transfers from mate for each resource class, and perceived frequency of shared time with mate in five companionate activities. Hierarchical complete-linkage cluster analyses of evaluation and frequency variables indicated questionnaire items selected to represent resource classes did cluster as theory predicts. Four~ cluster solutions were found for men's evaluation variables and women's frequency variables. Three-cluster solutions for women's evaluation variables and men's frequency variables fused variable clusters "love" and "status" which theory indicates are the most highly correlated resource classes. Validity of the cluster solution decision was confirmed using three clustering methods which found three—cluster solutions for all analyses. Only the complete-linkage method was able to separate love and status variable clusters. Fusion order of clusters did lend support to the model of structured rela- tionships among resource classes. Mean scores of men on evaluations of marriage and family life were significantly higher than women's mean scores. Both sexes were more satisfied with marriage than with family life or overall quality of life. The forward method of multiple regression was used to predict evaluation of marriage. The best prediction accounted for 81 percent of the variance in the women's analysis, 75 percent in the men's Kathryn Dalbey Rettig analysis. The best predictor of this variable set was evaluation of "your husband or wife." The Foas' theory successfully predicted evaluation of marriage, particularly for women. Significant predictors for both sexes were evaluations of: (l) love/affection; (2) sexual relationship (love- services); (3) how comfortable it feels at home (services); (4) Open, honest expression of feelings (information); (5) things you do together (shared time); (6) frequency of receiving love from mate; (7) shared time frequency with mate. Additional significant predictors of marriage evaluation for women were evaluations of: (I) respect received (status); (2) the way decisions are made (information). DEDICATED To Kari and Heidi Rettig ACKNOWLEDGMENTS There are so many people who have significantly contributed to the quality of my life during the completion of this degree that it will be impossible to name them in this space; however, they are remembered and I wish to thank them as important support systems. I am grateful for the assistance of my graduate committee which represented the finest scholars who are also persons it is a privilege to know. Dr. Margaret Bubolz, major professor and dissertation director, was always patient, optimistic, encouraging and so willing to contri- bute time and ideas. The Opportunities for learning and the insightful suggestions she provided were much appreciated. Dr. Beatrice Paolucci was the reason I found the motivation to begin attending a graduate school which would involve a trip of 170 miles to every class. She has been an inspiration and has always gone far beyond "the call of duty" to assist me in every way. Dr. Donald Melcer so effectively contri- buted to my understanding of the dynamics of family relationships. Dr. Norman Stewart was an inSpiration as a counselor-educator in the classroom and as a scholar and author. I have had valuable encouragement from Dr. Linda Nelson, Dr. Gertrude Nygren and Dr. Jane Oyer. I am indebted to Dr. Margaret Liston, Professor Emeritus, Iowa State University for initiating my interest in family economics and management by demonstrating the 111 integrative nature of the subject matter and its relationship to all social sciences. Critically important technical support was provided by Leonard Bianchi, Janet Vredevoogd, Dr. M. Suzanne Sontag, and Dr. Thomas A. Rettig. I received financial support from Mildred Ericson Fellowships, the Departments of Family and Child Sciences and Family Ecology for graduate assistantships, the College of Human Ecology for a dissertation fellowship and Dr. Thomas A. Rettig who carried all eXpenses not covered by the above sources. My family and friends were invaluable sources of support at all times. I would like to thank my parents Mr. and Mrs. C. W. Dalbey for the gift of life and the secure, happy childhood which provided the emotional strength which has made it possible for me to most effectively use my abilities. My sister, brother and children have also been encouraging and helpful. My friends have provided child care, food preparation, holiday entertainment, transportation, therapy and assorted other services without which life would have been considerably more difficult. iv LIST OF LIST OF CHAPTER I. II. TABLE OF CONTENTS TABLES . FIGURES INTRODUCTION . Statement of the Problem . Purpose of the Study . Scope . Theoretical Framework Summary of Central Ideas, Foa Theory . Properties of Resources Appropriateness of Environment . Motivational State of Exchangers . Application of Foa Theory to Present Study . Theoretical Definitions Assumptions . . . . . . . . . . . . . . . . . Assumptions About Methods and Measurements . Assumptions About the Nature of Behavior and Its Context Research Objectives . Research Questions and Hypotheses REVIEW OF LITERATURE . Exchange Theory Historical Development . Foci and Assumptions . Utilitarian Sociologists . Psychological Behaviorism . Collectivistic Exchange Theory . Generalized Exchange . . . . Page ix O‘U‘MN H on 11 11 12 17 26 26 27 28 29 33 33 33 35 36 37 38 39 CHAPTER Page Methodological Issues . . . . . . . . . . . . . . . 40 Issues of Conceptualization . . . . . . . . . . . . 40 Economic and Social Exchange Characteristics . . 41 Explanation of Reciprocity and Altruism . . . . 46 Two-Party and Multi-Party Exchange Models . . . 52 Equity and Distributive Justice . . . . . . . . 52 Application of Exchange Theory to Family Research . . . 53 Marital Quality . . . . . . . . . . . . . . . . . . . 60 Family Well- Being . . . . . . . . . . . . . . . . . . . 67 Quality of Life . . . . . . . . . . . . . . . . . . . . 75 Summary . . . . . . . . . . . . . . . . . . . . . . . . 86 III. METHODOLOGY . . . . . . . . . . . . . . . . . . . . . . 88 Sampling Procedures . . . . . . . . . . . . . . . . . . 89 Data Collection Procedures . . . . . . . . . . . . . . . 90 Description of the Study Sample . . . . . . . . . . . . 92 Instrument Development Procedures . . . . . . . . . . . 96 Examination of Literature . . . . . . . . . . . . . 100 Representation of Resource Classes . . . . . . . . 100 Questioning About Family Life Satisfactions . . . . 101 Preliminary Questions . . . . . . . . . . . . . . . 102 Pretesting . . . . . . . . . . . . . . . . . . . . . 103 Origin of Questionnaire Items . . . . . . . . . . . . . 103 Description of Variables . . . . . . . . . . . . . . . . 106 Dependent Variables . . . . . . . . . . . . . . . . 106 Evaluation of Life-as-a-Whole . . . . . . . . . 106 Evaluation of Family Life-as-a-Whole . . . . . . 109 Evaluation of Marriage . . . . . . . . . . . . . 109 Independent Variables . . . . . . . . . . . . . . . 113 Evaluation Variables . . . . . . . . . . . . . . 113 Frequency Variables . . . . . . . . . . . . . . 113 Data Analysis Procedures . . . . . . . . . . . . . . . 117 Statistical Methods and Assumptions . . . . . . . . . . 117 Crosstabulation . . . . . . . . . . . . . . . . . . 117 Correlation Analyses . . . . . . . . . . . . . . . . 121 Cluster Analyses . . . . . . . . . . . . . . . . 122 Multiple Regression Analyses . . . . . . . . . . . . 125 vi CHAPTER Page IV. RESULTS . . . . . . . . . . . . . . . . . . . . . . . . 132 Descriptive Data for the Major Variables . . . . . . . . 132 Perceived Overall Quality of Life . . . . . . . . . 132 Perceived Overall Quality of Family Life . . . . . . 133 Affective Evaluation of Marriage . . . . . . . . . . 145 Descriptive Data for Independent Variables . . . . . . . 150 Independent Variables, Affective Evaluation . . . . 150 Independent Variables, Frequency . . . . . . . . . . 154 Cluster Analyses . . . . . . . . . . . . . . . . . . . . 158 Objective 1 . . . . . . . . . . . . . . . . . . . . 159 Hypotheses l . . . . . . . . . . . . . . . . . . 159 Hypotheses 2 . . . . . . . . . . . . . . . . . . 161 Hypotheses 3 . . . . . . . . . . . . . . . . . . 162 Research Question 4 . . . . . . . . . . . . . . 163 Multiple Regression Analyses . . . . . . . . . . . . . . 166 Selection Procedures fer Independent Variables . . . 166 Objective 2 . . . . . . . . . . . . . . . . . . . . 176 Research Question 5 . . . . . . . . . . . . . . 176 Research Question 6 . . . . . . . . . . . . . . 178 Research Question 7 . . . . . . . . . . . . . . 178 Objective 3 . . . . . . . . . . . . . . . . . . . . 189 Hypotheses 8 . 189 Hypotheses 9 . 197 Objective 4 . . . . . . . . . . . . . . . . . . . . 197 Hypotheses 10 . . . . . . . . . . . . . . . . . 198 Hypotheses 11 . . . . . . . . . . . . . . . . . 199 Hypotheses 12 . . . . . . . . . . . . . . . . . 200 Hypotheses 12A . 200 Hypotheses 128 . 201 Hypotheses 12C . 202 Hypotheses 13 . 202 Hypotheses 13A . . 203 Hypothesis 138 . 204 Hypotheses 14 204 Hypotheses 15 206 Hypotheses 16 . . 206 vii CHAPTER Page Hypotheses 17 . . . . . . . . . . . . . . . . . 207 Hypotheses 18 . . . . . . . . . . . . . . . . . 207 Hypotheses 19 . . . . . . . . . . . . . . . . . 208 Objective 5 . . . . . . . . . . . . . . . . . . . . 208 Hypotheses 20 . . . . . . . . . . . . . . . . . 208 Summary of the Results . . . . . . . . . . . . . . . . . 209 Research Question 1 211 Research Question 2 211 Research Question 3 212 Research Question 4 . . . . 213 Research Questions 5— 7 . . . . . . . . . . . . . . . 213 Research Questions 8- 9 . . . . . . . . . . . . . . . 214 Research Questions 10-14 . . . . . . . . . . . . . . 214 Research Questions 15-19 . . . . . . . . . . . . . . 216 Research Question 20 . . . . . . . . . . . . . . . . 217 V. LIMITATIONS, CONCLUSIONS, DISCUSSION AND IMPLICATIONS . 218 Limitations . . . . . . . . . . . . . . . . . . . . . . 218 Conclusions . . . . . . . . . . . . . . . . . . . . . . 220 Discussion . . . . . . . . . . . . . . . . . . . . . 222 Implications for Education . . . . . . . . . . . . . . . 226 Implications for Research . . . . . . . . . . . . . . . 227 APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 APPENDIX A. Portions of Quality of Life Questionnaire Used in This Study . . . . . . . . . . . . . . . . . . . . . . 231 B. Sampling Procedures . . . . . . . . . . . . . . . . . . 251 C. Interviewers' Instructions . . . . . . . . . . . . . 254 D. Frequency Distributions of Variables . . . . . . . . . . 259 E. Reliability Analyses . . . . . . . . . . . . . . . . 266 F. Cluster Solutions for Evaluation and Frequency Variables . . . . . . . . . . . . . 278 G. Results of Predictions Not Used for Hypotheses Testing . . . . . . . . . . . . . . . . . . . . . . . 282 H. Correlation Matrices . . . . . . . . . . . . . . . . . . 293 LITERATURE CITED . . . . . . . . . . . . . . . . . . . . . . . . 295 viii Table 10. 11. 12. 13. 14. 15. 16. LIST OF TABLES A Contrast of Social and Economic Exchange Characteristics and Behaviors . Summary of Surveys of Perceived Overall Quality of Life . . . . . . . . Summary of Surveys of Marital and Family Life Satisfaction . . . . . . . . . Age Distribution of the Sample Number of Years Married of Families in Sample as Reported by Wives . . . . . . . . . . . Number of Children Per Family for Sample Age Range of Children Living in Household . . . Education Distribution of the Sample Family Income Distribution of the Sample Family Per Capita Income Distribution of Sample . Description of Dependent Variables Descriptive Statistics for the Dependent Variables Description of Independent Evaluation Variables . Description of Independent Frequency Variables Pearson Correlation Coefficients and Covariances fbr Women's Affective Evaluations of Dimensions of Family Life with Evaluation of Family-Life- aS’a-MIOIe--Fanllif 3 o o o o o o o o o o o 0 Pearson Correlation Coefficients and Covariances for Men's Affective Evaluations of Dimensions of Family Life with Evaluation of Family-Life- as-a-Whole--Famlif 3 . . . . . . . . . . . . Page 42 76 83 93 94 95 96 97 98 99 107 III 114 118 134 I37 Table 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. Means and Standard Deviations of Women's and Men's Affective Evaluations of Family Life and Life-as-a-Whole . . . . . . . . . . . Crosstabulation of Women's and Men's Affective Evaluations of Family-Life-as-a-Whole . . . Pearson Correlation Coefficients for Women's and Men's Frequency of Shared Time and Activities with Mate, with Affective Evaluation of Family-Life-as-a-Whole Pearson Correlation Coefficients fer Women's and Men's Frequency of Resources Received from Mate and Affective Evaluation of Family-Life-as-a-Whole . . . . . Pearson Correlation Coefficients for Women's and Men's Frequency of Resources Received from Mate, Created Variables, with Affective Evaluation of Family-Life-as-a-Whole Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Family Life by Evaluation of Marriage and Resources Received, Order not Specified--Women . . Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Family Life by Evaluation of Marriage and Resources Received, Order not Specified--Men . . . . . Crosstabulation of Women's and Men's Affective Evaluation of Marriage . . . . . . . . . . . Means and Standard Deviations of Women's and Men's Affective Evaluations of Dimensions Of Fmi 1y Life 0 O O O O O O O O O O O O O 0 Means and Standard Deviations of Women's and Men's Affective Evaluations of Selected Dimensions of Family Life, Listwise Deletion Means and Standard Deviations of Women's and Men' Perceived Frequency of Shared Time and Activities with Mate, Including Created Variables . . . . . . . . . . . . . . . . . . Page 139 140 142 143 144 146 147 149 151 153 155 Table 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. Page Means and Standard Deviations of Women's and Men's Perceived Frequency of Resources Received from Mate, Including Created Variables . . . . . . . . . . . . . . . . . . . . . . . 156 Summary of Cluster Analyses fer Women . . . . . . . . . . 164 Summary of Cluster Analyses fer Men . . . . . . . . . . . 165 Pearson Correlation Coefficients and Covariances fer Affective Evaluations of Dimensions of Family Life with Evaluations of Marriage-- Women . . . . . . . . . . . . . . . . . . . . . . . . . 168 Pearson Correlation Coefficients and Covariances for Affective Evaluations of Dimensions of Family Life with Evaluation of Marriage--Men . . . . . 171 Pearson Correlation Coefficients for Women's and Men's Frequency of Resources Received and Affective Evaluation of Marriage . . . . . . . . . . . 173 Pearson Correlation Coefficients for Women's and Men's Frequency of Shared Time and Activities with Mate with Affective Evaluation of Marriage . . . . 174 Pearson Correlation Coefficients for Women's and Men's Frequency of Resources Received from Mate, Created Variables, with Affective Evaluation of Marriage . . . . . . . . . . . . . . . . 175 Summary of Best Predictions of Marriage Evaluation by Selection Criteria . . . . . . . . . . . . . . . . . 177 Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Evaluation of Mate and Variables Selected by Statistical Criteria, Order not Specified . . . . . 180 Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage ‘ by Affective Evaluation of Mate and Resources Received, Order not Specified--Women . . . . . . . . . 181 Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Affective Evaluation of Mate and Resources Received, Order not Specified--Men . . . . . . . . . . 182 xi Table page 40. Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Evaluation Variables Selected by Statistical Criteria, Order not Specified . . . . . . . . . . . . . . . . . . . . . 184 41. Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Evaluation Variables Selected by Statistical Criteria, Including Decisions Made, Order not Specified-- Women . . . . . . . . . . . . . . . . . . . . . . . . . 185 42. Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Evaluation of Resources Received from Mate, Including Decisions Made, Order not Specified-- Women . . . . . . . . . . . . . . . . . . . . . . . . . 186 43. Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Evaluation of Resources Received from Mate, Order not Specified-~Women . . . . . . . . . . . . . . 187 44. Summary of Multiple Regression Analysis fbr the Prediction of Affective Evaluation of Marriage by Evaluation of Resources Received from Mate, Order not Specified--Men . . . . . . . . . . . . . . . 188 45. Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Evalu- ation Variables Selected by Theoretical Expectation, Order not Specified . . . . . . . . . . . 190 46. Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Evaluation of Resources Received from Mate, Order not Specified-~Women . . . . . . . . . . . . . . 191 47. Summary of Multiple Regression Analysis fOr the Prediction of Affective Evaluation of Marriage by Evaluation of Resources Received from Mate, Order not Specified--Men . . . . . . . . . . . . . . . 192 48. Summary of Standardized Regression Coefficients for Statistical and Theoretical Variable Sets fOr Women and Men . . . . . . . . . . . . . . . . . . . 202 49. Multiple Regression Analysis fer the Prediction of Affective Evaluation of Marriage by Frequency of Resources Received from Mate, Created Variables . . . . . . . . . . . . . . . . . . . . . . . 204 xii Table 50. 51. 52. D-2. E-2. E—3. E-4. E-S. E-6. E-7. Page Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Frequency of Resources Received from Mate, Created Variables, Order not Specified--Women . . . . . 205 Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Frequency of Resources Received from Mate, Created Variables, Order not Specified--Men . . . . . . 206 Summary of Standardized Regression Coefficients for Created Frequency Variables for Women and Men . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Frequencies and Relative Frequencies of Women's and Men's Affective Evaluations of Dimensions of Family Life . . . . . . . . . . . . . . . . . . . . 259 Frequencies and Relative Frequencies of Women's and Men's Perceived Frequency of Shared Time and Activities with Mate . . . . . . . . . . . . . . . . 262 Frequencies and Relative Frequencies of Women's and Men's Perceived Frequency of Resources Received from Mate . . . . . . . . . . . . . . . . . . 264 Summary of Reliability Analysis of Time and Resources Frequency Scales for Women and Men . . . . . 266 Summary of Reliability Analysis of Time Scale for Women and Men . . . . . . . . . . . . . . . . . . . 267 Summary of Reliability Analysis of Love Frequency Scale for Women and Men . . . . . . . . . . . . . . . . 268 Summary of Reliability Analysis of Status Frequency Scale for Women and Men . . . . . . . . . . . 269 Summary of Reliability Analysis of Services Frequency Scale for Women and Men . . . . . . . . . . . 270 Summary of Reliability Analysis of Information Frequency Scale fer Women and Men . . . . . . . . . . . 271 Summary of Reliability Analysis for Affective Evaluation Variables--Women . . . . . . . . . . . . . . 272 Summary of Reliability Analysis for Affective Evaluation Variables--Men . . . . . . . . . . . . . . . 274 xiii TabIe E-9. E-IO. F-I. F—2. F-3. F—4. 6-2. 6-3. 6-4. 6-5. 6-6. Summary of Reliability Analysis for Frequency Variables, Resources Received from Mate-- Women . . . . . . . . . . . . Summary of Reliability Analysis for Frequency Variables, Resources Received from Mate-- Men.................. Complete-Linkage Cluster Solution, Evaluation Variables--Women . . . . . . . . . . . . . Complete-Linkage Cluster Solution, Evaluation Variables--Men . . . . . . . . . . Complete-Linkage Cluster Solution, Frequency Variables--Women . . . . . . . . . . . . . . Complete-Linkage Cluster Solution, Frequency Variables--Men . . . . . . . . . . . . . . Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Evaluation of Resources Received from Mate, Order Specified--Women . . . . . . . . . . . Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Evaluation of Resources Received from Mate, Order Specified—~Men . . . . . . . . . . . . Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Frequency of Resources Received from Mate, Order Not Specified--Women . . . . . . . . . Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Frequency of Resources Received from Mate, Order not Specified--Men . . . . . . . . . . Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Frequency of Resources Received from Mate, Order Specified-~Women . . . . . . . . . . . Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Frequency of Resources Received from Mate, Order Specified-~Men . . . . . . . . . . xiv Page 276 277 278 279 280 281 284 285 286 287 288 289 Table Page 6-7. Summary of Multiple Regression Analysis fer the Prediction of Affective Evaluation of Marriage by Evaluation and Frequency of Resources Received from Mate, Order not Specified--Women . . . . 290 6-8. Summary of Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Evaluation and Frequency of Resources Received from Mate, Order not Specified--Men . . . . . 291 6-9. Multiple Regression Analysis for the Prediction of Affective Evaluation of Marriage by Combined Frequency and Evaluation Variables, Order not Specified . . . . . . . . . . . . . . . . . . . . . . . 292 H-l. Correlation Matrix, Evaluation Variables . . . . . . . . 293 H-2. Correlation Matrix, Frequency Variables . . . . . . . . . 294 Figure 1. 2. LIST OF FIGURES The Cognitive Structure of Resource Classes Hypothesized Relationship Among Variables Model of the Relationship Between Resources Received and Affective Evaluation of Marriage and Family Life . . The Delighted-Terrible Scale . xvi Page 14 16 104 CHAPTER I INTRODUCTION Quality of life depends upon finding balance between needs- goals of humans and resources of environment. Allocating scarce resources to promote individual well-being and simultaneously main- tain environmental quality is a global problem and an important goal of the modern state (Andrews, 1974). The task of resource allocation requires establishment of priorities and an assessment of the changing character of American life in terms of physical, social-psychological and cultural well-being in addition to observation of the changing Gross National Product and Consumer Price Index. The search for effective social indicators of quality of life has been long and difficult. There are disagreements on the factors to be included and unsolved problems of measurement. Social indicators have been defined as: Quantitative data that serve as measures of socially important conditions of society. These indicators may be "objective" con- ditions of society and persons (health, education, crime, mobility) and "subjective" perceptions of life experiences (satis- factions, aspirations, alienation) (Henriot, 1972, p. 3). There is general agreement that a combination of both objective and subjective indicators will most effectively measure quality of life. Toward a social report (USDHEW, 1968) specified a need for information about participation and alienation of Americans--objective 1 and subjective indicators of the functioning and change of social institutions--and indicated that Americans expect social institutions to protect individual freedom and also to satisfy needs for congenial social relationships and a sense of belonging. It matters whether group relationships in society are harmonious and satisfying. Statement of the Problem One of the most important social institutions in society is the family. Although it has been common practice to investigate "individual” well—being, the study of "family" well-being has been given minimal attention. Social indicators (1973) ignored the family and Social indicators (1976) gave attention to size, composition, stability, living arrangements, public perceptions, and international comparisons. Producing social indicators of family well-being must involve more than statistical tables of structural variables (Weitz- man, 1978). Information is also needed about family support and com- munication systems and behavioral patterns nurtured through affection, tradition and family duty. It does matter whether the family rela- tionships are harmonious and satisfying. Evaluations of satisfaction with family relationships are indicators of the functioning of the family as a social institution. The family provides the setting in which resources are created, allocated, and exchanged to meet physical, safety, and higher level needs of individuals. In the examination and measurement of both indi- vidual and societal well-being it is important to develop indicators of family functioning and the processes which contribute to quality of family life. The first step in assessing family well-being with subjective indicators has been to ask respondents about happiness or satisfaction with family life. The assumptions are: (1) A positive evaluation of family life is one indicator of family well-being. (2) Family well- being is an important aspect of perceived overall quality of life. (3) The family is an important component of our complex social system. (4) The family plays a critical role in the development of human resources--the production of human capital for the greater society. (5) What strengthens the family, strengthens society. Several studies of perceived overall quality of life have reported a correlation between positive evaluation of family life and positive evaluation of life-as-a-whole (Andrews 6 Withey, 1976; Camp- bell, Converse & Rodgers, 1976; Bubolz, Eicher, Evers 6 Sontag, 1980; Medley, 1976; London, Crandall G Seals, 1977; Sontag, Bubolz G Slocum, 1979; Wilkening & McGranahan, 1978). Family life is known to be an important domain of life to most people (Andrews 8 Withey, 1976; Stoeckeler G Gage, 1978). There is general agreement (Weitzman, 1978; Mancini, 1978) that the dynamics of family life and the resource exchanges taking place which influence family and individual well-being have not been conceptualized or mea— sured. Purpose of the Study The ultimate purpose of this research is to further delineate the dimensions of family life that contribute to people's satisfaction with it. This delineation is essential in order to arrive at a defi- nition of quality of family life and quality of life-as-a-whole. Some of the questions surrounding the issue are: (1) What is family well- being? (2) What are objective and subjective indicators of family well-being? (3) How can Americans measure change in this important social institution? (4) Does quality of family life refer to an out- put, a condition, or a degree of excellence? Assuming that a positive evaluation of family life is one sub- jective indicator of family well-being, the purposes for the next level of specificity are to determine the key variables which account for a positive evaluation of family life, and to describe what differenti- ates individuals who express positive, negative and neutral evaluations of family life. Campbell et a1. (1976) concluded that a major contribution to satisfaction with family life is the individual's relationship with his/her spouse. What is the relative contribution to evaluation of family life of feelings about spouse and feelings about the marital relationship? What are the dimensions of the marital relationship which contribute to people's satisfaction with it? The theoretical literature (Foa 6 Foa, 1973, 1974) suggests that persons who receive love, respect, personal services, and infor- mation in a small group setting over prolonged periods of time from particularly valued people have a high probability of being highly satisfied. The present research seeks to examine the relationships between the receipt of particularistic interpersonal resources and affective evaluation of family life and, particularly, affective evalu- ation of marriage. Scope The present study will be limited to the examination of social- psychological well-being as determined by subjective indicators. Family life is the domain of life-as-a-whole which is of central interest. Marriage is the domain of family life which will receive primary attention. The central areas of interest are the marital support and com- munication systems which influence marital relationships. The focus is on the interpersonal resources which are exchanged in face-to-face contact: love, status (respect), personal services and information. How effectively can evaluation of marriage be predicted by feelings respondents have about the receipt of these resources in the family setting from mate? The respondents are limited to married persons living together in the same household and having at least one school-age child in the geographical area of Oakland County, Michigan. The majority of respon- dents in the study sample are Caucasian, have a high school education, and are in the middle years of life. The analyses are limited to separate examinations of men and women. The survey method of information collection has limitations for understanding the dynamics of interaction in family life. Addi- tional limitations must be added for the problems faced by respondents in recalling frequency of activities, maintaining privacy of answers, and attempting to select a quantitative number for a qualitative dimen- sion of their lives. Theoretical Framework The theoretical framework fOr this study is based primarily in social, interpersonal resource exchange theory (Foa G Foa, 1973, 1974) with additional assumptions from family systems theory (Kantor G Lehr, 1975). The theoretical perspective of exchange provides the means for studying interaction of the individual and the near— environment, an ecological view of human social-psychological well- being. The Foa and Foa theoretical model was selected because it con- siders economic and psychological resources to be interdependent and equally necessary for evaluating quality of life. It provides a classi- fication of the events and conditions which make life pleasant and worthy, which offers parsimony, simplicity and is specific enough to pinpoint essential differences among people (Foa G Foa, 1973). The Foa model is one of the first attempts to integrate eco- nomic and sociological exchange theories by including both types of resources and clarifying a differentiation in the rules of exchange according to the structural position of resources on the dimensions of particularistic-universalistic, and concrete-symbolic (Figure 1). The theory discusses the consequent satisfactions of intercategory resource exchanges and the alternative resource substitutions of inter- category exchanges with resulting changes in levels of satisfactions obtained from the various substitutions. The Foa theory links the concepts of: resources, needs and satisfactions. It considers time, space, energy and institutional environment as important factors influencing resource exchange, but does not consider them to be resource classes as are love, status, More LOVE 0 STATUS SERVICES 0 O 2: 59, a .J :2 E2. E? << C... INFORMATION GOODS O o MONEY 0 Less Less CONCRETENESS More Fig. 1.--The Cognitive Structure of Resource Classes. Copyright 1971 by the American Association for the Advancement of Science. services, information, goods and money which are exchanged between persons in communication. These are the classes of resources which appear necessary and sufficient to account for the basic needs of human beings (Foa 6 Foa, 1973). The interpersonal resource exchange model offers a theoretical explanation for family life as a significant contributor to satisfaction with life-as—a-whole. It considers family as both an environment for human resource development and as an organism involved in the trans- formation of exchangeable commodities and communications. The theory makes it possible to examine intra-or inter-family resource exchanges in relation to individual and group satisfactions. The theory is help- ful in providing explanations for the resources exchanged in particu- laristic relationships such as marriage and the resulting levels of satisfactions expected from the various categories of resources exchanged. The addition of the assumptions of family systems theory of wholeness, interdependence and reciprocity make it reasonable to con- sider the study of resource exchange with the questionnaire method of information collection. The researcher is not able to observe the exchange, but the transfer of a resource from one person to another can be viewed as circular and reciprocally influencing. Supportive- ness begets supportiveness. It is both cause and effect--a simultaneous stimulus, response and reinforcement. Summary of Central Ideas, Foa Theory Man is a social creature who enjoys companionship and needs the support of a group for survival. Needs for love, status, services, goods, information and money cannot be satisfied in isolation. Since humans depend upon others for those resources necessary to well being, they therefore seek situations to exchange them through interpersonal behavior. The two basic mechanisms which motive social behavior are the strive to maintain Optimal levels of resources by giving what one has in abundance and receiving what is scarce; and the strive to maintain cognitive structure which insures the ability to exchange. The probability of an interpersonal exchange taking place depends upon: (1) the motivational state of potential exchangers, (2) appropriateness of the environment, and (3) properties of resources to be exchanged. Properties of resources. Resources are structurally ordered and interrelated. The rules of exchange vary gradually with position in the structure. Giving to self and giving to other are related posi- tively for love and negatively for money. Component elements of the structure must be considered in their structural relationship rather than in isolation (Figure l). The resources are structured on the dimensions of particularism-universality and concrete-symbolic forms of expression. Particularistic refers to the attribute which indicates the extent to which the value of a given resource is influenced by the particular persons involved in exchanging it and by their relationship. Love is the most particularistic resource and money is the most universal- istic since its value is least influenced by the person from whom it is received. 10 Position of a resource in the structure affects satisfaction with exchange. Like resources are preferred in exchange, with love as the most preferred resource of proaction and reaction. Resources proximate in order are more similar and interchangeable with one another. The larger the structural distance between resources being exchanged, the lower the satisfaction. Economic resources are more readily available in modern society but are distant from the needed, preferred particularistic resources so that their provision, even in increasing amounts, does not provide high levels Of satisfaction. The opportunity to exchange love with a highly valued partic- ular person in repeated encounters over a period of time offers the opportunity for highest possible levels of satisfaction. The actor, in giving love, simultaneously gives love to self and the object who receives love simultaneously gives to actor. A person who is unable to exchange love or who infrequently exchanges love can be reasonably happy if he acquires status or is pampered with personal services; however, satisfaction will be dimin- ished if goods or money are received instead of the needed love. Inappropriate resources of reaction leave an unbalanced state: strong residual aggression in exchanges of taking not offset by higher inten— sity of retaliation; and low degrees of satisfaction in exchanges of giving. Status is the second preferred resource of exchange and is simultaneously given to self and other. Love and status are closely related. It is possible to give respect without love, but difficult to give love without simultaneously giving respect and esteem. Self 11 esteem is giving status to self and is highly related to giving status and love to others. Appropriateness of environment. The probability of a resource exchange taking place is also dependent upon the appropriateness of the environment. The environmental properties of resources include: time for processing input, delay of reward, and optimum group size. The exchange of love requires time, repeated encounters, few persons and privacy of space. Shared time becomes the environmental condition necessary for transfer of the most particularistic resources. The institutional environment specifies what type of resources should be exchanged and for each institution there are certain resources which are more typical and more frequently exchanged: "In the family love and status are the crucial resources" (Foa G Foa, 1974, p. 151). Family is the social institutional environment ideally suited to the exchange of particularistic resources and the institution where . the widest range of exchanges take place. However, even in the family not every exchange is permissible or customary for one does not typically or frequently give money for services received from a family member. Motivational State Of exchangers. Completion of exchange also depends upon motivational states of actors. Motivational state involves the concepts of optimal range, actual and potential needs, and power. The Foa theory proposes that for each resource class there is an optimal range; when the amount of a given resource held by the individual is within the range, the person feels comfortable and is not motivated to initiate change. When the amount possessed falls below the lower bound of the range, the individual perceives a need 12 for the resource and will be motivated to increase the amount of the resource in his possession. When the amount of a resource exceeds the upper limit of the Optimal range the individual will be motivated to "get rid” of the resource through exchange behavior. Accumulation of an amount of resource in excess of need is the process of accumula- tion of power. The width of the optimal range varies with position of the resource in the structure, is most narrow for love and widest for money. The narrow optimal range of love causes frequent upsets of balance and requires constant restoration by increasing and decreasing amounts held by the system. Size of optimal range also varies among individuals. Individ- uals who have experienced deprivation of love in early childhood will be less capable of exchanging love which will lead to reduction in actual need for love despite very high potential need for love. The person may have a strong potential need for love but very little ability to actually absorb it. Application of Foa Theory to Present Study The basic premise of the present study which is based on the previously discussed theory is as follows: Under conditions where physical and safety needs of individuals are met and per-capita income is held constant, if there is satisfactory frequency of shared time between husband and wife and positive evaluation Of shared time, then there will be high frequency of particularistic resource exchanges, highly positive evaluations of resources received and highly positive evaluations of marriage and family life. 13 The hypothesized relationships among variables are indicated in Figure 2. If there is satisfactory frequency of shared time with mate, it is likely that there is a positive evaluation of shared time and a higher probability of frequent exchanges of love, respect, per- sonal services and information. The higher frequency of interpersonal resource transfers is likely to create a supportive emotional climate, generate positive feelings and result in positive evaluations of per- sonal needs for love, recognition and respect, comfort and assistance, companionship and shared meaning being met. The result is a marital and family systems balance (needs met) where the conditions offered (resources received) closely match the conditions required (resources wanted, needed, expected). The indicators of marital and family sys- tems balance are the respondent's positive evaluations of marriage and family life. Since the participant in an exchange gives what he has in abund- ance and receives what is scarce, the marginal utility of the receiver is higher than the marginal utility of the giver (Foa 8 Foa, 1974). The present study thus has respondent as receiver of resources while spouse is actor. It is an examination of the impact of the family, and particularly the marriage, upon the respondent. Family is con- sidered environment for the individual who is the unit of study. The hypothesized relationships among variables (Figure 2) include many reciprocal relationships. It is difficult to test theor- etical models with reciprocal relationships with statistical methods existing at the present time. An attempt to use LISREL IV for analysis and model testing had to be abandoned due to problems involved in 14 .mofinmfipw> wcoa< mafia >4~Zm m>HFWmLu< ucm m>aeumaa< om>um0mx mmumaommm UHPmHmm m>HHUmmm< mk<2 20mm nm>—mumm mmumaommm UfibmwmUzmzommm n ’F magm:0mumflom pONflmosuom»:--.N .mfim I w mFm m>~bummm< V ; wee: mes: mess ammm O>~uoowm< can po>woooz moou30mom coozuom mwzmcomummom ecu mo demo u-.m .wfim meamomm mmumzommm Aa4~z<; LO ZOHHM m>~bummu< x /, mum m>HPUmmm< m2_s ammm az< om>amumm mmumaommm Asm m>apumaa< 17 but also rewards and costs of children and the qualitative evaluation of the social, physical, economic, personal and cultural setting of the relationship and how that total "resource mix" has met individual needs over a period of time. The evaluation process involves a balanc- ing of rewards and costs of the relationship and its context as well as projections of future rewards and costs. It is hoped that in the future statistical methods will be able to examine more complex theor- etical models involving reciprocal relationships among variables. Theoretical Definitions This section includes theoretical definitions of concepts relevant to the study. Operational definitions are reported in Chapter III, Tables 8, 10, and 11. Quality of Life A person's sense of well-being, satisfaction or dissatisfaction with life, or happiness or unhappiness (Dalkey G Rourke, 1973). An individual's overall perceived satisfaction of needs over a period of time (Mitchell, Logothetti G Kantor, 1973). The subjective name for the well-being of people and the environment in which they live. For any individual quality of life expresses that set of wants which after being supplied, when taken together, make the individual happy or satisfied (Liu, 1975). Well-being Well-being refers to the level of life quality and is often used interchangeably with the term quality of life. 18 Quality of Family Life A person's overall perceived satisfaction/happiness or dissatisfaction/unhappiness with family life over a period of time. A person's internal response to perceived rewards received from the family members and the family environment over a period of time. A person's evaluation of the conditions offered in family life compared to the conditions desired. Quality of Marriage A person's qualitative evaluation of the marital relationship and the social, economic, personal and cultural setting Of the relation- ship over a period of time. A person's internal response to rewards received from mate, own role behavior and others in the near environment over a period of time. Objective Indicators of Life Quality or Well-Being Measures of external physical and social conditions of the individual existence not requiring personal evaluation of reporting individual (Sontag, 1978). Subjective, Perceptual Indicators Of Life Quality Personal, subjective evaluations of reporting individuals con- cerning life quality and well-being. 19 Affective Evaluation The assessment of life concerns involving both cognitive evalu- ation and some degree of positive/negative feeling such as affect. Affective evaluation is an internal response to the perceived environ- ment. Affective evaluation will be indicated in this study by a person's response selected from seven scale categories on the Delighted-Terrible Scale. Resource Any commodity, material or symbolic, which is transmitted through interpersonal behavor (Foa G Foa, 1974). The Foa theory specifies six classes of resources: love, status, services, infor- mation, goods and money. Any property of an individual which is made available to persons in the environment as a means for positive or negative need satisfaction (Levinger, 1959). Any reward that an actor can use in an exchange relation with other actors (Emerson, 1969). Love An expression of affectionate regard, warmth or comfort (Foa 6 Foa, 1973). The state of feeling which manifests itself in solicitude for the welfare of a person, delight in his/her presence, desire for his/her approval, and warm affection and attachment (Oxford English Dictionary, 1971). 20 Love is conveyed in verbal messages of affect, liking, enjoy- ment and physical messages of touch, expressions of the face, eye contact, body posture, and physical proximity. Love is more easily expressed in paralinguistic communication (Foa G Foa, 1974). The actor giving love to other simultaneously gives love to self. The value of the love resource is highly dependent upon the person from whom it is received. (Love is the most particularistic resource.) Particularistic The attribute which indicates the extent to which the value of a given resource is influenced by the particular persons involved in exchanging it and by their relationship (Foa G Foa, 1974). Concrete The attribute which suggests the form or type of expression of the various resources. Behaviors like giving an object or per- forming a service to the body or belongings of another are concrete. Language forms of expression are symbolic (Foa 8 Foa, 1974). Status An evaluative judgment that conveys high or low prestige, regard or esteem (Foa G Foa, 1973). In contrast to love, status is conveyed in verbal (symbolic) behaviors in messages of esteem, respect, and confidence in competence. Status conferred by self to self is called self esteem. The actor giving status to other simultaneously gives status to self. 21 The value of respect received is dependent upon the person from whom it is received, but status is less particularistic than love. The Foa definition of status is different from use of the concept by sociologists who describe status as a position or rank in a hierarchy of prestige, or a position in a social group relative to other positions in the group. The Foa definition more closely resembles the sociological definitions of prestige and esteem. Prestige is the evaluative judgment in the norms of society about the desirability of a particular status. The evaluation of an individual's role behavior in the status position occupied is called esteem (Young 8 Mack, 1962). Services Activities performed on the body, belongings, or environment of a person usually constituting labor of one person for another to increase physical comfort of the other or to save him/her energy (Foa G Foa, 1973). The actor giving services to other does not simultaneously give services to self. In contrast to status, services are conveyed in concrete ways--work, helpfulness, assistance--rather than in sym- bolic form. Value placed on services received is dependent upon the person providing the services, however, services are not as particularistic as love. 22 Information Information offered as advice, Opinions, instructions or enlightenment but exclusive of those behaviors that could be classi- fied as love or status (Foa G Foa, 1973). The actor giving information does not increase or decrease the amount of information possessed by giving it to other. Infor— mation is conveyed symbolically and is less particularistic than love 01' status . Tangible products, Objects or materials (Foa G Foa, 1973). The value of goods received is not as dependent upon the person who gives the object as is true with resources of love and status except in situations of gift giving. Money Any coin, currency, or token which has some standard unit of exchange value (Foa 6 Foa, 1973). Money is the least particularistic, the most universal resource of exchange, and is transferred in both concrete and symbolic forms. Shared Time Two persons engaged in one activity at the same place and time. The environmental condition necessary for the exchange of love, respect, services, and information in interpersonal relationships. 23 Optimal Range The range of each resource class in which the individual feels comfortable with the amount of the resource possessed and there is no motivation for change. When the amount of the resource possessed falls below the bound of the range, the situation will be subjectively perceived as a need for this particular resource and the individual will be motivated to increase the amount in his possession (Foa G Foa, 1974). The more particularistic the resource, the narrower the Optimal range and thus the more frequently the balance of the system is upset. Narrow optimal range requires constant restoration by increasing or decreasing the amount held by the system. The optimal range for love is more narrow for persons who have received less love in the early years of life (Foa G Foa, 1974). Need A state of deficiency in a given resource (Foa 8 Foa, 1974). Power The amount of a given resource available to an individual for eventual giving (Foa G Foa, 1974). Exchange The mutual giving and receiving of both material and non- material things. Thus it may refer to the transfers of services, goods, money, rights or benefits that are reciprocated by a transfer of something similar or different in return (Gould G Kolb, 1965). Transfer of values between two economic units. 24 Economic Exchange Behavior The two-way transfer of measurable resources between two per- sons who simultaneously agree upon the exact obligations of both parties and who complete the transaction in a specified period of time. The benefits of the economic exchange are independent of the supplier and independent of the relationship between the two persons. Social Exchange Behavior The transfer of a non measureable resource from one person to another with nonspecified obligations but expectation of reciprocity at some future time (not too soon). Obligations cannot be bargained or negotiated and must be left to the discretion of the receiver. Social benefits are not detachable from their source. The lack of neutral measurement of social benefits (resources) mean persons are always uncertain as to whether their "debts" are paid and tend to over-reciprocate to make sure. This leads to an expanding cycle of exchange (Diesing, 1962). Giving_Exchangg§ The actor increases the amount of the resource available to object (Foa G Foa, 1974). Taking Exchanggs The actor deprives the other of a given resource (Foa 8 F03, 1974). Intracategory Exchange An exchange where the same resources are being transferred (affection for affection). 25 Intercategory Exchangg An exchange where one type of resource is exchanged for another type (money for goods). Intrafamily Resource Transfer The transfer of resources within one nuclear family. Reinforcement The features of the environment which are capable of bestowing gratification upon an actor (Emerson, 1972). Reward The degree of value attached to a given type of reinforcement (Emerson, 1972). Pleasures, satisfactions and gratifications a person enjoys including statuses, relationships, interaction and feelings (Nye, 1979). Value The strength of reinforcers to evoke and reinforce behavioral initiations by an actor, relative to other reinforcers and holding deprivation constant and greater than zero (Emerson, 1972). Cost The magnitude and number of rewards of one type forgone to receive rewards of another type (Emerson, 1972). Any status, relationship, interaction, milieu or feeling dis- liked by an individual (Nye, 1979). 26 Assumptions Assumptions About Methods and Measurements l. The Delighted-Terrible Scale provides numerical responses which can be treated as interval-level data (Figure 4, p. 104). 2. Survey design with questionnaire response is an appropriate method for gaining some insight into the results of interpersonal resource transfers between husbands and wives. 3. Words such as "love," "respect" and "comfort” have similar meanings to all respondents. 4. Quality of family life and quality of marriage can be assessed by asking people directly about family life and marriage relationships. 5. Evaluation of marriage and family life can be predicted on the basis of information about giving exchanges only--without infor- mation about the costs or the taking exchanges. 6. Husbands and wives responded independently to the Quality of Life Questionnaire items. 7. Affective evaluation of marriage and affective evaluation of family life are indicators of well-being and of the quality of relationships. 8. The sampling methods have been adequate to provide randomly selected respondents. 9. Combination of the three sub-samples is appropriate for the examination of internal relationships among variables. 10. Multiple regression analysis is an appropriate technique for the research objectives and the nature of the data. 27 11. The respondent's evaluations of spouse are important in understanding the quality of marital and family relationships. 12. Respondents can accurately evaluate and report their feelings about marriage and family life. 13. The reporting and recollection of behavior is a reason- able representation of what has actually occurred. Assumptions About the Nature of Behavior and Its Context 1. Relationships in the family are circular and reciprocally influencing. Love produces love and this supportiveness is simul- taneous stimulus, response and reinforcement. 2. There are two basic mechanisms which motivate social behavior: (1) the strive to maintain optimal level of resources by giving what we have and receiving what we need; (2) the strive to maintain cognitive structure which insures the ability to exchange (Foa G Foa, 1974). 3. The goal of social interaction is mutual satisfaction of needs among exchange partners. Satisfaction of needs is accomplished by exchanges of love, status, services, information, goods and money. Time, space and institutional setting are not resources but factors influencing exchange (Foa G Foa, 1974). 4. All six classes of resources contribute to quality of life so when any one falls below a minimum level, quality of life is impaired (Foa G Foa, 1974). 5. The exchange of particularistic resources in an environ- ment facilitating prolonged, repeated encounters in a small group pro- vides the highest levels of satisfaction for humans. 28 6. Marriage is a highly particularistic relationship Offer- ing the opportunity for exchanges of love, status, and personal services in a manner which can produce high levels of satisfaction. 7. Husbands and wives experience the family environment in different ways and respond differently in evaluating marriage and family life. 8. Satisfaction is affected by level of involvement. Lower satisfaction is experienced by the partner with highest involvement (Foa & Foa, 1974). The person with highest involvement in family life is often the wife-mother and it is expected that this study, in con— cert with other quality of life studies, will find lower satisfaction with marriage and family life among women. Research Objectives The objectives of the research are to: 1. Explore the validity of Foa and Foa's resource exchange theory and the measured indicators of the theoretical model in the present study. 2. Select the best set of indicators to predict affective evalu- ation of marriage for women and men. 3. Investigate the credibility of the Foa's theory in predicting affective evaluation of marriage for women and men. 4. Investigate the contribution of each interpersonal resource (love, status, services, information) and shared time to the prediction of affective evaluation of marriage for women and men . 29 Describe the differences in evaluations of marriage, family life and life-as-a-whole for women and men. Research Questions and Hypotheses The research questions and hypotheses in the null form are: Questions and Hypotheses for Objective 1 1. H1: H2: H3: Do the questionnaire items which represent a resource class have greater proximity to each other than to variables repre- senting a different resource class for evaluation and fre- quency variables for women and men? Is there a pattern in the fusion order of clusters to support the theoretical model of structured relations between resource classes for both evaluation and frequency variables in both men's and women's analyses? Are there differences in cluster solutions for men and women? Will different methods of hierarchical clustering provide similar cluster solutions for the same data to support validity of the four particularistic resource classes? All proximity matrices are equally likely for frequency and evaluation variables in both women's and men's analyses. All orders of fusion are equally likely for frequency and evaluation variables in both women's and men's analyses. There are no differences in cluster solutions for men and for women for evaluation and frequency variables. Questions for Objective 2 5. Will evaluation, frequency, or a combination of evaluation and frequency variables give the best prediction of marriage evaluation? How will the variable selection method affect the prediction of marriage evaluation? Will the order specified by the Foa theoretical structure provide a good prediction of marriage evaluation? 30 Questions and Hypotheses for Objective 3 8. 9. H8: H9: Will affective evaluation of particularistic resources received predict evaluation of marriage for women and men? Will frequency of particularistic resources received from mate predict evaluation of marriage for women and men? Affective evaluations of particularistic resources received do not significantly contribute to the prediction of marriage evaluation for women or men. Frequencies of particularistic resources received from mate do not significantly contribute to the prediction of marriage evaluation for women or men. Questions and Hypotheses for Objective 4 10-14. 15-19. H10: H11: H12: What is the contribution to evaluation of marriage of the following: 10. Love evaluation? 11. Status evaluation? 12. Services evaluation? 13. Information evaluation? 14. Shared time evaluation? What is the contribution to evaluation of marriage of the following: 15. Love frequency? 16. Status frequency? 17. Services frequency? 18. Information frequency? 19. Shared time frequency? Evaluation of love and affection does not significantly con- tribute to the prediction of marriage evaluation for men or for women for the statistical or theoretical variable sets. Evaluation of respect received does not significantly con— tribute to the prediction of marriage evaluation for women or for men for the statistical or theoretical variable sets. Evaluation of services received does not significantly con- tribute to the prediction of marriage evaluation for men or for women for the statistical or theoretical variable sets. HIZA: Evaluation of sexual relationship does not significantly contribute to the prediction of marriage evaluation for men or for women in the statistical variable set. H13: H14: H15: H16: H17: H18: H19: 31 H128: Evaluation of how comfortable it feels to be at home does not significantly contribute to the prediction of marriage evaluation for men or for women in the statistical variable set. H12C: Evaluation of mutual helpfulness of family members does not significantly contribute to the prediction of marri- age evaluation for men or for women in the theoretical variable set. Evaluation of information received does not significantly con- tribute to the prediction Of marriage evaluation for men or for women for the statistical or theoretical variable sets. H : Evaluation of open, honest expression of feelings does 13A . . . . . . . not Significantly contribute to the prediction of marri- age evaluation for men or women on statistical or theor- etical variable sets. H : Evaluation of the way decisions are made does not sig- 138 O O I I O I nificantly contribute to the prediction of marriage evaluation for women in the statistical variable set. Evaluation of shared time does not significantly contribute to the prediction of marriage evaluation for women or men in the statistical variable set. Frequency of love received from mate does not significantly contribute to the prediction of marriage evaluation for women or men. Frequency of status received from mate does not significantly contribute to the prediction of marriage evaluation for women or men. Frequency of services received from mate does not significantly contribute to the prediction of marriage evaluation for women or men. Frequency of information received from mate does not signifi- cantly contribute to the prediction of marriage evaluation for women or men. Frequency of shared time with mate does not significantly con- tribute to the prediction of marriage evaluation for women or men. 32 Questions and Hypotheses for Objective 5 20. How do evaluations of marriage and family life differ for men and women? H20: There are no differences between mean scores of men and mean scores of women on evaluations of marriage and family life. CHAPTER 11 REVIEW OF LITERATURE The review of literature is organized in five sections. The first section discusses exchange theory, and the second the applica- tion of exchange theory to family research. The third section reviews selected studies relating to marital quality, the fourth section includes selected studies of family well-being, and the fifth section reviews quality of life studies that have included evaluation of family life or marriage. Exchapge Theory Exchange theory has roots in utilitarian economics, behavioral psychology and functional anthropology. The major foci and assump- tions differ among the various theories. As yet there has not been a successful integration of the theory in a way that resolves major difficulties in conceptualization and methodology. Historical Development There are two distinct traditions of social exchange theory in sociology; these vary in philosophical orientation as well as theoretical, methodological and geographical origins. Ekeh (1974) indicates that American sociologists have been unaware and virtually unaffected by the French collectivistic orientation, specifically the 33 34 work of Levi-Strauss (1969) and thus have been handicapped in the pro- cess of resolving the conceptual issues of exchange theory. The French collectivistic orientation to sociology has its philosophical roots in German Idealism and Catholicism. This orien- tation was opposed to centrality and autonomy of individual self- interests as a motivating force in social actions. Collectivistic sociologists advocated grounded theory construction as illustrated in the works of Durkheim (1951), Weber (1947), Mauss (1925), and Levi- Strauss (1969). The British individualistic orientation to sociology was influ- enced by utilitarianism and Calvinism and advocated a more atomistic world view by placing the individual as the motivating force of social action. Individualistic sociologists advocated logico-deductive rational theory construction as indicated in the work of Blau (1964). The interplay of ideas from theorists of opposing traditions encouraged the growth of exchange theory (Ekeh, 1974). Theorists from the French tradition developed ideas to refute those of the British individualistically oriented theorists and British theorists proceeded in the same manner. A contrast of the two philosophical views was reported by Stark (1958) and restated by Ekeh (1974). Catholicism Calvinism 1. Tendency towards an organic 1. Tendency towards an atomistic world-view world—view 2. Society conceived as prior to 2. Society conceived as posterior the individual to the individual 3. The community as carrier of 3. The individual as carrier all truth of all truth 4. Symbolism, artistic creativeness 4. Realism, sobriety S Emotionalism, mysticism 5. Rationalism 6. Cloistered contemplation as 6. Innerworldly observation as the ideal way to truth the ideal way to truth (Ekeh, 1974, p. 17) 35 The resulting collectivistic social exchange theory opposed the centrality of the individual as a motivating force in social action. Theory construction was based on the autonomy of society and the inability to reduce social processes to psychological ones. The time frame (compared to the individualistic orientation) was a longer one and focused on the processes of an individual's life-span. It led to grounded theory construction which can be seen in Durkheim's (1951) work on suicide from which he constructed a theory of integration. The researcher sets out to explain some social phenomenon by generat- ing theory from data. Individualistic exchange theory had its origin in utilitarian social science which had two different strains--economics and psychology--which were developed before sociology. Ekeh (1974) indi- cates that utilitarianism was a group of theories that placed individ- ual desires and wants at the center of concern and described an inherent tension between two incompatible trends: between exclusive concern with intrapersonal matters and matters of interpersonal rela- tions; between here-and-now pleasure-seeking hedonism and avoidance of pain with the inclusion of control in the maximization of benefits which sometimes involves delay of gratification; and between the con- sumption and production orientation. Foci and Assumptions _ The assumptions of individualistically oriented exchange theory will be discussed first. The assumptions of both utilitarian sociolo- gists and behavioral psychologists will be included as they developed from the common origin of utilitarian social science and influenced 36 the work of Homans (1961). Assumptions of collectivistic exchange theory as represented by Levi-Strauss (Turner, 1978) will then be dis- cussed followed by recent assumptions added by Nye (1979). Utilitarian economics assumed man was rational, that he would seek to maximize material benefits or utilities in a free and com- petitive market and that as "rational units” people would have access to complete information. Thus people could consider all alternatives by calculating costs and benefits of alternatives to make a decision which would yield maximum profit for minimum investment. Utilitarian sociologists. Utilitarian sociologists modified assumptions of utilitarian economics but continued to regard economic motives as the explanation for social action with the emphasis on maximum return for minimum investment (Homans, 1961). Exchange items were valued for their worth or for what they gained for the person giving them up and were amenable to the laws of supply and demand. Reformulation of utilitarian assumptions by utilitarian sociolo- gists involved the recognition that: (I) rarely do people attempt to maximize profits, (2) humans are not always rational, (3) their transactions with each other in an economic marketplace or elsewhere are not free from external regulation and constraint and (4) individ- uals do not have perfect information on all available alternatives (Turner, 1978, p. 202). The alternative utilitarian assumptions which developed were: (1) While humans do not seek to maximize profits, they always attempt to make some profit in their social transactions with others. (2) While humans are not perfectly rational, they engage in calculations 37 of costs and benefits in transactions with others. (3) While actors do not always have perfect information on all alternatives, they are usually aware of some which form the basis for assessments of costs and benefits. (4) While there are always constraints on human activ— ity, people compete with each other in seeking to make a profit in their transactions. (5) While economic transactions in a clearly defined marketplace occur in all societies, they are only a special case of more general exchange relations occurring among individuals in all social contexts. (6) While material goods typify exchanges in an economic marketplace, individuals also exchange nonmaterial com- modities such as sentiments and services (Turner, 1978, p. 203). Psychological behaviorism. The variant of utilitarianism in psychology led to psychological behaviorism, an exchange theory with emphasis on examination of overt behavior. In the assumption that humans are reward-seeking organisms, the concept of ”reward" was used as a restatement of the utilitarian concept of "utility." The concept of "punishment" was a restatement of "cost." Behavioral psychologists substituted "stimulus” for "demand" and "response" for "supply"; how- ever, there were no equivalent concepts for "investment" or ”profit" (Ekeh, 1974, p. 115). The assumptions of psychological behaviorism were: (1) In any given situation, organisms will emit those behaviors that will yield the most reward and the least punishment. (2) Organisms will repeat those behaviors which have proved rewarding in the past. (3) organisms will repeat behaviors in situations that are similar to those in the past in which behaviors were rewarding. (4) Present 38 stimuli that on past occasions have been associated with rewards will evoke behaviors similar to those emitted in the past. (5) Repetition of behaviors will occur only as long as they continue to yield rewards. (6) An organism will display emotion if a behavior that has previously been rewarded in the same or similar situation suddenly goes unrewarded. (7) The more an organism receives rewards from a particular behavior, the less rewarding the behavior becomes-~due to satisfaction--and the more likely the organism is to emit alternative behaviors in search of other rewards (Turner, 1978, p. 213). Collectivistic exchange theory. The collectivistic orientation to social exchange theory, as represented by Levi-Strauss (1969) rejected the values of economic man, emphasized exchange items as having symbolic rather than economic value, saw the benefits of social interaction to be the building of an interlocking network of social relationships, focused less on the internal psychological processes and more on the constraints of social structure and culture in explain- ing exchange behaviors. Attention was given to indirect exchanges, altruism, and reciprocity. There was less emphasis on man as a com- modity or a rational being in complete control of environment and little attention was given to the motivation of maximization of profit. Assumptions of the collectivistic social exchange theory of Levi-Strauss were: (1) Different social structures rather than indi- vidual motives are the important variables in analyzing exchange rela- tions. (2) Exchange relations in social systems are not restricted to direct interaction among individuals but include larger networks of indirect exchanges. (3) Exchanges do involve costs for people, 39 but these costs come from the society-—values, customs, rules, laws, which require certain behaviors. (4) The distribution of scarce and valued resources in society, whether material objects or symbolic, is regulated by norms and values. As long as resources are in abundant supply, or not highly valued in society, their distribution goes unregulated. (5) All exchange relations are regulated by the norm of reciprocity requiring those receiving valued resources to bestow on their benefactors other valued resources. There are different patterns of reciprocity--mutual and univocal-—which lead to different kinds of exchange relations (Ekeh, 1974). Generalized exchange. The addition of univocal reciprocity and generalized social exchange requires additional assumptions: (1) Groups, organizations, associations and even nations act to mini- mize costs and maximize rewards. (2) Humans are capable of antici- pating greater rewards and fewer costs from effective responsible governmental, educational, health and economic institutions. There- fore, they can invest time and other resources (costs) in attempting to improve these institutions and anticipate a profit from such invest- ments. (3) Humans are capable of conceptualizing a generalized reciprocity between themselves and society and its social institutions. Without investments in social organizations, social life with its rewards would cease. (4) Humans realize that alternatives they choose affect the rewards and costs of other members of groups to which they belong. Therefore, they can decline choices that would appear pro- fitable to them in the immediate sense because they can anticipate that, if the course of action they pursued increases the costs/ 40 reduces the rewards to other group members, they will reduce the rewards/increase the costs to the individual taking action (Nye, 1979). Methodolpgical Issues The methodological issues of exchange theories are: (l) the logico-deductive approach to theory construction of the individual- istically oriented exchange theorists differs from the grounded method of theory construction preferred by the collectivistically oriented exchange theorists. (2) There is the issue of tautology which involves the question: How are the "values" of actors to be defined and measured independently of the behaviors they influence? There is a tendency for the concepts of exchange propositions to be defined in terms of each other. As long as the major independent variable of exchange theory--value-—is difficult to separate from the principal dependent variable--behavior--it will be difficult to eliminate prob- lems of tautology (Turner, 1974, p. 284). (3) Exchange theories have been criticized for lack of deductive rigor and (4) exchange theories have been accused of failing to facilitate prediction but only to assist in the process of ad hoc reasoning. Issues of Conceptualization Some of the difficult issues of conceptualization for exchange theorists regardless of philosophical origin are the following: (1) integration of economic and social exchange concepts, assumptions, and behaviors; (2) inclusion of an explanation of reciprocity and altruism; (3) the inclusion of both two-party and multi-party exchange models within the same theoretical structure; (4) resolution of the debate over the centrality of the individual as a motivating force 41 in social action; (5) explanation and integration of the concepts of exploitation, power, distributive justice or equity in an integrated manner with the stated propositions of the theory (Ekeh, 1974). The first three issues are particularly important for the present study of family well-being and marital quality and will be discussed further. The issue of equity/distributive justice and power has been studied in connection with marital decision making but is less relevant to the present study. Economic and social exchange characteristics. The most per- sistently difficult conceptual issue for exchange theorists is the integration of economic and social exchange theories. A contrast of economic and social exchange behaviors and characteristics (Table l) combines the work of several authors (Bivens, 1976; Boulding, 1973, 1977; Diesing, 1962; Foa & Foa, 1974; Kuhn, 1975; Paolucci, 1977; International Encyclopedia of the Social Sciences, 1968). Social interaction within marriage or family life involves both economic and social exchange and the successful integration of the two theories is essential for the application of exchange theory to the study of family life. The Foa and Foa (1974) resource exchange theory is one of the first attempts to integrate economic and sociological exchange theories by including both types of resources and by differentiating rules of exchange according to the structural position of the resource on the dimensions of particularistic-universalistic and concrete-symbolic characteristics. Giving to self and giving to other are related posi- tively for the noneconomic (particularistic) resources of goods and money. Resources also differ in terms of their concrete and symbolic 42 .xufioocmwoon so :oflumo -wwfiumpw Oumwpoeew Owoe u no“: vowcmzoxo one mufiwocon osu .opme ma acoEoopwm one Ooco .uomcucoo x2 poflmwoomm mega Opwcwmop m as so mmooonm pagan m mm mhzooo owcmsoxo Owsocoom AquOOOQMOOA fimsuse .owcmzoxo uoouwp :o mammzan ouozv .fimoosmfloou >-moflumEOOD< .xuwooumfloou awoflfimxo .uomcucoo one mcfimoox so xufloonmfloou one com: ucomcflucoo mg owcmzoxm .wcficwwwcmn mo mmoo -ocm ozu smsonsu com: poonmm Ammsoocmufisewm one moduama :uon mo m:o«uwww~no womxm .codumOHmwumcw mo xmfiop new umanu mo acoEQO~o>op mommpsooco mmoooua zofim one .oumfiumonmmmcfi monopwmcoo ma mufimocon mo ensues Oumap -OEEH .mufiwocon manusm omsm -wwv moumoco omcmcoxo Hwfioom mxufisco .m> Hozom mo mozmmw can xufioopmfiooc ~moo>wcs mud: xnozuo: HmMOOm cowpm~ m cm owcmcoxo uoonwpcfi .Emfizcu~m :o mammcmeo ocoe .choflufiwcoo mmoqv .xOAOOpafioop afloMLQEM .AHMOOHQMOOA mo :oHumuoomxo cm :Oumo ma ocozu cwsogufim magnuou pocflmop com: acowcwucoo «on ma mumwocon mo x~mm3m oak .co>wooop on» mo :OMOOLomMO ocu ou umofi on umse pew .usonw pocwmwcmn on uoccmo .Oo:m>pm cw pewmwoomm Do: men m:0wumwfi~no :OMOmuao Auwooumwoom mucoeowfiscom Auwoosmwoom casuawasno co suaoacmomam omcmsoxm owEocoom owcmsoxm Hmfioom I: il- I'll-1|‘ OAOWMHOHOMHmzu .ms0w>m:om pew mofiumflcouomcmnu owcmcoxm OwEocoom pew Hmfloom mo ummhucou m amowmxsm An whoa pocfleaouop ma moousomoc oweocooo mo xuwoumow .o~ndmmon coflumoo-m pew =Omwcmmeoo pomxo moxme nuns: Hmpusoc xfifimmoe new .ofinmomcmcououcfl .o~nwmfl>wp Ohm .posammos x_uomxo on cue moopzomoc Oweocoom .:OmLOQ ecmzu m swooncu powcmzoxo on emu .owumflfimmco>ficz ocm new pofldamam ecu mo ucopcom -opcfi opm mufimocon ofieocoom .mfim:v«>fip:w mo moo:OwLomxo own; so we ~Hoz mm xuofloom mo m3m~ new mayo: ozu >2 nonwmov ma mOOMSOmOH Hmwoom mo xuwoumom .ouonocoo cmzu pecans owfionexm emumo Ohm mOOHSOmom .oumoocmwoounuo>o ow pcou pew pawn one :munoe: ma rampsoo Ho: Ohm o~aooa was» .ucoeousmmoe fiecusoc uomxo o: o>mz moopzomou fimwoom .xpmfipoewoucw cm cwzoccu powcmnoxo on yo: emu .AOMumwumfiaofiuummv awnm Icomumfios Amoco paw po>~o>cm mcompom one com: mpcoaop mufimocon fimwoom mo osfim> ozh .oocsom ocu Eopm ofinm icomuov uo: Ohm mufimocon fimwoom mOUHSOmmm MO QHDHwZ EmwfimeO>M=D .w> EWMHN~SUMHHMQ owcnsoxm ofieocoom owcmcoxm Mafioom owumwnouomhmnu .eoscaucou--._ «Home 44 .m_mom ofimwufise czocx w:OEm moop:Omoc museum oumoofifim ou 3o: mo>~o>cH .mpowcwsoxo ecu mo mosfim> no mooconomonn ca moocouowmwp Ohmscow meofiuommcmsh .coHumo«=:E -Eoo may 0» coca :ofiuommcmhu one on co>wm ma mwmmszo who: .osflu oeam osu um usooo uoccmo .mmoom .xocoe mo wcwxmu paw mcw>fiu .mHOm pom mcoow uozou mm; oco .uozuo ow mpoom mcfi>fiw :H .mflcmcoflumfi -maop O>Mumwo= m o>mz pocuo OH mcfi>ww mam mfiom ou wcfl>fio .pocwm some: EOHQOEQ osu new czocxcs mamow so“: mofioh no mmoflfion .mosfim> cw HOMMmcoo o>~o>=~ .mpowcmcoxo oz» coozuoc mosfim> mo Auwpmfiwewm may :0 m“ mammnmEm .cofluom umcmuu cmcu segues :OMOmowcse -Eoo ow =O>ww ma mammcmEo Ohoz .Aweaxaav Assume -mo; meow mo Oocomopa one a“ ssooo :mo w=fl>flwunooco~m>fiaEm oEOm mo>~o>cw o>o~ mcfl>wu uwcwxmu cam wcfl>wm coozuoa mangOMHmfiom .mfiom op mo>mw x_m:oo:mu~:Ewm oco .ponuo ou O>o~ m=fi>fiw :H .anmcofiu inflow o>fiuwwom m o>wn uozuo ow wcfi>mw pew m~om ou mcw>wu ”conpo pcm wfiom :oozaon magmcofluadom m:0wum:uwm Eofinonm wo chaumz COMOQOMCSEEOU .m> acauommcmnh Owcmzoxm mo mofism owcmzoxm Omeocoom owcmzoxm ~mfloom Owumflnouomnmno .emscaucoo--.a oases 4S .mcoflmMOOp ucowcwucoo AfimmSOSE OZH .mpco ow memoE mo coda -mcflpnonsm cam acoEumsmpm oca mo>~o>cH .ucosumo>:« Essacwe pom cpsuon onu mo :oflumnwwaoz .xumflwuz mo>fiumumaeoov gmcwwnme mo menou :H mcwoe oomoofifim ou compo a“ mumoo o>fiumccou~m mo meson :fi mpco mo :oHum:~m>m .m3o«> pew m~mow acowuo>wp so“: m:Ompom mo xuflofifin -fiofise m x: :onHoop o~mcwm < .momwwumfio mo3~m> paw mfimom .pocfimop sawmofio Eo_noma :fl mufismom .Eswunflmflsco paw xcospm; .:0wuapmoucw cw wcwufiomop EoHnOOQ m cm muouomw -m mo .moosmop mcmxcm> :« .coflu -mowmflwoe Humane .:o«u=~0mom .mofiufi_wnwm -moa oumcflmcw mo uzo :oMuom mo ompsoo m mewuco>cfi :0 new :OMHmwpoE :0 ma mammszm ooflocu new mcfixmz :oflmfiooo oEoouzo no :oflusfiom Eofinoum mommooond wcw>gom Eofinopm owcocoxm ofleo:0om owcmsoxm Hmwoom oflpmfipouomumnu .eozcwucou-u.~ oHnmh 46 qualities. The expression of respect or the giving of information involve a symbolic form of expression while the giving of services or goods involve a concrete form of expression. The Foa and Foa (1974) theory implicitly assumes that every human interaction involves exchange. It is unlikely there could be one-way transfers because some kind of verbal or behavioral message would be returned following the giving of any resource. There is an explicit recognition that love, respect and other noneconomic resources are exchanged between persons apart from economic transactions and simultaneously with economic exchanges. Explanation of reciprocity and altruism. The second concep- tual difficulty of exchange theories has been the problem of how to define and integrate the concepts of exchange, reciprocity and altru- ism. Ekeh (1974) incorporated the ideas of Gouldner (1960) and expanded the concepts of univocal reciprocity and generalized exchange conceived by Levi-Strauss (1969). Ekeh (1974) indicates that mutual reciprocity is a mutually contingent exchange of benefits between two or more units where each party has both rights and benefits. "A" expects to be benefited by "B" when "A" has helped "B." Mutual reciprocity operates in face- to-face situations and compels reciprocation only for what has been given or received. The norms of mutual reciprocity are: (1) help those who have helped you; (2) do not injure those who have helped you. In other words if you want to be helped by someone, you have to help them. 47 The norms of mutual reciprocity result in a model of restricted social exchange which operates only between partners in multiples of two (Levi-Strauss, 1969). Mutual reciprocity between partners who have no other potential partners is called exclusive restricted exchange. Inclusive restricted exchange is one in which restricted exchange partners are implicated in a larger network with other dyadic exchange relationships and the possibility for exchange of partners exists (Ekeh, 1974). The particular characteristics of restricted social exchange are: (l) unusual attempts to maintain equality and strong emotional reactions when the equality rule is breached; (2) a quid pro quo atti— tude that common investments and goods from which individuals can gain indirectly and ultimately are not workable; (3) a brittle nature of the relationship resulting in mechanical solidarity and frequent lack of trust (Ekeh, 1974). Univocal reciprocity exists when the reciprocations involve at least three actors and the actors do not benefit each other directly, but only indirectly. No party gives directly to the party from whom he received. There are two types of univocal reciprocity: (1) chain univocal reciprocity, (2) net univocal reciprocity. The norms of univocal reciprocity are: (1) People should help others who now need the type of help they themselves may need from some others in the future; (2) People should help others who now need help for which they were provided by some others in the past. The norms imply there is enough trust that the giver will be recipro- cated by someone somewhere in the future. It implies generalized duties to others and generalized rights. 48 The norms of univocal reciprocity result in models of general- ized social exchange which operate under the law of extended credit: The receipt of any benefit by one party is regarded as a credit to that party by all other parties and therefore reciprocation is regarded as a credit to all of them. Failure to reciprocate is regarded not just as the sole business of the cheated individual but of the group (Ekeh, 1974). The morality generated by generalized exchange systems differs from the quid pro quo attitude generated in restricted exchange sys- tems. There is a trust of others and in their ability to discharge obligations to the enrichment of society rather than for their exclu- sive narrow self-interests. There is a belief that persons are credit worthy and can be trusted to pay back what they owe. There are con- tributions to causes that do not yield immediate and direct benefits to the contributor, with the hope they will ultimately and indirectly benefit him or his family (Ekeh, 1974). Ekeh describes three types of generalized social exchange: (1) chain generalized exchange; (2) individual-focused net generalized exchange; and (3) group-focused net generalized exchange. Chain gene- ralized social exchange in a group of five members functions as follows: A + B + C + D + E + A (+'= gives to). Individual-focused net generalized exchange is where the group as a whole benefits each member consecutively until all members have each received the same amount of benefits and attention. Usually the members pool their social and economic resources to do benefit to each of the members. A five-party individual-focused net generalized exchange operates: ABCD‘+ E; ABCE + D; ABDE + C; ACDE + B; BDCE + A (Ekeh, 1974). 49 Group-focused net generalized exchange is where persons suc- cessively give to the group as a unit and then gain back as part of the group from each of the unit members: A + BCDE; B + ACDE; C + ABDE; D + ABCE; E + ABCD (Ekeh, 1974). Ekeh suggests that odd numbers of generalized exchange groups may manifest a greater degree of social solidarity than even numbered groups, but regardless of numbers the generalized exchange systems are relatively devoid of emotional loading and create a high degree of social solidarity with the major attribute of trust of persons in the system. Although sociological exchange theorists have given some attention to the integration of "reciprocity" into the theoretical structure of exchange theory, little attention has been given to the integration of "altruism.” It is the economic theorists who have recently contributed to the clarification and integration of the con- cepts of exchange, reciprocity and altruism. Economists have recog- nized that resource allocation is not accomplished by exchange alone, but must also be accomplished by giving (Arrow, 1975). In addition to studying the allocation of economic resources, some attention has been given to studying how love, fear, loyalty, trust, justice and altruism are related and interdependent in the allocation and exchange of economic resources. Boulding (1973) differentiated among the concepts of grants, exchange and reciprocity as well as specified the resulting conse- quences of these resource transfers. Boulding indicated that exchange, based on a conditional offer and a specific formality of contract between equals, does not have the power to create community, identity 50 and commitment and has the further weakness of being confined to a two-party relationship. The grant system of one-way transfers of exchangeables or commodities (sometimes separated by an interval of time) can be multi-party and can create loyalty, affection and commit- ment which fosters integrative relationships. Boulding further dis- tinguishes reciprocity from exchange by pointing out the status inequality and unconditional nature of reciprocity. The grant, as described by Boulding, is limited to economic resources (commodities and exchangeables) and thus does not achieve an integration of social and economic exchange behaviors despite the attempt to describe integrative relationships and the implication that noncommodity transfers of status and prestige may accompany the grant arising out of benevolence. The grant arising out of benevolence is an altruistic behavior, although Boulding does not explicitly label the grant as "altruism." The concept of altruism was examined in depth by Phelps (1975) who defined the concept as "behavior actuated by a sense of others, their desires and expectations" (p. 2). Phelps indicated that altruism is expressed in a variety of forms: individual, interpersonal, unilateral (as within the family), COOperative and multilateral (as in agencies of government, voluntary associations and private philanthropies). The task of the economic theorist, according to Phelps (1975), is to explain the allocation of resources and thus the analysis of altruistic resource use is an important task. The economic theorist must be able to answer the questions: Why, when and how do persons behave in a way that is apparently altruistic; for what motives, under what conditions, through what channels? SI Economists have been primarily interested in altruism as the giving of economic or measurable resources rather than the uncondi- tional gifts of love, respect, and personal services which are exchanged or given in the family setting. However, McKean (1975) likens altruism to unselfishness which he views as a concern for others without refer- ence to any agreement governing specific behavior, i.e., giving of personal services which may involve significant personal costs on the part of the giver. Hoffman (1975) defined altruism as "any purposive act on behalf of someone else that involves a net cost to the actor" (p. 137). Hoff- man and Nagel (1975) agree that empathy (the empathetic response to another person's distress, interacting with the observer's cognitive sense of other) provides the underlying basis for the altruistic moti— vation of humans. It is the empathetic abilities of persons which account for variations in altruistic behaviors (Hoffman, 1975, p. 138). Although Hoffman indicates a recent burgeoning interest in altruism by social scientists and discusses the developmental processes involved in the development of empathy and its resulting altruistic motivations, there has been little attempt to integrate the concept of altruism into the theoretical structure of exchange theory. Foa and Foa (1974) do not use the word "altruism" but specifi- cally describe giving exchanges of both economic and particularistic resources emphasizing the contribution of giving exchanges of particu- laristic resources to the satisfaction of individuals. The Foa theory also contrasts the behaviors and consequences of taking exchanges in contrast to the reactions of persons involved in giving exchanges. 52 Two:party and multi-party exchange models. The integration of two-party (micro) and multi-party (macro) exchange relations into a unified theory is a critical task if exchange theory is to be increasingly useful in family research. The two-party model of exchange is best exemplified in the work of Homans (1961) with empha— sis on the internal psychological processes. The works of Thibault and Kelley (1959), Blau (1964) and Emerson (1972) give greater emphasis to multi-party exchange models. Turner (1978) suggests that Emerson (1972) has developed a more viable exchange theory which is less vulnerable to previously mentioned methodological and conceptual problems. Turner (1978) sug- gests that the Emerson theory has more deductive rigor, uses prin- ciples of operant psychology in a way that allows fOr sociological theorizing, and does not have the vagueness in conceptualization present in previous theories. Emerson shifted the unit of analysis from the attributes of the actors to the form of the exchange relation- ship and the properties of the relation. Equity and distributive justice. The conceptual issue for exchange theorists of how to define and integrate the concepts of power, equity, distributive justice, altruism, reciprocity, and exchange into one theory has not been resolved. This issue is partic- ularly important for family research and will be discussed in further detail in the next section which is concerned with the application of exchange theory to family research. 53 Application of Exchange Theory to Family Research During the past decade there has been increasing interest in the application of exchange theory to family research (Edwards, 1969; Broderick, 1971; Rodman, 1972; Edwards 8 Brauburger, 1973; Burr, 1973; Foa 8 Foa, 1974; Turner, 1975; Safilios-Rothschild, 1976; Rollins 6 Bahr, 1976; Gottman, Notarius G Markman, 1976; Murstein, 1976; Traupmann, 1976; Johnson, 1977; Walster, Utne 8 Traupmann, 1977; Spakes, 1978; Johnson, 1978; Nye, 1978, 1979; Kersten, 1978; Osmond, 1978; Lewis 8 Spanier, 1979; Darling, 1979). Many of the studies cited above are applications of exchange theory to the study of kinship or inter-generational relationships. Darling (1979) reviewed exchange theory with a developmental emphasis applied to parent-child relationships. There are several studies and theoretical articles which apply exchange theory to the examination of marital relationships (Gottman et al., 1976; Johnson, 1978; Murstein, 1976; Safilios-Rothschild, 1976; Traupmann, 1976; Burr, 1973; Nye, 1979; Lewis G Spanier, 1979; Jorgenson, 1979). Gottman et a1. (1976) observed six couples with distressed marriages and six couples with nondistressed marriages in decision making procedures in the laboratory situation and found minimal support for the view that distressed marriages are characterized by less posi- tive or more negative reciprocity. The sample size and the knowledge of coders concerning the classifications of couples suggest caution in the interpretation of the results. Murstein (1976) applied exchange theory to creating a theory of courtship and mate selection including a discussion of rewards, 54 costs, and profit in relationships. Murstein suggested reasons for equity of exchange value not being necessary for a relationship. His reasons were called "alternative sources of rewards" and were similar to the resource substitutions discussed by Foa and Foa (1974). Without using the word "altruism" Murstein indicates that much of behavior is not exchange oriented because many persons have worked without equitable returns (mothers, Jesus, Albert Schweitzer and Pope John XXIII). Murstein suggested that rewards received by these indi- viduals were not external, but internal and thus by his definition cannot be considered exchange. Murstein hypothesizes that commitment to a partner is a form of internal reward. "If the individual believes he is in some way responsible for his partner's welfare or that he has committed himself to marriage by dint of continuing the relationship over a long period, he may remain despite inequities of exchange" (p. 112). Murstein (1976) defined "equity" as equal rewarding power. Two equitable persons can thus be totally dissimilar such as the example of the beautiful but poor woman who marries the ugly but wealthy bachelor, representing an equitable balance of beauty and wealth. Although Murstein discusses behaviors which represent reciproc- ity, altruism and distributive justice or equity; the concepts are not explicitly labeled or integrated into exchange theory. Contrasts between economic and social exchange behaviors are not discussed. The exchange model is limited to the two-party restricted exchange model with some recognition (implicit) of the existence of univocal reciprocity and generalized exchange. 55 Safilios-Rothschild (1976) in a macro and micro examination of family power and love specified a list of desirable goods (resources) that may be exchanged between spouses: 1. Socioeconomic: money, social mobility, prestige. Affective: affection, loving and being loved, feeling needed and needing the other. Expressive: understanding, emotional support, special attention. Companionship: social, leisure, intellectual. Sex Services: housekeeping, child care, personal services and linkage services. 7. Power in the relationship (p. 356). N 0014504 Safilios-Rothschild indicates that "special attention” refers to the entire range of special things one spouse can do for another such as make him/her feel good, appreciated, loved or to improve his/her mood. The category of "personal services" refers to washing clothes of spouse, ironing, dry cleaning, buying clothes. "Linkage services" include the services that link the family with other social systems such as bureaucracies, political and legal systems, insurance, and taxation. The central concepts of interest in the Safilios-Rothschild (1976) study are the economic and affective resources, their exchange, and the resulting balances of power. A central proposition of the study is: "The more a spouse has no direct access to a resource or no alternatives through which to receive the same resource, the greater value he/she accords to these resources and the greater the willingness to pay high cost to secure them" (p. 356). The author indicates there are few resources which can be con- trolled by only one spouse and also highly valued by the other with no alternative sources of reward. The socioeconomic and affective resources do fit this qualification in most societies. Unless women 56 inherit status and income from their fathers, they are dependent upon husbands for socioeconomic resources and therefore the socioeconomic resources are controlled by men. Women do have control over the reciprocation of the husband's love which can give women a basis for power despite their lack of control over socioeconomic resources. Safilios-Rothschild (1976) emphasizes the importance of inter- category resource exchanges in marital relationships such as: love/ sex for services, and status for all other resources. An observer of the marital relationship is less able to observe the inter-category resource exchanges than the intra-category exchanges and thus the rela- tionship may not appear equitable; however, satisfaction depends upon each spouse's perception of the nature and magnitude of the exchanges taking place. Although love is an extremely important variable for under- standing family dynamics, Safilios-Rothschild (1976) feels it has been neglected for several reasons: (1) Love has been considered a sub- jective, vague, elusive feeling that defies Operationalization and measurement; (2) It has been treated as a constant since Americans marry for love and stay in love until death or divorce. (3) Male family sociologists have considered love a "soft, feminine" variable that cannot be treated in the same way as "hard" variables, such as socioeconomic variables (p. 357). Exceptions include Scanzoni (1972) who discussed love in marriage, and Waller (1951) who originated the "principle of least interest." The Safilios-Rothschild hypothesis suggests that: "the spouse 'more in love' with the other is more anxious to maintain the relation- ship even at a high cost (by offering many resources and by receiving 57 few in return).” Affective resources and sex obtain such a high value for the spouse "more in love" that power may be willingly exchanged for these resources. The author looked at who makes important and infrequent decisions (power) by relative love involvement in order to test the hypothesis that the spouse "more in love" has less power in the marital relationship than the spouse "less in love." The hypothesis was consistently supported by answers given by Greek wives. Safilios-Rothschild attempted to integrate the exchange behav— iors involving social and economic resources and speak to the issues of power and distributive justice. Mention is made of reciprocity, but little attempt is made to integrate concepts of reciprocity and exchange. Altruistic behavior is implied in a definition given for love: "the willingness to make sacrifices for the loved person and the desire to please the other” (p. 358) but the word is not explicitly mentioned. Burr (1973) created and collected propositions that explain variation in marital satisfaction by incorporating propositions from symbolic interaction, balance, and exchange (Homans) theories. 'The following propositions were developed from exchange theory: 3.11 The amount of interaction influences the amount of sentiment and this is a positive, monotonic relationship. 3.12 The amount of sentiment influences the amount of interaction. 3.13 If the profit from interaction is rewarding, the sentiment produced by interaction tends to be positive, whereas if the profit is costly, the sentiment tends to be negative. 3.14 The value of interaction influences the amount of profit from the interaction and this is a positive, monotonic relationship. 3.15 The profit from interaction influences the amount of inter- action and this is a positive, monotonic relationship. 58 3.16 The amount of profit from interaction is related to the amount of influence interaction has on sentiment and this is a positive relationship. 3.17 The value of sentiment influences the amount of profit from the sentiment and this is a positive, monotonic relation- ship. 3.18 The profit from sentiment is related to the amount of influ- ence that variation in the amount of sentiment has on the quality of interaction (pp. 53-58). In addition to the propositions generated from exchange theory, Burr (1973) contributed a proposition regarding ”altruism" which was defined: "the tendency for individuals to respond in ways that favor the other person when there is a conflict of interest" (p. 59) and an additional proposition for empathy: 3.20 The amount of altruism influences marital satisfaction and this is a positive relationship (p. 60). 3.21 The amount of empathy influences marital satisfaction and this is a positive, curvilinear relationship with the influ- ence occurring in the low range of empathy (p. 61). Burr (1973) integrated thirty-one propositions into a theor- etical relationship to hypothesize variations in marital satisfaction. Nye (1979) reviewed exchange theories, stated foci and assump— tions, and discussed applications of exchange theory to various domains of family life such as: paid employment of mothers; communication as rewards and costs; choice, exchange and marital dissolution; social networks and the family; violence in the family. The discussion of communication as rewards and costs is most relevant to the present study. Nye suggests several general hypotheses: 67. There is a curvilinear relationship between the amount of verbal interaction in which people are involved and the satis- faction they feel with the relationship producing that verbal interaction. 59 68. For husbands, the relationship of spousal communication to marital satisfaction is curvilinear, with a total absence of verbal communication associated with acute marital dissatis- faction, a low level of verbal communication associated with satisfaction, and a high level of spousal communication associ- ated with marital dissatisfaction. 69. Wives whose husbands frequently communicate verbally with them are more likely to be satisfied with their marital relation- ships. 70. For wives employed full time, the relationship between con- versation with husband and marital satisfaction is curvilinear, with no conversation associated with marital stress but, beyond a minimum level, no change in marital satisfaction as frequency of conversation increases. 71. Verbal communication that agrees with our opinions and values is rewarding and increases satisfaction with the relationships. 73. Husbands or wives whose spouses engage in verbal communication attempting control of them are more likely to be dissatisfied with the marital relationship. 74. Husbands or wives whose spouses belittle them in verbal com- munication are more likely to be dissatisfied with the marital relationship. 75. Husbands or wives whose verbal communication frequently involves attempts at control are likely to spend less time with spouse. 76. Husbands or wives whose spouses' conversation often belittles them are likely to spend less time with the spouse (p. 24). Traupmann (1976) reported the initiation of the study of equity in marriage. The hypotheses were: (1) As intimate relation— ships progress over time, they become more equitable. (2) Individuals in equitable relationships will be more satisfied, than individuals in inequitable relationships. (3) Individuals in equitable relation- ships will feel more love and less resentment in their relationships than will those in inequitable relationships. The study was planned to include 100 randomly selected couples. Information was to be obtained by indepth interviews. 6O Traupmann (1976) reported that information was to be obtained from respondents concerning feelings about the amount of passionate love, compassionate love, resentment in the marriage and feelings about contributions made to the marriage by respondent and spouse. The respondents were to be questioned about several inputs and out- comes of the marital relationship including: 1. Personal contributions: Physical appearance, intellect, social grace. 2. Emotional contributions: Communication, understanding, liking and loving, respect, physical affec- tion, security, sex, acceptance. 3. Daily contributions: Finances, day-to-day maintenance, decision-making responsibility, sociability. Traupmann (1976) reported the theoretical work of the study, questionnaire development, and plans for the future. There were no reported findings. Lewis and Spanier (1979) after developing a proposition inven- tory relating to marital quality suggest: "Continued development of a social exchange theory of marital quality and stability appears to be one of the most fruitful tasks that could be undertaken by family theorists. In other words, the potential for a social exchange theory in this area is very promising indeed, as viewed from our inventory of propositions" (p. 285). Marital Quality This section will discuss the two major reviews of literature on marital quality which have been completed in the last decade, as well as the studies which have been completed during 1979-80. During the past twenty years there have been numerous studies using subjective feelings about marriage as the dependent variab1e(s). 61 The dependent variable has had various labels: happiness, satisfac— tion, affective evaluation, marital success, adjustment, and has recently been labeled marital quality. Hicks and Platt (1971) reviewed research on marital quality from 1960 to 1969 and reported a long list of independent variables positively correlated with marital satisfaction: 1. High occupational status, income, and education for husband. 2. Husband-wife similarities in socioeconomic status, age, religion and attitudes. 3. Older ages at the time of first marriage, duration of marriage. 4. Employment of wives and nonemployment of wives. 5. Male role performance. 6. Congruence of husband's self-concept and concept of husband held by wife. 7. Congruence of husband's self-concept and husband's concept of his father. 8. Greater similarity between self-concept and perception of spouse. 9. Women with kind, loving fathers whose husbands now met these needs. 10. Higher involvement in family activities. 11. Conventional styles of life. 12. Low child density (number of children/number of years married) 13. Frequency of expression of affection, understanding. 14. Personality traits of spouse: moderately managerial, docile, cooperative, responsible, considerate, helpful, tender, big- hearted, warm, friendly, neighborly, adaptable, flexible, emotionally stable. 15. Communication patterns: frequency of verbal communication, disclosure of feelings, sensitivity, empathy, personalization of language symbols, increase in supplementary nonverbal tech- niques of communication. There is also evidence to suggest that some of the above vari- ables have a negative correlation with marital satisfaction. The most contradictory results exist with the following variables: frequency of communication, amount of self disclosure, length of marriage, occu- pational status and education of husbands, and employment of wives. One of the most significant problems of research on marital quality involves the incomparability of results. Researchers have 62 used many different measures of variables with the same name, or have used a particular measure for variables with different names. In addition to the lack of consistency of measurement, most researchers have allowed respondents to define marital satisfaction by asking for self reports on feelings and attitudes. There have been few attempts to observe marital interaction or to use the experimental design with control groups. Research on marital quality has been characterized by difficulties in conceptualization and measurement. In the words of Snyder (1979): "the entire area of marital assessment has suffered from the lack of a comprehensive multidimensional measure with well- constructed norms that permit the simultaneous assessment of a broad range of dimensions in marriage as these relate to global marital satisfaction" (p. 813). Research completed after 1969 has been less general, included a greater number of variables and has focused on particular domains of marital interaction such as communication and traditional vs. non- traditional sex role orientation. There have been attempts to inte- grate findings by developing propositions and theories. The work of Burr (1973), reviewed in relation to the application of exchange theory to family research, was one of the first attempts to develop prOposi- tions relating to marital satisfaction. Lewis and Spanier (1979) examined findings of several researchers who studied quality and stability of marriage and devel- Oped a propositional inventory which was then integrated to form a theory of marital quality and stability. The propositions are orga- nized in the following major categories: premarital factors and 63 marital quality, social and economic factors, interpersonal and dyadic factors. Information obtained from respondents in the Oakland County, Michigan Quality of Life study from which the present data were obtained does contain information for testing the majority of propo- sitions suggested by Lewis and Spanier (1979). The present study is most closely related to the area labeled by Lewis and Spanier as interpersonal and dyadic factors affecting marital quality, particu- larly the propositions listed under the categories of emotional grati- fication and interaction. The following propositions of Lewis and Spanier (1979) are particularly relevant to the present study: 38. The greater the ease of communication between spouses, the greater the marital quality. 40. The more positive the evaluations of the other, the more the marital quality. 43. The greater the expression of affection, the greater the marital quality. 44. The more the esteem (respect) between the spouses, the more the marital quality. 47. The more equalitarian the marriage, the more the marital quality. 51. The more the sexual satisfaction, the more the marital quality. 69. The greater the companionship, the greater the marital quality. 70. The more the shared activities, the more the marital quality. 73. The more effective the problem solving, the more the marital quality. 85. The more effective the communication between spouses, the more the marital quality (pp. 282, 283). 64 The previously discussed reviews suggest the following vari- ables to be positively related to marital quality: love and affection, empathy, respect, sexual satisfaction, companionship and/or shared activities, problem solving ability and communication. Research reported since Lewis and Spanier (1979) support the positive corre- lation of these variables to marital quality. Snyder (1979) developed a marital satisfaction inventory con- sisting of 280 true-false items representing dimensions of marital interaction which discriminated between couples involved in marital counseling and a control group. The variables representing affective and problem solving communication as well as the items representing common interests and the amount and quality of shared leisure time were the best predictors of marital satisfaction. Next in importance were the partners' satisfactions with their sexual relationship and the extent to which they experienced agreement about finances. Jorgenson (1979) examined the contribution of five types of socioeconomic rewards to twelve indicators of marital quality. Marital quality included three major dimensions: (1) perceived role competence of spouse in ten specific roles, (2) marital satisfaction with five areas of the marital relationship and an item of global satisfaction, (3) dyadic commitment which consisted of a score result- ing from pooling three separate items. Results indicated husbands' perceptions of marital quality were unaffected by varying socio- economic reward levels. For wives the socioeconomic reward levels were moderately related to perceptions of husbands as competent pro- viders and satisfaction with income of spouse. 6S Jeries (1979) examined the relative importance of specific sociological, economic and psychological variables in explaining marital satisfaction. The findings indicated that going places together as a family and joint marital decision making regarding the best place for the family to live were positively related to marital satisfaction for wives. The frequency of borrowing and exchanging favors with neighbors and the frequency of money problems related to savings were negatively related to marital satisfaction for women. Marital satisfaction was measured by responses to one question on a four-point satisfaction-dissatisfaction scale. Albrect (1979) created a comparative indicator of marital satisfaction in order to examine correlates of marital happiness among remarried persons. Three indicators of marital satisfaction were used: (1) Respondents were asked to compare the present marriage with the former marriage which had ended in divorce, (2) to rate the degree of overall satisfaction they felt with their current marriage compared with that of other couples they knew, and (3) to compare their present marriage with the expectations they had had fOr that marriage prior to its occurrence. Responses were given on five-point scales. The independent variables in this study were primarily socioeconomic factors and were not good predictors of marital satisfaction for remarried respondents. Gilford and Bengtson (1979) used data from 1,056 married mem- bers of three-generation families to develop a two-dimensional measure of marital satisfaction reflecting positive interaction and negative sentiment. Respondents were asked to evaluate on a five~point fre- quency scale how often the events occurred between spouses. The 66 positive items were modified from Spanier's (1976) dyadic cohesion subscale (discussing, working together, laughing, exchanging ideas and good time). The negative sentiments included sarcasm, abnormal talk, disagreement, criticism and anger. Appropriateness of the two conceptual dimensions was verified using principal component factor analysis with varimax rotation. Results of the study showed the youngest generation highest on both positive and negative factors. The oldest generation showed moderately low levels on positive interaction but even lower scores on negative sentiment. There was a linear decline in the relationship of age and negative sentiment and a curvilinear relationship between age and positive interaction. Ammons and Stinnett (1980) studied 72 rural, middle class, middle aged individuals qualifying as having a vital marriage by the Cuber and Haroff (1965) criteria operationally defined as a score of 25 out of 35 possible points on the Vital—Total-Relationship Scale (VTMRS). The VTMRS measured: (1) degree of satisfaction derived from the marriage relationship, (2) degree of emotional involvement the couple had with each other, (3) the degree to which they enjoyed living their lives together, (4) the degree to which the couple did things together. A modified version of the Edwards Personal Preference Schedule (Constantine, 1971) was used to determine the degree to which respondents possessed each of fifteen personality needs. Results of the Ammons and Stinnett (1980) study indicated respondents had moderate to high: (1) needs for sexual activity, (2) needs to be understanding and supportive, (3) needs fOr achieve- ment and endurance, and (4) ego strength. It was concluded that 67 respondents' needs to be supportive and understanding would lead to encouragement of reciprocity; that needs for achievement and endurance affected the high interest found in developing and sustaining the relationship; and that high ego strength contributed to the ability to form intimate, caring relationships. McNamara and Bahr (1980) examined the dimensionality of marital satisfaction by factor analysis and concluded that role satis- faction is a separate dimension from role dissatisfaction and that the apropriate model of marital satisfaction is unipolar rather than bipolar. Research on marital quality completed since the Lewis and Spanier (1979) review indicated that socioeconomic variables were not very successful in predicting marital quality and that interpersonal and dyadic independent variables were of primary interest. Support was given by at least two different studies for the importance of the following variables as indicators of marital quality: (1) affective communication including understanding and support; (2) satisfaction with and high interest in sexual activity; (3) joint decision making or satisfaction with problem solving communication; (4) agreement on finances or low frequency of money problems related to savings; (5) amount and/or quality of shared leisure time. FamilyfiWell-Being Family well-being has been studied from both economic and social-psychological viewpoints. Although the emphasis in the present study is on social-psychological well-being, it is recognized that the health of the family system is dependent upon both economic and 68 noneconomic resources. The basic needs for food, clothing, shelter, and physical health must be met before it is possible to consider meeting higher level needs. Studies selected for review in this section emphasize social- psychological well-being of families. The assumption is made that basic physiological needs of individual family members have been met. The studies selected for review placed emphasis on family rather than individual as the unit of study. Several techniques have been used for studying healthy family systems: (1) observation of interaction processes in the laboratory or the natural setting, (2) information obtained by testing or inter- view of individual members and/or the family group, (3) reviews of the literature, (4) questioning family professionals, (5) studying control groups which have been asked to respond to the same stimuli as patient families, (5) combinations of the above techniques. A study combining all of the above techniques was completed by Lewis, Beavers, Gossett & Phillips (1976). The study originated with the hope that qualities of families which produce capable, adap- tive, and healthy individuals could be understood. Several ways of defining health were identified: (1) health as the absence of overt pathology (reasonable rather than optimal functioning); (2) health as optimal functioning as determined by a theoretical system; (3) health as average functioning which is a statistical concept that views the midrange of the majority as healthy; (4) health defined as process which takes into account changes in the system over time; (5) health defined as any combination of the above definitions. 69 The study sample was selected by the following criteria: (1) biological intactness of the family, (2) oldest child in mid- adolescence, (3) no family member in psychological difficulty (absence of overt pathology). Thirty-three families were selected for study and twelve of these families were selected for additional intensive study. Information was obtained from respondents by interview, video- tape of interaction processes, rating scales, evaluations of clinicians, and microanalytic counting methods. The information obtained enabled researchers to rank families on a continuum of average to optimal functioning. Results of the Lewis et a1. (1976) study indicated seven char- acteristics that distinguished Optimal from adequate families: (1) affiliative vs. oppositional attitude about human encounter, (2) respect for one's own and the subjective world-view of others, (3) openness in communication vs. distancing mechanisms, (4) firm parental coalition without evidence of competing parent-child coalitions, (5) belief in complex motivations, (6) spontaneity vs. rigid stereotyped interactions, (7) encouragement of unique vs. bland human characteristics. Both adequate and optimal families had high levels of initiative vs. passivity. Additional results from the Lewis et a1. (1976) study indi- cated that those families designated as optimal expressed far greater degrees of marital and family satisfaction. Compared to adequate families the optimal families had: strong affectional bonds, higher satisfaction of wives, husbands were more directly supportive of wives and showed less interpersonal distance, there was increased capacity to communicate thoughts and feelings, shared adult leisure 7O pursuits, community involvement, and a prevailing mood of warmth, affection, and caring. The mother in less than optimal families was the first to become dissatisfied, distressed or symptomatic. A study which was stimulating to the Lewis et a1. (1976) research group was the social psychological study of family health completed by Westley and Epstein (1969). This study investigated the relationship between the emotional health of 96 college students and the organization of their families. Information on reported interaction was obtained from interview and testing. Results of the study indi- cated the following family characteristics were highly correlated with high degrees of emotional health in older children: (1) The parental marriage was successful with high degrees of shared responsibility in the home. (2) Parents had continuing high levels of sexual interest and activity. (3) The power pattern was father led. (4) Problems were approached early and effectively. (5) Communication was open and direct. (6) There was a balance of autonomy and dependence. (7) There was a centrality of importance to the relationship between parents. Kantor and Lehr (1975) also studied healthy or "normal" families to develop a "descriptive theory of family process." The study differs from the Lewis et a1. (1976) and Westley and Epstein (1969) because the nineteen families were studied in their natural settings and not in the office or laboratory. Information was obtained by participant observation; tape recording; videotape; inter- views of the whole family, subgroups, and individuals; projective tests; self reports of family members; and microphones in all rooms of the house to record verbal communication during waking hours. It 71 was concluded that the principal activity of family process is dis- tance regulation--the manipulation of objects, events, and individ- uals in order to increase or modify associations or distances between one another. Kantor and Lehr (1975) identified strategies for regulating access to a personal subsystem by suggesting that persons on both sides of the interpersonal interface ask certain questions: "How close do we really want to get to each other? How much time do we want to spend together? Do we really like the other person(s)? Do we feel comfortable together? Is (s)he the kind of person we want to be associ- ated with? What do we have to gain?" (p. 29). The study suggests that the core purposes of the family system are to maintain stability through tradition, to achieve adaptation through consensus, and exploration through intuition. The distance regulation issues are the following: Access distance-regulation issues Space: closeness - distance Time: in phase - out of phase Energy: balance - imbalance Target distance-regulation issues Affect: joining - separating Power: freedom - restriction Meaning: sharing - not sharing The research then identified three distance regulation styles: the closed family system, the open family system, and the random family system. Each target distance-regulation issue was identified as having a target ideal for each type of system. Affect ideals for the closed system were durability, fidelity, and sincerity. Affect ideals for the open system were responsiveness, authenticity, and latitude. The 72 affect ideals for the random system were rapture, whimsicality and spontaneity. Fisher and Sprenkle (1978) surveyed Opinions of 310 randomly selected members of the American Association of Marriage and Family Counselors concerning their perceptions of healthy family functioning. Respondents were asked to rate thirty-four items in the questionnaire as to whether they were important, very important, or crucial to healthy family functioning. The respondents were then asked to rank order their first seven choices from among the thirty-four items. By com— bining data from both rating and ranking, the researchers identified a list of variables ranked high and low by both methods. The follow- ing variables ranked high by both methods: 1. Consider the sender of the message, the message, and own self as important and worthwhile, even if there is disagreement. 2. Attentively listen and observe while another speaks. '3. Family members can generate new ideas and change patterns of behavior and/or interaction in the face of new situations or modified assessments (flexibility). 4. Speak for self; use "1" messages: self responsible communi- cation. Avoid speaking for others (over-responsible) or appeal to authority (under-responsible). 5. Express feelings Openly and clearly (express feelings). 6. Family members validate and nurture each other verbally and nonverbally with regard to emotional needs (supportiveness). 7. Family members feel security, safety, and trust in one anothers presence (psychological safety). 8. Family members can successfully negotiate differences; that is, reach a decision that is acceptable to all, as opposed to limiting negotiation or being involved in endless negoti- ation (negotiation). 9. Attend to the affect and the content of a message. 73 Of the nine variables listed as important to family functioning as perceived by therapists, five were considered by the researchers to be indicators of family communication. Supportiveness and psycho- logical safety were indicators of family cohesion; and flexibility and negotiation were indicators of family adaptability. The key theoretical dimensions of cohesion, adaptability, and communication were identified by Olson, Sprenkle, and Russell (1977) as having excep- tional unifying and organizing potential. Physical caretaking--family members aid or assist one another in meeting physical needs-~was listed last by therapists in their priority list for variables important to healthy family functioning. Barnhill (1979) reviewed the theoretical literature on family therapy in order to isolate and integrate concepts of the healthy family system. The resulting dimensions of healthy family functioning included the following: 1 Identity Processes 1. Individuation: Independence of thought, feeling, and judg- ment of individual family members including a firm sense of autonomy, personal respon- sibility, identity and boundaries of the self. 2. Mutuality: A sense of emotional closeness, joining, or intimacy which is possible between individuals with clearly defined iden- tities. Change 3. Flexibility: The capacity to be adjustable and resilient in response to varied conditions and to the process of change. 4. Stability: Refers to consistency, responsibility, and security in family interactions. 74 Information Processing 5. Clear Perception: Refers to undistorted awareness of self and others, clear joint perceptions, and validation of shared events. 6. Clear Communication: A clear and successful exchange of information between family members, including checking out communication in order to clarify meaning, intention. Role Structuring 7. Role Reciprocity: Mutually agreed upon behavior patterns. 8. Clear Generational Boundaries Ferber and Birnbaum (1977) created a model of family well- being from an economic perspective which indicates that the well-being of the family is a function of the well-being of the husband and wife. The well-being of the partner is dependent on several conditions: family consumption, the direct (dis)satisfaction derived from work and leisure, the extent to which the individual could function inde- pendent of the family, the status the individual enjoys within the family, and on the well-being of the other spouse, weighted by the value this individual places on the other's well-being. Studies of family well-being are in agreement that the marital relationship is of central importance in determining health of the family system. The marriages in healthy family systems were charac— terized by strong affectional bonds and emotional support, shared responsibilities and leisure time, high levels of interest and satis- faction with the sexual relationship, open communication, and com— petence in problem solving. Healthy family systems were described as balanced on the dimensions of adaptability-stability, cohesion-individualism, and power. The communication skills facilitate open expression of 75 feelings and successful negotiation of differences with maintenance of respect and affection among all members. Quality of Life This section will first review survey research of subjective indicators of perceived overall quality of life (POQL) as the dependent variable in which evaluation of marriage or family life were included as independent variables (Table 2). The second section reviews studies having evaluation of marriage or family life as one of the major dependent variables (Table 3) and includes Mancini (1978) and the studies resulting from the Oakland County, Michigan Quality of Life Project (Jackson, 1979; Torres, 1979; Vliet, 1979; see also Sontag, Bubolz & Slocum, 1979). Surveys of perceived overall quality of life consistently found feelings about family life to be highly correlated with feelings about life-as-a-whole. Campbell et a1. (1976) reported that feelings about spouse and children explained the majority of variance in feel— ings about family life. Love and marriage were significant predictors of POQL in studies in which they were included. Evaluations of fun, leisure time, work, income and financial security, standard of living and health were also predictive of feel- ings about life-as-a-whole. If feelings about spouse explain the majority of variance in feelings about family life, then additional infOrmation about dimen- sions of the marital relationship which contribute to satisfaction is important to the explanation of feelings about life-as-a—whole. 76 .mcomsom poouo>wp can ommcwm mam .mxom~n .oocovwmou coop: .msumum oweocooo-OHoom uozo~ cu“: mucoecommop cow venom mm: coduommmflumm owdfi nozoq .:o«umo:po so .ommum ofioxo omMfi xfiweom .oEoocw .xom .oomu .owo mo :ofluappm ocu so pocfloHon mm: codumfium> fiwcofiufiepo ofiuufiq .xopcfi socos ocu pew .omwfi >~H5mm .csm mo u::OEo "Ono: :ofiuommmfiumm omflfi ou umoe wcwpsnwuucoo moHnmwpm> oak .mcomuom oHMCMm coco powmmMOMm opoe opoz mcomuom vownumz .msoquMpcfl o>wuoonpo xn cosy muoumOMpcw o>fiuoonn=m an pocwmfimxo mm: :ofiuommeumm cw oocmfinm> osoz .pommohocw Aomscopm oeoo:« mo commopocfl xuwfimsc omfifi paw :ofluQESmcoo .oEoocw x_asao can: coupoacmmpam .mucop -coamou mo o>fluawpomop pew Hmcfip3ufluom soon mofinmwhm> ucopeomopcw xcme ouo: ouogh .oocwumooom .ucoezmMHaEOOON wfiom .xuommm .Aozuon Eonm Eopooum .oocopcom nope“ .w:w>w~ mo pumpcmum .csm ”mfiuouwuo exudo x: poum3~m>o sumo .xfiwewm ma“ upzfiocw .om«_ mo memeOp xfim .xomsoopm oeoocw o>fluoon imam paw o>wuoomno coozuon Aucosnwcoo new xomacopm o>fiuoohno .xomscopm o>fiuoom loam ”zooswop< oEoo:H xfiwsom mac.“ - z mLH.H z kmN.H z seesaw soaps: w mzmoee< oeo.a u z Axnmav caspoaua mwcfipcwm fimmfiocwcd mofinmflpm> ucopcoaoch spaum .owwa mo xuw_m=o HHmpo>o po>woouom mo mxo>psm mo Ahoassmuu.m oHnMH 77 .xuwcseeoo use .xawemm .zufimo: Mono: cos you :ofluommmwumm omwfi mo thuoflpona umon och .owfia mo memo; o>fiuo3p0ha xHHmoflmo~oHn ozu magnum ucwusomsw umoe mm: owwa xgfleam cum: :owuowmmwumm .zufimo: .xuo: .wcw>wa mo pumvcmum .oofls sassao ”son o» emu; soum poxcms cofiuommmwumm omflfi mo mHOOOMpopm «moo och .mcomuon cowpume pom oEOOCM no“: commohocw mmwsmcoMu -mfiop o>o~ mo oocwuuome och .coeoz pofiupms op ucmunomEM Afiumfisowuumm mm: owmflpume \o>oq .mcomuon vowuamecs Op cozy vofipume O» can .modme coca mofimeom ou oocmupomefi Houmouw mo mm: owmfiuume\o>oq .mmo:o>fluoommo o>wuowpoum umon ocu we: owmfiuume\o>ofi mo :wmsov osh .u:oEcho>ow fimco«uw= .ucoe:0pm>=o fimhsumc .u:oEo>Ho>=fi fimcofiumNMcwwno .moflufi>wuom oemu opmam .omfifi genumuwmm .muouume socoe paw oeoocw .cowumospo .w:w>w~ mo ppmpcoum .moom .wcfimso: .xOM n=:EEoo .xuoz .xfiweom .nufimo: .ucoecuo>ow HmcoHum: .ucoecho>ow Homocw> loam .ucoecho>ow Hmoofi .xpuczoo .oo=w>osa .xuwo .voozuonzwwoc .wcfimsos .nufimo; .xuozomso; \xpoz .mowuw>wuom ousmwofi .mmwcm upcowpm .coupfiwso .owmwuuwe\o>o~ "oocowuomxo mo moose no mcflmeoo .mmooo:m\u:oso>ofi:om can xufiuaoom oweocooo .xuflhommoun .mwcmpcofium .xuflpsoom zaweom .o>o~ uofimom moafim> anewEhou mmnmfiv :omoxoz mo cofiuoOMMMpo: Ham.~ u z Anemia massaging w mazeaLasm ww~.m u z “mamas comcaxu< mwcflpcwu ummflocfihm mofinmwnm> ucopcoaopcm vaOm .emaesucou--.~ sagas 78 .cOOd new on mnocuo vac: coax :ofiumwoommm wcouum "mo>M3 you .mpommcmau cum: :0wuommmflumm pew muommcmnu mo mowocoacoum .coom -mpo>o pew whoMmcwuu cow: :OMO -OOMmfiowm ":oozuoc voumwxo maficm:Owum~ou o>wuwmom .cOOd mo muouowpoum umoc ocu mo oco mm: omfic xcfismm .poMMmfiumm umoe opoz oEoc um couvcwco cow: mcompod .con .xuomwm .eoom .auasc macawacow .moa~ inseam "mcofiuommmwumm umocww: .xuwusoom Hmwocmcwm .ucoEunwmm\om=oc .xuommm .cucooc .omMH xfifiemm "mcpoo -coo omfic acmupome umoe och .m:w>w~ mo pumpcmum .cummoc .xuoz .omfia sawemm ”:oEoz pow mnoqucoun umoc och as_aeac soc oe mwocuo .mnocuo you moop xcwsmm omcz cow: =o«uommmwumm "moccm -wpw> acopcomop cm:0muwppop .u:oE:co>om cacofium: .oEwu oumam .mpcowpw .oocopcomopcw .omMc wcwumououcfl .poom .moofim .omfiH x_wemm .xusmoc .xufiusoom cowocmCMM .ucoEuhmmo\omsoc .wcficuofio .cofimwcou .cam .cucmoc .xuoz .mcficuoEOm mcficmwfineouom .ucoe .3ow>poucw icouw>=o Hanson: mo acmuapcoo vcooom .xczum cmcwpsuwmcoc .xuommm ”mesoocoo omfifi HN me u 2 mo oocwuuonem can :ofiuommmmumw Amnmfiv nocowm w Ncocsm fl.c=ouc mascoxaaz w “assesses mmcwpcfim Hmmwocwpa moficmwuo> acopcomopcu >p=um .eoscaucou--.m manna 79 .m:0msom ocmcflm :mcp ococzumummuomfic cuflz podmmwamm opoe ouo: mCOmpom pomphmz .owwflnwms cow: cofloommmwumm cc oocmwsm> ocu mo umoe pocflmcmxo omsomm cum: m:o«umcou wcwcpoocoo moccmficm> .owmfiscms cucz pofiwmwumm ouoe ouoz coppcflco nozom cud: oamoom pew o_aoom sopfio .coupaflco can omsomm cow: mwcmcoflumcos Hmong x_ssac can: someone -mwpmm on whouscwpucoo Home: .owmdnhme .xcoz .w:H>fiH mo puopcmum .oMMH xcfieow .mowud>wuom wcwcpozcoc "cofiuoommcumm omfic w:MHOMp -opm cw acmuuomefi umos mcwoeoo .xocoE unocm mucoeoonwwmwp mo xocoscopm .omsomm mo wcwpcmumwopcs mo ucouxo .omsomm cum: nwcm:0w:mm quo mo ucsoem .omsoam xc mac upcmumnopcz mo acouxo ”omsomm cud: m=0wum~os mo m:0wuaoosom .uconcommos mo :ofiumozpo .oomp .oEoocw xfiweom .owmcnpme houmc no umcww .xuflowcmcws .cowm nwcou .xom .:o«uwm:ooo .oEoc um coupcwco mo pocesc .owm ”mowumfipouomnmco cocompom .pccco mm :OMwMHos .cofiumospo m.uocumm .xopcw oEoc coxouc .czou mo oNfim .moocmumESOHMO xcpmo ”ommwspme cow: :owuomm -mfluem uoflpoum Op pom: moccmfism> wev.fi u z commas muompoz w ompo>=oo .HcocmEmu mmcdpcflm Hmmflocdpd moccowpm> peopcomoch stum .cozcwu:OUun.m oficmb 80 .oEoc um xnoz ocz :oeoz pcm xma nom.xho3 ocz coeoz :oosuoc muss saasao co maesuaoom cm moocopommwp o: osoz ouoch .coe pom :mcu :oEo: pom coca mo Houowp noun powcouum a ma omwc Afiweom .sooa mo uouowposa umoc ocu ma can :oe new :oeoz pom cecuommmwumm mo «noose umoomonw mpcofix some: cameos ma moa_ scasaa .mo>«3 wcflxmumioeoc ou ucmuuomefi opoe ma chmcoflumcoa confine: .coE Op coco :oEo: powpumecs use compass cuoc co osoe memos xmemm .coEoz can :oe mofisume magnum Oceocooo -OMOOm 30c you xcwwHSOMH upon .:0wuomeHumm omw_ cud: woumcophoo xccwwc omoe mm: moss scmsac nos: concomOmsuam .momsomm mo monoum cuoz .owwomoha Hm:0wummaooo .cofiu -mospo .oEoocw .Eooumo mcom .cosucoo mo mSOOH .oomh .con .omflfi Acweom mo :oflum3_m>m .Auw>wuom oHSmfloH Lo moMpzum .xpoz can :oflu -mazooo .omflc xcwemm pen oEo: mocmsoo owfiznpcmcmsc mmm u z Amsmac :omxuan poxofimsocs .coeo: powuume on .poonmEo ono: mcomuom o>ocm ocu mo Hum coEoz coaches Nu :oe coaches soc :oEoz ofiwcfim mHH cos oawcmm so u z ”sumac acaceaz-oa>aa= mwcwpcwm Hmawocmua 101. I'll 1:111 v'vllll 11 mo~cmmwm> unopcomopcc sezm .eoscaucou--.~ oases 81 .wcflcuosom wcwcmficmeooom .csm .Ausmoc .hocuoc Eocm Eoeoouw .o>o~ .wcwmsoc .oEoocfi xcwemm .mfiom .omwfi xfifiemw ”:oEoz pom coon unoco mwcwfioom ou mcoMuzcmsucoo acmowmwcwwm .mcomxe .xuwusoom cmfiocmCMM .mcficu -oEom wcmcmwcn50oom .cu~moc HmCOmhoa .cofluoommm mam o>o~ .coupcwco .omMc xcdeom ”soc Ou cwwc Eouw monopno mcuoocoo omMc mo oocmuponEH .:ow»mo:po owoccoo .vownume .msumum oflEocooo -OMOOm cc cwfic omocu pom umocmwc coon .poumnmmom no .vozopwz .poouo>wp ouo: oc: mucopcommou how umozoc coca .xuwusoom paw .mucmococ owcfihm .xma .xwo: .mecofihm ass: .ssssao ass: aces macaEO ”opoz oocmcum> ucmofimMc Imam pouscwuucoo coacz mEoo_ .mEoufi occampm -oeop .wcwocnccoz ou ucmu -AOQEM oc Op ucwsocu moflcmeom mo acoecouw>co woo: ocu mo when moouomm mo xquum> m use wcagoo_u .ocas s_aeao co muoumowpcw .mo:~m> ucmuuomem .mwhouflno acmmo xc roam uscm>o como oucc mo mcfioeop xam finesse sagas: w mzoue=< whocquOO op amfisow nzox can so» mwcwcu ocu unocm coow so» on so: .xuoz can oszmwoc usocm mwcflcoom wCMHDmmoE meow“ Hanumoocom :oouuwcu new mEouw Owcmmpwosop co>om mhwmm omfizupcmcmsc 5mm u z ”mamas asuocm w Ncocsm .wmucom Am~.c u z finesse maaom w caaecaco .coecos mwcflvcflm Hmmflocficd moccmwnw> ucopcomopcm ‘11.! ’l" sszm .emacsucou--.~ scams 82 .mcsox ocu Op acmupome xccm -fioommo ouoz copsco wccpcouum ecu .mo>damcop cam mucowum cud: uomucoo wcfl>mc .powhume mcwoc .msocuo cum: mem>wc .wcw>w~ mo ~o>oc new msumum cocoflumnsooo .cofiu noospo .oeoocfl Ac poefiwcmxo mm: cowuommmwumm omMc mo oocmwum> ocu mo ofiuufic .oocopflmou can wcwocuccoz cmoflmxcm .c0n .mowu couches mo m:0wumswmflp Ac pocwm_axo mm: :0wuommmwumm owwfi cw oocmflum> ocu mo anchonmz .xucpsoom «wwocmcwm .o>o~ .uocuoc Eouw Eonooum .czw .wcficuoEOm wewcmwcmeooom .mofiuc>fiuom oecu opmmm .oEoocfl scaeac .QOHH saaaac .ocmm ”cos pom coon unocm mwcflcoom ow muouscwuucoo acmowmflcwwm .mo_0p can momSOmum mo mcodumzcmflp cm:Omuoa pew .coHuwmwofiu than cocoom .msumum Acflemm mmv.~ u z .msumum ofieocooouOMOOW Annmcv :mcmcmpooz w wcwcoxfiwz A.ucoov esoo~m w NHOcSm .wmueom mwccccfim concocfipd mo_cmflum> ucopcomoecc xcsuw III-|I--.-!I- IIIIIJ .emscaucoo--.~ «Hams 83 .omflz mo msuwum ucoexo~QEo ou woumcou z~o>fiumwoc "woumum ucoexocmEo :30 can xcwemm ocu cw muocumouommz mo pocesc .owmwunme cam xcoz .mmwcm -pcofium .ohsmwofi cow: mcofiuomm umprm ow poumaoh xHo>MoMmom mm omHH Afiflewm mpcmcmsc mcoe< .ucoecmficmeooom can mmoco>fio nomnoum pew xusmoc .wcfi>w~ mo pumpcmum “cos poxocQEo com sewuomwmfiumm omfi_ xfifiemw mo mpOHOMpohn umom .coeoz poxocmEo pom ucoe -cmwcmEooom meoeoz poxOHQEocs pom :oMOOOMmMumm owwc xfifleom mo mpouofipoum ucmofimficmfim ouoz cow was acoecmflcaeooo< .coE pozocmeo pom Eooumo m_om .cos can :oEoz poxocmeo com coduoommcumm xcfiewm mo nouowpopm ucmoflmw: -wwm m was copucoo mo msooc .mowufi>~uom cowoom .xhoumwc confines .owmumoum cacoHummsooo .ucoexo~a5o .owm ”mowumwpouomumco cm:0mwom .pHocowzoc ca pocesc .oocopwwou mo onm «omsomm mo =0wumospo .omsomm mo ucosxocaso .cocpcfico mo cocesc ”pouomcmco xmwsmm .:mesco:fi paw oocmuaooom .czm .pocuoc Eopm .Eopoopm pew oocopconopCM .ucoEcmflanooom .xpommm .mmo: no>wuompuum new xuamoc .wcfl>wc mo vhmpcmum Newpouwuo o3~m> .coflu -mosco pew .oeoocw .owflumowm HmcofiquSOOO .owmum ocoxo omfifi .Eooumo-w_om .ooos .copucoo mo msooH "moficmwsm> amouxopcou Amecmnmsc mme .mo>a: ammo «as u z ”messy accuse: mocmsoo owwzuvcmcmsc mmm u z mmsmav comxoan mwcfiecsa Haemocspa moccmfipm> acopcomopcfi -ll Ill"- 91.1! 1311.1. Acoum .cofluommmcumm omMc xcfieom pew confine: mo mxo>usm mo Ahmseamuu.m oacmh 84 .owmum Loomc ou cocoa Eonm commopocfi ommom ocoxo omfiq mm Acucchm pommococw so mmo po~o>o~ nocuwo pcm cocoa ou Acumo Eoum commopocw omem ofioxo omMH mm Hcom ommwpume pcm .omsoam .coupcwco .ocAE saaeac can: consume -mwumm pom mouoom :moz ”co: .Hoo. v.m ucaoacacmflm who: coca ecu moss inseam .coupcfico cog: :ofiuomwmfiumm can :wmwm cohpmfico wcw>mc mo mw=w~oom :oozuoc mcofiumfiophou .mo>fi3 coca sewuommmcumm coumonw commoemxo mpcmcmsz .SOOa co moss saasac accuse coco coppcflco ufiocu coax :0mu -ommmwumm woumouw commohmxo mo>w3 cam mpcwcmsc ouoz .pcmcmsc mo maumum «coexocaeo .cooa new ousmfloc .owmfihums .mmflcm lecoflpm cow: m:o«uowmmfiumm ou coumfiou xHo>qumom ma :0wuomm -mwumm omwc xcfiewm mo>wz wcoe< .mcmox wccuaoc upcfico oumfi .mhmox wcfismoc -pcflco ocppws .mpmo» wcfihmoc -e_a;u sagas Nassau amuse mo cupwc um ucopcoamoh mo ow< .mwcficoom couconmm .coupcwco mo cones: .oEoocfi xfifieww .oowh .ow< .cmmwm :ouvcfico mew>mc Ozocm mwcflcoom Hmocoumm mo camcosum .cfimwm coupcwco wcw>mc usocm mwcflgoom coucosmd .owmwpums .moocmcwm .mmficm upcoflum .oocopflmon .opzmfiofi .xcoz ":OMOONMmHumm mo mcwmeoa mocmsoo omfizupcmcmzc «mg n z cmkmcv mmacos mocmsoo owflzuvcmcmsc wnH u z ”mamas soa~> A.u:ooU Mcwocmz momcmwum> ucopcomoch xpsum .eoacwocou--.m manna 85 .coumc Op Hmpoe Eoum pommococfl owwum mm mmo po~o>o~ no awmwm omou cocuco pew canoe Ow Acumo Eonm commopocfl ommum ocoxo omc_ mm omou owmwpnme can .omsoam .conpcflco .co«uommmwumm omc_ xcwsmm com ouoom :moz ”:oEoz A.u=oov mounoh mwcfipccu Hmmfiocfind moccmdpm> «concomon:~ xv3um .eoscaucou--.m magma 86 There have been few studies using family life satisfaction as the dependent variable. The existing studies have used different independent variables. Jackson (1979) and Mancini (1978) both found fun/leisure time as significant predictors of family life Satisfaction. The studies did indicate that friendships, marriage, leisure, locus of control, work, employment status, standard of living, and accomplish- ments were all identified as significantly correlated to family life satisfaction. Summary Examination of the literature on marital quality, healthy family systems, quality of life, and family life satisfaction includes a diversity of independent variables and theoretical orientations. However, there are certain parallels in the findings which appear to fit within the F03 and Foa (1974) theoretical model: 1. Love Strong affectional bonds, affective communication, fun, friend- ship, emotional support, altruism, warmth, caring, enjoyment of company. 2. Love-services Satisfaction with or high levels of interest in the sexual relationship. 3. Status Respect, esteem, competence in role behavior, accomplishments. 4. Services Shared responsibilities, cooperation, working together, work. 5. Information Communication effectiveness, communication openness, frequency. Competence in, satisfaction with, joint, effective problem solving and/or decision making. This implies a balance of power, an equality in relationships. 6. Money Amount of income, financial security, agreement on finances. 87 7. Goods Standard of living. 8. Environmental conditions influencing resource exchange Companionship, shared leisure time. The Foa and Foa theoretical model does appear to provide a classification of the events and conditions which make life pleasant and worthy, which offers parsimony, simplicity and is specific enough to pinpoint essential differences among peOple (Foa G Foa, 1973). The theory states that resource classes are ordered. Love is the most preferred resource of proaction and reaction. It should then be possible to predict marital and family life satisfaction from the structure of the resource classes. CHAPTER III METHODOLOGY Data used in this study were collected in Oakland County, Michigan during November-December 1977 and January-February 1978 as a part of the Quality of Life Research Project by the Departments of Human Environment and Design and Family and Child Sciences at Michigan State University. The project was funded by Michigan and Minnesota Agricultural Experiment Stations.1 "Family" was the survey unit and was defined as a husband and wife living together in the same house- hold having at least one child between the ages of five and eighteen years. Information was obtained from self-administered questionnaires completed by husbands and by wives. This chapter describes the instrument development and pretest- ing procedures; sampling, data collection and data analyses procedures; describes the study sample; and includes descriptions of variables including operational definitions. 1Michigan Agricultural Experiment Station Project numbers: 1249 "Clothing Use and Quality of Life in Rural and Urban Communities," Dr. Ann Slocum, Director; 3151 "Families in Evolving Rural Communities," Dr. Margaret Bubolz, Director. Minnesota Agricultural Experiment Station Project number: 53-086 "Clothing Use and Quality of Life in Rural and Urban Communities," Dr. Joanne B. Eicher and Dr. Gloria Williams, Directors. 88 89 SamplinggProcedures A nationally known marketing research firm was employed to draw the sample and distribute and collect questionnaires and consent forms. The sampling department of the firm drew a two-stage system- atic sample with clusters and probability proportionate to size. (Larger census tracts had a greater chance of being selected.) Stage one of the sampling procedure involved the selection of census tracts and blocks identified as sampling points. Stage two of the procedure was the random selection of a household at each sampling point to be the first designated interview. Selection of census tracts was accomplished by dividing Oakland County into geographic areas based on the distinctions of rural-urban and racial compositions of the population. The research design required both black and white respondents and urban, suburban and rural resi- dents. An additional criterion for selection of census tracts was the specification of a 1970 median income of at least $12,000 to insure a higher probability of obtaining respondents having at least a high school education. The education requirement was necessary due to the complexity of the questionnaire. Details of the process for selection of census tracts have been described by Slocum (1979). After census tracts were selected, a two-stage systematic random sampling procedure with clustering was implemented. First the sampling points were selected from a list of numbers of occupied dwelling units with probability proportionate to household count. In the second stage a randomly selected household at each selected sample point was designated for the first interview. A specific walk 9O pattern was used by interviewers to designate the additional three households in the sampling point cluster. Interviewers were instructed to make an original call plus three additional callbacks to designated households in order to estab- lish contact with the family. If no contact was made or the household did not meet eligibility requirements, substitution was made of the house on the right, and then to the house on the left. The instruc- tions for interviewers are reported in Appendix C. The resulting total sample consisted of the following: The Rural Sample. The rural sample consisted of five townships which had a 1970 median income of $12,000. (Fifty-nine families were inter- viewed.) The Urban-Suburban Sample. The urban-suburban sample included the balance of Oakland County. a. The black sample included Pontiac City and Royal Oak Township. The 1970 median income for these areas was $6,000. The criterion of a $12,000 median income was not possible to maintain. (Fifty-four families were interviewed.) b. The balance of sampling points were chosen using only census tracts with 1970 median income of $12,000. (One-hundred twenty-four families were interviewed.) Data Collection Procedures Trained interviewers hired by the research agency completed the data collection in a four-month period of time. The interviewers had two briefing sessions conducted by the field work supervisor of the agency and members of the Quality of Life Project team before beginning to screen households. The information reviewed with inter- viewers is included in Appendix C. After determining eligibility of the household, the inter- viewers were instructed to obtain written informed consent from one 91 or both spouses at the time of placement. If only one spouse was at home, his/her written consent was obtained and the other spouse was able to sign before questionnaires were returned. Interviewers explained the questionnaire and left it with the family for completion. Several days later the interviewers were to telephone the family to arrange for pickup of the questionnaire. At the time of pickup the interviewer was to check for completion of all items. Families who returned completed questionnaires from both hus- band and wife received $10.00 and a summary of the findings. Families were assured by interviewers and in writing from Michigan State Uni- versity that their responses would be anonymous and their privacy would be protected. The possibility of collaboration between husband and wife was considered and independence of response was encouraged. Questionnaires were distributed inside envelopes to assist respondents in maintaining privacy of responses. Each section of the questionnaire was later examined for evidence of collaboration and coded accordingly. During the data collection procedure interviewers were having difficulty placing questionnaires and modifications were made in the screening procedures. The research firm's director of sampling felt the modifications were necessary due to the number of filter require- ments for eligibility and the probability calculations which indi- cated the necessity of covering nearly two-hundred homes to obtain fOur interviews. The modifications did place some limits on general- izability of the results beyond the present sample (Appendix B). 92 Description of the Study Sample The final sample for the Quality of Life Project consisted of 237 husband-wife couples and seven single parents. The sample for the present study did not include single parents or the thirteen fam- ilies with codes indicating high collusion on the family life sections of the questionnaire. The decision to eliminate families for reasons of collaboration was made by agreement of three investigators. The final study sample consisted of 448 persons or 224 husband-wife couples. These married couples ranged in age from 25 to 65 years with an average age of 37.5 for women and 40.2 for men (Table 4). Table 5 indicates the number of years married for families in the sample. Twenty-three couples had been married more than twenty-five years; the median was 15.5 years. Families were similar to the "average American family" with an average of 2.5 children living at home (Table 6). Table 7 reports further data on ages of children. One family had a youngest child under the age of five years and an oldest child 23 years or older; however, the age spread of children in most families was less than eighteen years. The years of education for women ranged from six to nineteen, the average was 12.8 years. Table 8 reports five men with twenty- two years of education, the mean was 13.5 years. Tables 9 and 10 describe the income distribution of the sample. Per capita income was calculated by dividing the total family income by the number of people who were dependent upon that income. The mean income per person for the year of 1977 for this sample was $6,055 93 Table 4.--Age Distribution of the Sample. Women Men Age in Years N % N % 25-30 41 18.2 28 12.6 31-35 58 25.9 44 19.6 36-40 53 23.8 54 24.2 41-45 29 12.9 38 17.0 46-50 27 12.1 31 13.9 51-55 11 4.9 15 6.6 56-65 2 .9 13 5.7 Missing Data 3 1.3 l .4 Total 224 100.0 224 100 0 .M 37.5 40 2 Mdn 36.4 38 8 Mo 31 33 94 Table 5.--Number of Years Married of Families in Sample as Reported by Wives. Years Married Number of Families % 0-5 16 7.1 6-10 39 17.5 11-15 52 23.3 16-20 48 21.5 21-25 36 16.1 26-30 16 7.2 31-35 5 2.0 36-40 2 .8 Missing Data 10 4.5 Total 224 100 0 ‘M 15.9 B12 155 Mo 10 SD 7.6 95 Table 6.--Number of Children Per Family for Sample. Number of Children Number of Families % 1 30 13.3 2 81 36.2 3 60 26.8 4 34 15.2 5 11 4.9 6 5 2 2 7 1 4 8 2 9 Total 224 100 O __ 2.7 Mo 2 0 96 Table 7.--Age Range of Children Living in Household. Age of Youngest Child Age of Oldest Child Total 1-5 Years 6-12 Years 13-18 Years 1-5 years 10 0 0 10 (4.5%) (4.5%) 6-12 years 49 34 0 83 (21.9%) (15.2%) (37.1%) 13-18 years 18 47 25 90 (8.0%) (21.0%) (11.2%) (40.2%) 19-22 years 3 12 19 34 (1.3%) (5.4%) (8.5%) (15.2%) 23 years and over 1 2 4 7 (.4%) (.9%) (1.8%) (3.1%) Total 81 95 48 224 (36.2%) (42.4%) (21.4%) (100.0%) and the median was slightly lower. sample for 1977 was $27,034. The median family income for this Respondents in the study sample are primarily white, middle aged, high middle income, high school educated couples who have been married about fifteen years and have 2.5 children living at home. Instrument Development Procedures An extensive questionnaire was developed to meet the needs of an interdisciplinary research team. Some items were developed by the Quality of Life Project members and some were used by permission from other researchers. Several sections of the questionnaire were not used in the present study. are reported in Appendix A. The sections relevant to this study 97 Table 8.--Education Distribution of the Sample. Education in Years Women Men (Midpoint of Category) N % N % 6 l 4 6 2 7 8 4 .8 8 3 6 10 28 12.5 30 13.4 12 113 50.4 60 26.9 14 38 17.0 54 24.1 16 17 7.6 21 9.4 17 14 6.3 18 8.0 18 6 2.7 18 8.0 19 2 .9 2 .9 22 S 2.2 Missing Data 1 .4 2 .9 Total 224 100.0 224 100 L M. 12.8 13 S 51913 122 136 M9 12.0 12 0 §D_ 2.3 3 2 98 Table 9.--Family Income Distribution of the Sample. Midpoint of Family Income Category in Dollars for 1977 N % 1,500 1 . .4 5,500 3 1.3 6,500 2 .9 7,500 4 1.8 9,000 7 3.1 11,000 4 1.8 13,500 9 4.0 17,500 35 15.6 22,500 45 20.1 27,500 43 19.3 32,500 28 12.6 42,500 32 14.3 62,500 (or more) 9 .4 Missing Data 1 .4 Total 224 100 0 14. 26,982 M92 27,034 Mo 22,500 99 Table 10.--Family Per Capita Income Distribution of Sample. 1977 Family Per Capita Income Number of Families % $2,000 and under 17 7.3 $2,100 - $4,900 80 35.6 $5,000 - $9,900 99 44.5 $10,000 - $20,000 24 10.8 $20,000 and over 3 1.3 Missing data 1 .4 Total 224 100 O __ $6,055 Mg 5,625 .USE 5,500 There were several steps in the process of developing the instrument: (1) examination of the literature on marital, family life, and overall life satisfaction to find the significant corre- lates; (2) study of the Foa and Foa resource exchange theory for words which would accurately represent resource classes; (3) asking people to indicate things about family life which give them the most satis- faction and dissatisfaction; (4) summarizing information obtained for the creation of a preliminary set of questions; (5) discussions with team members of the salient dimensions and the wording of questions; (6) pretesting at two stages of development followed by (7) modifica- tion of items for the final questionnaire. 100 Examination of Literature The process of literature review involved examination of a variety of sources which included magazines, newspapers, television interviews as well as the professional journals reporting research. A particularly relevant and thought-provoking newspaper article (Lee, 1977) reporting the questions asked by Family Service Association to measure quality of family life was the inspiration for the items: "how comfortable it feels to be at home," "how openly and honestly you can express feelings," the "way money is used" and the frequencies of showing affection and respect. The above items also represent resource classes in the Foa and Foa theory of services, information, money, love and status. The literature search involved examination of different theor- . etical perspectives and methodological approaches to the study of marital and family satisfaction, stability, happiness, health, quality, and well-being. An effort was made to represent as many dimensions of family life as possible and to include items which would facili- tate testing the Foa and Foa resource exchange theory. Some of the influential sources of information in the litera- ture review were: (Alexander, 1973; Burr, l 73; Locke, 1958; Kantor & Lehr, 1975; Kimmel G Van der Veen, 1974; Miller, 1976; Moos G Moos, 1970; Orden 8 Bradburn, 1968; Rollins 8 Cannon, 1974; Satir, 1972; Spanier, 1976; Strumpel, 1972, 1976; Hicks 8 Platt, 1971). Representation of Resource Classes The words used by Foa and Foa (1973, 1974) to represent love included: love, affection, warmth, tenderness, liking, enjoyment, 101 friendship, humor. Status messages use words of esteem, competence, respect, and approval. Services are conveyed in the labor of one person for another and are associated with the words of comfort, help, assistance, and work. Information is offered in the form of advice, instruction, opinion, enlightenment and with the process of verbal exchange between persons. The words associated with goods and money are not as difficult to describe or at least are more obvious. Items were created for the questionnaire to represent each resource class and also shared time. Respondents were given the oppor- tunity to evaluate resources received and to estimate the frequency of resources received from spouse. The items were standardized to have respondent as Object, spouse as actor, the exchanges were giving or increasing the amount of the resource available to object rather than taking or decreasing the resource available to the respondent. The summaries of items representing each resource class are reported in Table 13, pages 113-116 and Table 14, pages 118-120. Questioning About Family Life Satisfactions Three questions were developed in order to acquire opinions from persons about dimensions of family life which contribute to satisfaction: 1. What are the things about your family life which give you the most satisfaction? 2. It is also normal to be dissatisfied with different aspects of family life. What are the things about your family life that give you the most dissatisfaction? 3. The most important things about family life are: Responses to the questions were received from a group of 24 adult women between the ages of 28-48 years with a college education, a 102 group of 20 undergraduate students at Michigan State University, and a group of persons known by the research team who responded informally in the course of social interaction. A summary of the responses to the above questions indicated a primary source of satisfaction was labeled by people as "together- ness," "unity," or "cohesion." A secondary source of satisfaction indicated by number of responses was for "communication" and "sharing"; "shared activities" (vacations were often mentioned); and "love" and "enjoyment.” The third level of satisfaction sources was associated with the words "cooperation," ”helping” and "support." Summary of the responses to the question about dissatisfactions indicated problems of time management such as time demands and meshing of time schedules for family members. A secondary source of dissatis- faction was problems of communication. The third level of concerns mentioned quarreling of siblings or spouses over money issues, and the division of household work. Responses to the most important things about family life indi- cated in order of priority: love, sharing, togetherness, caring, respect for individual differences, and personal growth. Preliminary Questions The preliminary questions were created by combining informa- ition from the literature, the Foa theoretical model, and the responses of persons concerning sources of family life satisfaction. Two investigators COOperated on this effort. An attempt was made to include the salient dimensions of family life, the items to test the 103 theoretical model, and to state all items with simplicity and clarity. An informal testing of the questions by several interested persons resulted in modifications of the question format and clarifications in question wording. The resulting questions (after modification) were given to a group of five husband-wife couples who were asked to give comments and ask questions about the items. Some items were then modified before pretesting. Pretesting The questionnaire was pretested in one rural area and two sub- urban areas in Ingham County, Michigan and one suburban area in Oak- land County, Michigan in October 1977. Eighteen husband-wife couples who were married, living together, and had at least one school—age child completed the questionnaires. Minor modifications to the instru- ment resulted from the pretest. No changes were made in the questions related to evaluation of family life. Examination of the frequency distributions for individual items did not reveal serious problems. Origin of Questionnaire Items Andrews and Withey (1976) granted permission to use the Delighted-Terrible Scale (Figure 4) which was used to assess perceived overall quality of life, overall quality of family life, and evalu- ation of marriage in addition to the following items in the family life section: 6.1a Your husband or wife 6.1b Your children? 6.1g Your marriage? 104 .mm .a .aoao x_:s .xooconou .cmama .o>flcon< oocofiow cocoom cosmomom cocoom how ousufiumca ”Hz .cocu< ::oo .msonwwumo>:H oamflocfipd .xocuflz .m cocmoum pcm mzocpc< .2 cameo ”oousom .mcofiumscm>m o>fiuoomw< mo ucoammomm< pom pom: xocuwz paw mzoppc< xc monoco>oc ocmom occmccoh cowcwflcoc oce--.e .wdm oE Op sauna uo: moon 0 pa unocm acmsocu Oo>oz m powmmflummmflp so: poMmeumm cocufloclncmpusoz < avowmmwummmwp paw pocmmfiumm Accmzco poflwmflumm usocmv podmmfiummmflp poocmflfioa commocd acumoz pox“: zcumoz xmmmcc: occfippok n o m e m N H “~QOM H 105 The following items were used by Andrews and Withey but the wording was changed for the present questionnaire to read as follows: 6.1e 6.2e 6.3f The amount of respect you receive? The things you do together? Your sexual relationship? Several of the items have been used previously in a long his- tory of marital evaluation research: How often do you and your mate: 7.1a 7.1b 7.1c 7.1d 7.1e 7.1f 7.1g 7.1h 7.1i 7.1j 7.1k Spend time together-~just the two of you? Spend an hour or more just talking? Discuss personal feelings? Work together on a project? Take a drive or a walk? Eat at a restaurant? Entertain friends at home? Visit friends? Go to a movie or other entertainment? Attend a sports event? Attend a party? The above items were used by Orden and Bradburn (1968) as indicators of marriage companionship and marriage sociability and later by Miller (1976) who used five of the above items in an eight-item companionship scale. Miller asked respondents how often in the past month the respondents had engaged in companionate activities, which would pro- duce a different kind of evaluation than the scale used for the present study which involved a longer period of time. The following items were created by the Quality of Life Research Project team: How would you feel about your family life if you considered only: 6.1c 6.1d 6.1f 6.2a 6.2b 6.2e 6.2d 6.3a The love and affection you experience? The closeness and sense of belonging you feel? How comfortable it feels to be at home? The way money is used? The amount of money for personal use? The material goods it enables you to own? The way decisions are made? The mutual helpfulness of family members? 106 way household work is divided/accomplished? openly and honestly you can express feelings? kind of communication you have? amount of time the family spends together? time you spend with your children? time you spend with your husband or wife? friends it enables you to enjoy? How often does your mate: 7.2a 7.2b 7.2c 7.2d 7.2e 7.2f 7.2g 7.2h 7.21 7.2j 7.2k 7.21 7.2m 7.2n 7.20 7.2p Make Tell Let Tell Let Enjo Give you feel like an important person? or show you that he/she admires and respects you? you know he/she has confidence in your abilities? or show you his/her love? you know he/she enjoys your company? y a laugh or a joke with you? you a hug or a kiss? Do an errand for you? Make Do 5 Give Give Give Give Help SUPP himself/herself available to do work for you? omething to save you energy or make you comfortable? you some new information? you his/her opinion? you something you need or want? you money for personal use? you solve a problem or make a decision? ort you with discipline and guidance of children? Description of Variables Dependent Variables The dependent variables in this study are subjective indicators of perceived overall life quality (Life 3), overall quality of family life (Famlif 3), and quality of marriage (Table 11). Respondents were asked for their evaluations of life-as-a-whole, family life-as-a-whole, and marriage measured on a seven-point scale (Figure 4). Primary attention was given to evaluation of marriage as the major dependent variable. Evaluation of life-as-a-whole. The question "How do you feel about your life as a whole?" was asked as the first question and again after respondents evaluated various life quality domains such as work, 107 scocpcfico nsox new .owmfiuume uzox .omfiz no pcmcmsc nsoxuquM~ Acwsom poo» unocm coom :ox Op 30: A~.m can mm.~v momcommos ozu mo :mos ocu u m mflcewm wofioczumummuomfl~ unox usocm doom no» op so: n~.m use ~.Hv momcommoh ozu mo :moe ocu u m omMc owns x_asmc co COM Hm3~m>o ®>w uuomw< oHoczumnmmiomflc mo :ofiumscm>o o>wuoomm< .oefiu mo poflpom m Lo>o omflc xaflemm cud: mmocfimmmc::\:ofiuomm -mfiummmfip no mmocflmmmc \cofluommeumm po>woo Icon cmmpo>o m.:Ompom < .Amsmc .coocag w muuocuomoc .HcocuOsz memo co poduom m so>o mpoo: mo :OMOOOMmHOmm po>fioopom Hfimuo>o m.~m3pfi>wpcfi c< .Amamc .mxcsom w xoxcmov mmocwmmmccs no mmocwmomc no .omflc cud: :0wuommmfiommmflp no :OMO -omefiumm .wcwocn~co3 mo omcom m.:Omuoa < owwc inflame co somcmso .N muss co suscaso .2 moccwum> «concomon acoesnomcm hOHmoHch =0wuwcflmoo Hmowpouooce oaaasca> .moccoflpm> ucomcoaoo mo :oflumflhomoonu.ca oacmh 108 .ON mwmocuomxc pom pom: moccmficm>m wowmwuume usox Afico vosopfim ucoo so» m“ omflc xfifisow .oeflu mo nachos m Ho>o mflcm neoflumcoh ocp mo wcfluuom cou:u_:o paw Hm:Ompom .oweocooo .Hmfioom ocu mam aflcmcofluocou confines :3o use» unocm Hoom owmauume mo ocu mo :oflumzco>o so» pazoz 3o: w~.o coflumocm>o o>fluoomm< o>flumuw~msc m.:omsom < mommwppwe mo xpficmso .m acoESHHmcm HOumOMch :ofluwcwmoa amofluonooch oficmwhm> .emscaacou--.- scams 109 neighborhood, and family life (item 9.2). Life 3 is the simple average of the responses to the same question asked at two different points in time. Andrews and Withey (1976) found this measure of global well- being provided a more reliable and valid indicator than the single question and was one of their best indicators of global evaluation of life-as-a-whole (POQL). Evaluation of family life-as-a—whole. The question "How do you feel about your own family life-~your husband or wife, your marri- age, and your children, if any?" was asked as the third question and again after respondents had answered specific questions about family life. Famlif 3 (POQFL) was the simple average of responses to the same question asked at two points in time. There was a separation of approximately thirty minutes response time. The measure was an indicator of global evaluation of family well-being. Evaluation Of marriage. Respondents were asked: "How would you feel about your own family life if you considered only your marriage?". It was not possible to calculate the short-term test— retest reliability coefficient for marriage evaluation since the item was asked only once. However, the summary of reliability analysis for evaluation variables includes marriage evaluation and is reported in Appendix E, Table E-7. Multiple regression is a linear additive model and requires dependent variables to be normally distributed and at interval level of measurement. 110 The seven-point Delighted-Terrible Scale has been treated by Andrews and Withey (1976) as interval data. The decision was made by comparing the scale with two other scales: The findings suggest that respondents tend to use all of the most promising scales in approximately the same way, that the meaning they attach to scale categories seems not to be much influenced by what is being evaluated, that most of the categories of the D-T Scale seem to be separated by roughly one-step intervals on the other comparison scales, except for the most positive cate- gories where the separation may be less, there is a reasonably close correspondence between the seven categories of the D-T Scale and the Faces Scale (p. 227). Andrews and Withey also state that they had remarkably con- sistent results in examining several hundred sets of concern measures and not once encountered a "marked deviation from the suggestion that a linear additive model is the right one” (p. 121). The use of mul- tiple regression is appropriate for data measured on the Delighted- Terrible Scale. A variable is normally distributed if the mean, median and mode have the same value. Skewness and kurtosis for a normal distribu- tion will be zero. Examination of Table 12 indicates mean and median values of all dependent variables. There is a lack of symmetry since there is a range in skewness from -.54 for women's evaluation of family life to -l.70 for men's evaluation of marriage. The median values are larger than the mean values which is an additional indicator of the negatively skewed distributions. A larger proportion of men responded with highly positive evaluations of marriage than would be true of a normal distribution. Andrews and Withey (1976) had similar results in four national studies with only 4 percent of respondents who felt ”mostly dissatis- fied," "unhappy" or "terrible" about their marriages. This 111 m~.~ mo.~ -.m No.- me.m mc.m mcmoousx mm.u wm.: w~.H- em.- on.cu mv.c- mmoczoxm xuuoeexm co. co. co. co. we. mo. sonno ppmpcmum mm. ow. mm. mm. m~.~ cm.c zoMumw>ov pampcmum on. em. mm. mm. cm.~ mw.~ oocmfipm> suscanaaca> mm.m em.m mm.m ow.m N~.o mm.m :mwpo: om.m mm.m wn.m wm.m mm.m Hm.m :moz xocopcoh Monacou so: :oeoz so: :oEoz co: cosoz moss co suaaazo HHoHo>o po>woOHod compasaa>m woes inseam :oHum:~m>m owmflhpmz mofipmdumum o>Humfipomoo I. .moficmwnm> ucopcomoo ocu pom mofluwflumum o>Humflhomoonu.NH occmh 112 corresponds to 6% of the respondents in the present study who evalu- ated their marriages negatively. Kurtosis measures the peakedness and flatness of the distribu- tion defined by the distribution of cases. Negative values mean the curve is flatter and wider than the normal distribution and indicates there is more variance in the study sample responses than would be true in a normal distribution. Positive values indicate the distribu- tion is more peaked (narrow) than would be true of a normal distribution. The kurtosis of the marriage evaluation variables have the most highly positive values. The F test is a robust statistic which resists violations of assumptions particularly when the sample size is large. If a test statistic is robust, the actual probability of a Type I error is in agreement with nominal probability (as read from the F-table). Although the F-test is robust, Lindquist (1953) cautions that the F-distribution is slightly affected if measures of the criterion are very flat or very peaked. In these cases the probabilities read from the F-table are too small to represent the true risk of a Type I error and allowances should be made for this in the interpretation of results. In such cases . . . when risk read from the F table is 5%, the true risk may be as large as 8% and when the risk from the F-table is 1% the actual level of significance may be approximately 2% (p. 81). In this study it will be necessary to interpret with caution the results of hypothesis testing on the prediction of evaluation of marriage. 113 The F-test is not robust to violations of the assumption of independence. Separate analyses of husbands and wives were conducted in order to avoid violation of the assumption of independence. Independent Variables The independent variables in this study are indicators of inter- personal resources received (needs met) in the family environment. There are two basic types of independent variables: (1) evaluation of resources received as measured on the Delighted-Terrible Scale and (2) perceptions of the frequency of resources received from spouse as measured on a behavior-per-unit-of—time scale. The resources received are love, status, services, information, goods and money. Evaluation of and frequency of shared time was also an independent variable since it is a necessary condition for the transfer of inter- personal resources. Emphasis was placed on the more particularistic resources of love, status, services and information which the theory states are the best predictors of satisfaction. Evaluation variables. Evaluation variables are summarized in Table 13 which describes the variable, the theoretical definition, the indicator, and the way the question was asked in the questionnaire given to respondents. The complete set of evaluation variables can be seen in Appendix A. Only the variables used in the cluster and regression analyses for this study are included in Table 13. Frequency variables. Respondents indicated their perceptions of frequency of resources received from spouse on an eight—point scale. Values of the scale were: "never," ”about once a year," "about six 114 .o>Hooon :ox ooommoa mo oeso5m och o~.o scoom :ox wag uwcococ mo oncom ecu mmocomoco och p~.o wooco nanomxo so» coauoommm can o>oH och OH.O "xcco monopwmcoo 3oz we owed xcfleom usox oaocm Hoom 3oz pcsoz so: .oEHu mo eofipom m no>o uooamon use cofiuwc -mooon pom poo: mac mo :ofluomMmdumm po>fioo upon m.~m3pfl>wpcw :< .oefiu mo pofiuom m m ho>o :ofiuoommm pew o>oH you poo: mac mo sewuomMmHumm po>woo upon m.~o:v«>wp:fl :< .ANN .a .memc .mom w momv Eooumo no phmwou .owqumopa 30H so cmec mxo>coo umcu ucoempsn o>flum3~m>o :< .ANN .a .mnma .mom w momv anomaoo no .cuEpmz .pnmmou Oumcocu loommm mo :oMmmocmxo :< .AxnmcoMHOMQ ppomxov acoecomuum .:0wuoommm sum: can ~m>owamm mac com opwmop .oocomoum mac :H ucwflfiop .uoonco ca mo oumm~o3 ocu how oe:OAOHHOm cw maomow mumochmE coficz wcficoom mo opmum och . mwuhfiommh wouooaxo .popoo: .poucmz ocu cow: po>wooou moOH:Omop mo oocmfimc ocu mo :oflum:~m>o :< cowumscm>m woumum :0wum3~m>m o>oc .ucoecosfi>co xcflemm ocu :« fluoe mpoocv po>woooh moonsomou cocomhoapoucw mo :oflumscm>m .v moccmfium> ucopcomoch ucossnomcH acumOMch :ofipficfiwoa fimofiuouooch magmaem> .moHcmem> :ofiumsfim>m ucopcomopca mo :oflumwnomoonu.mm occmh 115 homes no pcmcmsc poo» cums pcomm nos mafia one am.o wuocuomou on so: mmcwcu och om.o hopes one mcocm -aume as: was e~.e wo>mc so» :OMOmowcseeoo mo pcfix och pm.o hmwcwcoom mmonmxo :mo so» saumococ can Aacono so: om.o wmflcmcoflooHon Hmsxom coo» mm.o hpocmflcmeooom \popfi>flv we cuos pcoc tomsoc xms och cm.o wmwoc -Eoe xfifismm mo mmo: ufiammcoc Heaps: mm.o soeoc om oc ow mcoom we oacmuuomeoo so: m~.o .oEwu mo powsom m uo>o mwcmcoflcmo -Eoo wow poo: mac mo cowoommmfiumm po>floo upon m.Hw:pw>wp:M =< .oefiu wo powhom m uo>o wcficmoa popmcm pow poo: mac mo :ofiuooMmHumm po>woo upon m.~m=pw>wpcfi c< .oEHa mo pofiuom m po>o ooceumfimmm pew whom neoo pom poo: mac mo :ofluommmwumm po>woo upon m.~m:p«>wpcw :< .oefiu oeam ocu an ounce oEmm ocu um Auc>fluom oco :« pommwco m:OmpoQ osh .mmficmcoflumcou —w:ompom though cc codqupomcm pew .moo«>HOm .uoommou .o>o~ mo nommcmcu ocu pom summmooo: :ofiump -coo coucoecocfi>co och .ANN .a .mems .aoa w mom: maumum no o>o~ mm powmwmmmco oc pcsoo umcu mpOM>mcoc omocu mo o>wmacoxo usc ucosco tucwflmco no mcofiuoauumcw .mcoficwmo .oo«>pm mm pouommo :ofiu85pomcc .ANN .a .mema .aoa w mom: xmpoco sac o>mm no wocuo ocu mo uuom neoo Hmowmxcm ommowocfl oo Hocuocm now common oco mo coca: accusufium ucoo xccmsm: :Omuom a mo mwcwwcococ no spoc ocu so possomuoa mofiuw>fluo< :ofiuwsfim>m oEHh pocmcm :oHumscm>m :odumeuowcm :ofium=~m>m moofi>nom ucoezuumcH .HOHMUHUCH :o«um:wmoo cmowuouooch ocnmaam> .eoscsucoo--.mc scams 116 .vcuoc .w .m .m ._ momocuomxc pow pom: moccmwsm> ”oooz meow: ma socoe has och mm.c wow: cw:Ompom 93oz pow oacmcfim>w socoa mo H.305. och c~.o wcso o» no» moccoco Ow mpoow cmfiuoume och om.c .oefiu mo eOMHom m co>o mcofiuwmwzcom now poo: mwc mo cofiuommmcomm po>woo upon m._m:pfi>flvcfl c< .oeflu mo pofiuom o po>o :ofiuwmwscom pom poo: mfic mo accuomwmfiumm po>fioo -eoa m.Hases>se=a =< .ANN .a .msmc .moa w some o5~m> owcmcoxo mo owe: pampcmum oEOm mmc cuccs coco» co .xocoehso .cwoo xc< .ANN .a .mhmc .mom w mom: mcmwcoume Ho muoonco .muozpocm occcwcmh cofiumscm>m socoz :owumscm>m mpoou ucoezuumcm Houmofich :0wuficwmoo cmofiuouooch occaapa> .eoseaucou--.ms osaae 117 times each year," "about once each month," "about once each week," "about 3-4 times each week," "about once each day," and "about 2-3 times each day." Primary attention was given to the frequency of receiving love, status, services and information from spouse and the frequency of shared time with spouse. The complete set of frequency indicators is reported in Appendix A. The summary of frequency vari- ables used in cluster and regression analyses is reported in Table 14. Data Analysis Procedures The Control Data Corporation 6500 model computer at Michigan State University Computer Laboratory was used to perform all analyses. Programs for statistical procedures were from Statistical Package for the Social Sciences (Nie, Hull, Jenkins, Steinbrenner, 8 Bent, 1975) and Applications Programming Group at Michigan State University (Allard, 1978). The significance level .05 was set for all statis- tical tests. Statistical Methods and Assumptions The statistical methods used for data description and analyses were: crosstabulation with gamma measure of association, Pearson product moment correlation, the agglomerative method of hierarchical complete-linkage clustering, and the forward method of multiple regression analysis. Crosstabulation Crosstabulation is a joint frequency distribution of cases according to at least two variables. It can be summarized by measures of association such as gamma which describes the degree to which the 118 wmofluwcwcm poo» cw oocopcwcoo mac oc socc so: uoc o~.h wsox muoonmou paw moncEpm oc socm\~coh c~.h ~:Omwom acounome« cm ocfic ~oom so» oxmz mm.h .m~.hnu~.h momcommou mo ocoom Hmuou och wmmfic \wac a so» o>fio mm.h woo: cogs ocom m so amass a soncm c~.n wxcmeoo nsox m>0mco oc soc: so: uoc o~.h wo>oH mac socm\ccoh p~.m nouns use: moon :oumo so: .oecu mo codcom m uo>o oocouom usoo mo :ofluwcmooop pew ~m>opomm .uooamou mo>flw oume mac cowcs socosc noun ocu mo :ofiumoo upon m.cm3pfi>wpcw c< floccofium> poumouov mucoscocm o>oc own: .osfiu mo weapon a co>o owns Eopm mcofimmosmxo o>o~ owcocexm use ouonocoo mo Aocosc nocm ocu mo COMOQoo upon m.~o:pw>wp:M c< .w:H>Hooop ma ucopcommop ocu OcoEoOhomcwon mo u:SOEm ocu mo HOumowpcw :< .ucomQSm cacoMOOEo mo mcfl>fim m .ucopcommou ocu pewsou pcoeo>oE mo socoscouw po>woouom och .oeflu mo ace: com m:o«uom mo cones: ocu no c0w>mcoc mo oumm socozcocm msumum socoscocm o>oc .omaomm Eocm po>wooou moOHSOmoH mo socoscopm coupomom acoeaupmcm powwowch :oMuwcwmoo caucuouooch wflnmeta> .moccmflcm> socoscocm acopcomopcc mo :OMHAHuomooun.v~ occmh 119 .o~.hnxm.h momcommoc mo ouoom Hmuou och scofimwoop m ocmE Ho Eoccoua m o>~Om cox aco: o~.h scoficflmo we; so» o>eu -.s scofiumenomcfl so: oEOm so» o>Ho c~.h .n~.hucm.h momcommou mo ouoom Hmooo och hoccmuuom -eoo so: ome no swuoco so» o>om ow mascooeom on n~.n woo» sow xwos op o» oficmcwm>m mcomec ocmz wN.h woo» pom pemupo cm on c~.h .o~.hnom.h momcommou mo ouoom Hmuou och moccmwum> poomonov socoscocm :ofiumEcowcm one: .oEcu mo pofieom m po>o omsoam Eonm :oflooecowcfi wcfl>wooou mo socosc nosm ocu mo :oflumoo upon m.cm:pfi>wpcw :< moccmwhm> poumoeov socosconm moofi>hom oumz .oEAO mo pofluom m co>o oume mwc Eoum oocmumwmmm mcfi>fioooh mo socosc noum ocu mo :ofioaoo -eoa m.~msea>eeca =< floccmflpm> poumosov socoacosm msumum oomz socozvoum cohumEpomcc xocoscohu mooM>Hom ucossuumcm poumoflpcc :oHuwcwmoo caucuouooch manaeca> 1.3111. 1'Ill0u1lll] .l' .eoscaucou--.sc magma 120 .mcumc .m .m .m .c momocuomxc pom pom: moccmcnm> ”ouoz .oc.himc.n momcommos mo onoom cmuou och scams \o>wpp m oxmh o~.h «poonoun m :o nocuomou xnoz pc.h wmwcficoom cacomnom mmaomwa o~.h wwcwxfimu umsn ouoE ho uaoc co pcomm c~.h woos mo osu ocu Oman hocuowow mafia ecoam ac.a floccmcum> poumouov macm:0wchEou .xocoscohm oech oumz .oefiu mo powhom m uo>o oume cufls mofiuc>cuom ouocoflcmmeoo mo socosc nouw ocu mo cofiwmoo upon m.mm3pw>wpcfl c< socosvocm osch poewcm HGOE—ahflm: H HOHNOMpcc cocoaeecoo smaspmcooce osaasta> .lll'll. .eosccocoo--.sa essay 121 values of one variable predict or vary with those of another. Gamma is an appropriate measure of association when both variables are mea- sured at least at ordinal level. Gamma is the number of concordant pairs (P) minus the number of discordant pairs (Q) divided by the number of united pairs (P + Q). A positive value indicates predomi- nance of concordant pairs and a negative value indicates predominance of discordant pairs (Nie et al., 1975, p. 228). Crosstabulation was used to describe women's and men's evaluations of family life-as-a- whole and evaluations of marriage. Correlation Analyses Correlation analyses provided the input matrices for cluster analyses, the inter-correlation information for selection of the inde— pendent variables for regression analyses, and provided an additional method for examining the theoretical model. The Pearson correlation coefficient“: is a measure of associ- ation between two continuous variables and indicates the strength and direction of the relationship. Values of.£ range from -1, a strong negative linear relationship, to a +1 representing a strong positive linear relationship. The use of Pearson correlation requires making assumptions of linearity, random sampling, bivariate normal distribution, and interval level data (Nie et al., 1975). The squared correlation coefficient describes the percentage of common variance between two 122 variables. The significance level of correlation coefficients indi- cates whether the value of_£ is significantly greater than or less than zero. Cluster Analyses Cluster analyses were used in order to determine whether the underlying structure of the data would validate expectations of the resource classes of love, status, services, and information as distant categories. The clustering procedure groups variables which are similar and thus is a data reduction and explanation procedure which is appropriate when the objective is model fitting or the generation of hypotheses (Everitt, 1970, p. 3). Anderberg (1973) indicates the only clustering technique which is appropriate for clustering variables is hierarchical clustering based on a similarity matrix (p. 210). Hierarchical clustering begins with a similarity matrix of product moment correlation coefficients among entities to be clustered. The process organizes and pictures the proximity matrix as an evolu- tionary tree called a dendrogram. The dendrogram is a two-dimensional diagram illustrating the fusions made at each successive level and provides a picture of hierarchical structure. The agglomerative method begins with all variables as separate entities and proceeds by a series of successive fusions which result in a merger of all variables. When two variables fuse, they are per- manently joined and become a building block for later mergers. Ander- berg (1973) explains the procedure for agglomerative clustering as follows: 123 1. Begin with n clusters each consisting of exactly one identity. Let the clusters be labeled with the numbers 1 through n. 2. Search the similarity matrix for the most similar pair of clusters. Let the chosen clusters be labeled p and q and let their associated similarity be Spq, p > q. 3. Reduce the number of clusters (entities) by 1 through a merger of clusters p and q. Label the product of the merger "q" and update the similarity matrix entries in order to reflect the revised similarities between cluster q and all other exist- ing clusters. Delete the row and column of S pertaining to cluster p. 4. Perform steps 2 and 3 a total of n-l times (at which point all entities will be in one cluster). At each stage record the identity of the clusters which are merged and the value of similarity between them in order to have a complete record of the results. Different agglomerative methods are implemented by varying the procedures used for defining the most similar pair at step two and for updating the revised similarity matrix at step three (p. 133). Complete-linkage clustering varies the procedure at step three in updating the similarity matrix. At each stage after clusters p and q have been merged, the data matrix is updated by examination of the correlation coefficients between the new cluster (labeled t) and all other variables. The lowest correlation coefficient is chosen. When Sij is a correlation measure: Str = min (Str, Sqr). The similar- ity Str is the similarity between the two most dissimilar entities in clusters t and r. If clusters t and r were to be merged (because they have the highest correlation), then every variable in the cluster would have a correlation of at least Str with every other variable in the cluster. The method is called complete—linkage because all variables in a cluster are linked to each other at some minimum simil- arity (Anderberg, 1973, p. 139). The objective is to form tight, homogeneous clusters. Several factors affect positively the validity of hierarchical clustering techniques: the existence of hypotheses, prior conceptions 124 or goals for the clustering method; adequate sample size; choice of appropriate method; and caution in generalization beyond the sample. Caution in generalization is necessary due to problems in determining validity of the existence and number of clusters present in the data. Different techniques of clustering have varying assump- tions and are likely to produce different solutions. Decisions regard- ing appropriate levels to stop fusions are made by the investigator who must ask if the clusters achieved are significant enough to provide evidence for the hypotheses studied. Several methods of validating results of imposing structure on data using hierarchical clustering techniques have been suggested (Dubes, 1977). Three of these methods are appropriate for the present study: 1. Global fit criterion: Measures the degree to which the desired structure describes the data. 2. Isolation criterion: Measures the distinctiveness, separation or gaps between two clusters in a particular environment . . a cluster is real if it forms early in the dendrogram and lasts a relatively long time before merging. 3. Use of several clustering methods on the same data. The global fit criterion was used by applying theoretical a priori criteria as a basis for decisions regarding cluster solutions. The null hypothesis under consideration was: all proximity matrices are equally likely. The alternative hypothesis: The number of clus— ters are four and represent the resource classes of love, status, services, and information. This would mean, for example, that vari- ables representing the resource class love will have greater proximity to each other than to variables representing the resource class information. 125 The isolation criterion was examined by calculation of an iso- lation index fOr each cluster. If Cl is the birth level of cluster Ci and c2 is the lowest level before Ci becomes a subset of another cluster, the isolation index for Ci is: I (Ci) = cl - c2. This index of isolation is also called the "lifetime" or "survival time" of the cluster (Dubes, 1977, p. 50). The Foa and Foa theory expects resource classes to be highly correlated. Love and status are the most highly correlated resource classes. It is therefore probable that the lifetimes of clusters will be short. On an absolute scale the complete-linkage method will give clusters with longer lifetimes than clusters achieved with single- linkage clustering techniques. Complete-linkage clustering updates the similarity matrix by choice of the minimum correlation. Single-linkage clustering chooses the maximum correlation and a third method available on the same com- puter program called UMPA updates the similarity matrix by calculation of the average correlation. The three methods were used for validation of the cluster solution decision. Examination of cluster survival time and patterns of merger will give some indication of relationships between resource classes. The agglomerative method of hierarchical complete-linkage clustering based on a similarity matrix will be useful in determining whether data fit the Foa and Foa theoretical model. Multiple Regression Analyses Multiple regression analyses were employed to search for the best set of independent variables to predict quality of marriage and 126 to test the Foa theory that exchange of the most particularistic resources produces the highest levels of satisfaction. The hypotheses predicted that high frequency and high satisfaction with shared time and interpersonal resources received, particularly love and status, would best predict quality of marriage and quality of family life. An ideal regression requires high correlation between inde- pendent and dependent variables but low correlation among independent variables. The presence of too many highly inter-correlated independent variables adds little to predictive power of the regression equation, detracts from descriptive abilities, and makes explanation of variance difficult. The close relationship of particularistic resources included in the Foa and Foa theory created problems for selection of variables for regression and in the interpretation of results. The search procedures for independent variables included two stages. First the correlation matrices of evaluation and frequency variables were examined in order to eliminate redundant indicators. The selected group of variables was then submitted for computer search. The forward selection search procedure of multiple regression was used. This search procedure is a simplified version of stepwise regression, omitting the test at each step of whether a variable once entered into the model should be dropped (Neter 8 Wasserman, 1974). The search procedure selects a minimum number of variables to explain maximum variance of the dependent variable. Variables are entered into the regression equation one at a time and only if they meet statistical criteria. The order of inclusion is determined by the respective contribution of each variable to the explained variance (Nie et al., 1975). 127 Forward regression begins with the calculation of all simple regressions for each of the potential independent variables resulting in‘F values. The independent variable with the largest 5 value and the largest zero-order correlation with the dependent variable is a candidate for the first addition. If this 5 value exceeds a pre- determined level, the variable is added. The second variable to be entered is the one with the highest squared semi-partial correlation with the dependent variable after partialing the variable already in the equation (Kerlinger 8 Pedhazur, 1973). The squared semi-partial indicates the increment in theR2 attributed to the second variable. The process continues until all variables are entered or until there are no variables which have an F to enter meeting the pre- specified level of significance. The multiple regression model (Neter 6 Wasserman, 1974) is: p-l Yi = B0 + Z Bk xik + ei k=1 where: Yi is the value of the dependent variable on the ith trial; BO is the Y intercept which is identified as (constant) in the multiple regression table; Bk represents a change in the mean response of the dependent variable with a unit increase in the independent variable xk when all other independent variables in the model are held constant; xil, . . ., xi, p—l are the values of the independent variables in the ith trial; ei are random error terms. The assumptions for multiple regression are: 1. The scores of the dependent variable Y are normally distributed at each value of the independent variable X and have equal variances at each X point. 128 2. The errors are random, independent, normally distributed at each X point and have constant variance. 3. There is random sampling. 4. Linear relationships exist between independent and dependent variables. 5. Additivity. 6. Independence of observations. 7 Interval level of measurement (Kerlinger 8 Pedhazur, 1973). The F statistic was used to test whether there is a relationship between evaluation of marriage and the entire set of particularistic resources (Hypotheses 8 and 9). The test of overall goodness of fit of the regression equation tests the null hypothesis that the sample of observations being analyzed has been drawn from a population in which the multiple correlation is equal to zero and that any observed multiple correlation is due to sampling fluctuation or to measurement error. This null hypothesis (Ho: R = 0) is equivalent to the null hypothesis that all k regression coefficients are equal to zero in the population (Ho: Bl = 82 = B3 = . . . Bk = 0). The test statistic is: SSreg./k RZ/k F - SSres./N-k-l (1-R2)/(N-k-l) where: SSreg. is the sum of squares explained by the entire regression equation; SSres. is the unexplained sum of squares; k is the number of independent variables in the equation; N is the sample size. This calculated 5 value was compared to a tabled value of_§ at a = .05 with k and N-k-l degrees of freedom. If the calculated .E value exceeded the tabled f_value, the null hypothesis was rejected. When the null hypothesis is rejected it can be concluded that one or 129 more of the population regression coefficients has an absolute value greater than zero (Nie et al., 1975). The probability of an'F ratio this large occurring by chance is less than .05. The test for a specific regression coefficient (hypotheses 10-19) involves decomposition of the explained sum of squares into components attributable to each independent variable in the equation. The test statistic is: Incremental SS due to Xi/l F = SS res / (n-k-l) This is equivalent to theLR2 change” after a given variable X1 is added to the equation containing all others. The degrees of freedom fOr the F ratio are l and (N-k-l). When the null hypothesis is rejected it can be concluded that the population regression coefficient has an absolute value greater than zero and the probability of an'F ratio this large occurring by chance is less than .05. Objective 2 (research questions 5, 6 and 7) required selection of the best prediction of marriage evaluation. The criteria for selec- tion involved examination of the mean square error, adjustedIR2 values, F to enter, and the number of significant predictor variables in the variable sets. The multiple‘R2 is also called the coefficient of multiple determination and is defined: 2 SS regression SS residual R = or 1 - SS total SS total 130 .5? assumes the value of zero when all Bk = 0 and takes on the value of 1 when all observations fall directly on the fitted response sur- face (Neter 8 Wasserman, 1974). .52 measures the proportionate reduction in total variation of the dependent variable associated with the set of independent variables. The closerR2 is to the value 1, the greater is the association between the set of independent variables and the dependent variable. '32 is a measure of prediction accuracy and the strength of linear association since it is the ratio of the explained variation in the dependent variable Y to the total variation in Y. ‘32 varies inversely with $8 residual, but 55 residual can never increase as additional independent variables are added to the equation. Therefore it is necessary to find the point where adding more independent vari- ables is not worthwhile because it adds little increase in 32. Since-R2 does not take into account the number of independent variables and can never decrease as variables are added, the use of mean square error is used as a criterion. Mean square error does take into account the number of parameters through the degrees of freedom. Mean square error can increase as the number of independent variables increase if the reduction in sum of squares error becomes so small that it is not sufficient to offset the loss of an additional degree of freedom. The objective is to find a set of independent variables which minimizes mean square error, or a set for which mean square error is so close to minimum that adding more independent variables is not worthwhile. 131 The adjusted R2 is the coefficient of multiple determination adjusted for the number of independent variables in the model: Adjusted R2 = l - [(n-l)/(n-p)] [SS residual/SS total] The coefficient of multiple correlation also accompanies the analysis of variance tables for hypotheses and is called Multiple R = (R2)%. Tables reporting the complete regression analyses report: (1) The standardized regression coefficients called beta weights com- puted on standardized values of the independent and dependent variables; (2) the unstandardized regression coefficients which estimate popula- tion parameters; (3) the standard error of the regression coefficient; (4) the computed 5 value for individual regression coefficients; (5) the probability for each of the F tests. When independent variables are correlated, the regression coefficient of any independent variable depends upon which other inde- pendent variables are included in the model. There is no unique sum of squares which can be ascribed to an independent variable as reflect- ing its effect in reducing total variation in Y. The reduction in total variation ascribed to an independent variable must be considered in the context of the other variables included in the model. The order of entry is also critical in situations of multicollinearity. Inter- pretation of results must be made in light of the above limitations. Selection of the best set of predictor variables for marriage evalu- ation must be made in regard to the present study sample only. CHAPTER IV RESULTS The results of data analyses are reported in five sections: (1) descriptive data for major variables, (2) descriptive data for independent variables, (3) results of cluster analyses, (4) results of multiple regression analyses, and (S) summary of results. Each research question or hypothesis is stated under the analysis technique and results are reported. There are twenty research questions but not all of them have hypotheses. Hypotheses have been numbered to corres- pond to research questions. The summary of results is organized by research questions. Statistical methods are discussed in Chapter III, summaries of reliability analyses are reported in Appendix E, frequency distribu- tions for all variables in Appendix D, and correlation matrices in Appendix H. Descriptive Data for Major Variables Perceived Overall Quality of Life Respondents in the study sample evaluated their lives posi- tively. The mean POQL (Life 3) score for both men and women was 5.3 (SD = .9). The results are consistent with findings in other quality of life studies (Andrews 8 Withey, 1976; Campbell et al., 1976). 132 C 9 133 The reliability of Life 3 was evaluated using Pearson correla- tion between Life 1 and Life 2 which resulted in a coefficient of .67 for women and .68 for men. Similar results were found by Andrews and Withey (1976) who indicated a reliability range of .61 to .71 in four national surveys. Pearson correlation coefficients in Tables 15 and 16 for POQL with evaluation of family life-as-a-whole (Famlif 3) were .68 for women and .60 for men. Other quality of life studies support the strength of the relationship between evaluation of family life and evaluation of life-as-a-whole (Andrews 8 Withey, 1976; Bubolz et al., 1980; Campbell et al., 1976; Medley, 1976; London, Crandall G Seals, 1977; Wilkening 8 McGranahan, 1978). The strength of the relationship between evaluations of family life and quality of life prompted the present study which has as its ultimate purpose the delineation of dimensions which contribute to positive evaluations of family life. Perceived Overall Quality of Family Life Respondent evaluations of family life (Famlif 3) were more positive than evaluations of POQL, particularly for men. The mean scores for Famlif 3 were 5.6 (SD = .9) for women and 5.8 (SD = .9) for men. The one-tailed dependent t-test (200) = 2.90 compared to the tabled value of 2.60, p < .005 indicated a decision to reject null hypothesis 20 that there is no difference in mean scores of men and women on evaluation of family life. The mean scores for the present study were similar to the Campbell et a1. (1976) study which employed a seven-point satisfaction scale and found a mean score for 134 scocuow0u oz vmo. cmm. com. vac. ocm. mom. cmm. mom. so: mwccco och .m mo>wc so: can. mum. Nco. sew. New. moo. com. moo. co-mo~=35§oo mo new: och .h wo>flooou who. woe. omo. th. mew. one. mam. hmo. so» pooamoe we pesosm och .c . . womfls ~ce _eo co pcmcmsc c3o> .m hoeoc on men. emm. emu. Ncn. mom. Nah. hme. coo. oc oO mcoom pd occmuuomsoo so: .e scoom . . . . . so: mcwmcococ mam cwm Nah cow was mo omcom pew mmocomoco och .m woocoflhomxo com. Nam. ems. Asa. Nmm. men. sow. oHA. so» coaouocca can o>oH och .N com. mom. mun. cmm. wcc. own. mom. mch. wowmwcume coo» .c BNA u 2 «me: u z ONN u z ”xcco eocoemm oocmasm>o u coeoe oo oocmecm>o H coeoa oo acoppo xcmmv icoo so» mfi omwc . u m o . .mm o . u m u . .mm o my ucoonmmoou sawemw coo: unoco :Omnmoa :Omcmod COmcmod coom so» pcsos so: .m wwcsmmuiococzumummuomfic-xc«so: mo :ofiumscm>m cuws omMc xcfieom mo m20wmcoefim mo mcofiumscm>m o>fluoomm< m.:oEo2 pom moocmwnm>ou vcm mucoflowmwoou coflumcocpou :omhmodu-.mc occmh 135 wpocmwcmeooom \eoec>ep moo. Hem. Hcm. one. cem. cme. mom. cme. ma ccos ecoc tomzoc xms och .0: «mpochE schemm mmm. 0mm. owe. mom. omm. oom. ecm. moe. mo mmoccsw -aam; _a:u:s one .m_ mam. mew. scoepcdco czo> .ec scocuowOO mpcomw eon. smm. xaaeac one sees mo u::OEm och .mc . . . . . . . . sesam:0aum~oc mmo owm mmm mmh mum mco com mmm Hmsxom pso> .mc . . . . . . . . sopme ohm m:0wm cmo mom moo mac eme amo mom mmm -Hoov xms och .Hc wmmcficoom . . . . . . . . mmoumxo :mo Hem mmm mwe heo hmm mum mam eom so» scumococ can xccomo so: .o~ hecmcmsc coo. mmm. mom. woo. cam. mum. mam. mom. use» ace: ecoam so» oEHp och .m cmh u z mmcc u z omm u z "Acco pouopwm oocmapm>o c coeoa oo oocmeum>o a ucoaoH oo c muoppo xcmmv -coo so» u“ omflc . o N u . .mm o . o m . .mm o N ucowoflmmoou sawemm use: unocm comnmod :omcmod :Omumom Hoow so» pcsos so: .eoscaueoo--.m~ «some 136 uaooxo momma can poumcflswco :ofipocop omflsumfic .Ho>oc mo. ocu um acmowmccwcm oce mHCONOMMWooo :ocumcoccoo cc< ”Ouoz .mocoom owcumoco cococumosooo cucs omocu .pcocomsoc ocu opqmuzo ooxocoeo one ocs :oEoz c .mump wccmmfis sew cows mommo ace moumccefico :ofluocoo omwsumficm cm coca. mace: Noe. owe. a ma omcc poo» .NN «soaco ou . .. so: moccmco Noo eom ow mpoow cmfleouee och .cN Now: . . cmCOmpom com woo mcm occmcfim>m socoE mo «:305m och .oN . . Npom: ma mcc mem socos has och .mc Nxohco omc. owe. Ow so» moccmco «a mpcocnm och .oc scocpcwco poo: ems. owe. can: ecoam so: oEwu och .hH awe u 2 «me: u z omm u z ”sane emnoeem acopuo xcmmo -coo so» we owe“ oocowum>ou H acoflonmoou oocmficm>ou u ucowowmmoou u x x N N N acoMONmmoou cwemm moo unoco :omumod :omcmoo :omcmoo coom so» pcsos so: .easceueou--.mc magma 137 moN. ome. Neocpcwco 930» .mc ceN. Hoe. mocoom schemw ocu oefiu mo ummwuwowmw .ec mNm. coN. ccm. oeN. ome. Npovw>wp ma cuos pcocomzoc ans och .mc mew. mas. eschew as no» massage as meeaatc 0:9 .NH mom. mom. omm. ooN. cem. No>woooc so: oooomoc mo ucsoem och .cc New. me. oem. ocm. hmm. Newmcococ mo omcom cam mmocmwwww Mm“ .oc . . . Nmmcwcoom omm wem oom. cem wwm. mmopoxo :mo so» scumococ can accomo so: .o omw. mum. Nae. th. oco. NmficmcoNumcou cmsxom ~30» .w cmo. eon. mNo. hum. eco. Noocownomxo so: :owuoommm one o>oH och .N mes. mmm. emm. mum. mac. am>aa so.A aoaomuacsseoo mo eeex one .o mco. com. oNo. Hon. Ago. Now“: coo» cues pcomm so: oeflu och .m mmk. Nme. Nee. Nam. owe. scocuomoa oe :os mousse use .e mac. ome. fine. mNe. Nmo. Nosoc um oc o» mcoom ON occmonomeoo so: .m mee. one. Novas no ecmcmsc poo» .N oow. ewm. eon. mwe. moo. Nowmwpeme coo» .H memH u z memc n z oNN n z c a anopno ccmmo "xcco oouopflmcoo so» oocmflum>oo N acoNONmmoou N ucofiofimmoou we xcweom nsox usocm Hoom so» pcsos so: :Omcmoo compooo I- ‘1’ .m on.eaa--mco;z-a-mm-aoas-s_seam co macsumsam>m can: aces schema co meoamcosso mo maceumscm>m o>wuoomw< m.:oz you moocmwum>ou can mucomofimmoou :owumcosuoo :omuooduu.o~ occmh 138 .mo. 0 N .cofiumEhomcfi Newmmfie ace cufis mommo ccm moumcwecco cofluocoo omfisumcco com. ooo. cm «was. mace: a ma moss use» .NN . . «cso wcc mem Op so» moccmco um mooow Hmwcooma och .HN eNH. Nmm. aceoesaau pace can: ecoom so.A cane .om Noc. moe. Now: cocomuom you socoe mo ucsoem och .mc ohc. Nce. Noomz ma socoe so: och .wc ewe. ems. new. mac. Nee. seems mom meonmsooe Na: age .NH . . . . . choche moe om: one moN ome xcfleom mo mmoccswococ Hmsuse och .oc oemc u z mem~ n z oNN u z c a Acoopo xcmmo ”xcco oouoowmcoo so» oocmfiuo>ou N ucoflofimwoou N ocofloflmmoou mfi xcweom poo» ozocm coom so» ocsos so: :Omuooo comcmoo 1111.. 1,111.11. .1. ‘ 1.1 I All. 11 11.1.11. '11"- III. 1... I III. .emscoocoo--.o~ oases 139 satisfaction with family life to be 5.9 for 2,074 respondents. Andrews and Withey (1976) indicated a mean of 5.7 using the Delighted-Terrible Scale. The reliability of Famlif 3 was evaluated by Pearson correla- tion of Famlif l with Famlif 2. The Famlif 1 question was presented to respondents prior to the family related questions and Famlif 2 was answered by respondents after they had completed the questions related to family life. Correlation results indicated coefficients of .81 for women and .70 for men. Comparison of the variable mean scores for Famlif l and Famlif 2 in Table 17 indicates no change for men and slight change for women when asked the second time about overall evalu- ation of family life. Table l7.--Means and Standard Deviations of Women's and Men's Affec- tive Evaluations of Family Life and Life-as-a-Whole. Women Men Standard Standard Mean . . Mean . . Dev1ation Dev1at1on 1. Family Life 1 5.6 5.8 1 2. Family Life 2 5.5 l 0 5.8 l 3. Family Life 3 5.6 5.8 .9 4. Life 1 5.3 .9 2 .9 Life 2 5.3 .9 3 1.0 6. Life 3 5.3 .9 5.3 .9 Crosstabulation of women's and men's perceived overall quality of family life presented in Table 18 indicates that an obvious majority of people give positive evaluations of family life. The study 140 .Nmmoo. u meson .xocmmouomwo ucHoq oco cho one Ho oocwm woo .owHH xHHeew Mo :oHuwzHo>o co oonwm mo>Hs ocm mocmcmsc mo NN.He Houoz HNo.ooHo HNN.HHo NN.Neo HNe.NNo HNm.NHo Hem.o Hem.o HNoo HNo ANN mN em Ho mm N N o Haooe casHou st.eHo HNm.eo mm oH HN N o o o o eoocwHHoo HNH.Neo HNa.NNo moH NH Hm 0N eH o H o eomaoHa Ne.NNo HNm.oHo eoHomHHam Ho N oN mN oH o o o HHHmoz No.No HNH.mo NH o H N N H o o emtz Nw.Ho HNe.o eoHcmHHammHo e o o H H H H o HHHmoz Hem.o HNOo N o H o H o o 0 season: Hoe.o Heoo H o o H o o o o oHHHHHoH eoHENHHoa eommoHa eomcmHHam eotz eoHHmHHammHo Homans: oHoHHHoH Homo: HHHmoz coo mococms: HmHOh so: mo>H3 3" .oaoc3lmlomiochleHEmL mo :oHumsHm>m o>Huoomm< m.:oz ocm m.coeoz mo coHumHscmummOHUun.mH oHcoh 141 indicated 7% of the respondents felt "terrible," "unhappy,” or "mostly dissatisfied" with their family lives compared to 4% of respondents in the Andrews and Withey (1976) study. Less than half of the husbands and wives (41.2%) agreed absolutely upon evaluation of family life. However, 89% of the respondents agreed or had dis- crepant scores varying by only one scale unit. The ggmmg of .60 indicates a high predominance of concordant pairs. Tables 15 and 16 report Pearson coefficients for marriage with Famlif 3 (a = .71 women and .70 men). The correlation was highest of all independent variables. This finding supports the conclusion of Campbell et a1. (1976) that a major contribution to satisfaction with family life is the individual's relationship with spouse. Tables 15 and 16 indicate the feelings about spouse and evaluation of family life were highly related (3 = .64 women and .67 men). Feelings about chil- dren were not as highly related to evaluations of family life as were indicators of shared time, love, status, and information. The correlation analyses shown in Tables 15, 16, 19, 20, and 21 provide support for the usefulness of the Foa and Foa resource exchange theory in predicting affective evaluation of family life. The correlations are higher for evaluation than for frequency vari- ables. However, both sets of variables indicated the love and status indicators to be highly correlated with overall quality of family life. Table 21 shows that the frequency variables created by com- bining indicators of the various resource classes are ordered in the expected direction. Frequency of receiving love from mate shows the strongest rela- tionship to evaluation of family life and is fOIIOwed by status. The 142 Table 19.--Pearson Correlation Coefficients for Women's and Men's Frequency of Shared Time and Activities with Mate, with Affective Evaluation of Family-Life-as-a-Whole. Women Men Variable 2 2 r r r r How often do you and your mate: 1. Spend time together just the .211, .045 .285* .081 two of you? 2. Spend an hour or more just .202, .041 .190, .036 talking? 3. Discuss personal feelings? .226* .051 .299* .089 4. Work together on a project? .221* .049 .195* .038 5. Take a drive or a walk? .257* .066 .119* .014 6' Mate time freauency’ .289* .084 .327* .107‘ companionship 7. Eat at a restaurant .104 .011 .173* .030 8. Entertain friends at home .034 .001 .096 .009 9. Visit friends .071 .005 .004 10. Go to a mov1e or other .062 .004 .105 .011 entertainment 11. Attend a sports event .102 .010 .034 .001 12. Attend a party .029 .001 .082 .007 13' Mate tlme ffiequency’ .137* .019 .102 .010 sociability aVariable created by combining variables 1—5. bVariable created by combining variables 7-12. :p.< .05. 143 Table 20.--Pearson Correlation Coefficients fOr Women's and Men's Frequency of Resources Received from Mate and Affective Evaluation of Family-Life-as-a-Whole. Women Men Variable 2 2 r r r r How often does your mate: 1. Make ygu feel like an important .473, .224 .375* .141 person. 2. Tell/show sh: admires and .380* .144 .343* .113 respects you. 3. Let you know he has confidence .327, .107 .327* .107 in your abilities? 4. Tell/show you her love? .412* .170 .367* .135 Let youoknow he/she enjoys your .327* .107 .315* .099 company. 6. Enjgy a laugh or a joke with .317, .100 .375* .141 you. Give you a hug or a kiss? .430* .185 .314* .099 Do an errand for you? .189* .036 .224* .050 9. Make himself available to do .206* .040 .185* .034 work for you? 10. Do something to save you energy .263* .069 .269* .072 or make you comfortable? 11. Give you some new information? .176* .031 .193* .037 12. Give you his opinion? .128* .016 .034 .001 13. Give you something you need or .259, .067 .292 .085 want? 14. Give you money for personal use? .182* .033 .129* .017 15. Help you solve a problem or make .173, .030 .180* .032 a decision? 16. Support you with disc1pline and .227* .052 .168* .028 guidance of children? fp < .05. 144 Table 21.--Pearson Correlation Coefficients for Women's and Men's Frequency of Resources Received from Mate, Created Vari- ables, with Affective Evaluation of Family-Life-as-a- Whole. Women Men Created Variable 2 2 r r r r Mate time frequency, companionshipa .289* .084 .327* .107 Mate love frequencyb .465* .216 .398* .158 Mate status frequencyC .429* .184 .376* .141 Mate services frequencyd .263* .069 .256* .066 Mate information frequencye .192* .037 .119* .014 aVariable created by combining variables 1-5 on Table 19. bVariable created by combining variables 4-7 on Table 20. CVariable created by combining variables 1-3 on Table 20. dVariable created by combining variables 8—10 on Table 20. 8Variable created by combining variables 11-13 on Table 20. * B.< .05. 145 correlations indicate greater strength of relationship between the most particularistic activities and evaluation of family life. The evidence lends support to the Foa and Foa (1974) theory. All coefficients in Tables 15, 16 and 21 are significant at the .05 level suggesting the basic assumption of the theory that all resources are necessary for quality of family life. The importance of feelings about marriage in predicting affec- tive evaluation of family life can be seen by examination 0f.52 values in Tables 22 and 23. Addition of marriage to the equation reduced the variance of Famlif 3 by 54% for women and 48% for men. Addition of "love and affection" to the variable set contributed an additional 7% decrease in the variance of Famlif 3. The set of five independent variables accounted for 63% of the variance in the women's analysis and 61% of the variance in the men's analysis. The importance of marriage in explaining the variance in family life evaluation is evident in this study and also in the Campbell et a1. (1976) study. Affective Evaluation of Marriagg Mean scores for evaluation of marriage were higher than for evaluations of life-as-a-whole or evaluation of family life with men having consistently higher mean scores than women on all three major dependent variables. The mean scores for evaluation of marriage (M = 5.95 men and 5.71 for women) were lower than the mean scores of 6.2 in the Andrews and Withey (1976) and 6.3 in the Campbell et al. (1976) 146 .m0. V ma. .mHmsHmcm :onmonoH Eoum mHo>oH oocmonocme so oouoooom moHooHHo>m smohHoooH zos ememo.os oNeoo. HmsNo. mons. mooom.N Hooomoc mo Hcaoem och .m Hone¢mmo sec; on on 0H oHsom mHH msmHo omeo oseos wNwoom s mHoow ac oHomuHomeoo so: .m . . . . . sm>0oo oocoHHooxo :os mmNoH ooH emooo eHNoo wmmss asson mm :oHooomwm ocm o>oH och .N waoo.seN ommmm. owmmm. mons. «waoo.seN omoHHHmE Hoes .H m HHmHo>o owcocu N: N: a onHuHsz ououcm on m oOHoucm oHooHHm> noum .coeos uuoowmfloomm no: Hoopo .oo>Hooom moopsomom one owmopumz mo :oHumsHm>m so omHo sHHEom mo :ofiuosom>m o>Huoowm< mo :oHuoHooHo ocu How mHmsHmc< :onmonom oHoHuHoz mo sumeesmun.NN oHooh 147 .mo. 0 m. .mHmsHmcm :onmonoH scum mHo>oo oocmoomocme so oouooHom moHomHHm>o sm>oco oocoHHooxo :os mNoew.eo onoo. Hmooo. momNN. mmsoo.N coHHooooa ens o>oH use .m NHNNH.om mHeHo. mHmoo. moons. rmmNem.N aHameoHHaMwmoMWMWMmmwmww .e omme.HoH mmoeo. wawm. eeNoN. .HNHom.oN wwwww>mwmemmmmowmowozmm .m Noomm.mNH Nmsoo. eemem. Hmoes. .mNmmN.Hm HmzHHoWHMMMmWNmmmwowmw .N HomNe.me NHHwe. NHHme. momma. .HowNe.mmH onHHHms Haas .H m onHo>o omcmco N: N: a oooHquz mooucm OH o ooHoucm oHomNHo> moum Iv [1:11.111]- l1! .cozunoonHooam no: Hoooo .oo>Hooom moOH30mom oce owmwuumz mo :oHHmsHo>m so owwc usEmm mo :OMHmsHo>m o>Huoowm< mo :oHuoHooHo ocu How mHmsHmc< :onmonom oHoHuozz mo semeezmnu.mN oHomh 148 studies. The one-tailed dependent t-test (200) = 2.52 compared to the tabled value of 2.35, p_< .01 indicated the decision to reject null hypothesis 20 that there is no difference in the mean scores of men and women on evaluation of marriage. The reliability of marriage evaluation could not be examined using correlation since the question was asked once. However, the descriptive statistics discussed in Chapter III and frequency distribu- tions reported in Appendix 0 indicate results similar to the Andrews and Withey (1976) study in which evaluation of indicators was of central concern. Therefore it is reasonable to assume there are no particular problems in using evaluation of marriage as a dependent variable since the F test is robust to violations in the assumption of normality when the sample size is large (Nie et al., 1975). Crosstabulation of women's and men's evaluations of marriage in Table 24 indicates a 39.3% absolutely agree on assessment of the marriage relationship. The ggmm§_of .512 indicates a predominance of concordant pairs. One-hundred forty-one of the respondents (36%) were "delighted" with their marriage, 34% of respondents were "pleased," and 7% were "mostly dissatisfied,” "unhappy" or felt "terrible" about the marriage. Only 6% of the respondents indicated mixed feelings about the marital relationship. Com- parison of the crosstabulations for evaluations of family life and evaluations of marriage indicate greater agreement in evaluations of husbands and wives for evaluations of family life (ggmmg = .60) than for evaluations of marriage (gamma = .51). I49 (I; HNNHm. n mesa: .omeHHmE mo :oHuosHo>o co oonm mo>cs oce moceomzc mo 9m.om HoHoz HNo.ooHo Nm.Hmo No.omo HNN.oHo HNo.No HNo.mo Hoe.Ho HNN.No HNo NNN oN ow NN NH w m m Haooe cssHoo HNo.Heo HNN.NHo Ha Ne om N e H o H eooawHHoa HNe.Nmo HNm.mHo NN HN om mH N H m a eommoHa No.wHo Hoo.mo eoHomHHam as N HH HH N m o H sHHmoz NN.mo HNm.o N o H N H H o N eotz HNN.No HNm.o echmHHmmmHa m o o o m N o o sHHmoz HNN.Ho HNoo e o H H N o o o seamen: Ne.Ho HNm.o m o H H o o o H oHoHHHmH H.o eoHcNHHoo eomaoHa eomwmwwmm eotz eonwwwmmmHa Nuance: oHoHHHoH a monsoon: Hooch so: mo>Hs .ommHHHmz mo :oHumsHm>m o>ouoowm< m.=o2 oco m.:oeoz mo :oHumosomummocu--.eN oHomh 150 Descriptive Data for Independent Variables Independent Variables, Affective Evaluation Mean scores for the affective evaluation variables in Tables 25 and 26 ranged from 4.5 to 6.0 but were generally above 5.0. The exceptions were for the resource class indicators of services, goods, and money. Women indicated greater dissatisfaction with the "mutual helpfulness of family members" (4.8) and the "way household work is divided/accomplished” (4.5). Both husbands and wives had mean scores of 4.7 fOr feelings about the "amount of money available for their personal use" while husbands were slightly more dissatisfied with the material goods they were able to own (M = 4.9 men, 5.0 women). High mean scores were found for evaluations of marriage, children, comfort at home and the indicators of love. The use of listwise deletion in regression analyses which eliminates all cases with any missing data, produced a slight increase in mean scores and a slight decrease in standard deviations as shown in Table 26. The present study did not attempt to examine differences among types of similar respondents within the total sample, however, Table 26 does report mean scores of women employed outside the house- hold compared to mean scores of the total number of women in the sample. The item "how comfortable it feels to be at home" had consis- tently high mean scores for both men and women. The item seems to represent the overall emotional and physical climate of home and family life, the combination of particularistic love and services family life provides. Alternatively, it may be an indicator of what 151 Table 25.--Means and Standard Deviations of Women's and Men's Affective Evaluations of Dimensions of Family Life. Women Men Variable (N = 224) (N = 224) 'M S.D. ‘M S.D. 1. Your husband or wife? 5. 1.4 5. 1. Hour children? 1.1 5.9 1.0 3. The love and affection you 5.7 1.3 5.8 1.2 experience. 4. The closeness and sense of 5.7 1.3 5.8 1.2 belonging you feel? 5. The amognt of respect you 5.4 1.3 5.6 1.1 receive. 6. How comfortable it feels to 6.0 1.1 6.0 1.1 be at home? 7. Your marriage? 5. 1.4 6.0 1.3 8. The way money is used? 4.9 1.2 5.0 1.1 9. The amount of money available 4.7 1.4 4.7 1.3 for personal use? 10. The material goods it enables 5.0 1.2 4.9 1.2 you to own? 11. The way decisions are made? 5.2 l. 5. l. 12. The things you do together 5. 1.4 5. 1.3 13. The mutual helpfulness of 4.8 1.2 S. 1.2 family members? 14. The way household work is 4.5 1.4 5.0 1.1 d1V1ded? 15. How openly and honestly you 5.2 1.3 5. 1.2 can express feelings? 16. The kind of communication 5.2 1.5 5. 1.3 you have? 17. The amount of time the family spends together? 5' 1'3 5° 1'3 18. Your sexual relationship? 5.4 1.5 5. 1.5 19. The time you spend with your 5.4 1.0 5. 1.2 children? 152 Table 25.--Continued. Women Men Variable (N z 224) (N = 224) 'M S.D. .M S.D. 20. The time you spend with your 5.2 1.3 5.2 1.4 husband or wife? 21. The friends it enables you to 5.1 1.2 5.1 1.2 enjoy? 153 Table 26.--Means and Standard Deviations of Women's and Men's Affective Evaluations of Selected Dimensions of Family Life, Listwise Deletion. Women a Men Variable (N = 165) (N = 72) (N = 154) ‘M S.D. .M S.D. 'M S.D. 1' The 1°Ve ind affe°t1°n 5.7 1.3 5.7 1.3 6.0 1.1 you experience. 2. The closeness and sense of belonging? 5 8 1 3 5.8 l 2 5 9 1 l 3. Your sexual relation- ship? 5 5 l 3 5.5 l 3 5 3 l 5 4. The amount 3f respect 5 4 1 2 5.6 1 2 5 7 1 2 you receive. 5. How comfortable it feels to be at home? 6 O 1 1 6'0 1 2 6 1 1 O 6. The mutual helpful- ness of family 4.9 1.2 5.1 1.2 5.2 1.1 members? 7. The way household work is accomplished/ 4.6 1.4 4.8 1.4 5.0 1.1 divided? 8. The way deCISlonS 5 2 1 1 5.3 1 1 5 4 1 O are made? 9. How openly and honestly you can 5.2 1.2 5.3 1.2 5.5 1.0 express feelings? 10. The kind of communi- - ’7 cation you have? 5 3 1 4 5'4 1 ° 5 4 1 “ 11. The time you spend . with husband or wife? 5 2 1 3 5'1 1 ° 5 2 1 4 12. The things you do 5 4 1 3 5.3 1 4 5 4 1 2 together? 13. Your marriage? 5.8 1.3 5.8 1.4 6.1 1.2 14. Your family-life-as-a 5 6 9 5.7 9 5 8 9 whole [FAMLIF 3] Note: Listwise deletion eliminates all cases with any missing data. aN = 72 women with occupational prestige scores which indicates paid employment outside the household. They are a subset of the group of 165. 154 Reiss (1960) suggested was "rapport" which he defined as a felt presence of ease and relaxation related to the development of posi- tive regard. Further investigation is necessary for the clarification and refinement of measures regarding this dimension of family life. Independent Variables, Frequengy Frequency variables were converted from whole number scale values (Appendix A, page 238) to decimal numbers based on the common denominator of 365 days in a year. The following conversions were made: 1 = 0; 2 = .003; 3 = .016; 4 = .032; 5 = .142; 6 = .499; 7 = 1.000; 8 2.500. Mean scores for frequency variables were calculated using the addition of decimal numbers and therefore have a different appearance from mean scores on affective evaluation variables. Examination of the range of mean score values on frequency variables in Tables 27 and 28 indicates highest values for love indicators, followed in order by status, information and services indicators. Indicators of shared time and activities in Table 27 show highest mean scores for the companionate activities of husband and wife. Four variables were created by combining indicators of each of the resource classes of love, status, services and information (Table 28). Two variables were created by combining indicators of shared time (Table 27). For example, the created variable ”mate love frequency" was created by adding scores of the four indicators of love frequency for each respondent. Summary of the reliability analyses of the time, love, status, services and information scales is reported in Appendix E. Cronbach's (1960) alpha coefficient was used as the 155 Table 27.--Means and Standard Deviations of Women's and Men's Perceived Frequency of Shared Time and Activities with Mate, Including Created Variables. Women Men Variable isles. rails How often do you and your mate: 1' ifizngwgigz £3figther’ jUSt .7 .7 224 .7 .8 220 2. Spend an hour just talking? .6 .6 221 .5 .6 222 3. Discuss personal feelings? .5 .6 218 .4 .5 220 4. Work together on a project? .2 .4 216 .2 .5 222 5. Take a drive or a walk? .2 .2 223 .2 .4 220 6' Mate time ffefluency’ 2.2 2.0 208 .0 2.0 213 companionship 7. Eat at a restaurant? .1 .l 224 .1 .2 222 8. Entertain friends at home? .1 .1 223 .1 .1 222 9. Visit friends? .1 .l 223 .08 .2 221 10. Go to a movie or entertainment? .1 .1 224 .08 .2 223 11. Attend a sports event? .02 .08 222 .05 .2 222 12. Attend a party? .01 .02 221 .03 .1 219 13' Mate time ffiequency’ .43 .3 219 .44 .9 217 sociability values: 8 = 2.5. Note: Transformation of the eight point scale to decimal numbers based on the common denominator of 365 produced the following aVariable was created by combining variables 1-5. bVariable was created by combining variables 7-12. 1:0; 2=.003; 3 = .016; 4 = .032; 5 = .142; 6 Mean scores were calculated using decimal numbers. .499; 7 = 1; 156 Table 28.--Means and Standard Deviations of Women's and Men's Perceived Frequency of Resources Received from Mate, Including Created Variables. Women Men Variable .M S.D. .E ‘M S.D. .E How often does your mate: 1. Tell or show his/her love? 1.1 .9 223 .9 .9 222 2. Let youvknow he enjoys your .7 .7 224 .7 .8 222 company. 3. Enjgy a laugh or a joke with .8 .7 223 .9 .8 221 you. 4. Give you a hug or a kiss? 1.5 1.0 224 1.3 1.0 220 5. Mate love frequencya 4.1 2.8 222 3 8 2.9 219 6. Make you feel like an .6 .7 224 .6 .7 222 important person? 7. Tell or show he admires and .7 .7 224 .6 .7 220 respects you? 8. Shows confidence in your .6 .7 224 .6 .7 222 abilities? 9. Mate status frequencyb 1.9 2.0 224 1.8 2.1 219 10. Do an errand for you? .6 .7 223 .7 .7 221 11. Make himself available to .5 .6 224 .9 .9 221 do work for you? 12. Do something to save you energy or make you com- .6 .6 222 .9 .8 222 fortable? l3. Mate services frequencyC 1.7 1.6 221 2.5 2.2 220 14. Give you new information? .5 .5 222 .7 .7 220 15. Give you his opinion? .9 .8 224 1.1 .9 220 Table 28.--Continued. 157 Women Men Variable _M S.D. .N 'M S.D. .N 16. Help you solve a problem or 4 5 222 4 6 222 make a decision? ' ' ° ° 17. Mate information frequencyd 1.8 1.5 220 2.2 1.7 218 Note: Transformation of the eight-point scale to decimal numbers based on the common denominator of 365 produced the following values: 1 = 0; 2 8 = 2.5. aVariable bVariable CVariable dVariable = .003; 3 = .016; 4 = .032; 5 = .142; 6 Mean scores were calculated using decimal numbers. created by created by created by created by combining variables combining variables combining variables combining variables = .499; 7 = l; 158 reliability estimate. Alpha coefficients ranged from .66 for the infOrmation scale to .95 fOr the status scale. Reliability of the information scale might have been higher if items which involved more particularistic exchanges of information between husband and wife had been used. Two items which were in the time scale (Table E-2) could have been used as indicators of the resource transfer of particularistic information between husband and wife: How often do you and your mate "spend an hour or more just talking" and "discuss personal feelings." This change in the infor- mation scale is suggested for further investigation. The next section will discuss results of hypothesis testing and answers to research questions. The study has twenty research questions. Questions four, five, six and seven do not have hypotheses. The research questions beginning with number eight were stated in Chapter I and will not be repeated, but rather are stated in the form of hypotheses. Cluster Analyses The agglomerative method of hierarchical complete—linkage clustering was used to answer research questions and hypotheses for Objective 1. Cluster analysis is used as a descriptive tool for data reduction, explanation and model fitting. It is not used for a statis— tical test of hypotheses. However, the existence of theoretical prior conceptions or goals for the cluster-solution strengthens its use in model fitting and positively affects the validity of the cluster solution decision. Objective 1 159 Explore the validity of Foa and Foa's resource exchange theory and the measured indicators of the theoretical model in the present study. Hypotheses 1 H0: H1: All proximity matrices are equally likely. There are four clusters. They represent the particular- istic resource classes of love, status, services and information for frequency and evaluation variables for women's and men's analyses. For evaluation variables the following clusters will form: Love: The love and affection you experience. The closeness and sense of belonging you feel. Your sexual relationship (between love and services) Status: The amount Of respect you receive. Services: The mutual helpfulness of family members. The way household work is divided/accomplished. How comfortable it feels to be at home. Information: How openly and honestly you can express feelings. The kind of communication you have. The way decisions are made. For frequency variables the following clusters will form: Love: Tell or show his/her love. Let you know he/she enjoys your company. Enjoy a laugh or a joke with you. Give you a hug or a kiss. Status: Make you feel like an important person. Tell or show he admires and respects you. Let you know he/she has confidence in your abilities. Services: Do an errand for you. Make himself/herself available to do work for you. Do something to save you energy or make you comfortable. 160 Information: Give you some new infOrmation. Give you his/her opinion. Help you solve a problem or make a decision. Results of the complete-linkage clustering appear to confirm the hypothesis of four distinct resource classes with a close relation- ship between love and status. Four-cluster solutions were found for men's evaluation variables and women's frequency variables. Three- cluster solutions were found for men's frequency variables and women's evaluation variables. Both three-cluster solutions fused love and status which are the most highly correlated resource classes. The single indicator of status evaluation probably contributed to the lack of separation in love and status resource classes for evaluation variables. Appendix tables F-l and F-2 indicate solutions for evaluation variables. The four-cluster solution in the men's analysis occurred at the fifth level in the dendrogram with a minimum correlation between items in clusters of.£ = .529. The women's analysis gave a three- cluster solution at the sixth level with minimum correlation between items in clusters of‘g = .569. The following evaluation variables fused for both men and women: £222: The love and affection you experience. The closeness and sense of belonging you feel. (Isolation index: women = .184, men = .219) Services: The mutual helpfulness of family members. The way household work is divided/accomplished. (Isolation index: women = .273, men = .217) 161 Information: The kind of communication you have. How openly and honestly you can express feelings. (Isolation index: women = .286, men = .150) Status: The amount of respect you receive. ‘ (Isolation index: men = .065, women = emegence at .631 but is fused with love) Appendix tables F-3 and F-4 indicate cluster solutions for frequency variables. The women's analysis provided a four-cluster solution at the sixth level of the dendrogram with a minimum correla- tion of .495 between items in clusters (Table F-3). The solution for the men's analysis was at the fourth level in the dendrogram with minimum correlation of .707 between items in clusters (Table F-4). The following variables fused for both men and women: Status: Make you feel like an important person. Tell or show he admires and respects you. Let you know he has confidence in your abilities. (Isolation index: women = .126, men = .185) Services: Make himself available to do work for you. Do something to save you energy or make you comfortable. (Isolation index: women = .236, men = .142) Hypotheses 2 Ho: All orders of fusion are equally likely. H2: The fusion order of resource classes for both evaluation and frequency variables for women's and men's analyses will be: (1) love-status, (2) love-status-services, (3) love-status-services-information. The alternative hypothesis was confirmed for frequency vari— ables in both women's and men's analyses (Tables F-3 and F-4), but not for evaluation variables. The order of fusion for evaluation variables is love-status-information-services if one ignores "comfort at home" which clusters with love for women and with status for men. The 162 analysis confirms the validity of the theory and suggests problems of measurement which require additional investigation. Information indi- cators were more particularistic than services indicators fOr evalua- tion variables. Hypotheses 3 There are no differences in cluster solutions for men and for women for evaluation or frequency variables. Null hypothesis 3 was rejected for both evaluation and fre- quency variables. Differences on evaluation variables were found with "sexual relationship," "the way decisions are made," and "how comfort- able it feels to be at home." "Comfort at home" merged with love for women and added to status for men. ”Sexual relationship" did not merge with love or with services, but with information for men at.£ = .529, fifth level and for women at‘E = .449, eighth level. This eighth-level merger was after love, status and information had merged, indicating evaluation of the sexual relationship to be less related to variables representing other resource classes. "The way decisions are made" fuses with information for women (.569, sixth level) but adds to ser- vices for men (.367, eighth level). Differences in cluster solutions for men and women on fre- quency variables were found with love and information clusters. The isolation index for the information cluster was .236 for women and .142 for men. The solution for women (Table F-3) fused ”give you new information" and "give you his opinion.” "Help you solve a problem or make a decision" fused with services at the ninth level of the den- drogram. The solution for men shows "new information" adding to 163 services at the eithth level. "Give Opinion" and "decision" merge at the eleventh level just before all variables merge. The isolation index for the love frequency cluster was .265 for women and .031 for men. The solution for women fuses the concrete and symbolic expressions of love: ”tell/show love” and ”hug or kiss." "Enjoyment of company" and "laugh or joke” merge with status. The solution for men shows a fusion of ”tell/show love” with "enjoyment of company" which are both symbolic expressions of love. "Hug or kiss" does not enter until after the fusion of love, status and services. Research Question 4 Will different methods of hierarchical clustering provide similar cluster solutions for the same data to support validity of four particularistic resource classes? A summary of the results from three hierarchical clustering methods is found in Tables 29 and 30 and confirms the validity of the cluster solutions. The three techniques produced similar results with complete-linkage providing clusters with the longest lifetimes as expected. The three methods found three-cluster solutions for women's evaluation variables and men's frequency variables. Only the complete- linkage method found four-cluster solutions. The other clustering methods were unable to separate love and status resource classes for men's evaluation and women's frequency variables. The three methods indicated more consistent results for women than for men. The frequency indicators of status, which had the highest alpha coefficient in the reliability analysis, had the same isolation index for all three clustering techniques in both men's and women's analyses. The combination of reliability, correlation and cluster 164 Table 29.--Summary of Cluster Analyses for Women. Evaluation Variables Frequency Variables Cluster a a Birth Merge Life Birth Merge Life Complete-Linkage Solutionb Love .897 .713 .184 .725 .460 .265 Status .631 .799 .673 .126 Services .713 .440 .273 .673 .435 .238 Information .855 .569 .286 .495 .233 .262 Single-Linkage SolutionC Love .897 .769 .128 .725 .652 .073 Status .764 .799 .673 . .126 Services .713 .661 .052 .673 .495 .178 Information .855 .769 .086 .495 .435 .060 Special Method, Currently UPGMAd Love .897 .713 .184 .725 .639 .086 Status .709 .799 .673 .126 Services .713 .564 .149 .673 .381 .292 Information .855 .709 .146 .495 .435 .060 aAlso can be labeled the isolation index. b . . . . . . Correlation coefficients represent minimum correlation between variables. c . . . . . Correlation coefficients represent max1mum correlation between variables. d I I I 0 Correlation coeffic1ents represent average correlation between variables. 165 Table 30.--Summary of Cluster Analyses for Men. Evaluation Variables Frequency Variables Cluster Birth Merge Lifea Birth Merge Lifea Complete-Linkage Solutionb Love .748 .529 .219 .707 Status .594 .529 .065 .892 .707 .185 Services .617 .400 .217 .724 .582 .142 Information .744 .594 .150 .504 Single—Linkage SolutionC Love .748 .744 .004 .707 Status .675 .892 .707 .185 Services .617 .724 .661 .063 Information .744 .621 .123 .599 Special Method, Currently UPGMAd Love .748 .744 .004 .707 Status .626 .892 .707 .185 Services .617 .525 .092 .724 .622 .102 Information .744 .557 .187 .557 aAlso can be b . Correlation between variables. c . Correlation between variables. dCorrelation between variables. labeled the isolation index. coefficients represent minimum correlation coefficients represent maximum correlation coefficients represent average correlation 166 analyses lend support to the validity of the Foa theory. The reli- ability and cluster analyses suggest some possible problems of measure- ment which require further investigation. Multiple Regggssion Analyses Selection Procedures for Independent Variables Several sets of independent variables were selected for the prediction of affective evaluation of marriage. The first step involved choosing sets of: (1) evaluation variables, (2) frequency variables, (3) a combination set of evaluation and frequency variables. For each set of variables there were two selection methods. The first selection procedure was examination of the corre- lation matrices in order to choose indicators which were highly corre- lated with the dependent variable, but not as highly correlated with each other. The objective was to select one indicator from each of the particularistic resource classes (love, status, services, infor- mation). This procedure was used for the selection of sets of evalu- ation and frequency variables. The submission of these statistically selected variable sets for computer search did not include specifica- tion of order of entry. The second selection procedure involved choosing variables which most closely and explicitly represented the resource class as defined by the Foa theory. The decision was made by agreement of the researcher and another investigator familiar with the theory. The objective was to select one indicator for each resOurce class including goods and money. The procedure was used to select sets of evaluation 167 variables. The variable sets were submitted for computer search with and without instructions fOr order of entry. The specified order of entry for evaluation and frequency variables was: love, status, services, information, goods, money. The set of combined evaluation and frequency variables was submitted for computer search (1) without specification of order, (2) specification of an alternating order, and (3) specification of three evaluation variables followed by three frequency variables. The resource classes of information goods and money were not included in the set of combination variables due to their low correlation with affective evaluation of marriage (Tables 31 and 32). The variables selected by statistical criteria were first rank ordered from high to low according to correlation with affective evalu- ation of marriage. Tables 31 and 32 and 33, 34, and 35 indicate the rank order of variables. A matrix was constructed for evaluation vari- ables having coefficients above .70 and for frequency variables having a correlation of .30 or above with feelings about marriage. The four matrices were examined for the purpose of eliminating redundant indi- cators. An example of the process can be given using choice of an indicator to represent the resource class of love. The two best indi- cators of love were "the love and affection you experience” and the "closeness and sense of belonging you feel." Examination of the corre- lation matrix for wives indicated "closeness and belonging" to have six coefficients above .70 while ”love and affection" had five coef- ficients above .70. However, some of the five highest correlations were with variables which would not be included in the analysis, such 168 hoone ooo NwH.H mom. oss. ooo.H mwe. moo. ooe. wso. mconHooo sos och .o . . . . . . . . NHocoowou NHN H owm eNo oNH H Nme smo Hoe oso oo nos mwchu och .w . . . . . . . . N283; .Sos 5H: owo H mmm oom moo H Noe owo ose Noo ocoom nos oEHu och .N . . . . . . . . $38.: 2; ooo Hsm ooo Hoo H woe ewo ewe ooo Hooomoe mo acnOEm och .o .oHcmcoHu . . . . . . . . . e . . mNH H Nee moo ohH H Nse Nwo mom ooh -mHoH Hmnxom snos .m . . . . . . . . Noeoc um oo ou mHoow omN H wmo ooh meo H msm NmN Hmm NeN HH oHoouHomEoo so: .e «Hoom nos Noe.H hoN. Hew. emm.H Hho. on. wow. mss. wcchoHoo mo omcom one mmocomoHo och .m . . . . . . . . NoocoHHomxo nos :oHu ooe H owo wNw woe H ewo st Noo oNN noommm one o>oH och .N oso. NNw. Nost no osmomnc Hnos .H Hcho ooHooncoo nos mH owHH sHHewm unos unoom Hoom nos oHnos so: II. oNN n 2 II. oon n 2 II. oNN u z oocmHHm>ou H I. oocmHHo>oo NH I. NH HHoon.c:mmo oHomHHm> N H H H .coeosnuomeHHmz mo mcoHomnHm>m cHHs owHH sHHEom mo mconcosHo mo mcoHomnHm>m o>Huoomw< How moocmHHm>ou ooo mucoHonwoou :oHumHoHHou :OmHmoo--.Hw oHomh 169 smcooEoE sHHEnm mo weo. ooH. ooe. mwo. eoH. oee. msH. wHe. mmo:_nooHoc Hnnone och .oH . . Nsonco ou nos NNH NNe moHoaco HH mecoHHo use .oH . . Noomn owH Hme mH socoe sms och .mH NoocmHHnEooom oow. NON. oee. NoN. oON. Nee. moN. mme. NeoeH>Ho mH coo: oHocomnoc sms och .eH mNN. Nse. N:OHoHHco Hnos .mH NHocuoon NHN. mom. menace sHHeao was osHH co wagons och .NH wacHHoom mmoHoxo oew. NmN. Now. ooo. emm. wsm. me. NHo. coo nos sHumonoc one chooo so: .HH No>oc nos :oHuoo mwo.H eom. wNo. osH.H mme. wNo. sme. oso. -cheeoo «O och och .oH Hcho ooHooncoo nos mH omHH sHHeom Hnos unoom Hoom nos oHnos so: .1! oNN u z .11 mmoH u 2 II. oNN u z oonoHHm>ou H oocmHHm>ou H H HooHo xcnmo oHooHHm> N .H N M. N M .‘1'511-l'.1|1|1|.'|1l I I- .emscHocoo--.Hm oHoaH 170 .moH mo onOHw ocH mo Homonm m .oHocomnoc ocu oonuno oosoHoEo oHo ocs :oEoso .mumo wchmHE scm cuHs mommo HHm moumcHeHHo :oHHoHoo omHsHchm . . Hm ooHHH mHe eeo oHocsuoummuomHH Hnos .NN . . Hm oHHHsaHH mHon-a mom NHh immuomMolehEmm H30» .HN . . NcoHoHHco Hnos cHHs mNH wmm oconm nos oEHH och .oN oeH. owm. Ncso oH nos moHomno HH mooow HmHHone och .wH sown HmHOm ooH. woe. -Hoo How oHomHHo>m sonoe mo unnoeo och .NH Hcho ooHoonnoo nos NH owHH sHHemm Hnos Hnooo Hoom nos oHnos so: NN u z amoH u z .1: oNN n z oocoHHm>ou H oocmHHm>ou H H HHooHo cnmmo oHomHHm> H “I .eoscHocoo--.Hm mHoaH 171 NoocmHHnEOooo mHo. mHN. Hoe. oNH. mHe. NoooH>He mH noon oHonomnon so: onH .mH ooN. Nee. HnoHoHHco Hnos .NH woN. owe. Nsomno OH nos moHoono HH monoHHm och .HH mHo. ooN. eem. NwN. oww. NomHs Hnos cHHs onoom nos oEHH och .oH woN. woN. oem. me. msm. No>oc nos noHunoHnnEEoo mo oan och .o Now. wNm. wsm. Nwm. ohm. No>HoooH nos HoonmoH wo annoem och .w Hmo. Nme. NNo. oNN. woo. Noonoonoo oe nos mmnHoo ooH .N . . . . . NHooo nos mow Now Noo mwm oHo wnHwnoHoo mo omnom one mmonomoHo och .o . . . . . NoonHooo wos Nom Noo owm HNo mmoano nos sHHmonoc one anooo so: .m wNm.H emm. HmN. mme. wNo. NoHcmnoHHmHoH Hmnxom Hnos .e oww. oee. oNo. Hse. owo. NoonoHHonxo nos noHHoommo ono o>oH och .w . . . . . Noeoc mHo Hom oos Nwe eoo Hm oo OH mHoom HH oHomHHomeoo so: .N meo. Now. Nost Ho onmomnc Hnos .H Hano ooHoonnoo nos mH omHH sHHEmm Hnos Hnoom Hoow nos oHnos so: memH u 2 .II memH u .1! oNN u 2 H H HHooHo xnmmo oHomHHm> ooanHm>ou N M. N M .nozuiomeHHm: mo nOHuonHo>m cHHs och sHHEom mo mnonnoEHo mo mnoHHmnHm>m o>HHoomm< How moonoHHo>ou one mHnoHonmoou noHHmHoHHou nomHooouu.Nm oHomh 172 .muno wnHmmHE cuHs mommo HHo moumnHEHHo noHHoHoo omHsHmHom moN. mme. Hm omHH_ oHocsumummuomHH Hnos .NN oow. ewm. eos. mwe. moo. _m mHHEom_ oHOcsnmlmo-omHH-sHHeom Hnos .HN . . NnoHoHHco NNo moH Hnos cHHs onoom nos oEHH och .oN . . Nsohno on oso mNN nos moHonno HH mooow HmHHonE och .oH . . Nomn HonOmHon eoo oom How oHooHHm>m sonoe mo unnoem och .wH .Ho owo m nonm NHH. Nem. s o go o o . HHEom ocH oEHH wo unnoen och hH HNH. wem. Noomn mH sonoe sos och .oH eoe. mNH. emm. meH. wNw. Noose ooo moonHooo so: ooH .mH . . . . . NmHooEoE New mmH eom moH eoe sHHeom mo mmoanonoc Hmnune och .eH Hano ooHoonnoo nos NH omHH sHHEmm Hnos Hnoom Hoom nos oHnos so: oemH u 2 II. memH u 2 ll. oNN n 2 H H HHoohm cnnmo oHooHHm> ooanHo>ou N H. N H .oonnHonoo--.Nm oHooH 173 Table 33.--Pearson Correlation Coefficients fOr Women's and Men's Frequency of Resources Received and Affective Evaluation of Marriage. Women Men Variable 2 2 r r r r How often does your mate: 1. Make you feel like an important person? .421* .177 .359* .129 2. Tell or show he/she admires and .409, .167 .365* .133 respects you? 3. Let you know he has confidence in .380* .144 .348* .121 your abilities? 4. Tell or show you his love? .423* .179 .426* .181 5. Let youoknow he/she enjoys your .400, .160 .374, .140 company. 6. Enjoy a laugh or joke with you? .391* .153 .394* .155 7. Give you hug or a kiss? .463* .214 .385* .148 8. Do an errand for you? .311* .097 .304* .092 9. Make himself available to do work .257, .066 .237, .056 for you? 10. Do something to save you energy or .310, .096 .266* .071 make you comfortable? 11. Give you some new information? .309* .095 .175* .031 12. Give you his opinion? .221* .049 .041 .002 13. Give you something you need or want? .322* .104 .297* .088 14. Give you money for personal use? .196* .038 .088 .008 15. Help you solve a problem or make a .300, .090 .207, .043 deCiSion? 16. Support you with disc1p11ne and .244, .060 .163* .027 guidance of children? *p < .05. 174 Table 34.--Pearson Correlation Coefficients for Women's and Men's Frequency of Shared Time and Activities with Mate with Affective Evaluation of Marriage. . Women Men Variable r r2 r r2 How often do you and your mate: 1. :pepgugime together just the two .306* .094 .313* .098 2. Spend an hour or more just talking? .340* .116 .218* .048 3. Discuss personal feelings? .290* .084 .262* .069 4. Work together on a project? .259* .067 .168* .028 5. Take a drive or a walk? .303* .092 .154* .024 6. Mate time frequency, companionshipa .400* .160 .325* .106 7. Eat at a restaurant? .164* .027 .141* .020 8. Entertain friends at home? .077 .006 .027 .001 9. Visit friends? .064 .004 .097 .009 10. Go to movie or other entertainment? .008 .097 .009 11. Attend a sports event? .102 .010 .063 .004 12. Attend a party? .091 .008 .066 .004 13. Mate time frequency, sociabilityb .157* .025 .116* .013 aVariable created by combining variables 1-5. bVariable created by combining variables 7-12. fp < .05. 175 Table 35.-~Pearson Correlation Coefficients for Women's and Men's Fre- quency of Resources Received from Mate, Created Variables, with Affective Evaluation of Marriage. Women Men Created Variable 2 2 r r r r l. Mate time frequency, companionship .400* .325* 2. Mate love frequency .507* .461* 3. Mate status frequency .439* .378* 4. Mate services frequency .357* .301* 5. Mate information frequency .338* .156* *p_< .05. as your ”husband or wife" which did not represent a resource class, and "closeness and belonging" which would be eliminated if ”love and affec- tion" were chosen. It was therefore decided that ”love and affection” was the preferred indicator because it was less rudundant, it explicitly used the word love, and it would best represent the resource class. This procedure was used to select the best indicator for each resource class. The selection procedures resulted in ten sets of independent variables with variations in resource class indicators and specifica- tions of order. A total of eleven evaluation variables and ten fre- quency variables were used in the analyses as well as the set of five created variables representing shared time and the resource classes of love, status, services and information. The selection procedures and resulting variations were used because of the high intercorrela— tions among independent variables which present problems in the 176 interpretation of multiple regression, and the objective of selecting the best set of indicators fOr predicting evaluation of marriage. Objective 2 Select the best set of indicators to predict evaluation of marriage for women and for men. A summary of the five best predictions is reported in Table 36, criteria for selection discussed in Chapter III. The best prediction was obtained using (1) evaluation variables selected by statistical criteria, followed in order by (2) evaluation variables selected by theoretical expectation, (3) combination of frequency and evaluation variables, and (4) created frequency variables. All of the best pre- dictions were selected by the forward method of inclusion. Research Question 5 Will evaluation, frequency, or a combination of evaluation and frequency variables give the best prediction of marriage evaluation? The evaluation variables were better predictors of marriage evaluation than were frequency variables. The three best predictions were accomplished in women's analyses with evaluation variables. The best prediction accounted for an 81% reduction in variance of the dependent variable, had a mean square error of .356, and included five variables which were significant predictors of marriage evaluation. The best prediction for frequency variables was accomplished in the women's analysis with created variables. Table 36 indicates the high mean square error of frequency variables with few significant predictors of marriage evaluation. There was a slight reduction in . . . 2 . mean square error, increase in adjusted R and F to enter by uSing a 177 Table 36.-—Summary of Best Predictions of Marriage Evaluation by Selection Criteria. Variable Set and SMean Adjusted Highest SNumIEI 0f Selection Method quare R2 F to Enter lg“? icant Error Variables 1. Evaluation Variablesa Statistical Criteria, Spouse Women .356 .81 494.999 5 Men .396 .75 404.937 5 2. Evaluation Variablesb Statistical Criteria Women .462 .75 327.614 6 Men .526 .66 210.903 4 3. Evaluation VariablesC Theoretical Women .607 .68 322.617 4 Men .694 .56 190.036 2 4. Combinationd Theoretical Women .647 .63 287.867 2 Men .739 .53 187.507 3 5. Created Frequencye Women 1.198 .28 64.251 2 Men 1.160 .26 60.240 3 Note: All variable sets were computer ordered. arabies 37, 38, 39. bTables 40, 41, 42, 43, 44. CTables 45, 4o, 47. dTables Appendix Table G-9. eTables 49, 50, 51. 178 combination set of frequency and evaluation variables. However, the combination of variables with different scales creates an additional consideration in the interpretation of results. Research Question 6 How will the variable selection method affect the prediction of marriage evaluation? The best predictions were achieved with variables selected by statistical criteria. The adjusted}:2 in the two women's analyses were .81 and .75 for the variable sets selected by statistical criteria. The evaluation variables selected by theoretical criteria accounted for 68% of the variance and had four significant predictors of marriage evaluation. Research Question 7 Will the order specified by the Foa theoretical structure pro- vide a good prediction of marriage evaluation? The theory specified order did not achieve the quality predic- tion that was accomplished with the forward method search procedure. When the forward method search procedure was used, the love indicators were entered in the first step of the analysis which the theory would predict (Table 45). The status indicator (respect received) enters at second step for women but at the third step for men. "Open honest expression of feelings” (information) precedes the services indicator (”mutual helpfulness of family members”) for both sexes. The theory states that services are more particularistic than information trans- fers; however, in this particular study, the service indicator was not as highly relevant to the husband-wife relationship as was the "open, 179 honest expression of feelings." Results of the theory specified order are reported in Appendix Tables G-1 and G-2. Selection of the best predictors was a necessary step before hypotheses testing. The theoretical model which predicts high inter- correlation among independent variables was confirmed. The high correlations create particular difficulties in the interpretation of multiple regression analyses, therefore it was necessary to select the best predictions for examination of the contribution of particularistic resources to evaluation of marriage. The best prediction of marriage evaluation is reported in Tables 37, 38 and 39. This prediction had the lowest mean square error (women .356, men .397), highest F's to enter, and greatest number of significant predictor variables. The adjusted 32 for women indi- cates the 81% reduction in variance by the set of six predictors. The variable indicating evaluation of mate contributed a 70% reduction in variance of marriage evaluation for women and the other variables contributed little additional reduction in variance as can be seen in the_R_2 change. The adjustedR2 for men showed the set of five predictors explained 75% of the variance in marriage evaluation. The evalu- ation of mate variable contributed 65% to the explanation with minimal additional change in._R_2 made by the addition of the other variables in the set. A problem occurs because of the high correlation between "evaluation of mate” and evaluation of ”love and affection" (g = 0.74 for women, 0.64 for men). Both variables are also highly correlated with evaluation of marriage (see Tables 31 and 32). The use of the 113(1 .oo.n u HoHN.HH H eoHooHo .oN.N u H=_N.m. H ooHonso .mc. v .v_.N u Ho_N.cH a eo_o N. 5 Ha wa. oHH. ch. a... w—c. $3.. 93:. NmH. mes. 09v. Geo. HNc. won. amN. xNH. 53:. mo.. Hes. an. N_=. 93.. sec. Hue. a I . ".|||llllll H:;.n.:oonu ::.ryonuoz one . 7.31:3; NHHHHoaooto mHm. wON. «co—.m .om. cNMh.m wme. .ch.s— oow. «cv~.w an. com—.mh .mm. v~v.m oNN. wNn._ chm. comn.n oow. oomn.o_ w_v. eNon.o ohv. ccmm.mH Hoe. conc._s Nam. HnoHOHHboou nCHmnoHuoz o. no Loan: oHnonnum mo_o=.Hn> ona oun! nnHmmonnz vo~.oHno:aHzn: . iii! I‘D'IIIIIUI ll‘.nl.| ill :anmonoz oHnHH_:I 0 . .ooHoHuoom on: Hoocc .nHHoHHHu HaOHHmHuaHm so oouoo_om on noHHnn_n>: so oun.cnax on no..=:_n>c o>HHuouu< a: :c_Ho_ooH; 9c» no: mHmsHan< nocmmono: o_nHu_nz--.sm oHoah woN. u N: eoomnHe< Noo.- H.:oom=ooo emN. u N: oHoHHHns NNH. ooHHoocca eon noon now. a n oHoHHan o... Hogoonoo on anoHHH mHN. ooo: on Hoocaon oHN nee.nmn HuHoH «NH. oHcmcoHonHoH HnnHon Hnos Hoc=.H Non. oHN nnN.nn HoneHmog one. ocHx Hnos o.NNo.NNH Hmo.Hm m Nm_.mmN ooHnmoouon no: on.- Hoooonnoo. on. a Na eonnHe< ooo. ooHooocHo eon o>oH mHn. . N: oHoHHHn: mHH. eo>Hooot Hooomoc ogH Noo. « z oHoHoHn: mnH. chmnoHunHoH Hanson one» nNH. one. ooo moonHooe so: oHN Nno.noe HaHoH NHN. one; on .Hocuoo Hooo.o omm. oHN NHn.VN HonoHnon eNo. ennomn; HnoH n.nno.om_ woN.em o ANN.NNH :onnutuua mommy Hooon unwumdwwwuw oHonHHa> H oHanow .u.= monnom nnHomHHn> . . . . nno: ho anw Ho ooHnow HonaHH2> so mHms_::< ..n .1....ud.¢l.l3 H.......-.¢. ..H VIVIHI 181 .m9. V Wk .mHmsHmnm nonmonoH EOHw mHo>oH oonmonHanm so ooHooHom moHomHHm>n Hm>0oo oonoHHooxo nos Nowno.emH NHHoo. Nome. NHmoo. onNm.H noHoooHco ooo o>oH ooo .o o>HoooH nos omNmm.ewH oeooo. meeHN. NeNoo. .ooHeo.oH omwmwwwmwo Honono onH .m Neon.NHN Honoo. momow. eNNow. :Nommm.o onnmnonoomwmoflwmwwwmwmww .e m ME ememH.oNN HmHHo. mNooN. NmNow. *NoooN.HH oHonmemmmmmzmwz wnH .m HmmoH>nmmo HNmoo.wwm mNoNo. NoeNN. oommw. :NomHo.mw ooo; on on on oHooo oH oHooooocsoo no: .N oNooo.eoe NHwoo. NHNoo. ommmw. .oNooo.eoe eooomnn Hnos .H n— HHQHO>O owH—MED mm mm m OHQMHHSE houcm O» m mHoOHQHHHm omnmhhm> moum .nonos -uoonHoonw Hon HooHo .oo>Hooom moOHnOmo: one ooo: mo noHHman>m o>Huoomm< so omeHHmz mo nOHHonHm>m o>Hoooww< mo noHHoHooHo ocH How mHmsHmn< nonmonom oHnHHan mo sHmnnnw--.wm oHooh 182 .8. 0 .HH. .mHmanno nonmonoH EOHM mHo>oH oonmonHanm so oooooHom moHomHHo>m Hm>ooo oonoHHonxo nos omNNo.NNH moooo. NommN. onon. .NNooH.m noHooooco ooo o>oH onH .m ooHNe.omH HmNoo. NNNeN. oweow. .nmeH.o anHHommmmmmwmmmuwommw .e wNwHN.HoN ooemo. omoeN. omoow. .eemNN.NN MMWWW>wmeHMmmmowmowonmm .m eeNoo.mmN NNHmo. oomoN. NNoew. .mHNoe.Nm oHnmooHooMmmowwmmwmmwmww .N owNmo.eoo mNemo. mNemo. mwwow. .oNNmo.eoe ooHn Hnos . .H n HHoHo>o ownmcu N: No : oHnHHan Hounm o» n oooHoHnm oHonHHo> noum .Ii‘! 5.. 1 .noznuoonHooow Hon HooHo .oo>Hooom moOHnOmo: one one: mo noHHman>m o>HHoomm< so oonHHmz mo noHHmnHm>m o>HHoomm< mo noHHOHooHn ocH How mHmsHmn< nonmonom oHnHanz mo sHmnnnw--.om oHooh 183 "evaluation of mate" in the variable set causes "love and affection" to be added to the equation at the last step where it is no longer a significant predictor of marriage evaluation for women. Even though some of the predictive power is lost, the addition of "love and affec- tion" instead of "evaluation of mate” facilitates testing the theor- etical model. The decision was made to eliminate "evaluation of mate" from the variable set for purposes of hypothesis testing. The results of removing "evaluation of mate" from the analysis are reported in Tables 40 through 44. The set of independent variables accounts for 75% of the variance in marriage evaluation by women and 66% of the variance for men. This is a decrease in the adjusted R2 of 0.06 for women and 0.09 for men. There is also a decrease in the highest F to enter for both sets of subjects, but an increase in the number of significant predictor variables for women. "Love and affection" now has the highest F to enter and contributes 60% to the explanation of marriage evaluation of women. The men's analysis (Table 44) indicates "love and affection" is a significant predictor of marriage evaluation but contributes only 2% to the 53 change when added to the equation at step three. The "way decisions are made" was an information indicator selected for the analysis because of its high correlation with marriage evaluation for women (5 = 0.68 for women, 0.38 for men). Tables 40 and 43 show the analysis for women with the same set of predictors used for men. Tables 41 and 42 indicate the results of the women's analysis when adding the "way decisions are made." Despite the attempt to eliminate redundant indicators, the correlation matrix (Appendix H) 184 owH..- 23.. ~._.H_ 4H .71... ::_rrocxoz TON—1:32.... Hr s.. _..._ n H...», 3.3....1...H.> :H.. H...._..> _ .H.. .on.m u HmHN._H H eoHnnho oHo. em. one. HHN. com. on. New. oww.- :nn. n_.m oHo. n:_. «no. .Nq.n NHH. owH. Hon. .xN.NH ooo. .NN. non. .Hm.o_ Nae. Hx_. o .o_.=o Noo. one. :nN. mo._ nmN. NHN.- NxN. o... now. New. Hon. .H_.n .14. mo_. :nn. .No.n_ Nxo. cmN. :nn .oo.nH nNm. _n_. = .mo..m oHo. NHN. Non. .Nn.o Nmo. Hn_. H=Hon .:n_HH:.o:u 9 2E. .1 ::.rrnza;z 3 u .1.. .29— 3.5.9:; :CHILUH2D: 19M _—v.__ 319:“ 5:2: sHHHHonan; 3: Ho .3... 1... 9.7:: :1. :SHIxocunz o.;_H_:2 .x1HHH1I .: ::._r:_1>c 22:: .._ H535. .Hnonnno. oo>HoooH Honors: nn.nnonnno HmonOc non: HHnnuwnOOH nonmocnxo Hnonoc non: HocHox:~ no wwanh . .._.N . HH_N.oH H noHHaHn Noo. " .n nozanne . HNo. . N: oHoHHan nNn. u n oHn.H_nz Hocuowou oo muncch :o_u99usa on: o>=c .NN .mm.mem Haunh chmnnHHnHoH Hanna: H==:.. on. mHN mn:.n__ _anoHao: 98:: an HHouEnu oooeo.ns o_s.on o o_m.NmN nnHmmoHao: mum NoN. u N: eoHnan< omN. n .n oHoHHan s now. u n oHoHHan oioc Ha HHOueou oo>HoooH Hanson: oHN st.o:e HaHoo anznnHH=HoH Hannow H:=:.. ooe. NHN ooo.oo .ononoz nnHHnoows on: o>Oc now—w.wn_ owo._w o wNm.o=m nnHmmonox mwamz a 1HH1 oHnnom . . monnnow noHHnHHn> .Ho. .> _ can: a n on In: so oocnnw nonsHH=> on mstHnn< .oo.:.on;: .3: HooH: .oHHHHHHH HoaHHAHHnH: .H>.HH.H,:< .7. :3. .oH—5..— .HH: .3. /.rx_H.:< Elf/(:23: D_H_.~_:Z--.H=. .H_£n._. 185 .ow.m u _N_N.__: ooHcahc .o:.N n .NHn.h_ : oo_onhs mm. . TEN SN. 35.. . - :nnunnnu. Ne... 25. 3... N3. in: 5.323... 7.2.2. 2:. non: 3N. . No. 3.921 2:. £5. .oNHHé o3. 2:. ..of...a3 on LEE... 3N. u N: 25:2. :2. Non. .HoN.o moo. NHN. 3.9. 3 :ocsoo HNHH. u x 232:: H... N3. 23.... Non. o.... 2...... o: £332. .3: as_. _:6. .xmh.=_ chm. ::_. oo>HounN Hungry: :HN Nwe.o=v Hnuch :oN. one. .cm_._n one. oeN. ochnoHHn_oH Hanna: H==:.H Noe. ~_N moo.ho Hanono: oN_. eon. .mee.n :mo. ew_. :c_Hooo:: on: o>oH .mne.wo an.ee N NNw.o=m nnHmroHaoz muss! Hno.n_oHoou Hno.oH:o“nu ooHMHHNMcH 7. . . 7. :n...2.o..woz 3:239... c... 5......3HMHHHHH. 2.33....5: Scat; ...._ .oHHHMHonzw .no oww....“nm ”MHMMMMMN .HN. :3 =3 . «a ._ 1 ~.f 1.» 35:... “T. :n.z:onwsz ooNHocnonnan: no.11oczoz o_L.H_nI «.HH—:2 oonnHHa> he meann< 4 y ..H.- ..u _::oHn;;: H3:.THCH..;1:2 r::.r.n;: u:_EHT:: .nHHoHHHJ HauHHAHHon so ooHHo_om no_on_cn> ::.H::_:>c sc ouoHHHez on ::.H2:_1>: o>HHooos< s: ::HHn.o;H; cc. H3; 7.as_n:< nnHzmono: o_:_H_nz-u._e oHoah I150 186 .mo. \ n. .nHHoHHHo HnOHHmHHnom so ooHooHom moHomHHo>Hw HonH<2nonHo oweoo.mo Hmooo. monmN. NNHNN. HmNee. mnnHHooc moonoxo nos sHomonoc one anono so: .N H H Ho owo mmon.HHH eoeoo. emme. eoONw. .memmm.e HHzHHomHMMmewnnuw on“ .o n . GEO HQ 0 OH eeNoo.omH eHHHo. oommN. onow. .owNNo.o Mm“w“>mwmeooomocsooozo: .m H . . 0 GE eowmo.emH NNHNo. oeNeN. ooHow. :NeHmH.wH ononmwwmqwmwzmwn moH .e o>HoooH nos omon.mmH Nmmeo. mNoNN. Nowew. :Nommo.mm ohmmmwwmwo Honono onH .m mNmNN.mNN woeNo. memNo. ewHNw. :NNNNH.om oHnoooHoonwoflwuwwwmwmww .N . . . . . Hm>0co oonoHHonxo nos Nmeo NNN meooo meooo oweNN .NNNHo NNm nonooocco ooo o>oH onH .H mm HMNHO>O OMCNSU NM Nm m Omn—MHHDZ hwucm OH L GUOLQHCM ®~DNMRN> QQHW .noeoz--ooHuHoonw Hon HooHo .oooz mnonHooo wnHonHonH .oumz EOHm oo>Hooom moOHnomom mo noHHmnHo>m so omeHHm: mo noHumnHo>m o>Huoomm< mo noHHoHooHo ocu How mHmann< nonmonom oHoHHan mo sHmEEnw--.Ne oHooh 187 .m0. v m. .nHHoHHHo HnOHHmHonHm so oouooHom moHonHHn>m HonHnmmo ooo; on on oo omooH mmH NmoHo oNoeN omoow .NHomH oH mHooo HH oHoooHoHeoo no: .e . . . . . HmnHeHmo osHoooH nos omon me Nmmeo mNONN Nowew 4Nommo mm oooomoo co Honoso ocH .m . . HmmoH>nwm-m>oHo mwaN.mNN woeNo. memNo. ewHNw .NmNNH om oHnmnoHonHoH Honxom Hnos .N . . . . . Hm>oHo oocoHHooxo nos Nmeo NNN meooo meooo oweNN «Nmeo NNm :oHoooooo ooo o>oH ooH .H m HHGHON’O owH—NED N”— Nm m OHQMHHDZ hvucm OH & ofiwhmucm mmnmwhm> moum .noEosuioonHoonw Hon HooHo .oun: Eon oo>Hooom mooHnomo: mo noHHmnHm>m so omeHHmz mo noHHmnHm>m o>HHoomm< mo noHHoHooHn ocH How mHmsHmn< nonmonom oHnHHan mo anEEnwuu.me oHomh 188 .nHHoHHHo HnoHHmHHnHm so ooHooHom moHomHHo>Ho .mo. v m. HmnHoHo oonoHHooxo nos Newwe mmH moNNo ooomo wNooN tomeeH eH noHooooco ooo o>oH onH .m Hmmmo.mNH Hwan. mNNNo. NoNoN. .oeHNN.Hw onnmoonoomwwowwmwmwmwmww .N OED m 0 O memoo.oHN eeowe. eeome. ooooo. .memoo.oHN meww>mwmeooomownooonom .H "— HamHO>O OMGNSU Nm N: m QHQHHHDZ hwucm Cu L afithHfim odnmhhm> moum .noZuuoonHoonw Hon HooHo .oum: Eon oo>Hooow moOHnomo: mo noHHmnHm>w so oonHHoz mo noHHmnHm>m o>HHoomm< mo noHHOHooHn ocu How mHmsHmn< nonmonom oHoHanz mo sHmnnnwu-.ee oHomh 189 shows values ranging from 0.47 to 0.72 (love and status) for women's variables. Objective 3 To investigate the credibility of Foa and Foa's resource exchange theory in predicting evaluation of marriage. The credibility of the theory was studied using three sets of independent variables: (1) evaluation variables selected by statis- tical criteria reported in Tables 40 through 44, (2) evaluation vari- ables selected by theoretical criteria reported in Tables 45 through 47 and (3) frequency variables reported in Tables 49 through 51. Separate analyses were conducted for men and for women on each set of independent variables. The hypotheses have been written for resource classes which means that each hypothesis for evaluation variables actually is a combination of fOur separate hypotheses--statistical and theoretical variable sets with separate analyses for men and women on each variable set. Hypotheses 8 Affective evaluations of particularistic resources received do not significantly contribute to the prediction of marriage evaluation for women or men. The null hypotheses that all beta coefficients were equal to zero were rejected for women on the statistical variable set, F (7,212) = 95.405, p < .001 (Table 41) which includes "decisions made"; and the set which does not include "decisions made," F_(6,213) = 108.818, p < .001, (Table 40); and the theoretical variable set, F (6,209) = 75.822, p < .001 (Table 45). The adjusted R2 values of .75 (Tables 40 and 41) and .68 (Table 45) indicate the percent of variance accounted for by 1590 lll.‘l1.ll'.lll.cul. .I . I'll.-. III.“ I!!! III‘ 1 . 'n 1...: .1: I 1.1.10 11:1- :1 I.I.II 11.1.01. 1... In... I . tillil Ii .| .1 0 I i 4 u :. .olll i!!! ill-"Ill.l|nllll i .41.!)‘1! .nn.m u ooe.- osx. moo. no.. N:_. . NoH. eon. coo. ewe. one. wNo. was. ooo. oNN. NH..- Noo. ooH. ooo. oow. ooo. :_c. o H:;_u_..o:; ::_rzonm;: 1.4“. ._H._ 11...... ~f. sH_HHonooHn HnoN._. n esHEnHo so.e nem. ewo. wN. ooo. NNo.- .N._ owe. one. No.N see. enH. .e=.oN ooo. oNH. «mo.oV who. eoe. n==.. soN. woN. N... m.h. weN. she.. 93:. ewo. ..wm.h mom. mm..- .on.m. :Nm. ooN. cone.w. woo. eNN. .Nce.NN c.:. enm. Hs.sn ooe....Hooo - . 5122.123. 192...: Hz... . H:;.n. .ne 9. on Hons: .o. ; H....z...:u...z ononnum one—onnflnnHrnz .oN.N u .oHN.wH : oo_onho .e..N u .ooN.oo H ooHoan one. u N: noohnHe< ..nnHznoo. ooo. n oHnnom z whon.nuo.oc Hanan: mmN. u x o.oHu_nx own .nnoIHon How sons: oosHoooH Hoonmoz mHN nee.wnm .auoh eoHnmotono omens; non: Hnon.H one. oHN NoN.meH HoneHhoz nnHHoouua on: osno o.on.mm omm.wm m noo.No. nonmonoz no: Hunnanoo. oNo. u N: ooumnHo< Om: .onnmcon Hos sonoz mwo. u ocnnom z meson .oHHoHoz nNn. u n oHoHHan amoanon.oc .nan: nOHmmoano Hnonoc none mHN Nmo.m:e .ath nosHooo. Hooomo: Hcoo.o Nee. ooN NNx.oN_ HoneHhon eoHHoocce one one. n.NNn.mN NNe.oe o me_.eNN elosocnon Mofimx . oHnnnm . . mocannm :cHHnHH=> o_osHHn> : nnoz H... oH.=!m .o sonnet n:.rzocuo: ;_&.H.:Z oo:nHHn> so ers.nn< .oo...ooom Hz: coon: .::_H1Huo:xc HenHHoHnoo. so ooHuo.on ro.o=HHn> nOHH=:.1>o so n21HHHnZ .: ::.Hnn.=>c o>HHouo.< so n:.HsH7;H; no. H3: m.rs.=:< ncanHonx n.:.H_nZ--.m9 n.o:. 191 .m3. V mo .mHHoHHHo HnoHHoHoocH so ooHooHom moHomHHo>n Hsmzozo own oHNNw.mN wHooo. Hwao. Nsst. loHH. HmnOmHon How oHomHHm>o sonon mo Hnnonm och .o Hmooooo mNNem.Ho moooo. momwo. Nost. ehwmo.o :30 oH nos moHonno HH mooow HmHHonE och .m HmmoH>Nmmo HNomo.oHH mhwoo. womho. wHNNw. smomoo.m mHoonoE sHHeom mo mmoanmnHoc Honunn och .e HonHHoooH nos HoonmoH mo HnnOEm och .N loHo.eoH oemeo. Nooeo. oHeow. «Nosoe.NN Hm>0oo oonoHHonxo nos NHNHo.NNm mNHoo. HNHoo. NmmNN. .NHNHo.NNm eoHoooooe one o>oH onH .H n HHmHo>o ownocu N: N: a oHnHHan Houn: OH n nooHoHnm oHonHHn> non .nonosu-oonHoonw Hon HooHo .oHn: EOHw oo>Hooow moOHnOmo: mo noHHmnHo>m so omeHHoz mo noHHonHm>m o>HHoomm< mo noHHoHooHo ocH How mHmann< nonmonom oHnHHHn: mo sHmEEnwuu.oe oHooh .mO. V N... .oHHoHHHo HnoHHoHoocu so oooooHom moHooHHm>n 192 HmnoH>nmmo oewNm.mm moooo. omoom. omems. meNo. mHooEoE sHHEom mo mmoanmnHoc Hmnunn och .m Hsmzozo own HmnOmHon omst.oo wemoo. Hmoom. wmems. mHmoN.H Hnos How oHomHHo>n sonoe mo Hnnono och .e o>Hooo nos moooo.No Nomoo. memom. NNNmN. NHNHo.N HmnHoH ocH .H n HHmHo>o ownmcu N: N: m oHnHHan HoHnm cu m mooHoHnm oHoMHHn> non II||.I II II III II III.IIIII .noZIuoonHoomw Hon HooHo .ouw: Eon oo>Hooom moOHnomom mo noHHmnHm>m so ownHHan mo noHHnnHo>m o>HHoomw< mo noHHoHooHo ocH How mHmann< nonmono: oHnHHHn: mo anEEanI.he oHonh 193 the set of independent variables. Evaluation of particularistic resources does contribute to the prediction of marriage evaluation fOr women. The null hypotheses that all beta coefficients were equal to zero were also rejected for men, F (6,215) = 73.646, p < .000 (Table 40) for the statistical variable set and F (5,210) = 55.528, p < .000 (Table 45) for the theoretical variable set. The adjusted R2 values of .66 on the statistical set and .56 on the theoretical variable set indicate the percentage of variance accounted for by the set of independent variables. Hypotheses 9 Frequencies of particularistic resources received from mate do not significantly contribute to the prediction of marriage- evaluation for women or men. The results of statistical tests for hypotheses 9 are reported in Table 49. The null hypothesis that all beta coefficients were equal to zero were rejected for women, 5 (5,193) = 15.806, p.< .001; and for men, F (5,201) = 15.333, p < .001. This set of independent variables contributed a 27% reduction in the variance of marriage evaluation for women and a 26% reduction in variance for men. These adjusted 2 . . . . . ‘R_ values are con51derably lower than the predictions with evaluation variables. Objective 4 Investigate the contribution of each interpersonal resource and shared time to the prediction of affective evaluation of marriage for women and men. 194 Results of statistical tests for individual beta coefficients for evaluation variable sets are reported in Tables 40 and 4S, frequency variables are reported in Table 49. Hypotheses 10 Evaluation of love and affection does not significantly con- tribute to the prediction of marriage evaluation for men or for women for the statistical or theoretical variable sets. The null hypotheses that all beta coefficients for love were equal to zero were rejected for men and women on both variable sets. The statistical variable set fOr women, F (1,213) = 9.37, p < .002; .5 (1,215) = 12.28, p < .001 for men (Table 40). "Love and affection” had a standardized beta ranking third in magnitude after "comfort at home" and "sexual relationship" for both sets of subjects. The theoretical variable set reported in Table 45 indicates the probability that the regression coefficient is equal to zero is small, 5 (1,209) = 72.462, p < .001 for women and F (1,210) = 46.69, pp< .001 for men. The standardized regression coefficients for "love and affection” have the highest absolute value of all variables for both men (.434) and women (.510). Tables 42, 43 and 46 indicate that "love and affection" enters the regression equation first and accounts for 60% of the variance in marriage evaluation for women. Table 47 shows "love and affection" accounts for 47% of the variance in marriage evalu- ation for men in the theoretical variable set, but contributes only a 2% R2 change when added to the equation at step three in the statistical variable set (Table 44). 195 Hypotheses ll Evaluation of respect received does not significantly contribute to the prediction of marriage evaluation for men or for women for the statistical or theoretical variable sets. The null hypotheses that the regression coefficients for "respect received" were equal to zero were rejected for women, _F (1,213) = 10.94, p < .001 in the statistical variable set, and g (1,209) = 18.409, p < .001 for the theoretical set. Standardized regression coefficients for "respect received" in women's analyses were second in absolute value (.260) for the theoretical set following "love and affection" (.510) as theory would predict; and fourth in absolute value (.181) for the statistical variable set following "sexual rela- tionship" (.271), "comfort at home" (.205), and "love and affection" (.190). Table 46 (theoretical variables) indicates "respect received" is added to the regression equation at second step in the women's analysis and contributes 4% to the R2 change. The statistical variable set (Tables 42 and 43) indicates "respect received" added to the equation at step three with the contribution of 4% to the R2 change. The F tests for men failed to reject the null hypotheses that the regression coefficients for "respect received” were equal to zero, .F (1,215) = .84, p < .360 for the statistical variable set; E (1,210) = 2.62, p.< .107 for the theoretical variable set. The standardized regression coefficient for the theoretical variable set (.107) was third in absolute value following "love and affection" (.434) and "open, honest expression of feelings" (.304). The standardized regression coefficient for the statistical variable set indicated 196 ”respect received" to have the lowest absolute value of all variables (-.054) with a negative value indicating an inverse relationship between evaluation of ”respect received" and evaluation of marriage for men. Table 47 for the theoretical variable set indicates "respect received" enters the regression equation at step three, but con- tributes only 0.06% to the R2 change. The statistical variable set indicates "respect received" enters the equation last and contributes only 0.01% to the R2 change (Table 44). Hypotheses 12 Evaluation of services received does not significantly con- tribute to the prediction of marriage evaluation for men or for women for the statistical or theoretical variable sets. Three different indicators were used to represent the services resource class. The indicators of services used in the statistical variable set were: ”how comfortable it feels to be at home," and "your sexual relationship." The indicator of services for the theor- etical variable set was: ”mutual helpfulness of family members." The use of three indicators with separate analyses for men and women involves testing six hypotheses for the services resource class (in contrast to the four hypotheses for the love and status resource classes). These six hypotheses have been combined and stated as three. Each of the three hypotheses represents one services indicator, one variable set, and a separate analysis fOr men and for women. Hypotheses 12A Evaluation of sexual relationship does not significantly con- tribute to the prediction of marriage evaluation fOr men or for women in the statistical variable set. 197 Tables 40 and 41 indicate the results of statistical tests for significance of the individual beta coefficients for variables selected by statistical criteria. The null hypotheses that the beta coefficients for "your sexual relationship" were equal to zero were rejected for men, 5 (1,215) = 14.51, p_< .001; and fOr women, F (1,213) = 31.45, p_< .001. The standardized regression coefficient for women (.271) is greater in absolute value than for any of the other variables, and ranks second for men (.227) following ”comfort at home" (.356). Tables 42, 43 and 44 indicates "your sexual relationship" enters the regression equation at step two in both women's and men's analyses but contributes 7% to the R2 change for women and 14% to the R2 change for men. Hypotheses 12B Evaluation of how comfortable it feels to be at home does not significantly contribute to the prediction of marriage evalu- ation for men or for women in the statistical variable set. The null hypotheses that the regression coefficients for ”comfort at home" were equal to zero were rejected for women, F (1,213) = 13.42, p < .001; and for men, 5 (1,215) = 40.19, p < .001. The standardized regression coefficient for men was the highest absolute value of all variables (.356) and for women it ranked second (.205) following "sexual relationship" (.271). Table 44 indicates "comfort at home” enters the regression equation first in the men's analysis and accounts for 49% of the variance in marriage evaluation. Tables 42 and 43 indicate "comfort at home" contributes .01% to the R2 change in the women's analyses. The variable is added at step four (Table 43) and is changed 198 to step five (Table 42) when adding the variable "the way decisions are made." Hypotheses 12C Evaluation of mutual helpfulness of family members does not significantly contribute to the prediction of marriage evalu- ation for men or for women in the theoretical variable set. Table 45 reports results of statistical tests for individual beta coefficients for variables selected by theoretical criteria. The null hypothesis that the regression coefficient for "mutual help- fulness" equals zero was rejected for women, §_(l,209) = 7.381, p < .007. The standardized regression coefficient ranks fourth in abso- lute value (-.137) following "love and affection" (.510), "respect received" (.260) and "open, honest expression of feelings" (.193). The negative sign indicates an inverse relationship between evaluation of "mutual helpfulness" and evaluation of marriage. Table 46 indicates that "mutual helpfulness enters the regression equation at step four, but contributes only 0.09% to the R2 change. The null hypothesis that the regression coefficient for "mutual helpfulness” was equal to zero for men was not rejected, .5 (1,210) = .23, p'< .879. The standardized regression coefficient ranked fifth in absolute value (-.009) after all other variables, and had a negative value which indicates an inverse relationship with evaluation of marriage. Hypotheses 13 Evaluation of information received does not significantly con- tribute to the prediction of marriage evaluation for men or fOr women for the statistical or theoretical variable sets. 199 The information indicator for both theoretical and statistical variable sets was: "how openly and honestly you can express feelings." The additional indicator of information added to the women's analysis for the statistical variable set was "the way decisions are made." The presence of two different indicators requires clarification of the two separate hypotheses. Hypotheses 13A Evaluation of open, honest expression of feelings does not significantly contribute to the prediction of marriage evaluation for men or women on statistical or theoretical variable sets. Table 45 indicates the null hypotheses that the regression coefficients for ”open, honest expression of feelings” were rejected for men, 5 (1,210) = 24.04, p < .001; and for women, 5 (1,209) = 13.319, .p < .001 for the theoretical variable set. The standardized regression coefficient in the men's analysis lfOr ”open, honest expression of feelings” had the second highest absolute value (.304) after ”love and affection" (.434). The women's analysis indicated a value of .193 which was third in absolute value following "love and affection" (.510) and "respect received” (.260). Table 47 indicates that ”Open, honest expression of feelings” is added to the regression equation in the men's analysis at step two and contributes 9% to the R2 change. The variable is added at step three in the women's analysis (Table 46) and contributes 2% to the 2 R change. Table 40 indicates the null hypotheses that the regression coefficients for "open, honest expression of feelings" were equal to 200 zero were not rejected for the statistical variable set fOr women, If (1,213) = 1.14, p < .287; or for men, F (1,215) = 3.10, p_< .080. The standardized regression coefficient for women had the lowest absolute value (.050) and for men was second lowest (.100) pre- ceding "respect received” (-.54). Hypothesis 138 Evaluation of the way decisions are made does not signifi- cantly contribute to the prediction of marriage evaluation for women in the statistical variable set. Table 41 indicates that the null hypothesis that the regres- sion coefficient for the "way decisions are made" is equal to zero was rejected for women, 5 (1,212) = 4.426, p.< .037. Examination of the standardized regression coefficients indicates the value of .113 ranks fifth in absolute value; however, there are two variables with standardized regression coefficients of .179 (love and affection, respect received) and one with a value of .178 (comfort at home). The evaluation of "open, honest expression of feelings enters the ’3 regression equation at step four and contributes 2% to the 3‘ change. Hypotheses l4 Evaluation of shared time does not significantly contribute to the prediction of marriage evaluation for women or men in the statistical variable set. Table 40 reports results of the statistical tests for these hypotheses. The indicator for evaluation of shared time was: "the things you do together." The null hypotheses that the regression coefficients were equal to zero were rejected for women, F (1,213) = 9.11, p < .003; and for men, 5 (1,215) = 8.42, p < .004. 201 The standardized regression coefficient for women (.148) ranked fifth in absolute value; the coefficient fOr men ranked fourth in abso- lute value (.152). The variable was a significant predictor of marriage evaluation for both men and women but contributed minimally (1%) to the R2 change for both sets of subjects. A summary of the contributions of evaluations of interpersonal resources received is reported in Table 48. The theoretical variable set for women closely matches the order predicted by theory but there is a reversal in the order of importance for services and information resource classes and a negative regression coefficient for the services indicator. ”Mutual helpful- ness of family members" is not the most relevant indicator of particu- laristic services exchanged between husbands and wives. The theoretical set for men indicates different results. "Love and affection" and "Open, honest expression of feeling" are important predictors of marriage evaluation followed by "respect received" which is third in magnitude of importance. The variable indicating "goods" did not meet the statistical criteria for entering the regression equation. The statistical variable set indicates the importance of evalu- ation of sexual relationship for the prediction of marriage evaluation for husbands and particularly for wives. Differences between men and women are evident in the importance of "respect received” and the "way decisions are made" in predicting evaluation of marriage. The following hypotheses discuss the contributions of the frequencies of resources received from spouse to affective evaluation 202 Table 48.—-Summary of Standardized Regression Coefficients for Statis- tical and Theoretical Variable Sets for Women and Men. Variable Women Men Theoretical Variable Seta Love and affection (love) .510* .434* Respect received (Status) .260* .107 Open, honest expression (Information) .193* .304* Helpfulness of family members (Services) -.137* -.009 Material goods you own (Goods) .086 Amount of money for personal use (Money) .025 .065 Statistical Variable Setb Your sexual relationship (Love-Services) .268* .227* Love and affection (Love) .179* .215* Respect received (Status) .179* -.054 Comfort at home (Love-Services) .178* .356* Way decisions are made (Information) .113* Things you do together (Shared Time) .106* .152* Open, honest expression (Information) .032 .100 Note: *Indicates rejection of the null hypothesis. aTable 45. bTables 40, 41. 203 of marriage. Statistical tests fOr individual beta coefficients for frequency variables are reported in Table 49. Hypotheses 15 Frequency of love received from mate does not significantly contribute to the prediction of marriage evaluation for women or men. The null hypotheses that the regression coefficients for love frequency equal zero were rejected fOr men, F (1,201) = 25.19, p < .001; and for women, F (1,193) = 12.78, p < .001. The standardized regression coefficients for love were the highest absolute value fOr both men (.490) and for women (.336). The R? values reported in Tables 50 and 51 indicate ”mate love frequency" contributed a 26% reduction in variance for women and 23% reduction in variance of marriage evaluation for men. Although the-R2 values were much lower for love frequency predictions than for love evaluation predictions, the love resource class indicators had the highest F to enter for both men and women. "Love frequency" contributed the most significant and meaningful explanation of marriage evaluation variance for the prediction reported in Table 49. Hypotheses 16 Frequency of status received from mate does not significantly contribute to the prediction of marriage evaluation for women or men. The null hypotheses that the regression coefficients for mate status frequency equal zero were not rejected for women, F (1,193) = .25, p < .615; or for men, F (1,201) = .23, p < .633. The standardized regression coefficients (.045) for women and men (.045) indicated a rank of fourth in absolute value for a five-variable set. 204 .Nn.n u HmoH._. o eoHonHo .oN.N u H_:N.mo n es_neHn .oN.N . HHe_.m. H no_oeHs an. . N; noomnse< n m=.nNe_ oeH. NoN.m Hzenohsoo. oNN. . N; oHoHHan Hen. Nme. =N. sum. .oN. soeonsotc mooHstoh one: HNo. u x oHoHHHn: men. men. HN. .oH. NnN. soeonnoec nnHeHn one: oo_. one. .oH.e one. _n_. osoeoneotc on.. one: onN eNo._NH HnHoH ooN.- Non. .H=.=H HHo. on..- soeonootc nsHHnneoan one: H=e=.. no... HnN oNn.HNN HonoHroz nae. ooo. .oH.mN oHo. ooN. soeonnotc oso. one: n.men.m_ nnN.NH m Hon.nn eoHomotnon 3m NNN. u N: noonnsn< n eon.nmn .wH. eNe.o Hoenoheoo. HoN. u N: oHoHoan one. ooo. HN. moo. one. soeonnoec eoHon-toceH one: new. I n oHoHHan Hen. HHn. HN. new. Noe. soeonnotc eanHm one: Hon. eoe. om. oHo. Nee. soeonnotc hooHscom one: no. mo.mNH .oooH No_. Noo. .Nm.N ooe. eN_. osononnoNu oeHH oan .o:o.. wo... moH eN._wN .enonox own. one. .nN.NH eNo. oeH. soeonnotc one. one: o.eon.m_ amo.nH m oo.en elonotnon own“? . .. HnsnoHHHoon ...J.on.. “Hwhwwwwwyu sH...oaooHn : n:_mvoHuwz Hrhfluu“wuhu o_onHHn> : oHnnvm .o.o moHanom :nHHNHHa> ooN.—.2555: o .H: .HnHH... nirffiuox nnox .Ho :5 .Ho ooHnow oHntnnHm ooNHonoont.r=: ::_zonaoz o.:.~.nz ans: oo>HOnox rounnczoz on sononnoso so o2=HHszz .: :nHH=:_:>: o>..;e.:< on ::.~n.onH; oc. onnnHHn> mo mHms.nn< .no.o=.cn> nounoHu .anz Hz. [Hms.n:< ::_mmocxo= o.nHH.:z--.oe o.onh 205 .8. v N. .mmoHo oOHnomoH como How mHonoHonH wanHoEoo so ooumoHo moHomHHe>n somow.mH whooo. NwooN. oowmm. oNNHN. sononoon noHomEHomnH own: .m mwews.oH omHoo. msowN. NNwwm. mswmm. sononoon mnHmHm ooo: .e eHwem.oN NwNoo. eewwN. Nonm. omowh. sononoon mooH>Hom ooo: .m oHcmnoHnmneoo eszH.om moomo. NmmwN. owemm. nowmsw.oH .sononoon oeHH ooo: .N HNHmN.eo eomeN. eoweN. Nomoe. sHNHmN.eo sononoon o>oH ooo: .H n HHMHo>o ownmcu No No x oHnHanz Hounm o» n mooHoHnm oHonHHm> nouw .noeosinoonHoonm Hon HooHo .moHooHHm> ooHnoHu .ounz EOHm oo>Hooom moOHnomo: mo sononooHn so omeHHn: mo noHuonHm>m o>HHoomm< mo noHHOHooHo ocu How mHmsHon< nonmonom oHnHHHn: mo sHmEEanI.om oHonh 206 .m0. V We .mmmHo oOHnOmoH como How mHOHmoHonH wanHoEoo so oonoHo moHonHHo>n onmw.wH whooo. HHoNN. oemNm. owmoN. sononoon mooH>Hom oHo: .m NHHoH.oH onoo. Nmth. oheNm. smNom. sononoon mnuoom oHo: .e . . . . . anmnoHnmneoo onsm mN wmeo oNeNN mNmNm nwsmoo o .sononoon onHH ooo: .m . . . . . sononoon Nemwo em emmNo oeomN oeoom sonmw o noHumnHomnH ooo: .N meoeN.oo NHNNN. NHNNN. hmose. aweoeN.oo sononoon o>oH ooo: .H n HHnHo>o ownmcu N: No a oHoHHan Hounm o» n mooHoHnm oHonHHm> nouw .noz--ooHcHooom Hon HooHo .moHowHHo> ooHooHu .ounz EOHm oo>Hooox moOHnOmom mo sononooHn so oonHHmz mo noHHmnHm>m o>Huoomm< mo noHHoHooHo ocH How mHmsHmn< nonmonom oHoHHHn: mo sHmEnan-.Hm oHomh 207 The frequency of receiving status from spouse was not a significant predictor of marriage evaluation fOr men or for women. Hypotheses l7 Frequency of services received from mate does not significantly contribute to the prediction of marriage evaluation fOr women or men. The analysis for women, F (1,193) e .54, p_< .464 and for men, F (1,201) = .20, p < .652 failed to reject the null hypotheses that the regression coefficients for mate services frequency were equal to zero. Tables 50 and 51 indicate the variable is added at the third step for women and the fifth step for men with a minimal R2 change (.003 for women, and .0007 for men). Hypotheses 18 Frequency‘of information received from mate does not signifi- cantly contribute to the prediction of marriage evaluation for women or men. The null hypothesis that the regression coefficient for mate information frequency is equal to zero was rejected for men, F (1,201) = 10.03, p < .002. The standardized regression coefficient ranks second in absolute value (-.263) and indicates an inverse relationship between frequency of information received from spouse and a positive evaluation of marriage. The null hypothesis that the regression coefficient for "mate information frequency" equals zero was not rejected for women, F (1,193) = .21, p < .645. The standardized regression coefficient ranked lowest of all variables in absolute value (.034). 208 Hypotheses 19 Frequency of shared time with mate does not significantly con- tribute to the prediction of marriage evaluation for women or men. The null hypotheses that the regression coefficients for "mate time frequency" equal zero was rejected for women, F (1,193) = 7.32, ‘p < .007; and fOr men, P (1,201) = 4.36, p < .038. The standardized regression coefficient for women (.192) is second in magnitude (.164) following "love frequency" (.490) and "information frequency” (—.263). Frequency of shared time with mate is a significant predictor of marri- age evaluation for women and men despite the minimal contribution to the R2 change (4% for women and 2% for men). To summarize the contributions of frequency indicators of resource transfers to the prediction of marriage evaluation, the standardized regression coefficients for men and women can be compared by examination of Table 52. The similarities between men and women on the importance of "love frequency" and "shared time frequency" are evident. The most obvious difference between men and women occurred in the importance of "information frequency" in predicting evaluation of marriage. Objective 5 Describe the differences in evaluations of marriage, family life and life-as-a-whole for women and men. Hypotheses 20 Ho: There is no difference between mean scores of women and mean scores of men on evaluations of marriage and family life. H20: The mean score for men will be higher than the mean score for women on evaluations of marriage and family life. 209 Table 52.--Summary of Standardized Regression Coefficients for Created Frequency Variables for Women and Men. Variable Women Men Mate love frequency .336* .490* Mate status frequency .045 .045 Mate services frequency .055 .041 Mate information frequency .034 -.263* Mate shared time frequency .192* .164* Note: The regression coefficients were reported in Table 49. *Refers to rejection of the null hypothesis. The results of hypotheses 20 were reported at the beginning of Chapter III with the description of major variables. As predicted, the men reported a significantly higher evaluation of marriage than did women, 5 (200) = 2°52’.E < .01. The mean scores of men on evalu- ation of family life were also significantly higher than the mean scores of women, £_(200) = 2.90, p_< .005. The mean scores of men and women were equivalent (5.3) for evaluation of Perceived Overall Quality of Life. Summary of the Results The descriptive data for major variables indicated the results of this study to be consistent with other quality of life studies. Respondents evaluated their lives positively. The mean score for POQL (Life 3) was 5.3 for women and for men. Feelings about family life were highly related to feelings about life as a whole (:_= .68 for women, .60 fOr men). 210 Evaluations of family life were more positive than evaluations of life-as-a-whole Q! = 5.6 for women, 5.8 for men). A one-tailed dependent t-test (200) = 2.90, p < .005 indicated that the mean score for men was significantly higher than the mean score for women on evalu— ation of overall family life (Famlif 3). Crosstabulation of women's and men's evaluations of family life indicated a predominance of con- cordant pairs (gammg = .609). Correlation analysis showed higher relationships between the most particularistic resources of love and status with evaluation of family life, but all resource classes had correlation coefficients significantly greater than zero suggesting the basic premise of the Foa theory that all resources (love, status, services, information, goods, and money) are necessary to quality of family life. Feelings about marriage were highly related to feelings about family-life-as-a-whole (3 = .71 for women, .70 for men). Multiple regression analyses showed marriage contributed a 54% reduction in the variance of family life evaluation fOr women and a 48% reduction for men. The importance of marriage in explaining the variance in family life evaluation led to the decision to study dimensions of marriage which contribute to a positive evaluation. Mean scores for evaluation of marriage were higher than mean scores for evaluation of family life. A one-tailed dependent t-test (200) = 2°52:.B < .01 indicated that the mean score for men (5.95) was significantly higher than the mean score for women (5.71) on evalu- ation of marriage. Crosstabulation of women's and men's evaluations of marriage indicated a predominance of concordant pairs (gamma = .512), 211 but less agreement on marriage than was shown on family-life-as-a- whole (gamma = .609). Research Question 1 Do the questionnaire items which represent a resource class have greater proximity to each other than to variables repre- senting a different resource class? Results of complete-linkage hierarchical clustering appeared to confirm the hypothesis of four distinct particularistic resource classes with a close relationship between love and status. Four- cluster solutions were found for men's evaluation variables and women's frequency variables. Three-cluster solutions were found for women's evaluation and men's frequency variables. The three-cluster solutions fused love and status. The presence of both concrete and symbolic expressions of love in the frequency variables, and the presence of only one status indicator among evaluation variables may have con- tributed to the difficulty of distinguishing between love and status resource classes. Research Question 2 Is there a pattern in the fusion order of clusters to support the theoretical model of structured relationships between resource classes? The fusion order hypothesis (love-status—services-information) was confirmed for frequency variables, but not for evaluation vari- ables. The reversed order of information and services for evaluation variables seemed due to the more particularistic nature of the infor- mation indicator. "How openly and honestly you can express feelings" is more relevant to the husband-wife relationship than is ”helpfulness 212 of family members" or the "way household work is divided or accom- plished." Research Question 3 Are there differences in cluster solutions for men and women? There were differences in cluster solutions for both evalu- ation and frequency variables. The differences for men and women on evaluation variables were found with "comfort at home," "way decisions are made," and "sexual relationship." "Comfort at home" merged with love for women and with status for men. "Way decisions are made" fused with information for women and with services for men. "Sexual rela- tionship" merged with information for both men and women, but at a later stage in the dendrogram of women. Differences for men and women on frequency variables were found with love and information clusters. The solution for men on the love resource class fused two symbolic expressions of love: "tell/show love” and "enjoyment of company." The solution for women indicated a fusion of concrete and symbolic expressions of love: "tell/show love" and ”hug/kisses." "Enjoyment of company” merged with status for women. It is not clear whether the lack of agreement between the sexes on information indicators was a problem of measurement or an actual differ- ence in attitude. The solution for women fused "new information" and "opinion"; "decisions" fused with services at a late step of the analysis. The men's solution merges "opinion" and "decision" at a late stage of the analysis; "information" added to services. The lack of agreement on information indicators suggests a need for further inves- tigation. 213 Research Question 4 Will different methods of hierarchical clustering provide similar cluster solutions for the same data to support valid- ity of four particularistic resource classes? Complete-linkage clustering found four-cluster solutions for women on frequency variables and for men on evaluation variables. Three-cluster solutions were found for women on evaluation variables and for men on frequency variables. Both three-cluster solutions fused love and status. Single linkage clustering (based on maximum corre- lations) and a special method called UPGMA (based on average correla- tions) found three-cluster solutions for evaluation and frequency variables for both men and women. Only the complete-linkage clustering was able to separate the love and status clusters of variables to find four-cluster solutions. Research Questions 5-7 What kind of variable, which selection method and which entry order for variables will achieve the best prediction of evaluation of marriage? The best prediction was obtained using (1) evaluation variables selected by statistical criteria, followed in order by: (2) evaluation variables selected by theoretical expectation, (3) combination of fre- quency and evaluation variables, (4) created frequency variables. All best predictions were selected by the forward method search procedure. The best set of predictor variables was selected by the follow- ing criteria: lowest mean square error (.356 for women, .397 for men), highest adjusted‘R2 (.81 for women, .75 for men), highest F to enter (494.999 for women, 404.937 for men), number of significant predictor variables (five for both women and men). The evaluation of spouse 214 contributed a 70% reduction in variance of marriage evaluation for women and a 65% reduction for men. The variable set for women listed by order of entry included: "your husband," "comert at home" (services), "way decisions are made" (information), "sexual relationship" (love-services), "respect received" (status), "love and affection” (love). The last variable was not statistically significant. The variable set for men listed by order of entry was: "your wife," "your sexual relationships" (love-services), ”how comfortable it feels to be at home" (services), ”the things you do together" (shared time), "love and affection" (love). Research questions 8 through 20 were answered by hypothesis testing. Research Questions 8-9 Will evaluation of particularistic resources received and frequency of resources received from mate predict evalua- tion of marriage for women and men? The null hypotheses that all regression coefficients in the evaluation variable sets (statistical and theory) and all regression coefficients in the frequency variable sets would equal zero were rejected for both men and women in all three predictions. Evaluation of particularistic resources received and frequency of resources received from mate were significant predictors of marriage evaluation. Research Questions 10-14 What is the contribution to evaluation of marriage of the following: evaluation of love, status, services, infor- mation, and shared time for men and women? 215 The null hypotheses that each individual regression coefficient would equal zero were rejected for women fOr love, status, services, information and shared time. Evaluation of "love and affection" has the highest F to enter and the highest R2 (.60) for both variable sets. The addition of "respect received" added 4% to the R2 change. There were differences in the indicators for information and services resource classes in the statistical variable set and the theoretical variable set. Services indicators for the statistical vari- able set included "your sexual relationship" and ”how comfortable it feels to be at home"; both variables were significant predictors of marriage evaluation for women. The services indicator for the theor- etical variable set was "mutual helpfulness of family members" which was a significant predictor of marriage evaluation with an inverse relationship to the dependent variable. The information indicator "how openly and honestly you can express feelings” was used in both variable sets and was a significant predictor of marriage evaluation for women in the theoretical but not the statistical variable set. The information indicator "the way decisions are made" was a significant predictor of marriage evalua— tion for women in the statistical variable set but contributed only 2% to the R2 change when added to the regression equation at step four. The significant predictors of marriage evaluation for the men's analyses were evaluation of: "love and affection” (love); "Open, honest expression of feelings" (information); "sexual relationship" (love- services); "comfort at home" (love-services); and the "things you do together” (shared time). 216 The F-test failed to reject the null hypotheses that the regression coefficients for status were equal to zero in both variable sets for the men's analyses. "How openly and honestly you can express feelings" was a significant predictor in the theoretical variable set, but not in the statistical variable set. "Helpfulness of family members" was not a significant predictor of marriage evaluation for men. The statistical variable set for men indicated ”comfort at home" made the most significant contribution to the prediction of marriage evaluation (32 = .49) and "sexual relationship was added at second step two (R2 change = .14). The theoretical variable set showed ”love and affection" made the most significant contribution to the pre- diction of marriage evaluation (32 = .47) and "open, honest expression of feelings" was added at step two (R2 change = .09). Research Questions 15-19 What is the contribution to evaluation of marriage of each of the following: frequency of receiving from spouse love, status, services, information and shared time. The null hypotheses that the regression coefficients for fre- quency of receiving love and sharedtime were equal to zero were rejected for both men and women. The frequencies of receiving status and receiving services from spouse were not significant predictors of marriage evaluation for men or for women. The frequency of receiving information from spouse was not a significant predictor of marriage evaluation for women, but was significant for men. The negative regression coefficient indicated an inverse relationship between 217 frequency of receiving information from spouse and a positive evalua- tion of marriage by men. Research Question 20 How do the evaluations of marriage and family life differ for men and women? The mean scores for men were significantly higher than the mean scores for women on evaluations of marriage and family life. CHAPTER V LIMITATIONS, CONCLUSIONS, DISCUSSION AND IMPLICATIONS This chapter includes limitations of the study, conclusions, discussion of results, and implications for education and research. The general purposes of the study were accomplished. Results supported validity of the Foa and Foa (1974) resource exchange theory both in terms of the six distinct resource classes and also credi- bility of the theory in predicting marriage evaluation for men and women. There was support for the structured relationships among resource classes and for the order of classes in predicting satis- faction. However, results should be interpreted with respect to some possible limitations of the study. Limitations Limitations of the study include those of design, sample and sampling procedures, measurement, problems of respondents, and problems of statistical analyses techniques. Survey design is not the most desirable way of gaining infor- mation about dynamics of resource exchanges and transfers among family members. However, if one assumes the transfer of a resource is circular and reciprocally influencing--a simultaneous stimulus, 218 219 response and reinforcement--then it is reasonable to assume that infor- mation obtained from questionnaires can provide indicators of resource exchanges. Modifications in sampling procedures made by the research firm did raise questions about the degree of randomness in selection of respondents. However, sampling procedures were not a problem for this study with its primary interest in theoretical relationships among vari- ables. Generalization of results beyond the present sample was not an objective. Typical measurement difficulties of survey research were present in the study. It was necessary to assume respondents had similar defini- tions of "love," "comfort" and respect. "How comfortable it feels to be at home" was a significant predictor of marriage evaluation for men and women. It was thought to be an indicator of services, but the meaning to respondents is uncertain. There was a need for indicators of the ser- vices resource class which were more particularistic and more relevant to the specific husband-wife relationship as well as to family-life-as- a-whole. It is recognized that respondents tend to answer in socially desirable ways and to remember positive experiences more easily than negative experiences. The tendency of respondents to give highly positive evaluations of marriage and family life provided skewed distributions on the dependent variables. Selection of a sample of married persons decreased variability of responses on the dependent variables. It is therefore likely that the actual probability of a Type I error is higher 220 than the stated alpha level of .05 which was used for statistical tests. Results must be interpreted with this caution in mind. Respondents also faced problems in recalling frequencies of activities, maintaining privacy of answers, and attempting to place a quantitative number for a qualitative dimension of their lives. The close relationship of particularistic resources included in the Foa and Foa (1974) theory created problems for selection of variables for regression and in the interpretation of results. The presence of too many highly correlated independent variables in multiple regression analysis makes interpretation of variance difficult. Testing theoretical models having reciprocal relationships or high correlations among inde- pendent variables is difficult with most existing techniques of statisti- cal analyses. Conclusions The most important conclusions of this research are that the Foa and Foa (1974) resource exchange theory is a valid and useful tool in predicting evaluation of marriage, family life and quality of life; and that interpersonal resource exchanges are significant predictors of satisfaction with marriage and family life. Several other conclusions are: 1. The significant predictors of marriage evaluation for both men and women were: "love and affection," "sexual relationship" (love-services); "how comfortable it feels to be at home" (ser- vices); "open, honest expression of feelings" (information); "the things you do together" (shared time); frequency of receiving love from mate; and frequency of shared time with mate. 221 "Respect received” (status) and the ”way decisions are made" (information) were significant predictors of marriage evalua— tion for women but not for men. The two indicators may suggest the degree of democracy in the relationship, success in negotiating differences, or the ability to cooperate. Evaluation of "love and affection" (the most particularistic resource) was so highly related to evaluation of "your husband or wife" that both variables could not be successfully used in the same regression equation for testing theoretical predictions. Evaluation of "husband or wife" was the strongest predictor of marriage evaluation for both men and women. Addition of "comfort at home" and "sexual relationship" to the variable set accounted for a major portion of the variance. Evaluation variables were more successful than frequency vari- ables in predicting marriage evaluation which was consistent with the hypothesized relationships among variables. Further investigation is required to explain the negative relationship among variables. Further investigation is required to explain the negative relationship between frequency of receiving information from mate and a positive evaluation of marriage for men. Marriage is the dimension of family life which yields the most satisfaction and is the strongest predictor of overall quality of family life for both men and women. 222 7. Men evaluate marriage and family life more positively than women. 8. A greater proportion of husbands and wives agree upon evalua- tions of family life than agree upon evaluations of marriage. 9. Men and women evaluate marriage more positively than family life or life-as-a-whole. The present research supports the Foa and Foa (1974) theory of structured relationships among resource classes. The interpersonal resources were ordered in their effectiveness for predicting marital and family life evaluations. Love and affection, recognition and respect, comfort and assistance, sharing and companionship, and shared meaning are the human needs which are satisfied through ex- changes of the resources of love, respect, services and information in close relationships such as marriage and family life. Satisfaction of these needs, under conditions where physiological and safety needs have been met, provides Opportunities for highest levels of satis- faction for humans. Discussion The Foa theory predicts that love and status should be the best predictors of marriage evaluation. Love indicators for both evaluation and frequency variables were significant predictors of marriage evaluation for both men and women, but the status indicators 223 were not significant predictors for men. The results may indicate that men have greater opportunities outside the family to receive respect and esteem than women who are more dependent upon husband and children to meet status needs. Alternatively, the results may be due to the high correlations among independent variables. A possible solution to the problem of multicollinearity suggested by Nie et a1. (1975) is to create a new variable which is a composite scale of the set of highly correlated variables and to use the new scale variable in the regression equation in place of its components (p. 341). The created frequency variables in this study are examples of this attempted "solution." The corre- lations continued to be high, particularly for men. The following correlations show relationships of love frequency with other variables: (1) status frequency (5 = .73 men, .71 women); (2) services frequency (3 = .66 men, .55 women); (3) information fre- quency (E = .58 men, .55 women); (4) shared time frequency (3 = .48 men, .41 women). Status frequency is also highly related to all other indicators: (1) services frequency (3 = .61 men, .47 women); (2) infor- mation frequency (3 = .52 men, .51 women); (3) shared time frequency {3 = .60 men, .45 women). It is possible that the high correlations among; independent variables caused problems in the explanation of variance in the multiple regression analyses. The theory states that sexual relationship falls between the resource classes of love and services. The proximity of sexual commtuiication to the love resource class in a particularistic 224 relationship predicts the results obtained in this study that evalua- tion of the sexual relationship was a significant predictor of marriage evaluation fOr men and women. The information indicators which are more distant from the love resource class were less consistent and effective in predicting evaluation of marriage. The importance to women of ”the way decisions are made" (information) and "respect received" (status) implies the importance of balance of power, an equalitarian relationship or the ability to respectfully negotiate change. The results suggest some of the issues addressed by Lewis et a1. (1976) in the discussion of differences between optimal and adequately healthy family systems. Optimal families demonstrated equalitarian marriages with complementarity and reciprocity where power was shared without conflict and problems were solved with respectful negotiation. Patterns of authoritarianism and dominance-submission were incompatible with the pervasive attitude of respect fOr the world-view of others. Compared to adequate families, the Optimal families indicated: strong affectional bonds; higher satisfaction of wives; husbands who were more directly supportive of wives and showed less interpersonal distance; increased capacity to communicate thoughts and feelings; shared adult leisure pursuits; community involvement; and a prevailing attitude of warmth, affection, and caring. "In less than optimal fami lies the mother was the first to become dissatisfied, distressed 0T S)nnptomatic. . . . The father with more outside sources of esteem is Ckften the last family member to become symptomatic" (p. 225). Results of the present study also suggest support for the fOIIcnving propositions of Lewis and Spanier (1979): 225 1. The greater the expression of affection, the greater the marital quality (43). 2. The more the esteem between two spouses, the more the marital quality (44). 3. The more the sexual satisfaction, the more the marital quality (51). 4. The more equalitarian the marriage, the more the marital quality (47). 5. The greater the ease of communication between spouses, the greater the marital quality (38). 6. The greater the companionship, the greater the marital quality (69). 7. The more shared activities, the more the marital quality (70). 8. The more effective the problem solving, the more the marital quality (73). 9. The more positive the evaluation of spouse, the more the marital quality (40). 10. The greater the rewards of spousal interaction, the greater the marital quality (90). 11. The more effective the communication between spouses, the more the marital quality (85). The present study indicated that frequency of information received from spouse was a significant predictor of marriage evaluation fOr men but had an inverse relationship with the dependent variable. The finding suggests that some attention should be given to further investxigation of Nye's (1979) proposition: Ixxr husbands the relationship of spousal communication to marital Satxisfaction is curvilinear with a total absence of verbal commun- ication associated with acute marital dissatisfaction, and a high lenrel of communication associated with marital dissatisfaction (68). The Foa and Foa (1974) resource exchange theory states that 311 Si)! resource classes (love, status, services, goods, money and infbrflurtion) are necessary for quality of life and when any one falls 226 below the optimum range, quality of life is impaired. Results of the present study indicated that goods and money were not significant pre- dictors of marriage evaluation. The results are probably due to the high median income ($27,034) for respondents in this sample. The lack of goods or money was not a problem to these people. Future research efforts could be directed toward investigating any differences which might be present in a low income group of respondents. The application of exchange theory to marital and family research will require the integration of economic and social exchange processes into one theory: inclusion of the concepts of reciprocity, altruism and equity; and a theoretical model which facilitates simul- taneous examination of two-party and multi-party exchanges. Future family research needs to examine generalized exchange behaviors with indirect reciprocities and also needs to study the contributions of altruistic behaviors to family and individual well-being. Implications for Education The importance of the family in meeting the status needs of women who do not work fOr income outside the household is suggested by the significance of "respect received" as a significant predictor of.marriage evaluation fOr women. The Lewis et a1. (1976) study also noted ‘the wife-mother as the first indicator of problems in the familjr system. Maintaining the self-esteem and meeting status needs 0f pelnsons at home is difficult in a society that gives recognition and identity on the basis of contributions to market rather than non- market labor. 227 Recent attention has been given to educating people about the value of household work and the problems of homemaking as an invisible occupation, however, more attention needs to be directed toward educa- tion of home-staying women. Women must be encouraged to deveIOp interests and become involved in activities which are uniquely their own in addition to the interests and activities they perform as services to other members of the family. The active pursuit of an individual interest contributes to the sense of self and esteem for self which is critical in maintaining mental health. Mental health of the wife- mother has a significant impact upon all family members. Educating persons about the importance of meeting status needs of home-staying women by increasing awareness of the economic and social contributions of women to the welfare of society is an important step toward strengthening family life. Enhancing the capacity to share time and love, to have a satisfying sexual relationship, and to create an atmosphere of comfort on a social-psychological and physical level in the home environment will help increase people's satisfaction with their marriage and family life, and will strengthen families. Implications for Research The present study was planned to be one of the first stages of fanuily data analyses fOr a larger quality of life study. Many interesting questions remain to be answered for which the existing data can provide information. One of the first questions which logically follows the present StUdYis: How do people who differ in evaluations of marriage differ demographically and perceptually (income level, occupation, years 228 married, number of children, self esteem, community embeddedness, homogamy, congruence of opinion)? The data can provide information on the above relationships and test the propositions which have already been stated by Lewis and Spanier (1979): Higher marital quality is associated with: l. Husband-wife homogamy of: 3. Religious affiliation b. Age c. Race d. Status 2. Resources a. Higher levels of education . Older age at first marriage Higher social class b c. d. More positive self concept 3. Socioeconomic factors a. Higher occupational status of husbands b. Higher family incomes 4. Community embeddedness a. Greater community participation b. Less density of residential population Some implications for research were suggested by problems with measures and analyses techniques. What results would be obtained from using different analyses techniques such as multiple classifi- cation analysis or discriminant analysis? Discriminant analysis could be used to statistically distinguish respondents with highly positive and lrighly negative evaluations of marriage and/or family life. The existing programs can proceed by entering all discriminating varitibles directly into the analysis or by a stepwise method which selexrts the ”best" set of discriminating variables. The use of discriminant analysis would be a way of validating results of the PTeSEHTt study. It would also be possible to include a greater number 229 of independent variables in the initial search for the variables which best discriminate between the two groups. The recent interest of social scientists in making causal inter- pretations of nonexperimental data has led to some new methodological approaches which may be useful in testing theories more exactly. One approach is general Linear Structural Relations (LISREL) by JOreskog and SOrbom (1979). The use of structural equation models requires statistics based upon, but go beyond, multiple regression and analysis of variance. The LISREL model is particularly designed to handle theor- etical models with latent variables, measurement errors, and reciprocal causation (simultaneity interdependence) (p. 3). Data for the present study have been submitted for analysis by LISREL IV but results are not yet available. Problems with measures in the present study indicated the following needs for further investigation: (1) clarification of the dimensions of "how comfortable it feels to be at home," (2) refine the indicators of information and services resource transfers in particularistic relationship, (3) change the frequency scale to refer to a shorter period of time required for memory, (4) consider elimi- nation of the frequency indicators and adapt the items for the evalu- aticni scale, (5) create an index of several items to represent each resorxrce class for evaluation variables. The use of hierarchical cxmu>lete-linkage clustering in combination with reliability analyses woulcl be helpful in making the refinements. The information frequency scale: could be improved by adding "spend an hour or more just talking,” and 'Niiscuss personal feelings." 230 Effective problem solving and decision making is frequently mentioned in the literature on healthy family systems as a variable of critical importance. Findings in the present study suggest that women may not be involved in decision making the way they would like to be. Additional information is needed to verify this conclusion and to determine reasons for different evaluations of the variable by men and women. The variable needs to be added to the men's analyses despite its low correlation with evaluation of marriage. The cluster analyses provided conflicting results which require clarification. The "way decisions are made" added to "how openly and honestly you can express feelings” and the "kind of com- munication you have" in the women's analysis of evaluation variables. Examination of the frequency variables indicated that how often your mate "helps you solve a problem or make a decision” added to the services indicators for women and to ”give you her opinion" (informa- tion) for men. The men's analysis of evaluation variables showed the "decisions” variable added to ”mutual helpfulness of family members" and the "way household work is accomplished/divided” (services indi- cators). Results of the present study indicate that the Foa and Foa (1974) resource exchange theory provides a comprehensive and concise guide; for studying quality of life, quality of family life, and is credilnle in predicting evaluation of marriage. Future research needs to EN: directed toward further testing of the theory by obtaining infor- matioqi from subjects with differing life circumstances. APPENDICES APPENDIX A PORTIONS OF QUALITY OF LIFE QUESTIONNAIRE USED IN THIS STUDY QUALITY OF LIFE Department of Family and Child Sciences Department d HmanEmirormentandDesign College of Human Ecology Michigan State University ,- Agricultural Experiment Station . I. Project numbers. 3151 and 1249 Fall 1977 231 232 IICHIGAN STATE UNIVERSITY mam-Inna! ' mums-Wow. fall l977 Dear Friend: Host of us are aware of the rapid changes taking place in our society today. As we face energy shortages and resulting changes in the naterial products we use. changes in the patterns of family activities and in the roles of men and women. it becomes essential to plan for change that will contribute to one's sense of well being and satisfaction with life. The College of lit-Ian Ecology at Michigan State University is concerned with the quality of life of families in the state of liichigan. Two departments within the college. family and Child Sciences and human Environment and Design. have under- taken the task of determining what components of life are important to the quality of life of Michigan families and to what degree they are satisfied with those aspects of their lives. You will find questions about various aspects of your life such as your spare time activities and your neighborhood. and any questions which focus on your family life. your clothing and your Job. Your participation in this study is very inortant. You will provide us with information necessary to understand the feelings people now have about their quality of life. and this will suggest possible ways to immove satisfaction with life in our changing society. This is a questionnaire on how you feel about your life. it is rather long. and it will take sore time to fill it out. Inst of the questions should be interest- ing. sane may be dull and tiring. many will be easy because it is about your life. but some questions will require nore thought. Answer them all as well as you can. There are no 'right' or 'wrong' answers. It is your experiences and opinions that are cost inortant. By signing the consent form you agree to couplete the entire questionnaire to the best of your ability. Our signatures guarantee you anonymity. lihen both of you complete separate questionnaires. we will send your family a check for $10 shortly after the interviewer picks up the two questionnaires. lie sincerely appreciate your participation in this study and thank you in advance for your time. effort and interest. A sunnary of research findings will be sent to you when the study has been completed. If you have any questions about the study. please call 5l7-353-5389 or 517-355-1895. Sincerely. WA”?!- Dr. liargaret ii. lubolz. Professor Fmily and Child Science's- ilr. Ann c. Slocum. Assistant Professor linen Environment and Design ~ 2233 W Please read the directions at the beginning of each section before answering the questions. It is very haportant that you answer each question as care- fully and as accurately as you can. Be sure to respond to all the questions on both front and back of each page. Both you and your spouse are asked to complete separate questionnaires. Please do not discuss your answers before both of you have finished the entire questionnaire. when you have completed the :uestionnaire. return it to the manila envelope provided and seal the - enve ope. YOUR FEELINGS ABOUT LIFE CONCERNS In this section of the questionnaire. we want to find out how you feel about various parts of your life. and life in this country as you see it. Please include the feelings you have now-otaking into account what has happened in the last year and what you expect in the near future. .All of the items can be answered by simply writing on the line to the left of each question one of the following numbers OR letters to indicate how you feel.. For example write in 'l' for terrible. '1“ if you have mixed feelings About some question_(that is. you are about equally satisfied and dissatisfied wdth some part of your life). and so forth on to '7“ if you feel delighted about it. if you have no feelings at all on the question. write in “A.” If ‘you have never thought about something. write in "B.“ If some question doesn't apply to you. write in 'C.‘ ' ‘for two of the questions we also ask you to write in some important reasons for why you feel as you do. Please finish this section before going on to the next section. . ' I feel: Leer-1. r-1 Llr 2 E] , u 5 LEJ E]— Terrible . Unhappy flostly Mixed iiostly Pleased Delighted _ . _dissatisfied < (about satisfied equally - satisfied and dissatisfied) E lieutraluneither satisfied nor dissatisfied ‘- lever thought about it I . Q Does not apply to me l.l how do you feel about your life as a whole? ___;;jl.2 how'do you feel about the freedom you have from being ‘ f bothered and annoyed? I feel: Terrible ihhappy I.3e I.3b ‘0‘ I.5 I.6 1.7 I.8 .qumIIIIII I.9 I.IO I.II I.I2 I.I3 I." 234 lil III III Ill lil liostly iiixed lastly Pleased Delighted dissatisfied (about satisfied equally satisfied and dissatisfied) E] hutralo-neither satisfied nor dissatisfied lever thought about it E] Does not apply to me how do you feel about your own fluily life--your lusband or wife. your marriage. and. year children. if any? that are sure of the most important reasons for LILY. you feel as you do about your fmnily? how do you feel abut the count of beauty and attractiveness in your day to day life? . how do you feel about your independence or freedom-the chance you have to do what you want? how do you feel about how much you are accepted and included by others? llow do you feel about your Job? liow do you feel about your standard of living-othe things you have like housing. car. furniture. recreation. and the like? llow do you feel abut your safety? llow do you feel about what our national goverrmrent is doing? iiow doyou feel about how Inch fun you are having? how do you feel about your house or aparbnent? ilow do you feel about what you are accuplishing in your life? llow do you feel about your particular neighborhood asa place to live? 235 I feel: [I B H I] B B I] Terrible Unhappy iiostly iiixed lipstly Pleased Delighted dissatisfied (about satisfied . equally satisfied and dissatisfied) E Neutral-wither satisfied nor dissatisfied lever thwght about it E Does not apply to me 1.3a liow do you feel about your own fanily lifeuyour husband or wife. your marriage. and. your children. if any? l.3b that are sane of the most important reasons for w_hy_ you feel as you do about your faily? 1.4 how do you feel about the mount of beauty and attractiveness in your day to day life? l.5 _ llow do you feel about your independence or freedaa--the chance you have to do what you want? 1.6 how do you feel about how much you are accepted and included by others? l.7 how do you feel about your job? 1.8 How do you feel about your standard of livingo-the things you have like housing. car. furniture. recreation. and the like? l. 9 liow do you feel about your safety? 1. lo liow do you feel about what our national goverment is doing? I. ll ilow do you feel about how arch fun you are having? 1. l2 liow do you feel about your house or apartment? “1. l3 how do you feel about what you are accurplishing in your life? .1." how do you feel about your particular neighborhood as a place to live? IIIIII 2236 ion: FELINGSMUT roua mini 1.le - CIRCLE THE NUMBER which best describes your feelings about your own family life. For example. circle '1' if you feel terrible about something. circle “4" if you have mixed feelings (that is. you are about equally satisfied and dissatisfied). and circle '7“ if you feel delighted about it. 6.1 How would you feel about your own ‘ family life if you considered only: 6.la Your husband or wife? l 2 3 4 5 6 7 6.lb Your children? : l 2 3 4 5 6 7 6.lc The love and affection you experience? . l 2 3 4 5 6 7 6.ld The closeness and sense of belonging you feel? 1 2 3 4 5 6 7 6.le The amount of respect you ‘ receive? l 2 3 4 5 6 7 6.lf How comfortable it feels to be at home? l 2 3 4 5 6 7 6.19 Your marriage? l 2 3 4 5 6 7 6.2 How would you feel about your own family life-~your marriage. husband or wife and children-oif you considered only: 6.2a The way money is used? 1 2 3 4 5 6 7 6.2b The mount of money available for your personal use? l 2 3 4 5 6 7 6.2c The material goods it enables you to own? l 2 3 4 6 6 7 6.2d The way decisions are made? l 2 3 4 5 6 7 6.2e Theithings you do together? 1 2 3 4 5 6 7 1237 qir +- 1' ‘.%? ‘2; e o “v, in, “fig. qafdb,l2L ). e d‘ o ‘55, gag, <5;.'hr' ‘fihb' ‘9' ‘Qér 6; ‘90 I ‘6’ "o ’o- ’4 s w ’v «a a, ’0 “t 6.3 Now would you feel about your own family life if you considered only: 6.3a The mutual helpfulness of family members? I 2 3 4 5 6 7 6.3b The way household work is ' divided/accomplished? l 2 3 4 5 6 7 6.3c how openly and honestly you can express feelings? I 2 3 4 5 6 7 6.3d.The kind of cmnication you have? . l 2 3 4 5 6 7 6.3e The amount of time the . family spends together? . l 2 3 4 5 6 7 6.3f Your sexual relationship? l 2 3 4 6 6 7 6.39 The time you spend with ‘ your children? I 2 3 4 5 6 7 6.3h The time you spend with your husband or wife? 'l 2 3 4 5 6 7 6.3i The friends it enables you to enjoy? l 2 3 4 5 6 7 6.4a Have you had any children born to you? 6.4b 6.4: 6.4d [ 1 no [ 1| YES —> llumber of children born to you: If you had it to do over again would you have children? [ ] N0 [ ] YES ?ow strongly do you feel about the answer you gave to the above question 6.4b ? I ] Very strongly‘ [ 1 Somewhat strongly [ ] hot strongly Uhat are some of the reasons you feel as you db about having children? 238 CIRCLE THE ilUIBER corresponding to the-category which most accurately estimates how often the followi events occur. For example. circle '1' if saliething never happens. circle '4' II; it Happens about once each month. and circle '8" if it happens about two to three times each day. I" o ’1 (‘5 e ‘3 4 a ‘15 0‘ 0,, c4, 9» *9 ’0 45L “‘2 4 0' it», 7.l liow often do you and your mate: 7.la Spend time together-"dust the two of you? l 2 3 4 5 6 7 8 7.lb Spend an hour or more Just talking? l 2 3 4 5 6 7 8 7.lc Discuss personal feelings? l 2 3 4 5 6 7 8 7.ld iiork together on a project? i 2 3 4 5 6 7 8 7.le Take a drive or a walk? 1 2 3 4 s 6 7 8 7.lf Eat at a restaurant? 1 2 3 4 5 6 7 8 7.1g Entertain friends at hone? l 2 3 4 5 7 8 7.lh Visit friends? 1 2 3 4 5 6 7 8 7.li Co to a movie or other entertaiimlent? l 2 3 4 5 6 7 8 7.l: Attend a sports event? i 2 3 4 5 6 7 8 7.li: Attend a party? 1 2 3 4 5 6 7 8 7.2 how often does your mate: ' 7.2a liake you feel like an important person? l 2 3 4 5 6 7 8 7.2b Tell or show you that he/she adulres and respects you? l 2 3 4 5 6 7 8 7.2c Let you know he/she has confidence in your abilities? l 2 3 4 5 6 7 8 mo rell'oi- show you his/her love? i 2 3 4 5 6 7 8 239 4 at ‘69 9% 69? at "a ’0 a l) e ‘ul 4' I 4. 0% ‘1’ a, a? e e 0g 6g. 1‘ #9 a $4 ‘3 ‘3 e «'04 $04 7.2 liow often does your mate: 7.2a Let you know he/she enjoys your crmrpany? l 2 3 4 5 6 7 8 7.2f Enjoy a laugh or Joke with you?‘ l 2 3 4 5 6 7 8 7.2g Give you a hug or kiss? 1 2 3 4 5 6 7 8 7.2n Do an errand for you? l 2 3 4 s 5' 7 a 7.2i liake himself/herself available to do sane work for you? 1 2 3 4 S 6 7 8 - 7.2.) Do smething to save you energy or make you canfortable? l 2 3 4 5 6 7 8 7.2k Give you sale new information? 1 2 3 4 5 6 7 8 7.21 Give you his/her opinion? 1 2 3 4 5 6 7 8 7.2!: Give you sane thing that you need or want? i 2 3 4 5 6 7 8 7.2n Give you money for personal use? I 2 3 4 5 6 7 8 7.20 Help you solve a problem or make a decision? l 2 3 4 5 6 7 8 7.2p Support you with discipline and guidance of children? l 2 3 4 5 6 7 8 7.3 liow often does your oldest child who lives with you: 7.3a Make you feel like an important person? I 2 3 4 5 6 7 8 7.3a Tell or show you that he/she ' admires and respects you? 1 2 3 4 5 6 7 8 7.3c Tell or show you his/her love? I 2 3 4 5 6 7 8 7.3d Give you a hug or kiss? l 2 3 4 5 6 7 8 Z40 ( (5% “a 4% 7.3 How often does your oldest child who lives with you: 7.3a Do an errand for you? 1 7.3f Do sanething to save you energy or make you canfortable? 1 7.3g liake himself/herself available to do sane work for you? l 7.3h Give you some new information? l 7.3i Give you his/her opinion? 1 7.31 Give you some thing that you need or want? 1 7.3k Spend several minutes Just talking with you? I 7.4 llow often does the entire fmnily group: 7.4a Sit together for a meal? I 7.4b Have a discussion of ideas? l 7.4c Discuss a decision or a problem? I 7.4d iiork on a project together? 1 7.4e Play a game? I 7.4f Go to a movie or other entertaiment? l 7.49 Attend church services or activities? 1 7.4h Go on a trip or vacation? - l 241 10st people have disagreuents in their relationships. Please CIRCLE THE NUMBER under the category that indicates the approximate extent of agreeuen or disagreement between you and your mate for each of the following items. for exailple. circle “i" if you always disagree on a subject. circle '3' if you freqiently disagree. and circle '6' if you always agree on it. ‘4 1 x Q; (a la 01% 4" '1 ‘5' ”I ‘5' 4;- ‘3‘ a a a, a, , , ‘9» h. 9 9,. 9,. 9,. ’e ’e 1i- «a 0a ’e 8.l liandling family finances 3 4 5 6 8.2 Hatters of recreation 3 5 6 8.3 Rel igim'is matters 8.4 Derlonstration of affection 8.5 friends 8.6 Sex relations 8.7 Conventionality (correct or proper behavior) 8.8 Philosophy of life 8.9 Iiays of dealing with parents or in-laws .8.l0 Aims. goals. and things believed important 8.“ Amount of time spent together 8.12 Making major decisions 8.l3 Household tasks 8.l4 Leisure time interests"and activities 8.l5 Career decisions 1242 Now that you have done some thinking about your family life and your life in general. we would like'to ask you how you feel about them. Please write an the line to the left of each question one of the following numbers on letters to indicate how you feel. For example. if you feel terrible about TE write in 'l." if you have mixed feelings about it (that is. you are about equally satisfied and dissatisfied) write in '4.“ and if you feel delighted about it write in "7.’I If you feel neutral about it (that is. you are neither satisfied nor dissatisfied). write in “A.” if you have never thought about it. write in '8.“ if it does not apply to you. write in 'C.“ I feel: '1 .8 fl 4 E 6 Terrible Unhappy 'Nastly Nixed Nostly Pleased Delighted dissatisfied (about satisfied equally satisfied and dissatisfied) [:J Neutral--neither satisfied nor dissatisfied lever thought about it [:J Does not apply to me 9.1 how do you feel about your own family life--your husband or wife. your marriage. and your children. if any? 9.2 ‘How do you feel about your life as a whole? 9.3 This study has asked you to tell us how you feel about various parts of life. Are there things which affect your quality of life which have not been included? If so. please write them below. [ IOU WULO BE A GOOD TINE TO TAKE AOREAK BEFORE GOING ON TO THE NEXT PAGE. 2445 YOUR fANlLY SITUATION This study is about the quality of life of family members. Therefore. we are interested in knowing some things about yourself and your family. As you answer the questions. please consider only yourself and the family members ggg_living in your household. fOR mu outsnmi. me: A cuccx NARK IN THE amms [J] on um: TliE ANSHER on THE LINE PROVIDED. I3.I I3.2a I3.2b I3.3 I3.4 I3.5 Nhat is your sex? [ ] Nale [ ] female how old were you on your last birthday? ______Age at last birthday what is the month. day. and year of your birth? Bat“ Fry '20? OT EITEFI Nhat is your>religian. if any? ] Protestant: [ [ ] Catholic [ ] Jewish [ I "(please specify) ] None ] Other: (please specify) Nhat is your race? [ ] Nhite [ ] Black/Negro/Afro-American [ ] Other: (please specify) Do you (or does a member of your family who lives with you) own your home. do you rent. or what? (CHECK ONE) [ ] Own or buying [ ] Renting [ ] Other: ‘- ‘(please specify) 244 MR FAMILY SITUATION This study is about the quality of life of family members. Therefore. we are interested in knowing some things about yourself and your family. As you answer the questions. please consider only yourself and the family morbers n_o_w_ living in your household. fDR EACH quasnou. PLACE a CHECK lulu: IN THE amms [I] on NRITE THE ANSHER on THE LINE PROVIDED. I3.I l3.2a I3.2b I3.3 I3.4 I3.5 Nhat is your sex? [ ]Nale [ ] female How old were you on your last birthday? Age at last birthday what is the south. day. and year of your birth? HIE“ m '08? 0' Birth Nhat is your'religion. if any? [ ] Protestant: [ 1 Catholic I ] Jewish [ ] None [ ] Other: _—(please specify) (please specify) Hhat is your race? [ ] Nhite [ ] 8lack/NegroIAfra-American [ ] Other: _’(please specify) Do you (or does a member of your family who lives with you) own your home. do you rent. or'what? (CHECK ONE) [ ] Own or buying [ ] Renting ’ [ ] Other: (please specify) 245 l3.6b Is this your first marriage? l3.7a 13.7b [ths——> [Jm-——> In what year uere you married? l3.6b In what year did your present marriage begin? l3.6c How did your last marriage end? CHECK ONE. [ 1 Death ———>rear of death: [ ] Divorce -——->Year of divorce: [ ] Annulment -——>Year of annulment: Nhat is the highest level of formal schooling that you have capleted? CHECK ONE. [ ] Less than 8 grades. of elmnentary school [ .] 8 grades of elmnentary school [ ] l-3 years of high school I ] Calpleted high school and received diplana or passed high school emivalency exam ] l-3 years of college J College graduate. bachelor's degree J Post bachelor's course work ] Post master's course work ] PhD. EdD ] Other professional degree (such as ND. DO. JD. DDS): I I I [ I) Master's degree I I I ’(please specify) Are you _iQ! attending or enrolled in one of the progrmns listed above? [Ives——> IJNO l3.7c If YES. is that full-time or part-time? [ J full-time student [ J Part-time student l3.7d Please specify in which one of the above programs you are now enrolled (such as high school. college. master“ s program). Type of school or progrmn 2465 l3.8a IN THE PAST. have you been enrolled in any type of educational program otEEr thin high school or college. such as vocational school? I3.8d l3.9a []YES-—) [HO l3.8b If YES. please specify your field of training (such as business. office work. practical nursing. beautician. mechanic. electrician). field of training l3.8c Did you complete the training program? [ 1 YES I 1'0 [ JDDES ual mu Are you NOH enrolled in any type of educational program other than high school. Eollege or graduate school. such as vocational training program. arts and crafts classes. or religion classes? [JYES—> [1'0 l3.8a If YES. what type of educational program s t? . field of training or type of program Are you presently employed. unemployed. retired. or what? CHECK AS HANY AS APPLY TO YOU. . . [ J Housewife or househusband [ 1 Student 60 TO QUESTION I3.IOa ON PAGE 38. [ ] Permanently disabled (unless you also check one of the categories below in which [ ] Retired case)go to l3.9b on the next page . [ J Unenployed (that is. previously employed for pay and/OR presently looking for a Job) [ ] Temporarily laid off OR on strike ' Ofilon sick leave GO TO QUESTION l3.9b ON THE NEXT PAGE. [ llgorking now I3.» l3.9c I3.9d l3.9e I3.9f l3.99 247 If you are working now on are temporarily laid off Oil on strike Oil on sick leave. what kind of work do you do? Iihat is your main occupation called? (If you have two Jobs. your main occupation is the Job on which you spend the most time. If you spend an equal amount of time on two Jobs. it is the one amich provides the most income.) Iain occupation iihat do you actually do in that Jab? Nhat are some of your main duties? Duties Ilat kind of business. industry or organization is your Job in? lihat do they do or make at the place where you work? Kind of business. industry or organization Uhat they make or do About how many hours a week do you dothis work? CHECK ONE . ] Less than 20 hours per week ] 20 hours per week ] 2l-39 hours per week I 4l-SO hours per week 1 .51-60 hours per week I I I I ] 40 hours per week I I I 1 Here than 60 hours per week Do you do this work inside your home. outside your home but on your own property. or away from your hone and property? CHECK THE ONE PLACE IN NHICH YOU DO lKJSI Of THIS HORK. I ] Inside my home I ] Outside ey home but on my property I 1 Away from u home and property Are you an hourly wage worker. salaried. on coamission. self-employed. or what? CHECK ONE. I ] Hourly wage worker {_Jaunw I J Hark on comliission. tips I )‘Self-ewloyed in own business. professional practice. or farm I ] Hark without pg: in family business or farm I3.9h l3.9i I3.9I I3.9m I3.90 248 How long have you been in your present Job? years and mnths Is this your first Job? [ J rro ——)h 3.9: what kind of work did you do in your first full-ti Job after completing your education or training? I 1 YES Irat was your occupation called? Occupation INJk Nhat did you actually do in that Job? lihat were ‘ sac of your main duties? Duties Uould you be satisfied to stay in your present position indefinitely? [° 1 YES [lilo Do you anticipate a change from your present occupation or your position within the near future? I ] YES —) 3.9n If YES. please describe your anticipated new I J . position. met your title will be and what you will NO do. Anticipated new position Title Duties Are you currently uployed in a second Job? [ Jrr:s—) 3.99 IfY§5.ebouthowmany hoursaweekdoyoudo this work i [ ]NO I JLess thanZOhaurs per week I J 20 hours per week I J 2l-39 hours per week I J 40 hours per week 249 l3.lla llhat do you estimate will be' yorr total family incone beforegx_e; in l977? Please include incone fron all source'sgbefore taxes. including incone fron wages. property. stocks. interest. welfare. Aid to families with Dependent Children. child support fron a previors marriage. and any other money incone received by you and all fmnily mmnbers who live with you. ESTIMATED 10m mun rum income. 1977 ] Under 83.000 I J 812.000 - 814.999 1 83.000 - 83.999 I: J 815.000 - 819.999 1 84.000 - 84.999 [ J 820.000 - 824.999 1 85.000 - 85.999 I J 825.000 - 829.999 ] 86.000 - 86.999 '[ J 830.000 - 834.999 1 87.000 - 87.999 I J 835.000 - 849.999 ° 1 88.000 - 89,999 I J 850.000 - 874.999 I J 810.000 - 811.999 I J 875.000 and over l3.llb About how much of this total fmnily yearly irlcrmre do you estimate that Mwill earn in 1977? ESTINEED NATION OF TOTAL FAMILY "CONE. I977I EARNED BY YOURSELF ] Does not apply. not errployed in l977 J Under 83.000 HHHHHHH I J 812.000 - 814.999 1 83.000 - 83.999 I 1 815.000 - 819.999 1 84.000 - 84.999 I J 820.000 - 824.999 ] ss.000'- 85,999 I J 825.000 - 829.999 1 86.000 - 86.999 I J 830.000 - 834.999 1 87.000 - 87.999 I J 835.000 - 849.999 1 88.000 - ”.999 [ 1 850.000 . 874.999 1 810.000 . 811.999 I 1 875.000 and over l3.l2 In the caning year. world yor say your financial situation will get worse. stay about the same. or get better? CHECK ONE. I ] Get worse I 1’rStay abort the sane [ 1 Get better HHHHHHHH H 250 ls.la He would like to know something about the people who live in your household. In the chart below. please list for = their birth date. age at last birthday. sex and mar ta status. m ist any person more than once. Please use the following nrmbers to indicate marital status: [I] Never married - I4] Separated I2] Nerried I5] Divorced. not remarried I3] Nidawed. not remarried [6] Don't know Date of Age at Sex I birth last (circle :33? mo.[dgy[yr. birthday N or F) sews: (husband or wife) CHILDREN BORN TO THIS MARRIAGE. LIVING IN THIS HWSEHOLD F PleeSe list in order from oldest to youngest CHILDREN BORN TO NIFE PRIOR TO THIS MARRIAGE. LIVING IN THIS HNSEHOLD Please list in order from oldest to youngest CHILDREN 001m TO HUSBAND PRIOR 10 nus mamas. LIVING IN THIS HOUSEHOLD Please list in order from oldest to youngest ADOPTED CHILDREN NOT BORN TO EITHER SPOUSE. LIVING..- IN THIS HOUSEHOLD " - Please list in order from oldest to youngest 3333333333333333333333333' mmmmmmmmnmmmmmmnfimmnfimmm CONTINUED-ON surpass. NOTE: If there are not enough spaces. please finish the list on the last page. APPENDIX B SAMPLING PROCEDURES APPENDIX B SAMPLING PROCEDURES Basic Sampling Design Area: Oakland County Number of Sampling Points: 75 Area divided into categories by type of area and racial composition: I. Rural, defined by named townships, using only areas with 1970 median income of $12,000. One-fourth of sampling points chosen as probabi1ity-proportionate-to-household count sample of these townships. II. Urban/Suburban--Balance of County: 3. Sampling points where black residents in high prOpor- tion using only tracts with 1970 median income of $6000 or above. These are in Pontiac City and Royal Oak Township. One-fourth of sampling points chosen as probability-proportionate-to-household count sample of these two places. b. Balance of one-half of sampling points chosen as probabi1ity-proportionate-to-household count of this remaining area of county not in I or IIa using only tracts with 1970 median income of $12,000. Eligibility Requirement for Household to be Selected for Interview Must have child/children age 5-18 Must have husband and wife living together Original Sampling Design for Selection of Household In each sampling point cluster, a randomly designated household was chosen as the site of the first interview and each fourth household 251 252 from it (using a prescribed walk pattern) was to be designated household for interview until four were selected. Original call plus three callbacks on designated households. If no contact, or household did not meet eligibility requirements, substitution of house to right, then house to left. MODIFICATION There are no modifications in selection of sampling point cluster areas. Modifications in screening and selection of households need to be made because of the imposition of filters to households with child age 5-18 plus husband and wife living together. This makes a skip interval of four households and heavy callbacks on designated households imprac- tical. At first designated household, if contact is made with an adult, interviewer may ask which houses in the group of 19-20 included in the originally defined sampling cluster (allowing for desig- nated and substitute households) have both children 5-18 and husband/wife living together. This includes, of course, asking about this first designated household. If only four households of the 20 qualify, then these four become the designated households. If eight quality, every-other-one becomes the designated household. If 12 qualify, then every third one (OBJECTIVE: Choose a random sample of households in the originally chosen area which fit the eligibility require- ments). If the first designated household at which inquiry is made is eligible, an interview is to be completed there. If no contact is made on the first call at the first designated household, the interviewer may proceed immediately to the right substitute household to try to reach someone who can answer whether the originally designated household meets the eligibility requirement. If it does, three callbacks will be required on it. However, if it does not, interviewers can proceed immediately at the substitute household, using the respondent there as source of information on other households. If in any sampling point cluster block there are not four eligible households, the interviewer adds additional households beyond the first 20, including proceeding to another block according to the original sampling instructions. 253 If information on households in the block cannot be obtained at the first contacted household, proceed with the skip interval as originally planned and ask for such information at second designated household. THIS MODIFICATION IN SCREENING HAS BEEN MADE TO: Preserve the original choice of geographic sampling point-by- probability methods. Preserve the random selection of households, but change that random selection to randomness of those which meet eligi- bility requirements, rather than of all households. THIS MODIFICATION IS NECESSARY BECAUSE THE NUMBER OF FILTER REQUIRE- MENTS FOR ELIGIBILITY GREATLY REDUCES THE NUMBER OF HOUSEHOLDS WHICH CAN FALL INTO THIS SAMPLE. The most extreme example is in Pontiac where: 40% Households with school age children Black households = 40% Sixty percent (60%) of black households with school-age children have a father present. This means that the probability of a household being eligible within the selected areas in Pontiac are: p = .4 x .4 x .6 = .096 Therefore slightly under one in 10 households can be used. Sticking with a skip interval of four means one would cover an area of nearly 200 homes (including those skipped) to obtain four interviews. This is clearly impractical. Source: Written communication from senior statistician of research agency hired to conduct survey. APPENDIX C INTERVIEWERS' INSTRUCTIONS MICHIGAN STATE UNIVERSITY (21W 00 nun-aw MY um LAW“. - mum.“ - «an November l5. 1977 This is to introduce an interviewer from Market Opinion Research Company. This interviewer is asking your participation in a study of the quality of life o’ families in Oakland County. Michigan. The research project and questionnaire have been developed by the Departments of family and Child Sciences and Human Environment and Design. College of Human Ecology at Michigan State University. The project has been funded by the Michigan Agricultural Experiment Station. You and your spouse's cooperation in granting a short interview and in completir; self-administered questionnaires will be sincerely appreciated. and yOur nascs will in no way be linked to your responses. Sincerely. o 722%,»); A“ Margar M. Bubolz. Professor Family and Child Sciences Mud... Ann C. Slocum. Assistant Professor Muman Environment and Design 25¢! 2555 Job # 7117 November. 1977 OAKLAND COUNTY LIFESTYLE Interviewer Instructions TYPE OF INTERVIEHING TECHNIQUE For this study you will not be doing any actual interviewing with a responde' You will. however. screen households within each area to determine eligi:ility for placement of questionnaires. and you will be required to return to thcse households to pick up and verify comp etion of those questionnaires. O b ELIGIBLE RESPONDENT/HOUSEHOLD In order for a household to be eligible for placement of questionnaires. the following criteria must be met: 1. The household must be occupied by a married couple. 2. The couple must have one or more children from five years of age through 18 years of age. 3.) The husband and wife must both consent to filling out a questionnaire. In order for a household to be considered complete. BOTH questionnaires are to be completely filled out and must be accompanied by a signed consent fcrr. RESPONDENT INCENTIVE In order to show their appreciation for respondent's co-operation. Michiga' State University will issue a $10.00 check to each family who participates in this study. These checks will be mailed directly to the household approxira: 7; four to six weeks after they have completed the questionnaires. Additionally. a summary report of the findings of this research project will be mailed to tre participating households upon completion (this will be a couple of months 5526‘ receipt of the check.) UOTA Each area has a quota of four completed households. This means that four husband/wife sets and consent forms will be completed for a total of eight questionnaires per area. SAMPLING PROCEDURE Standard sampling procedure is to be used for this study. Proceed to the corner indicated by a red X on your area mapsheet. Begin at the household indicated in the bottom right-hand corner of your mapsheet. this becomes your first designated household and should be written in on your first call record. If you are unable 2565 Oakland County Lifestyle Interviewer Instructions to place the questionnaires at the designated household. you will substitute by going to the residence to the right. then to the left. then by skipping four households from your designated one. and continuing this pattern until you have placed them with an eligible household. Please look at the following examp e: 11’ Desi . 9 d ‘g£::] [ii] [fillsza ‘5?) [ii] ;::I Egijljffl (if) [:21 This is the pattern that you will follow in covering your blocks to deterr‘re eligibility for placement. CALLBACKS There are three callbacks required on the first household attempted for each set of questionnaires to be completed. Let's examine some possible field situations. Since you can only place your questionnaires in households meeting certain criteria it would be futile to make three callbacks on a household containing a widow over 65. when you begin work in an area and run into a no answer at one of your designated households. check with the residence to the right. explain the purpose of your visit and ask if their neighbor meets the eligibility requirements. If they do. you should continue to call on that household; if not. ask the person you are speaking to if they meet the requirements and attempt placement. In other words, screen your neighbcrrccd efficiently for eligible households before attempting callbacks and you will minimize the number of trips made to an area considerably. INTERVIEHING HINTS * Make sure that at least one (either husband or wife) has signed the conse': form and is certain that the other spouse will do so before leaving tie questionnaires. ' Stress confidentiality. * Remind respondents that the $10.00 and the sunmary report will only be sent to households who successfully complete both questionnaires and sign the consent form. * State a specific date and time for pick-up of questionnaires and arrange for both spouses to be present if possible. ‘ Call your respondents before you return to your area to pick-up the questionnaires. 257 “I" U U“ WHY. 'fll I”? W Teammate iotawiew ”late call record. Teroinate (— intervie- Mlase call record (code 4 to net laava tumonnairas with spouses Ma cutest with use»! ular rife ‘6“. 7. I. v I ‘ “late cell record hm. seli ‘ 9 a 0) (code 0) tie", esplaie purposes of study d inactive parent IIII'I'1‘327 ‘ “'7'“, T ‘ Minna IIEOPwin . - ° ' cool I ll . e. am a child ope s u» i: “M '5; " "“7 ‘ Wily living in sample? ’ 8. new and um mu, living together in household’ ES. ELISISJ ls sparse also anal tive further uplaaasien to both Diva further ewolenasien to one spouses epusa sewifl? Gnome call racer: (code a) '8 b a; leave ouestir neires with spocse m 7 "pr the consent for! to inside ) 22”,,t'm‘m and "h'" of front cover of euestionnai'e of spouse whc is not hora. Plate with "IN (code I) and be“ inside 0'. a II' envelope. u not okafll record. plan pict up date seal. a we Diva spouse who is presert his! late signed consent fore with you If "gum". “a envelope. and paper clip to call record (“late front (code 3) and out 'y-——) Phone to date—inc ii . ~ ‘ of cell record. plan pick up one spouse eonsene: Phone to deter-in if both are and "a I ”lated and ready for pick up Plan another ”at “hath-an am we we Phone laser to deterrina on time r ”to" to pics in Questionnaires if bot: are mleted an- . road, for pics up . Pica up in:c‘:‘a'.ec lies , '0 Gear 0' a 12' mines and edit Questionnaires ”on men" nestionnaircs follo- roce-vnded pneodwu’ pics up date TES I: one ooe.-so :95?- if one or both Mitimiflt are and time Leniogsgsv: rt} r?“:‘ I, .T :u": ‘T:;[. ‘, fl ' "P ‘c.’..‘ ranserwas owns 'a ‘ ”‘ envelope. insert eovoIesed call ”a,“ u pick » .mmmu,“ Cavitie'cell T's}- roconlendtonsentfoasnse-e a ”preset large envelope. 1 “on I' a 12' envelopes and edit cant fore- it gig-'9. I ti iras ID: use Questioerei'az than for both signatures on again if no um a. consent for! or inside 1hr. If g" is twine. return :1' ' felle- reea-andod procedures. if one of both Questionnaire; are :fi::g’,::'l:g":;§“ partially or totally insulate record. Pepe? clip emcee fore and cow- plated to" record teat-w and 900%." the-so am in. ope-Jinn- oeire'. to ‘0' a l)‘ Wt‘lini' . Alia-en to In" unravels-r 12° outlier". In.“ -] ire Hun- aui tale-ewo “struck-r. eve-r- in “It“. an» n. I 25€i RICHIGAN STATE UNIVERSITY mace U m M? “I? MING ’ IICHICMN - .:a Fall 1977 He. the undersigned. willingly consent to participate in a study about the quality of life of Michigan families. we do so with the understanding that our responses will contribute to the goals of the research project being conducted by the College of Man Ecology at Michigan State University and the Michiga' Agricultural Experiment Station. The purposes of the study have been explained to us. and they are repeated in the letter attached to the questionnaire. Thus. we have knowledge of the aspects of the study. we agree to complete the questionnaires as accurately and completely as we are’able. we further understand that our names will in no way be linked to the answers we have given. and we reserve the right to withdraw from the study at any time. we desire to participate in this research and consent and agree. PLEASE SIGN YOUR FIRST AND LAST NAMES. Wife's Signature Date Husband‘s Signature Date Street Address City/Town. State Zip Code He. the undersigned. guarantee complete anonymity to the persons whose signatures are above. Their names will in no way be linked to the responses given. He further agree to pay the abovasignad family an amount of SlD.DO upon receipt of the two completed questionnaires. It will be happy to answer any questions they *wight have about completing the questionnaires. Please call 517-353-5389 or Sl7-355-l895. ”(WW/M X‘MO/ MM Dr. Margaret M. Iubolz. ProfeSEbr Dr. Ann C. Slocum. Assistant Professor family and Child Sciences Muman Environment and Design APPENDIX D FREQUENCY DISTRIBUTIONS OF VARIABLES .a.. 42...: “I _m.v. :— —¢.m. m— .o.m. c Ao.v. _— Ao.n. c .m... c. Ae._m. _h .e.ch _m .~.on. _o .n.mn_ an ac.a_. we Ac.m~. cm .a.nn. ca .n._n. an ._.~m. Nu an._m_ an _v.=m. :c .=.mn. en 2*...3 :c .:.»m_ on _r..._x ._..._ A..~m. A_.¢~. .m.a~. a~.n~. .q.=n. .a.cm. .~.mm_ -.~m. Am.am. Rm.xn. A=.=m. .x.=m. ~a.nn. ~m.sm. .e._m. Aa.mm. .:.ce. .e.=m. .m.nm. .~.:m. mm Nm ac cc Nu m: «a be ac ch 9: :5 an _x 1.7.7:; .;__4 x_.Et; .m.mm. ~c.vn. .v.qm. .a.mm. Am.nnv ~_.nn. .m.c_. Am.h_. fic.m__ .m.c_. .o.a_. .c.c~. .m.v_. .n.v_. .v.n_. .m.:~. .:._n. _a.n_. ._.;_. ~«.x_. .s on as us on we hm cv mm vv v0 Nm an cv we :9 CM _v 7 t;._:._er x...;s¢ .2 1:... ..:.==:_ .m.¢_. ._.=~. Aa.n__ Ao._~. .m.c_. .m.c_. Am.o. .c.n. Ah.c. .n.c. .c.__. .w.a. .5.9.. ._.~_. .~.__. .~.=_. 2.4: ._.h_ A...__. .x.m. _e we 9v he _v mm w. v. o~ _N 9N KN v~ :N c— :N .m.:. .:.n. .m.a. An.c_. .m.«. ._.n. .c.n. A~.~. .~.~. .m... Ao.m. ._.n. .v.m. .m... .e.m. .m... .c... .~.~_ .c.m. ._.m. _r: .1. affix :- >_.r:2 v. n— @— v~ _— o. ._.n. -.~. .a.m. .m.v. .m... _m.. _n.__ .m... .a.. .e._ .c.». .x... n. o— z x;;:::: .m... a .n-. cages .a.. ~ An-u col vocao mace» ~nmuouw! .o— .o.m. m A.-. coeds .m... m .n-. gel on: _ncomuoa no» o_nm_mu>w xenon mo ~=aon< .m _~.~. m ~v-. color .m... m A_-. col com: um xenon st .o .~.~. m Am-. coloa .m... n .m-. cow oua_wuaw use» .s .e.. _ An-v coco: Av._ _ ~n-v can use: an on On m_ouu a“ o-aeuccuwoo so: .0 .m... w AqNN. coco: .c... c ~m-. col ou>_ouon uuoamo» no neocl< .m .v.. _ AQNN. cola) .a.. ~ .-~. eoe u:_u:o_ua was nmocomo_u .v .=.. N _n-. color .q.. _ .mwm. cue odo:o_uonwo coquuouuu ace o>oa .n .v.. _ Av-u :oeo: .-~a eoe zone—_gu be :o_.w:_u>m .N _a.. n .v-_ codes .a.. n .-~. eoe emscam we =o_uw:_n>m .— 3: z ;_£..a;. _: 2.....24295... ..>.:..,:< 12...: .5... 11.2.2.3 _: z....:..:7...... 02.3.3. 1:: 10:35:93.--.7: 036... 259 215C) .3.. .<._. .—.. .:.. .a.. .a.. .e.. _=.. .q.. .a.. _e.. .a.. .e.. .m... .v.. .3.. .=a.= "J ‘I "J ...... :m ...9_. mm ...:_. em .e.:. _u .v._n. me .:.~_. 2v .m.~.. cm .a.n_. =9 .3.... an ...~_. n~ .~.e.. mm .a.w.. aN .m.;. 9. .x.m. m. ._.~. 0— .v.... —N .w.... an .n.n.. xw .x.m. m. .m.x. a. ,o z 1.....2 ._.... .c.nm. mm .a.a~. e: ....m. «a .:.a~. he .c.em. an .m.mn. mu .v.=n. xo .m.mm. m. .n..n. a. .q.en. .. .a.a~. .o ....m. m: .c.m_. we .o.a~. he .n.m~. mm .m.a~. co .o.cm. mm ...mm. :a .~.vv. am .m.an. xx .r.z..u.._ n.5m. .0 .m...n. 0.. .a..<. cs .m.mm. ma .~.mw. mm .9..~. a? .c.=~. we .v..~. av .v.m~. .m .c.m~. we .m..n. =5 .m.a~. no .o.v~. mm .x.=m. ac .c.nm. en .=.mm. vn .M.N~. am .=. .c. .m... m .m-. cows) .c.m. x .-~. :91 Logauxo. x..:n. 02.. .oo.c:ot< .5. ._... a .m-. coca: .m... n =9- o>az no. co.usu.::llou up 3:.a .o. .v.. . .n-. coeds .v.. _ .-~. can mac..uou mmocaxo :ao x_.mo=o: ace x.:oao so: .m. .~.~. m .nNN. coso: ..-. coo ao;:..=louoa \eou.>.s .cos a.o=am:ag .5: .9. .m... n .m-. case: .v.. _ .-~. col acuaeue A..lw. .o ma9:.:.g.~z .aaunx .m. .a... e .w-. eoaax .m... m AmNN. cos Logaoao. as so» na:.;h .m. .x... v .QNN. sues: .a.. n .m-. :9: was: one mco.m.uu¢ Aux .._ 3' z u.s...u+ xztlv a.ca-all...1l.. . .52... 2.3;- - . . A. 9...... 2(51 .a.. N .a.. ~ .A.~. o .m... n .v.. . .e.. . .a.. u .z.. N ..... m as 2 .=a_= .m.=.. m~ .=.x. a. .o... .. .a.v. ._ .._.. m~ ...e.. mm .o.«. a .c.n. a .v.nm. ax .:.vm. x. .x.o~. so .v.m~. .m .=.~v. ea .m.ce. ms. .m.=m. as .o.w~. cm .e.xm. on .e.:m. a; .m.m. m. ._.e. c. n z au.;u..u= 39:39.; .x.em. a. .v.em. he .e.vm. .. .=.mm. v. n. _e .c.nn. ~c ...ee. a: .=.mm. a_. .a.zn no .m.mw. an O: z 7:..1..=£ .A- ~I..Z .e.=.. mu .a.m_. _m .:.._. am ...e.. mm .e.n_. an .c... A. .=.=.. _v .=.m_. a~ .2.._. mm n.:_. no u z oow.: .o.m. a .o... n. .m.w. a. .m.v.. ~m .a.. ~ .5... Q .~.. v .v.. a .:.e. a .=.e. a a z .........z._ ....m.r..... A..¢:2 .2... v .~.~ c .3.2. .. .e.m. ~— .a.. N .a.. N .a.. N .a.. a an... e .v.. . oo 2 .3932... .-~. cues: .n-. eoe u=.>.. .a acaaeaam .~.~ a .v-. case: .a.. w a.-. cow assoc. x..lwa .auoh .o co.ua:_a>a .n-. coco: .e.. . .v-. eoe .cxcs 195::OL. n o... A..sa. .m-. coco: .v.. . .-~. eoe .cxoa 301639.. m u... .~.u. m .-~. cued: .~.~. m ..~n. col Acnco o» o.nmco macaqu. ._~ I! z a.:...;. nullitl it. .s .1c==..::.--._-: u.s:b 262 .m... .e.. .c.n. .m... ...~. .3.9. .m.o. .n.:. .m.~_. 1...... x... \.. .n .227. v. .c. .m... .e.. .m... .=.. .~.~. .a.. ..... .o.m. .a.v. .m .... .a.... ...:~. .m.-. .~..~. .m.a~. ....m. 3 . ._. x... :1. L“) N .— v. :v mv cm .0 co .5 2 .d 0...... ...:._< ...n c .a.. N ...m. n .N.N. an .m... m .m... m .a.». =~ .m.q. a. .m.v_. ~m .=.~.. aw .m.~_. m~ ...~_. .~ .a.n.. av .N.a.. n9 .n.a.. .v .n.MN. mm .:.a.. Nv .N.a_. we a z .aae.. .993 x.-. .::;< .m.n_. .:.n.. ...z.. m.:.. .x.m.. .m.e.. .m..m. .m.sm. .N.mn. .m.mN. .x.-. ..c.a.. .m.cu. .e.m~. .N.NN. .a..N. .m.:.. ...o_. .we... 3.4 3.. cc .m 9v an .0 av nm em .2..:.. .s._.. .n.am. .x.xm. .m.vm. .h.mm. .c.~n. .=.=m_ .c.a.. .m.=~. .a..~. .=.n.. .o.m.. .n.v_. .s.._. .9.3. .o... .9... z ‘3...) —_ :13. u.:: .;:;< .n.:.. 3.. ... xx 5. rm .5 mm ac vv an. he an mm mm ON .N PQ .- 2 .22... :1... :1... ....:.< .n..~. n: .m .94. a... :13...— .3 .m.an. c: .n.mm. a. .a.mm. on .x.a.. Nv .x.x.. Nv .m.a.. NV .a.m.. .m N.m.. vn .N.m.. vm .v.a. .N .m.m. a. .x.m. n. .=.e. a .o... .. .=.¢. x. a z 2...... ..rux ..;.. 1...: . . .. ......... .¢.x_. _e .m.e. :_ .wnn. :95 .....m.. .3. 3...... .. .71.. :92...) « .5352-3uhvu39 .3 .12}. = .5 .5 .fi. .4... x ~.~. . ._-. :se .m.c. e. ...m. n .mnw. cuss: emcee... ..z.> .a .¢.a. NN .n.~. c .NNN. :91 ...=_. an ...~. c .n-. coed: «use; «a macs... :.a.»ou:w .5 .m.o. a. .a.. N .NNN. :oi .e.m. N. .N.n. m .va. cute: eucacsauxu. a .a as. .c .a.x. :n .m.e. :. .:-. cos .a.x. :~ .m.=_. mm .nnw. coco: «3.31 a he o>..a a oxah .m .m.~.. =~ ...o. m. .-~. cow .m.N_. xn .m.N_. xN .o.~. :oeoz wuquOLa a co uuzaoxs. 4.03 .9 .o.m. a .a.e. _. .=-. :9: .m... m .m.v. ._ .c.~. cows: wmmc._uo. .n:om.on mmzum.: .m ...n. . .m.e. c. .-~. cue .a.. u .m.«. .. ..-. zoos: .m:...a. “nan 0.0: .9 mac; cs azoam .N .m.e. c. .m... m .=-. cos .w.c. e. ...m. n .vNN. cues: ezox no 33. oz. .msq cuz.ua:. 25.. 1:9;m .. v z a z .m::.. .2. ”9.35 .::x 3:3 :3x :3 :o..: as: .22. ....>..Z \.. z... ...:..< .LnlalaxivalliI. his... .2.) {1.2.2.}. .1... .5... 7:27 .: x._..:.x.... 1..>_.........._ 2.....2 .5... 2.2.2.5... .3 {1.13.25.— 31.23. 1.... n.u.u...u.....u.....--.N-.. 92...... 263 \».N .n.n. x21 .23.... ...m. n .x... v .a._ ~ ...m. . .q.m. n. ...n. c .:.m_ m. n 2 u z a z 2. 2...... .2... b... :15 4....) 4....) $1.... 9...... .......< 27¢. .:....< 9...: _......< .m.N.. aw .x.a. an .x.a. Nn .m.x. :. .we:.. :25... ...:.... 9...: ...:..< .a..». n: .x.xm. n: .x.x.. n.m_. ‘9 .:_:.. .39» IDS—u we QM .;: =.....< .m.mm. 35 .:.nm. m: .:.mm. an .a.:u. as .e::.. .21.; \..u.... ......... .8... N. .m.:_. mN ._.xn. a: .c.~m. ex .a.~. :95 co::: «x..an a 5=uu.< ._. .n-. =9- : 9.0) w.:o>o muuoam a sco.u< .:. ”9.35 .sox can so. as :o..o :9: {.4latl.!..‘...|,.“l4. .saac..:au--.~-: 9.33. 264 .x.:_. we .m.c. a. .5.... an .;.n. .. .x.a. wn .e.: _n .3 . 1' SN .x.a. mm .x.:. MN .x.:. an _n . m; v 7.. ... . s... a... ._.;_. cm ._.n.. nw .....,._. x... .w ...__ v...‘ ......a. r.— .a.:n. _; _.....— A-._. x. _ 2.2.3... .3.cu. xm ._.vn. em .:.m~. xm .:..~. he .a.:~. n: .x.n~ .c .x.c~. :c .71.... m ._.wn. em .e.:«. x; .e.xe. c: .e.:m. x; ...m«. :x .x.:.. 3: .x.:.. :: .._ A... .. 5.. .1... 222—r. ... .17. 5.... .:.n.. .;.a.. .e.mw. n.m~. ...c.. ...x~. n.:.. .m.;.. .e.~.. ....n.. .:.n.. ...:.. .n.ww_ ...n._ :...A_. ...m.. .x.w._ .m.:.. _..-.—..— J.» a) :9 av LG N h ¢z 0m av we hm an :e _ m; 0” \.. u .......< 1.9.... .2 .;.a.. we .m.:~. am .2.MN. _m .....M. v.5 .14... m... .e.m.. :m .m.:.. um .w.... an .m.e__ Nm .a.:_. 5m .m.:u. cc .v.m_. :. .:.._. x« .s.... :2 4.. .3 .. I.- .. I... 22:3» .m.x. a. .¢.a. NN .x.a. - .¢.m. m. .w.... an ...c.. cm .e.m.. :m ...=.. QN ...=_. v~ ._.~_. nw .~.~. m .c.n. u. .2... a. .¢.¢. mm .m.:.. «a .~.:_. cu .m.:. e. .e.:. _u .a.:.. .. 2...... :1... L z... _._.x.< .a.v. _— ...h. c. .c.«. x .v.m. w. .=.¢. x. .w.m. w. .m.x. :. .m.@. a. .c... .. .x.m. m. .a... .. .c.«. x ...w. s .s... x .x... m. .x.m. w. .:.e. a .m.:. w. n z .c.:.. ...... ...... 1...... . c .......< .3.. ~ .~.~. m ..-. :92 .m.e. c. ...~. 0 .v-. nose: «3:» .0. 4.53 ea c. o.n~..a>a u.omi_s emu: ... .e.. . .m... n ..-. can .a.. ~ .v.. . .m-. cots: «sex so. vans.» an on .c. .v.m. ~. ._.m. 5 .-~. :ul .:.9. a .m.m. n. .v-. coco: mo.u...aa .30. :. oucuv.ucou we: as 30:. :c» .0. .m .v.m. ~_ .v.m. ~. .cNN. col .¢.m. m. .q.m. ~_ .v-. coco: «so. vuuonrou was mou.eva u: zcgm go ._oh .5 .m.v. :— .=.v. Q .-~. col .o... .. .o... ._ .QNN. cane: peanuon acauuonl. an a... .v0. :0» 0.3: .o .~.~. m .c-. can .c.. N .~.~. m case: wmm.x .o as: a as» o>.u .v .:.m. a .v.. . ..-. :9: .a.. N .a.. ~ .n-..coaox «0.0. .0 guns. n .oncu .n ...m. n .c.m. w .Nuw. col .m.e. c. .m.e. c. .v-. nose: NunaQIOu use» muonco 0; roe. as. .9. .~ .x... e .m.v. :. .-~. :91 .m... m ...~. c .nNN. coco: «o>o. xozm .3 ..oh ._ a z n z .m::._ .:. .o.a£ .39. was: as..o no: .. .3; Lb>zz \.. z... ......_< .. ........ .5. ... >. ............. ....>_.:....._ [122 ...... 12:53! ... x... 1.2.7.... ..>..fi_...z .53 1...........=..0..._--.m-._ 97...... 2(35 .n.~. c .e._ _ ._.u_. RN ...». n a.e. :_ .a.e. :. n..w_ _; ....m__ an ...:_. e _:.v. x ‘J .m.x.. _w .e.:. n .3.2_ A... .. n .::;< .c... n. .:.x. x. .x.~n .m .:.~.. a~ ...~.. hm ...~.. KN .m..m. an .w.~v. a: ._.nm. Nu .m...~. .wc .:.mm. c. .:.n~. _c ... x... ....:. ...:.. .:.x.< .m.c__ .v.m_. .c.m_. .v.~_. ._.=~. ...-. .m.:~. _:.~_. .a._~. _:.an. .e.._. .¢.-. .7...” .— J...) \< m .::;< m~ Cm mm @m me No C? 3v m: .:.m~ cm .m.mv. ma .c.¢_. we .=..~. 5v .c.nm. mm ...v~. vm .m.v.. we .v.n.. cm .a.._. =« .m.~u. :m .m.o_. n» .v._~. x. .we... .;.3 ;.t. ;.:: .::;< .~.m.. en .n.v.. mm .~.v.. mm .m.v.. ~m n.m.. en .v.n_. on .c.n. a .=.«. a ...=_. v~ .c... n. .e.a. .N .¢.¢. a. n z .n»:.. .. 2...... 5...... ...c.. .:.x.< .n.c. a. ...n. c. .3... c~ .c._~. he ...o.. QN .n.o_ q. .q.. . .z... e .o.m. a .n.c. m. ~.~. m .n.c. v. a z _;_:._ .39x .2; 192.. c .::A< ...~_. BN .:.v. _— .c.v. a .m.¢. a. .o.m. a ...n. n .e.. _ .m... m .w.. _ .~.~. m .m... w .=.q. a 0. z ..::.. Leg» \;.:0 .:ca< .~.v_. mm .m.~. col .w.m. n. .m.~. evict «um: .u:0muon Lou xoch :0. o>wu .c. .a.. N ..-. cal .m... n .nNN. cone: «2...... .5 .60.. no» u:.:uocOm so. o>.u .n. .w.. _ ANN~. :0! .c.v. Q .-~. coco: co.n..vv a can: .0 so_ao.a a o>.cm so» n.o: .o. .v.. _ .=-. cos .q.. _ .QNN. coco: co.:.:c m.: =0» o>mu .m. .a.. N .v-. :9- .m... n .-~. cones wcc.u~l -uo.:. so: :0» v>.u .v. .1.. ~ AwNN. col .c.n. c .-~. coco: mo.nausc.s=u :0. 9.5! no xuuoco so» u>az o. u:.:.ol0n 0a .N. no 2 .c. ”can: use» moo: cou.o 39: .o>oz .1uy4. «.ul.. J... .J.... .n-qh‘lfllJ‘ll'l‘llu .vo:c..=ou--.n-a 9.55. APPENDIX E RELIABILITY ANALYSES .oume so.» wo>wouwp coflmehomcfi vcm mau.>pvm .msumum .o>o. mo accoscopm mmv:.o:Hm .oume.umo x...wnmw.wp 0:. mm vmm: m. mcm.m ucmfiufimmoou m.£uwn:opu ”ouoz wNo. m.c.. m.v.o. .v.~. :05 nam. nom.o mov.m .o.~. nose: moans scum mmop=OmmL .mcompunpoucw mo xocmscmum .o ooo. 5.... ww~.~ .v.~. :05 «mo. mvv.. New.. Ao.~. :mEoz meom zonescmum coflumspomcH .m mum. om..m .mv.~ .v.~. :05 mm.. wom.. woo.. .o.~. nose: onum xucoscmum moow>umm .v awn. moo.m mow.. .v.~. cos o.m. mom.. wmw.. .o.~. case: o.mom accoscmpm maumum .m wow. :vm.~ wmx.m .w.~. cos oww. .m..~ .mo.e .o.~. case: o.mum zucmscmhm w>oa .m vmn. Ncc.m mo..m mmHN. :05 mo.. qwm.. mm..~ .wom. case: onum n.5m:o«:mQEOUumeh .. m:@.< sewumw>wo upmucmum o.mum mo :moz onuw .coz van 2950: Low mo.=um xucmsco». moup30moz tam mEMH mo mwmx.m:< A....nm..om mo xpmesswun.finm o.nmh 266 267 .oumsflumo xHMHManHoH on. mm vow: m. asa—m acoMUMwwoou m.:umncouu ”ouoz emu. Noo.m wo..~ :05 «on. ewm.. mm..~ nose: m.mum 05.. you mu..m..m.m .o m... mmm. owe. ~m~.m aw... mam. oNN. .m.~. :05 o... m.m. wee. m.m.m .wm.. .NN. me.. .wom. case: x.m3 ho m>wuv m meH .m. .N.. wv.. mum. «N..m .hw.. mwv. 5mm. .m.~. cos was. mum. mom. .mo.m .mm.. can. vow. .mom. nose: Hammonm :o Hmcumwou xuoz .v ooo. mwm. mom. ow..~ ON... o.m. wwm. .m.~. cos moo. .nm. ooo. ovv.m mmo.. coo. Nwe. .mom. nose: mm:..mom .m:0muom mmsumwo .m v.0. va. woo. wmm.~ mam.. mmo. m.m. .m.~. cos wNo. w... .vo. .mm.~ mmm.. Nmo. .oc. .wON. nose: wcfimeu use: an ucmmm .N Non. mom. «.m. mm..m .om.. mwn. .v.. .m.~. cos owo. n... mvo. mvo.~ oww.. emu. moo. .mom. :msoz swsumMOu mafia wcomm .. wopo.og no.0.oo @mum.0O :Owumxuhhou =OMHN~QHHOU Eon: mm EOHH mm :Owuww>oo Emu. m. m.m.u.sz .mu0husmu. cam: m£n-< VOHmavw fiOHUOHhOU ®U=mwhm> cam—z VHGGCMHW m.mom w.muw Ill II- .:02 new cmeoz pom m.mum mEHH,mo m.mx.m:< AuMHMAmMHmm mo humeszwnu.mum oHnmP 268 .opmswumo xufl..nm..op ecu mm com: m. acowufiwmmou 059.0 m.:umncouu ”ouoz mowow. wmmmm.~ omwmw.m :05 m.mmw. mncmn.m .Hnmo.v 00503 0.00m accoscouu o>0. pow mowumwumum .m emmee. e..nv. eeeee. meeme.e .emve.~ amenm. «emem.. .e.~. gee ommnn. moomm. oevno. cvvno.v mmmom.m mo~.o.. mmvnv.. .e.~. 00503 mmflx no was 0 30> o>«o .v .mvmw. Nowmm. mom... .on.m.m noonm.~ woo.w. m.wmw. AQHN. cos VOBHw. onmnm. memom. mn.cm.m anmm.m n.0mn. mmwcw. Ac.N. 00503 :0» new: 0x0n ho swam. m xencm .m Nmomm. ovmwm. mmmmn. mmvmm.m m.oo..m .m.wn. mowmn. AVHN. :05 moown. m.wnv. mvooo. movm..m m.mmm.m mow... comes. .oHN. 00503 x:wm&0u 9:0» exonco 30:. so» we. .N emmow. whe.o. mmown. om.ow.v .w.nw.~ noomm. mmcmm. .v.~. :05 ammm.. «meme. ewe... m.oeo.e neeem.~ e..vm. eeomo.. .e.~. nose: o>0. 30:m\..oh .. 00.0.09 00.0.oo “www.0o :0..M.oppou 00.00.0upou 50.. w. Emu. m. coflumfl>0o .. m. o. .u.:: .m.oh-Ee.. awe: m:&.< cmpwscm wouoougou oucmwpm> :00: vymvcmum 0.00m 0.00m .00: 00m 0050: now 0.00m zucmscmgm m>04 m0 mfimx.mc< xufl.wnmw.om m0 zumsssmuu.num 0.20h 269 .oumsfiumo xum.finmfi.ou 0:. mm com: m. u:0.u.mm000 0:0.0 m.:000:0~0 ”0.02 evmvm. mommo.~ ommow.. :05 o.o.m. nmmm.. Howaw.. 00503 0.00m xocoscmpm magnum pom muflumfiumum .v nmmvm. am.cn. mm.nw. om.~o.~ coca... oHan. ono.o. .wHNV :05 oHeow. momma. mummw. wmwcw.. evowm.. NmmHN. mmnfio. no.~. :meoz mmfiuwmwnm 030» a. mocovflwcou mm: o: 30:. :0» pm. .m mmvmm. ovvfiw. oHoaw. mm.~o.~ cmomm.. .HHNN. owwnm. .wHN. :05 mmomw. emu... wHVQw. mn..m.. mowvm.. mmwmn. mmmve. Ae.N. 00503 :0» muuonmmu 0:0 mwquum m: 30:m\..09 .N e~e.m. e.mmm. .mq.m. memmm.. ommm... memes. emm.e. .e.~. :ea cemmw. mummc. omomn. mnmww.. omco~.. mmnmn. H..mo. moHNV 00503 comhom acmuuomafi :0 mx.. .00m :0. 0.02 .H 0.0.0: www.mo 00.0.09 00.00.00000 :0..m.opuou New. m. “mu. m. cowpmfl>oo Emu. m. 0.0.».32 .muokusoum :00: m:a.< vmuwzcm wouuoppou oucmfipm> :00: upmvcmum 0.00m 0.00m .00: 000 00502 pom 0.mum xocmscmpm magnum m0 mMmAHmc< A...«nmw.mm m0 xamsszmuu.vum 0.20H 270 .00000000 00000000000 000 m0 000: 00 000.0 000000mm000 0.0000000u "00oz 00000. 0m0.m .00.~ 00E cmmmn. wom.0 000.0 00003 0.00m 0000:0000 m0o0>00m 00m 000000000m .0 ~00Nn. nwmmm. «onus. 00N00.N n.00m.. nwmvw. 0~0N0. avHNV 00E mmomm. .omov. eNovo. .m.~... 0m~00.0 ommmm. 00mmm. .00N. 00003 00000000000 00» 0x00 00 xw0000 so» 0>0m 00 000000000 00 .m mmmmn. me.mm. .mm... .eomo.~ ove~e.. momnm. .oeem. .e.~. :00 mwwnm. 00000. 0N000. m0mwo.0 wmvv... ovomo. omem. .oHN. 00003 :0» How 0003 00 00 0.00000>0 m000000 000: .N 0.0mm. monvv. mmmmo. ommmm.m mmnwn.0 Nwmvn. 0.000. AVHN. 00E woman. mwww.. . nommv. 00~m~.. mnmno.0 Nwwvo. mvwwm. ~00N. 00003 :0» 00m 000000 00 00 .0 0000.00 0000.00 0000.00 00000000000 00.00.00000 0000 w. 0000 w. 000000>00 000. 00 00000.0: 00009-000. 0002 0000< 00000m 00000000 0000000> 0002 0000000m 0 0 0.00m 0.00m .002 000 0000: 00m 0.00m 0000:0000 0000>00m mo 0000.00< x0000000000 mo 000EE=mnu.mum 0.00% 271 .00000000 000.0000.00 000 m0 000: 00 000.0 00000000000 0.00000000 ”0002 00000. mm0.0.. 00000.0 000 00000. 00000.. 00000.. 00003 0.000 0000:0000 0000000000. 000 0000000000 .0 0N000. 0000.. 000.0. 00000.. 0..00.. 00000. 00000. 00.00 000 000m0. 0.000. 00000. 00000.. 00000.. 00000. 00000. A0.~0 00003 00000000 0 0x00 00 00.0000 0 0>.0m :00 0.0: .0 00000. 00000. 00000. 0000... 0.00... 00000. .000... 00.00 000 .0000. 00000. 00000. 00000. 00000. 00000. 00000. 00.00 00003 0000000 000\m00 0>0u .N 00000. 00000. .0000. 00000.. 00.00.. 000.0. 00000. 00.00 000 00.00. .0000. 00000. .0.0N.. 00000.. .0000. 00000. 00.00 0000: 00000000000 300 :00 0>0o .. 0000.00 0000.00 0000.00 00000.00000 00.00.00000 000. m. 000. m. 000000>00 000. w. 0.000.32 .0000.000. 000: 000.< 0000000 000000000 0000000> 0002 00000000 0.000 0.000 .002 000 00002 000 0.000 0000:0000 0000000000. 00 0000.00< 000.0000.00 mo 00000:mn:.0um 0.000 272 mmm. «we. owe. www.mmm ovw.mo~ awcmCOMHNHwH Hmsxom Hno> .wH vmm. mfin. omo. vmv.omm oon.mo~ mwcomm xfiweam mafia .5“ mmm. mgw. mow. mmh.oqm www.moH cowumUficsssou mo wcfix .0“ «mm. emu. _qu. ~mm.mmm omm.mo_ cofimmopnxo “mono; ammo .mfi 0mm. mvo. woo. ~vw.wmm nom.oo~ wovd>flw xhoz cfiozmmsoz .wH 0mm. «no. m_o. mvm.~om mom.mo~ mmocfismmfio; Hmsusz .m~ qmm. wfin. . mun. omH.omm aom.moH Hocuomou ow mmcfinp .NH mmm. chm. wow. mfio.mmm mmo.mo~ wcme mpm mcoflmfiumu >52 .fifi omm. awn. mum. mfim.vom www.mOH mwoow Hmfiumumz .c~ 0mm. om“. mac. _am.mmm wm~.oofi mm: Hmcomgom pom xmcoz .m 0mm. mmo. cam. wmv.mom vom.moH cam: mfl xocoe km: one .w mmm. oww. mfiw. mon.wvm co~.moH ommflppms mo coflua=~m>m .5 «mm. mm“. on“. Noe.wmm mmw.vo~ use; an “Homeoo .0 «ma. non. mmh. o~5.mmm chm.mofi vm>fiwuou “commom .m mmm. mum. mmw. nfio.omm moo.mo~ mcwmcofion nan mmocomofiu .q mmm. flew. New. oon.omm Hm_.mo~ cofluuommm can m>og .m 0mm. mfim. oww. Ho~.oom ooo.moH cogufiflgo mo coflumsfim>m .N qmm. non. awn. omv.~mm mNN.mo~ omsoam mo :ofiumsfim>m .H fimuofima COMHN—PHHOU COMHNTMVHLOU UGHOHOQ woumfimo saga MH e_mflu~:z _muoe-sou~ emufi “H soufi m_ ofinmfipm> azafi< vmumscm couuoupou oucmwpm> mdmum ewo: ofimom li'llll’t-.l -9]!!|.I-A“'-,lllli'llllll.|lll‘ all'l' ‘11'l‘ ‘1'IIIIII'. tin!!! .cosozirmo_nmfigm> coflumzfim>m o>muuowm< how mflmxfimc< xuwfifinmwfimm mo AhmEEsman.n:m o~nmh 273 .mumawumw Auwmwnmwfiop mm wow: acmmuwmmmoo «gnaw m.:umn:ouu "ouoz 5mm. nov.~mm Nfiw.fiofi wzmfi< mucmwum> ewo: ”ofimum pom mowpmfiumum owo. “mm. moo. www.mom mac.moH xoncm so» mwcowum .HN «mm. «an. mmn. Hem.vmm mfio.mofi ucmnmss spa: asap .oN 0mm. oHo. 5mm. ~mo.mom omv.moH couufifigu :9“: mafia .mH umuomoa :OMUmeppou :ofiumdmupou woumfim: wouwfimc soufi “H o_aflu_=z Hmuop-souH smu_ mH souH m_ «fiamfipm> acm~< coumsvm wmuuouhou oucmflhm> mfimum :moz ofimuw " I: . illl'll‘ 0.1!.-l| ‘i‘ttl .uoscflucou--.n-m ofinme 274 ova. ooo. o_n. ooh.on~ noo.no_ moomcofiomooh Hoaxom use» .oo ooo. och. ~_o. ooh.oo~ Hev.noo monomm a~o5mo meg? .NH No“. man. man. ~H_.mnm oom.ho~ cooooUocsasou mo ocox .oH mom. ooh. ooh. Nov.eom mofi.no~ coommmumxo “mono; cone .o_ com. eNo. ooo. ovc.oo~ ooo.no_ oooo>oo goo: ofioommaoz .oo ova. ooo. ooo. ~o_.om~ oov.noo mmo=H=omooo aways: .mo mom. «no. con. oom.oo~ oo~.noo “monomoo oo so» masons .No com. who. moo. ooo.no~ oom.hofi come «on mcoomoooo hm: .HH ova. Boo. Noo. oom.~m~ mmo.hoH mooow “mohoooz .o_ ova. ooo. mmo. Noo.mw~ NNw.~o~ om: oncomoog poo zone: .m ova. Hno. ooo. ooo.mo~ moo.noo ooo: mo xocos an: «op .o mom. ~55. ooh. ooo.~o~ “no.ooH ommfluoms oo cooumsfim>o .5 ova. ooo. ooo. vmo.mo~ owo.ooH use; on “homeou .o mom. “Nu. ooh. oo~.~o~ vom.ooH oo>oouou “commoa .o «om. ooh. ooo. ooo.oo~ moh.ooo wcowcoooo on“ mmocomoou .v ooo. “on. “on. onm.oo~ ov~.oo~ coopoomom ocm o>oo .o ova. moo. ooo. 5mm.om~ ooo.oo~ copooooo oo cooooson>m .N com. Non. «mo. h__.oo~ mmo.oo~ monomm mo coouoafim>m .o wouo~mo coflumfioupou cofiumnmuhou cmuo_@a wouofioo sooo mo moaoo_=z omooP-Eooo aooo oo sooo mo oooowom> mamfi< cokmscm couumpuou oucmwum> m~mom :mmz ofimum (I. zit]! .1. I: l.l!!! El!!!“ .coz:umo_nmfipm> :owums~m>m o>wuuomm< pom mwmxfimc< xuMHHDmMHQm mo xhwaszmun.wam mfipmh 275 .oumawumm xufifiwnmflfimp mm com: pcowuwmmwou «gnaw m.;umn:0pu ”muoz mum. Hov.vfim ovm.~HH mzm~< mocwfipm> :mm: ”ofimom pom mowumflumum ooo. ooo. ooo. ooo.ow~ ooo.noo xonco sex mocoopm .oN mum. mmn. c_n. ovo.owm Hum.no~ who: :u«3 wEMH .om nvm. «fin. oow. www.~m~ mov.no~ cmpwfiwco cu“: msflb .mH wouofioo sewumfimhpou acoum~oppou woum~oo wcuono swam 9H ofiafiugzz ~ma0huaoam EouH mH EouH mH mfinmflpm> mam~< vopmscm wouoopuou mucmfium> mfimum cam: o~mum .ooscoocou--.o-m ooooe 276 .oumewumo xuwfiwnmfifiou on too: acmMUWmeou acmfim m.;umn:ouu ”ouoz 5mm. www.mm mov.m Axocoscmum causeway .cosozu NLQ~< oucmwhm> cam—z umfiwum MO muwumwumuw Noo. moo. owo. ooo.oo noo.m mcoomfiuoo .mfi ooo. ooo. a_o. ooo.oo ooo.o mcoocfino o>oo .NH ooo. Noo. ooo. Bmo.oo _om.o cooomspooco 302 .HH ooo. wNo. ooo. ovq.oo oom.o xwpmco o>om .ouooeou .oH ooo. ooo. ooo. ~o~.oo Hom.o sex poo goo: .a com. com. «no. ooo.oo sno.o mocmnpm .o ooo. ooo. owo. ooo.~o ooo.“ moo; ho ms: .5 moo. coo. moo. non.oo moo.o oxen go omamo .o ooo. Boo. ooh. omm.~o ~o~.o xcmnsoo mo “amazohco .o ooo. ooo. ooh. oa~.Hm o_o.o o>oH zoom\oooe .e ooo. own. own. ohm.oo moo.o xooooow no oucoofiocou .o Noo. Non. ooh. ooo.oo ooo.o ouoamop ocm cooumgfiao< .N woo. ooo. _Oh. ooo.oo ooo.o compoa ocmuhonso .H fiwuo~®o COHumdoo—MOU :Owumxoghoo _umuomva UCHQHOD sooo oo moaoo_:z oooo+-eoo_ scum oo sooo o_ ooomogm> mam~< vopmzcm couomupou oocmwpm> mfiwum cam: mfimum .:mEoz -uoumz Eoum uo>dooom mmupsomom .mmfinmwpm> Aocoscmgm uow mfimxmmc< xu-wnmw_wm mo thsezmnn.mum manmb 277 .oumeflumo xuwfifinmflfimh mm vow: ucowowwmoou msafim m.:umn:ouu "ouoz wmm. ~mo.mm m_v.o_ A»0:03605m mURSOmmh .coev mam~< mucmwpm> cam: uofimum mo momumfiumum omm. ooo. ~5o. omo.oo oHo.oo mcoomoooo .oH ooo. 5cm. ”Ho. ooo.oo oo5.m acoomepooco o>oo .NH eNo. oHo. Noo. ooo.oo 5oo.m cooumEHoocfi 302 .HH HNo. ooo. e55. ooo.o¢ 5mv.m 5w5o=o o>mo .puoosou .oH 55a. 5mm. NO5. oO5.ov 5vo.m so» you one: .a «No. .eoq. moo. oo5.oo mo5.m mocmuum .o omm. ooo. Noo. ~N~.mv 55H.m mmox go as: .5 “mm. ooo. oo5. owo.mv voo.m oxom yo swam; .o ooo. o5o. vow. o5o.o¢ ooo.m 5=mmsou mo ucvs5o5:m .o omm. oo5. No5. ooo.5v 5ov.m o>o~ zoom5H~op .e mom. om5. om5. 5mo.mo Noo.m 5ofloooo =5 oozooomcou .o owo. owo. oo5. ~o5.mo «mo.m ouoamop ooo cooumpoao< .~ omm. Noo. mo5. omv.mo 5m5.m compo; «coupoaso .H wouofioo sewumfimpuou codum_opgou woumfioo acuofimo sooo oo oomooosz "mooy-soo~ sooo oo sooo oo ooomoom> mzmfi< woymscm mmuuouhou oucmwhm> mfimuw :moz mfimum .coz -uoumz 505w ©o>woumz mouusomom .omfinm55m> xocosvmpm you mfimxmmc< Auwmmnmwfimm mo theesmuu.o~um mfinmh APPENDIX F CLUSTER SOLUTIONS FOR EVALUATION AND FREQUENCY VARIABLES Table F-1.-Complete-L1nkage Cluster Solution, Evaluation Variables - Women. f8?§b85$2&E§‘cEt3?EEflES’ EVALUATION VARIABLES WIVES uFFEQTIvi MILLJKD\ OOnuooo-uuoouunnqoouugu ‘x‘ QQEuOMV— X".x °o$~«-~~uu..-u~~-xxx XXX QJOth-ZUMU) Xxx LU XX XX xxx oofiuu-««~~««~-xx xx xxx «ammo -‘<‘ Z O H .— d D -J <1 > Lu LJ > .4 y. \) LU '4- u. q \n O 2 <1 3 U') ”D T. mxaq...’ OHO..."Q.”OO~OO”~””-”“ xxxx wOZDZHQ(Hu ' xxx 00-3\coooo¢.ooooouoo.ooou-OXXX' KXK oww.;H'zom XXX XXXX oouoooooguooflooooflu..xxx XXXK Dermoah xxx actuoooongooenoouooflflfloox YXXX dwmamok x XK< oomuOO”..””~OO-”~“”OOx XXXX agowwzmmm xxx ooNuoooouuooOO-Ouuflxxxx XXKX Jo>m xxxx XXXX oou~~~..~~~~..~~xxxx XXX" 30 X X oakuuu..uoocnu.gno¢~-xx XKX'K xuaamagzwm xx xxx cowocuflooooOONOOnoou“”xx XKV’K' DLUQHWHOZW XX oom-..~..~oonuqoo”~“00~ ”“xx Table Fb3.-Complete-Linkage Cluster Solution, Frequency Variables - Women. RIVES RESOURCE FREQUENCY VARIABLES 280 OUUr-W—ooz 0"” on to 00 e. on oo oo no no on o. I. oo o. u p. a. tuxmstZQ 00“" too- 0000 ooo. no on no ooo... ooo. ooo. W4>UJW2LUK0> 01-Oct 0000 o. to no so ooo. no no ooo. "xxx 000.000... 0000 0000 one. oo oo ooo-eoxx_" 13L: cor-n u" oouu u u" u u Coxhgxx MIC) .aQ>|u xxxxx 00* lo 00 on 0.00 00 0. one. o. .0 00%); XX.\ '3wa 00“.. 00 to o. 00 00 no It 00 Co on 00 00 O. 00 0. be W230>ZW2- oow.. nu ooo. u no 000000 on no oo 0. or. x uoz h-homzum as: 00”.. can ooo...” 000000 oofixXfixx .‘XXXXX 00“.. .000 .0000... 00000009 AK KKKK aumANUb ~£¢L0m~. xxx'ax asa-“(xxx xx xxxx XXXKXK xxxxxx xxxxxx :- xxxxx xxxxxx x xx; xxx :- xAAxA x): A X‘- >~ Okmmm Momma U~JZUJ¢U>~ XXKXK Qfl‘aufiflquOMUQQuoouux XXXXXXX‘ . :Oxfi x xxxxxxxx omuunquOlOU“-QOOQOOK ,(XxXXXxX madczo xxxxgx OCco..-..-..-..oo......au~ uuxxx X Xxxx ”Kg-(XXKKX XKZ£LIJZP Gamuuuuuuuuuuuuua £0103! Ao>m OO-f-o-~~«~u~«..u-u.. OOZmHouzom xxxXXxxx conuuoounoouuuguuxs" °'<:'<‘<’m . A>mzozv om: flmcompom uzox mnmvo.5o ooeoo. w5moo. m55Hw mmmoo. sow o~nm-a>w mocos mo «cacao ask .o oom5o.so ooooo. o5ooo. m55so. .~o~_o.e =05 mosooco no oowwwowwwpquEowoe .o Hm5oe.mm oNeoo. oNNoo. o5o_o. .ooomo.o~ AmmwemumomawmeMMHwmwosmmwmmxmoz .e om5so.ess . ooooo. ooowo. memm5. ooomo.o 5susao oc mommwwwmwwmo~MMMMMowoe .o Naomm.oos ooooo. oesmo. oooo5. .ooeom.o~ osoooos =05 woodman mo ”wwwmwowop .N ooooo.5om womoo. oomoo. 5eoo5. .ooooo.5om -utaaxo =05 cooooooom5wwmow>wmcwoe .H & Hfimhw>0 mwcmr—U .Nm NZ m OHQMHASE hvucm OH nu mvwhmwfim 0Hmehm> Qwum ill... I: . 1 III 'I- 'I III.- I .coEOZulpowmwooam poopo .oumz seam po>flooom mooasomom mo sewum5~o>m x5 owmwuumz mo cowumsfim>m o>wuoomm< mo :0wuowooha ecu new mmmzymc< cowmmohmom oamwuasz mo thEEswuu.Huo manMH 285 .oo. v my .cowuwuooaxo Hmowuohoogu An wouoofiom xpuco mo povpo wcm mofinmwpm>m m>mzozv om: Hmcomnoa use» ooo~m.~e ooooo. omNoo. 5Noo5. osooo.s poo «Hoosuasm 5ocos we assess «an .o . . . . 5oooooo ago on Nomoo.so o5soo oomoo om5e5 oo5m5 :05 mmsomco on mecca Houuooma ooo .o . . Hzooemmoo memosos swoon.m5 ~5ooo oooso ooos5 ooomm 5Huaao mo mmocssoamoo “means one .o . . . . . 5o=aooo 00:0 Noooo 5os N555e ~555e 5s_mo .Noooo 5os -ssoaxo :o5 cououooom ecu use“ use .H a fifimuo>o owcmgu mm mm m ofimmufisz poucm ow u mvohoucm oHnwwhm> ovum .co2-nvo«mfiooom gonna .ouwz scam co>wooom moopzomom mo :owumsfim>m xn owmflppmz mo newumo~m>m o>wuoowm< mo :ofluofipoua may now mwmxfimc< :oflmmopwom ofimwuaaz mo thsszwnu.muu oHnmh 286 .mo. V Na. .:o«um~opuoopou:w x2 wouoofiom mofinmwhm>m owwov.w~ omv~o. :ox o5o5N. o_omo. .ooooo.o Ao=ommmv so» How vcmnpo cm on .m wnmmm.fim mwmmc. ommvm. Nmmmo. *vavm.mfi mm>oav :ox nu“: oxon o no swam~ m xoncm .N ooso~.oo o5oos. o5oos. ooooo. somsos.oo Am>ooo omux o so mo; o oo» o>oo .H a _Hmao>o owcmcu mm mm x o_mflu_:z ooucm cu m ovohoucm ofinmwum> moum l.lv -. ll . .[l‘ i. II .I lif'bl'll‘l' '.- II ‘II 9D.-v D -l'l‘ll' v' 0. - -Ollell'l‘llli II ‘1 .coeozunvofiw_ooow Ho: noose .oumz Eopm wo>wooom moUASOmom mo xocoscoam xn ommwuumz mo :omums_m>m o>«uoomm< mo :OMuowvoum may now mfimxamc< cofimmoamom omnfiumsz mo xumassmuu.muo omnmh 287 .oo. v m. .cowum_ossoosou:fl x2 vouoofiom mofiamwsm>m fiomus>mmmo womw~.- havoc. mmmmm. nmmwv. vooom.~ osnmusomEoo so» oxms so xwuoco so» o>mm o» wcfisuoEOm on .m ommww.v~ mnmoo. monmm. cannv. mmmom._ Ammup>mmwv so» you vcmsso cm on .v . . . . . . 5o=s04v mmflx a so an: a so» o>wu .N . . . . . mm>ooo 5oos5 om ommos osmos oomoo .5oos5 om oos cos: oson o so zooms somom .s & -mhm>0 mwcmr—U mm mm m owgmumsz How—um OH u— mflohoucm Qfinmwhm> mmum I III ‘ ‘SOnIlli. ‘llll;“.! .coz-nvommsoodm no: goose .oumz Eosm vo>mooom moos30mom mo xocozcosm >9 owmwhhmz mo cesumsfim>m o>fluoomm< mo :owuowcoha ocu how mwmz~m=< :ofimmohwom ofimmufizz mo xumsssmnu.v-u ofinmh 288 .oo. v.ms .=0wuouoomxo amowuouoocu x5 vofimwoomm xsuco mo sovpo pom moaamwsm>m ooo5o.~s o55oo. oooom. o5oso. oo5ss.~ ooo socomsoo sow smooEAMMMOWWso .o osooo.os oooso. ~N5om. 5s5oo. ooooo.m so ooo: so» mcsommmwmowomownso .o ooooo.os o5oNo. oooom. ooooe. .oseos.5 owossooo smumwwwmmmmnwwso .e ooooo.os oo5oo. oo5sm. soooo. 5o5os.s Aowws”mmwwommmowmoouwmmsmammoz .m oooos.5~ Noooo. oooom. oo5oo. .omsoo.5 ooo MmmwmwmoomonomwoMMmmmoo .N omooo.oo ooo5s. ooo5s. oommo. .oNooo.oo Am>ooo osos ms; zoom so ssoe .s m Hampo>o ow:m:u mm mm m ofimfiufisz poucm cu m wwohoucm ofinmwhm> Roam 9.". L .coEozn-vosm~ooam sopho .oumz Eosm vo>fioooz moossomom mo Aocoscopu xn owmfissw: mo sewumsfim>m o>fiuoomw< mo cowuowwopa ox» sow mwmxfimc< scammoswom ofiawufisz mo Asmassmnu.mno ofinmh 289 .oo. v.m. .COMkuooaxo fisowuouoozu »n vommmooom xsuco mo sovso can momnmflum>m Nooo5.o oesoo. ssomm. soo5o. ooooo. oos sosoosos soo »osoE»»MMOWWso .o o5soo.ss ooooo. s5omm. ooo5o. sos5s.s so ooo: so» wossmwmwmowomswmso .o oooos.ss ooooo. soosm. soooo. 5oooo.s owossooo smnmwwwusmMMWWso .o osmso.os osooo. oomsm. oosos. Nsseo. oo mmmw”wwwwwsmowswmwonsmnoz .o om55o.5~ os5so. Nomsm. mosoo. .ooooo.o ooo mMmmwwwowsmoMoMWUWmemos .N ooooo.oo oooos. oooos. oomoo. .ooooo.oo so» mosos ooo sosofimwowwoo .s a Hawso>o omcmcu mm mm m osawu_:z soucm o» m moosoucm o—nmwhm> noum ‘3'. » Q: III (In. .cozu-wosm»ooam sovso .ousz 50pm mo cofiuw3~m>m o>wuoomm< mo :ofiuowooha osu vo>fioooa mooh30moz mo xocoscosa x2 owmwsswz sow mfimxfimc< :OMmmosmom ofimfiufisz mo theasmuu.onu ofinmfi 290 .mo. v Na, .coHumuoomxo Hmofluosoocu »n wouoofiom mofinmwum>m . Amahoso os5oo s5 ooooo ooooo oooo5 Noo5o osos ozoso so ossos o: souso so: .o flomus>sooo msooEoe mo~o~.oo ooooo. oo5oo. Nsoo5. sosso.o »sseoo oo ooosssssso; sossse one .s oosos.oss mmooo. ooooo. oNoo5. soo5os.o oo os MwmwwwmmwoomomssMWswwono: .o Nsoom.oo_ ooooo. oosmo. omoo5. .ooooo.om ossooos so» sooooos oo mmnwmwowss .m ooooo.5om oomoo. oomoo. 5ooo5. .ooooo.5oN -ssooxo so» sossooooonwstw>wwswoe .s m fiamso>o owcmcu mm mm m mammu~sz sopcm OH x woosoucm omnmwpm> moum .coeozuuvofimfiuomw so: sowso .ousz Eoso vo>wooom moos20mom mo »o:o:cosa was cesum3~m>m »n owmflnsmz mo cosum=_m>m o>suoouw< mo cosoowuosa ocu sow mwm»~s:< cogmmoswom ofimfiufisz mo xumsasmuu.nuu ofinmh 291 .mo. V Ms .:0wsmuoomxo HmuHuoHoocu an wopoomom moanmwhm>w »o:us>s:oo mmoom.ov wvcoo. o~nvm. oommm. nofifim. sex how xhoz ow Cu mfinm Imam>m mfiomuo; moxme osm coumo so: .0 nomus>smoo msoosoe No5oo.os o5ooo. Noooo. sooo5. 5oomo. »sseoo so moosssssso: sossss use .o ~ossoso Noooo ow ooomo Noooo so5o5 .ooem5 ms osos ososo5mssos ooo sosso so: .o oosoo.oos No5oo. oooso. 5o5s5. .ooooo.os ossooos so» soosoos so Mwnwmwowss .N . . . . .. so>o:o ooso Noooo 5os N555o N555o 5ssoo .Nooom 5os -ssosxo so» sossoooso oso osos use .s s ssososo owsosu N: N: : ossssssz ososso so a oosossm ossossos soso .ooz--oososooso so: sowso .osmz Eosm wo>fioooz moosSOmom mo xocosvosu was :oflsm:~w>m »n owmwuumz mo cosum3~m>m o>fluoomw< mo sewuowvoha ozu sow owo»fimc< :OMmmosmom ogmwufisz mo xpmesswnu.wuu ofinmp APPENDIX H CORRELATION MATRICES 2512 .oo. v a. .oo.o u soo~._o a sososs: .s_.~ u soo~.oo a oo_ooss ooo. ooo.o_ oo.. e5o._ .ssssosso. coo. u N: oosmsfls< o~=.- soc. __~. 3.x. mom.- xsoz ucsos mo »o=o::osa wen. u N: o-nwu_:: ooo. soo. soo. e~o. o5o. moss—sos_o: so so_sos_o>o ov5. u s ossssssx . . . . . . . 32.2.. NB: .va _M— — NN— CN_ 3°£m\——Ou HO AUCOSQYULK . .- . . . . 0so. as. o.o .soo o ooo __~ so:o\__os so »scos:osa ooN oo5.o~n _osop .oN. ~=s. ooo.o 5so. ooN. sossooos . sooaoos so sossss_s>: sooo.o oo5. oo~ soo.5c_ _ssosous . o o. . . . ._ . :o_~oouua o5s .c. ._vo mm 535 e o ass use. me :o_sss_s>: .oo~.ov om5.m~ o ~55.o5_ comomosuoz so: 25. sss.v_ ss~. No... .soososooo oNo. u ~s oesosfls< . . . . . . sooooos ooo. u s assasss: 5.: we» v.3 m:— a_m 3o;m\__os mo xocoasosu man. u ~¢ o~mmu_:x . . . .. . . - .. o>o. ms: _oo so. oss so» so:o\__os oo »o=ossoss 5oo.- 5os. soo.m ooo. oo..- ooos_sss_o: so sossss_s>: 5ss. “so. soo.o ooo. s5_. :sos moses oo »ssossoss 5o~ 5~o.oon sassy oo-. sss. .o_.- ooo. -_.. osssooos .. .. .ss suoooos oo so_sss.s>: sooo.o 59o. _oN o5o.oo. sssosous N5o. s 55..55 ooo. oxo. :o_sooooo . . . ooo o>o. so sosssssoss ._5_.oo ~o~.oo o .o5.o- cosmoosuos cola: .srs_u_.r.osxo _.s~;: . .co.o.s_ssu cosmossus: 1os=s_ss:. osmsvm mohwscm cosuausa> :csmxosssz ».__.2:2:s; ;; as sess: aco_o_ooo:u o_na_s=> on cue: .m.o we law &O.ouwaow 7o~.7s=1:=.m :o_mzosuoz asstcssx . 7.5 _ 1.51:: 7:: I . I u . . . :- xlltl I'l.-.’ IIIT’IIIIQ I‘{!¥y 01ll’."llllal ::_rmosuoz o_:.~.:z oucsmsm> mo mam»_ac< . All ‘0 lqflutnluauvl'j .sosusuoam yo: souuo .mo_nu_so> :3..::_:>; 1:: xocszvos; coc.cscu »£ ou:.ss:2 do :c_~=:_=>: o>_uuouo< mo :o_uo_1osn ecu sou m_m»_a:< :ommmouuom o—nwu—zzuu.muu o—nah Table H-1.- Correlation Matrix, Evaluation Variables. 293 NVOFNQcoNNNO M)‘0mfi~~&roc rOMNONQrNNON OnMNcfiNNv-OEC nmnnmPpnummq 00.00.000.00 c OHNvONFOGNNQO 0N0000dfioo¢vw «dug: 613‘N ”IV-4's.- NmrNSOOONOOOQ omronfi‘vv-MMQ 0000000000... oonmnchONcmon moxnhcesq¢3w~o omaocokrcoosco OOLnONw0annc~w cm~M~r4~pm~mng 00.000.00.000. NP‘IfiNr-Cc'schomm o-mo-OM~IONONM¢.NO '-oc.nn.~.n~:~nc.a.r