H 7‘ a; I. U . . . . I I . . c I . . \ I ‘ v ' a o . o . a v ‘. a - \ \ o u ..I . ., -~ I - any I ‘ o v ‘V .x . V .‘ I .--0-°' -~-‘\.\.“‘ ~ ———— .._. FAMILY DEBT PATTERNS AS RELATED TO THE FAMILY UFE CYCLE AND SOCIOECONOMIC STATUS Thesis for the'Degree of M. A. MICHIGAN STATE UNIVERSITY ~ - ROSE TAYLOR SALSB’URG . . _ ‘ o . . ‘ . . . _.'. . _ . . _ l O .. I " o ‘ 2 T ‘ ' ~ . 77 ' I . _ . I . " . ~ -. a . . . . v ' ‘ . , - . O . . . .. o ' , 0 a _ ,. 1- . ‘. ~ . ‘ ' ' . -.‘. I t ’ ' ‘ - ' " - 1. _ . ' ' ' 7 . - - —. I .- _, . . . , ' . . . . . . . , , . ~ to n . 4 .. a I .-v ' ‘ - .. ' ' la . . ,. , ' - u. ' -. ‘a . . .. . . . u . . . . ,_ . . n u. -. - . I . . - - r ~v .. . '- _ ‘ ' ' ' I - .,,'vv... - . -r-- - .n ~ u . . . . .. ... .. a . .. . . '1 . I . .. , . I . 0' 4. a . .a- . . .-. .. - . .T o ‘ v , 1 - . . . s . . - ~ 0 . . 0 .~ I - o ‘ u .‘ . .', . . v . — .r . ar’p a- ,, ..,. Y « .y. ('F .fl. III I IIII II I III’III 29 IIII IIII 0751 76 [8092 g: This is to certify that the thesis entitled FAMILY DEBT PATTERNS AS RELATED TO THE FAMILY LIFE CYCLE AND SOCIOECONOMIC STATUS presented by ROSE T. SALSBURG has been accepted towards fulfillment of the requirements for M.A. degree in Family Economics and Management W Major professor Date 2/75.; 0-7 639 Go a?» ABSTRACT FAMILY DEBT PATTERNS AS RELATED TO THE FAMILY LIFE CYCLE AND SOCIOECONOMIC STATUS BY Rose Taylor Salsburg The purpose of this study was to look for family debt patterns as related to stages of the family life cycle and socioeconomic status levels by analyzing the variations in the ratios of debt to income utilizing the components of installment and noninstallment debt. The debt components were classified by incurrence (purpose for which incurred), type and their respective totals. The dependent debt variables thus formed were: installment, noninstallment and total incurrence variables for each of five purposes-~additions and repairs, car, durables, other debt, and medical and dental; totals for installment, noninstallment, and the (grand) total debt. Family was defined as all persons living in the same dwelling who are related. A single person unrelated to the other occupants or living alone is a separate family. Installment debt was defined as private, non-mortgage debt subject to two (2) or more regular payments. Noninstall— ment debt was defined as generally private, non-mortgage debt not subject to more than one payment. The debt measure Rose Taylor Salsburg (degree of debt) was the ratio of outstanding dollar amount of debt to the family's disposable income. The source of data for this study was one year (1968: Wave II) of the 1967-1970 panel study on consumer durables and installment debt conducted by the Survey Research Center (SRC), University of Michigan. The SRC life cycle classification was modified for this study and operationalized into five (5) stages of the family life cycle. The Duncan Socioeconomic Status Decile scale was modified and operationalized into low, middle and high levels of socioeconomic status. The total sample population numbered 1252 families (primary family units). The noninstallment incurrence variables were eliminated from the major statistical analysis test because of the very small number of families who bad debt in this category. The statistical method used was a two-way analysis of variance (ANOVA). The level of significance was at the .05 probability level. Where there were significant main effects as a result of the two-way ANOVA, a post hoc analysis was made to determine where the significant differences were occurring. The method used was an ANOVA for each possible contrast of levels for the significant variable, socio- economic status (SES), holding the other independent variable, family life cycle (FLC), constant. The results of the two-way analysis of variance data analysis showed a significant effect of SES levels on the following debt variables: durables installment debt, total Rose Taylor Salsburg car debt, total installment debt, total noninstallment debt, and the (grand) total debt. The post hoc analysis showed that none of the significant differences were between low and middle levels of SES; but significant differences occurred between low and high SES levels, between middle and high SES levels, or both--depending on the individual debt variable. The basic direction of differences was linear with the largest debt ratio at the low SES level and the smallest ratio at the high SES level. Exceptions were: total installment debt where middle SES level had the largest ratio; and total car debt where middle and high SES levels were almost the same with the smallest ratio. There was only one significant interaction effect of family life cycle and socioeconomic status. This occurred for total other debt. There were no significant (main) effects from FLC stages. The results of this study indicate that credit/debt patterns are influenced by SES level. In addition, although FLC stages did not show statistically significant patterns of degree of debt for most debt categories, a pattern was reflected in the frequency distribution. The interactive effect of stages of the family life cycle and socioeconomic status levels with respect to total other debt may indicate a changing pattern of installment debt from traditional to newer types of credit instruments. FAMILY DEBT PATTERNS AS RELATED TO THE FAMILY LIFE CYCLE AND SOCIOECONOMIC STATUS BY Rose Taylor Salsburg A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Family Ecology 1976 DEDICATION To the memory of my mother To the patience and understanding of my daughter and son To the love and inspiration of my husband ii ACKNOWLEDGMENTS I would like to express deep appreciation for the encouragement, understanding and inspiration received from the late Dr. Carol Shaffer until her untimely death. I owe a particular debt of gratitude to Dr. Dennis Keefe for stepping in so ably to provide ongoing guidance and encour— agement in the completion of this thesis and Dr. Linda Nelson who helped me bridge the gap over many obstacles. I would also like to thank the other members of my committee, Dr. Jean Schlater for her valuable suggestions and Dean James Rainey for his patience and assistance. To the Survey Research Center, Institute for Social Research, University of Michigan for permission to use their data and for providing clarification and assistance, I extend my sincere thanks. To Dr. Mary Andrews and the Office of Research Consul- tation and Data Support my thanks for valuable assistance in data processing and statistical analysis. Use of the Michigan State University computing facili- ties was made possible through support, in part, from the National Science Foundation. For this, I wish to express appreciation. iii A special thanks to Carol Adam, Gary DeLorme and Dave Piekarski who were available with their expert assistance at critical moments. I am especially grateful and indebted to my husband for his encouragement, understanding and inspiration which supported me throughout this time-consuming, difficult period. And to Barbara and Frank, whose confidence that I could "make it" gave me the needed assurance. iv LIST OF LIST OF CHAPTER I. II. III. IV. V. TABLE OF CONTENTS TABLES . . . . . . . . . . . FIGURES . . . . . . . . . . INTRODUCTION . . . . . . . . Theoretical Framework . . . Definitions . . . . . . . . Objectives . . . . . . . . Hypotheses . . . . . . . . REVIEW OF LITERATURE . . . . METHODOLOGY . . . . . . . . . Data Source . . . . . . . Procedures for Sampling and Collection . . . . . . Data Study Design and Operational Definitions Independent Variables . . Dependent Variables . . . Procedures for Analyzing Data FINDINGS AND DISCUSSION . . . Frequency Distribution . . Debt in the Sample Population Tests of Hypotheses . . . Debt Incurrence Categories Type of Debt Category . . Comparison of Means for Significant Main Effects . . . . . . SUMMARY, CONCLUSIONS AND IMPLICATIONS . . . summary 0 O O O O O O O O 0 Conclusions . . . . . . . . Implications from This Study Suggestions for Further Research V Page vii xi ll 12 15 27 27 28 30 31 34 37 40 4O 45 47 48 53 58 65 65 67 7O 71 APPENDICES A. Frequency Distribution of Individual B. Mean Ratios of Individual Variables Not REFERENCES Debt Variables Significant vi 75 75 84 88 Table LIST OF TABLES Frequency Distribution: Total Sample Population by Family Life Cycle Stage and Socioeconomic Status Levels (N=1252) . . . . . . . . . . Installment Debt: Percent of Families With No Debt (Tables A.l - A.6) . . . . . . . . . . Noninstallment Debt: Percent of Families With No Debt (Tables A.7 - A.12) . . . . . . . Total Incurrence Debt: Percent of Families With No Debt (Tables A.13 - A.18) . . . . . . Type of Debt Group: Percent of Families With No Debt (Tables A.6, A.12, A.18) . . . . . . . All Debt Variables by Percent of Total Popu- lation and Grand Mean Degree of Debt for the With Debt Population . . . . . . . . . . . . Two-Way Analysis of Variance of Family Life Cycle and Socioeconomic Status for Installment Debt Incurrences . . . . . . . . . . . . . . . Two-Way Analysis of Variance of Family Life Cycle and Socioeconomic Status for Total Incurrence Debt . . . . . . . . . . . . . . . Two-Way Analysis of Variance of Family Life Cycle and Socioeconomic Status for Types of Debt 0 O I O O O O O O O O O I O O O O O O O 0 Post Hoc Analysis of Variance Contrast of All Socioeconomic Levels for Significant Debt variables 0 O O O O O I O O O O O O O I O I O Durables Installment Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies) . . . . . . . . . . . . . . . . vii Page 34 42 43 44 45 46 49 51 54 57 62 Table Page 4.11 Total Car Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies) . . . . . 62 4.12 Total Other Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies) . . . . . 63 4.13 Total Installment Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies) . 63 4.14 Total Noninstallment Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies) . 64 4.15 Total Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies) . . . . . . . . . 64 A.l Additions and Repairs Installment Debt Frequency Distribution: Families With No Debt . . . . . . . . . . . . . . . . . . . . . 75 A.2 Car Installment Debt Frequency Distribution: Families With No Debt . . . . . . . . . . . . . . . . . . . . . 75 A.3 Durables Installment Debt Frequency Distribution: Families With No Debt . . . . . . . . . . . . . . . . . . . . . 76 A.4 Other Installment Debt Frequency Distribution: Families With No Debt . . . . . . . . . . . . . . . . . . . . . 76 A.5 Medical and Dental Installment Debt Frequency Distribution: Families With No Debt 0 O O O O O O O O I O O O O O O O O O O O 77 A.6 Total Installment Debt Frequency Distribution: Families With No Debt 0 O O O O O O O O O O O O O O O O O O O O 77 A.7 Additions and Repairs Noninstallment Debt Frequency Distribution: Families With No Debt . . . . . . . . . . . . . . . . . . . . . 78 A.8 Car Noninstallment Debt Frequency Distribution: Families With No Debt 0 O O O O O O O O O O O O O O O O O O O O 78 A.9 Durables Noninstallment Debt Frequency Distribution: Families With No Debt 0 O O O 0 O O O O O O O O O 0 O O O O O O 79 viii Table A.10 Other Noninstallment Debt Frequency Distribution: Families With Debt . . . . . . . . . . . . . . . . Medical and Dental Noninstallment Debt Frequency Distribution: Families With D8bt........‘......... Total Noninstallment Debt Frequency Distribution: Families With Debt . . . . . . . . . . . . . . . . Total Additions and Repairs Debt Frequency Distribution: Families With Debt . . . . . . . . . . . . . . . . Total Car Debt Frequency Distribution: Families With Debt . . . . . . . . . . . . . . . . . Total Durables Debt Frequency Distribution: Families Wtih Debt . . . . . . . . . . . . . . . . . Total Other Debt Frequency Distribution: Families With Debt . . . . . . . . . . . . . . . . . Total Medical and Dental Debt Frequency Distribution: Families With Debt . . . . . . . . . . . . . . . . . Total Debt Frequency Distribution: Families With Debt 0 O O O O O O O O I O O O O O O 0 Additions and Repairs Installment Debt: Page No . . 79 No . . . 80 No . . . . 80 No No . . . . 83 Mean Ratios of Debt to FLC and SES (Cell Means and Frequencies) . . . . . . . . . . . . . . . 84 Car Installment Debt: Mean Ratios of Debt to FLC and SES (Cell Means and Frequencies) . . . 84 Other Installment Debt: Mean Ratios of Debt to FLC and SES (Cell Means and Frequencies) . . . 85 Medical and Dental Installment Debt: Mean Ratios of Debt to FLC and SES (Cell Means and FrequenCieS) O O O O O O O O O O O O 0 ix . . . . 85 Table Page 8.5 Total Additions and Repairs Debt: Mean Ratios of Debt to FLC and SES (Cell Means and Frequencies) . . . . . . . . . . . . . . . . . . 86 B.6 Total Durables Debt: Mean Ratios of Debt to FLC and SES (Cell Means and Frequencies) . . . . 86 B.7 Total Medical and Dental Debt: Mean Ratios of Debt to FLC and SES (Cell Means and Frequencies) . . . . . . . . . . . . . . . . . . 87 LIST OF FIGURES Figure Page 4.1 Total Other Debt: Mean Degree of Debt (Percent) Interaction Between FLC and SES . . . 52 xi CHAPTER I INTRODUCTION The family has become increasingly accepted as a meaningful and basic unit for the study of economic behavior. Individual disciplines in the social sciences and multi- disciplinary groups more frequently utilize the family as a logical unit of research and study. An indication of this trend is a recent conference on "Social Structure, Family Life Styles, and Economic Behavior," sponsored by the Institute of Life Insurance at Williamsburg, Virginia, in January, 1972. The participants represented various disciplines, including economics, home economics, sociology, and education. The discussions and presentations, as reported in a volume entitled Family Economic Behavior (Sheldon, 1973), seem to emphasize a needed differentiation from the concept of consumer behavior. Ferber, in his presen- tation, reinforces the interrelationships of the various dis- ciplines by pointing out that individuals act, not only on economic factors, but also on those which are the focus of the other social sciences including sociology and social psycholoqy. He expresses the need for: . . . bringing together these various dimensions of consumer behavior within the framework of the family to provide more realistic explanations of economic behavior (Ferber, 1973:29). There is a need for more empirical research on family economic behavior to assist families not only in the alloca- tion and use of their economic resources but to gain insight into possible economic problems they may face. Bonde, for example, believes that: We know relatively little about many aspects of the problems of consumer credit or the larger field of investigation, family economics. [We need to look for]answers to innumerable queries regarding consumer credit and other economic pro- blems. The . . . available data, both qualitative and quantitative, both macro and micro, relevant to family economics . . . need to be analyzed, relation- ships ferreted out, and findings published for appraisal (1967:149). To reinforce the above-expressed needs in the area of consumer credit, Schlater states that: Today, special consideration must be given to credit as a resource. With credit so universal a medium for purchase of goods and services, the need emerges for better understanding of the credit function, of ways in which credit can help individuals and families achieve their goals, and of the factors involved in over-extension of credit use (1970:48). While there has been a substantial growth in dollar volume and use of credit in recent years, consumer credit in this country is not a relatively recent phenomenon. Morse relates that: We have evidence of the use and misuse of credit in the codes or laws reflecting the judgment of society . . . Consumer credit was part of the Colonial family's way of life. Benjamin Franklin's sage ad- vice against borrowing was based not only on his sad experience with uncollectible credit, but on his observations of the use of credit in Colonial America (1967:20). Although the institution of credit/debt is an old one and the dynamics of debt utilization are constantly changing, the changes have been more radical in the most recent decade. Assessment of the changes and their impact should enable us to understand more fully family economic behavior in general as well as specific areas of family financial management such as consumer credit/debt and the threat of over-extension. There are many facets to debt dynamics, not only quantita- tive--amounts, ratios, categories--but also societal and in- stitutiona1--social acceptance of debt, credit instrumenta— tion, and legal consequences. Available historical data on the use of consumer credit show its tremendous growth. From 1950 to 1970, total con- sumer credit measured in current dollars rose from $21.5 billion to $127.2 billion--almost a six-fold increase. It continued to climb in the early 19703 reaching $190.1 billion by the end of 1974 (Federal Reserve Bulletin: May, 1975 and October, 1972). Dividing the totals into installment and non- installment credit shows the proportions of the total and the differences in the growth pattern. Using the same years for comparison as above, 1950 to 1970, installment credit rose twice as fast, from $14.7 billion to $102.1 billion--a seven— fold increase; while noninstallment debt rose from $6.8 billion to $25.1 billion--a three and one-half fold increase. During the same period, the BLS Consumer Price Index rose 61.3 percent. The Federal Reserve Board subdivides the consumer installment credit into 1) automobile paper, 2) other con— sumer goods paper, 3) repair and modernization loans, and 4) personal loans. In 1950, the largest component (41.5 percent) was automobile paper; other consumer goods paper was second largest (32.7 percent). In 1970, automobile paper was still the largest, (though not as great in proportion) with 34.5 percent; other consumer goods paper at 30.9 per- cent was almost matched for second place by personal loans (29.7 percent). Personal loans increased sharply as a per- cent of the total installment credit during the 20 year period. Statistics on aggregate trends between 1950 and 1970 indicate not only a rising level of total outstanding install- ment credit but also a steady increase in the ratio of out- standing installment credit to disposable income (Hendricks, 1973:6; original source: Economic Report of the President, January 1972). The data are consistent with the findings of the Survey Research Center (SRC) Panel Study which found that "outstanding installment debt balances have tended to increase at a much faster rate than income" (Hendricks, l973:7). In addition, "the upward shift in the debt—income function in the 1ate'603 seems especially significant since it occurred despite rising costs of borrowing which persisted throughout the period" (page 29). There has been a growing trend toward social and cul- tural acceptance of the use of consumer credit. Contrasting an earlier period with the present, Bonde found: The attitude toward the use of credit has changed from one of disdain to one of general acceptance. At the turn of the century those who borrowed for personal consumption were considered improvident. This is not so today (Bonde, 1967:148). Related to social acceptance of credit use is the attitude toward its use. The Survey Research Center has been collecting data on attitudes toward installment debt, its impact and extent of change over a period of time. From its documentation,* it concludes that: There can be little doubt that the attitudes of the American peOple toward buying on credit have become more favorable over the past decade (Hendricks, 1973:138). The SRC believes that favorable changes in attitudes will continue to be an important force in the growth rate of consumer credit in the 19703 due to several factors including . . the current trend on the part of lending in- stitutions, especially credit card agencies, to grant credit that is not associated with the acquisition of a specific asset (page 156). Expansion in the area of credit instrumentation--that is, the devices by which debt is made available--has made it easier and faster to obtain credit. Building on the availa- bility of credit through the use of captive credit cards and their revolving credit feature has been the introduction and growth of bank credit cards which have added other "conve- nience" features such as overdraft checking privileges and *For more in-depth coverage of this aspect see Chapter 7 on "Attitudes Toward Installment Debt," in Hendricks et al., Consumer Durables and Installment Debt, 1973:119-144. cash advances on such cards, various devices for instant— aneous or pre-approved cash advances, and "line of credit" form of loans in connection with checking accounts. The expanded use and, for some, overuse of credit has increased the danger of overéindebtedness. The effects of excessive debt can range from financial overcommitment re- sulting in the need for professional counseling (Huber, 1965) to complete family disruption, from garnishment to actual personal bankruptcy (Hermann, 1966). In the chapter on "Difficulties With the Repayment of Installment Debt" (Chapter 9) the 1968 Survey of Consumer Finances states: It is well known that not only the number but also the proportion of families entering non- business bankruptcy or Title 13* has grown faster than the pOpulation. From 1954 to 1967 the number of personal bankruptcies grew from slightly over 40,000 to over 160,000 a year, a fourfold increase, while population increased only 22 percent from 163 million to 199 million (Katona et al., 1969:153). Although bankruptcy may not be an absolute remedy and is not without risk, the trend is significant. These various changes in the dynamics of debt utiliza- tion make family financial planning more essential, the need for information about credit and credit management more critical, and any resultant problems more complex. The family uses of credit are many--for large, discre- tionary expenditures, for unusual or unexpected expenses, to pay old debts or consolidate loans, and to enable a family *Title 13 is different from bankruptcy in that the debtor maintains an obligation to repay his debts. This is usually called the Wage-Earners' Plan. to utilize consumer goods such as cars and household durables while paying for them by adding the cost of credit to the price. However, a family using credit, whether planned, impulsively, or in an emergency, needs to understand the relationship of credit to income over time. "Although credit makes possible more purchases in the present, it does not increase income but rather borrows from future income" (Gross and Crandall, 1963:477). The SRC Panel Study points out that "studies of install- ment debt are less common than studies of expenditures on major consumer durables" (Hendricks, l973:vi). The limited information available on installment debt is, with a few exceptions, generally in aggregate data form (see above dis- cussion of available statistics), or is cited as a contribu- ting factor to a problem but not measured separately by categories of use or type. Noninstallment debt information, the little that is available, is classified by source, that is, single-payment loans, charge accounts, service credit from the perspective of the family unit. Data on noninstall- ment debt have been rarely, if at all, analyzed. What are the family patterns of use? Where are the debt burdens likely to be? An empirical investigation seems worthwhile undertaking and would be of value. There is insufficient data analysis regarding the extent to which debt utilization is related to stage of the family life cycle and socioeconomic status. This lack leaves a gap in the understanding of family debt behavior or credit utilization, limiting the generalizations that can be drawn and thus limiting the ability of the profession to make pre- dictions about this area of behavior. Theoretical Framework The concept of stages in the family life cycle has been used as an important tool for analysis in social research. It offers a useful framework for providing empirically-based information. Its use for this study is based on two related theories. The first postulates that variations in income and expenditures follow basically characteristic patterns as a family progresses through the stages of the life cycle and at each stage there are a number of characteristic financial problems (Bigelow, 1931). The second postulates that at each stage of the life cycle there are developmental tasks or growth responsibilities, "the successful achievement of which leads to satisfaction and success with later tasks" (Duvall, 1967:49). One of the basic tasks of families is the alloca- tion of resources, including allocating money resources for various needs and costs. Lack of satisfactory handling of this task at any stage can cause disorganization at future stages. This is particularly important in the early family stages (Duvall, 1967). The concept of socioeconomic status is a "composite of social and economic attributes that tend to cluster together" (Kahl andlxnds, 1955:321); it has been used by social scientists as a "significant variable in measuring and compar- ing behavioral traits" (Lawson and Boek, 1960:149). It is generally believed that living patterns, attitudes and goals, as reflected in an individual's or family's use of resources, for example, money income, are influenced by his status group (Gross and Crandall, 1963:192). Research in family economics in general, and the manage- ment of its credit resource in particular has not fully utilized this concept. For example, Brown states: "Motiva- tions for family economic behavior are complex and varied, and we can be sure that economic motives are not the only critical ones. A family's position in the social structure creates certain orientations which are undoubtedly reflected in spending patterns" (1969:127). Definitions The term credit when used in this thesis means consumer credit--that is, private, non—mortgage, short- and inter- mediate-term credit of the family unit or household. Indivi- dual and family credit--that is, private, nonpublic, non- business credit--has been traditionally labeled "consumer cre- dit" to differentiate it from "business credit." In addition, the terminology of consumer credit is interchangeable with consumer debt—-the former having a more positive connotation, the latter, a negative one. Hereinafter, the terms consumer credit (debt) and credit(debt)vdll be used interchangeably. 10 Debt patterns for this study refer to incurrence and type. a. Debt incurrence is debt classified by purpose for which incurred, that is, additions and repairs, car, durables, "other" debt, and medical and dental. b. Type of debt is debt classified as installment or noninstallment. The SRC Panel Study defines installment debt as private, non-mortgage debt subject to two or more regular payments. Noninstallment debt is generally private, non-mortgage debt not subject to more than one payment. (See Chapter III, "Methodology," for more specific and complete operational definitions.) Degree of debt is the term referring to the dependent measure, debt to income ratio. The family unit means all persons living in the same dwelling unit who are related by blood, marriage, or adoption. A single person unrelated to the other occupants in the dwel- ling unit or living alone is a separate family unit. Stages of the family life cycle is a construct repre- senting succeeding patterns of family composition and age in the life of a family, by which families are placed in num- bered stages. ll Objectives The purpose of this study is to analyze the variations in the ratios of debt to income for the components of installment and noninstallment debt in relation to stages of the family life cycle and socioeconomic status, for the year 1968, to determine if debt patterns exist, and to inter- pret the findings in respect to implications for family financial management. (The year 1968 is taken from the 1967— 1970 SRC Panel Study, see Chapter III below.) The determination of whether discernible debt patterns exist would be an important contribution to family economic research. An understanding of the existence and nature of family debt patterns, as they may relate to family life cycle and socioeconomic status, empirically supported, can be used by counselors or educators to help families in their financial management, in a preventative planning approach to anticipate needs, and hOpefully to ease or prevent possible overindebted- ness from occurring. The specific objectives of this study are to: l) extend the Survey Research Center's analysis of total installment debt by disaggregating it into categories of purpose for which the debt is incurred, 2) seek new information on noninstallment debt use and degree of debt, 3) add empirical information on debt patterns to the growing body of behavioral information based on the family life cycle concept, 4) 12 explore the usefulness of the socioeconomic status concept in the family economic behavior area specifically through credit/debt utilization and management, and to 5) identify possible debt burden areas. Hypotheses The specific hypotheses to be tested for the effect of stages of the family life cycle and socioeconomic status levels on family debt patterns are as follows: I. The degree of debt (ratio of debt to income) will vary for each of the debt incurrence categories both for the installment incurrence debt group and the total incurrence group. A. Stages of the Family Life Cycle 1. Additions and Repairs From Stage I to IV, the ratio will increase, peaking at IV, then declining with V. Car Starting at a relatively high level, the ratio will increase to its peak at Stage II and slowly decline with V being lower than I. Durables The same pattern will emerge as with Car. Other Starting low at Stage I, the ratio will peak quickly at II, then decline to its lowest level at V. l3 5. Medical & Dental Starting low at Stage I, the ratio will peak quickly at II, decrease slowly through III and IV, then increase slightly at V (bimodal). B. Levels of Socioeconomic Status 1. Additions & Repairs A linear pattern will emerge, with the largest ratio at the Low level and the smallest at the High level. 2. Car An inverted uneven V pattern will emerge, with an intermediate ratio at the Low level, the largest ratio at the Middle and the smallest ratio at High. 3. Durables The same pattern will be found as in Additions & Repairs. 4. Other A linear pattern will emerge, with the smallest ratio at the Low level and the largest ratio at High. 5. Medical & Dental An inverted uneven V pattern will emerge, with the smallest ratio at the Low level, the largest at the Middle and intermediate ratio at High. II. 14 Interaction There will be no interactive effect between family life cycle and socioeconomic status. The degree of debt will follow the same patterns for the three types of total debt--installment, noninstallment and (grand) total--with respect to family life cycle and socioeconomic status. A. Stages of the Family Life Cycle Starting at a moderate ratio, the ratio will increase to its peak at Stage II and slowly decline to its smallest ratio at V. Socioeconomic Status Levels A linear pattern will emerge, with the largest ratio at the Low level and the smallest ratio at the High level. Interaction There will be no interactive effect between family life cycle and socioeconomic status. CHAPTER II REVIEW OF LITERATURE The first empirical study of consumer finances using stages in the life cycle as the independent variable rather than age classifications was by Lansing and Morgan (1955). Their data analysis of income, assets and debts, and selected expenditures included the findings that the wife's income was the factor which accounted for the bimodality of family in- come, young marrieds were buying relatively large amounts of durable goods when income was fairly low, and the proportion of families with debt reached a peak for young families with children and did not drOp substantially until the children left home. In addition to showing that the family life cycle is related to important economic behavior variables, they mention the possibility that many of these patterns are culturally and socially determined, and suggest further re- search in this area. Lansing and Kish (1957) found the stages of the family life cycle to be of greater explanatory value than age classes (as measured by the variance using rho) in six economic characteristics including indebtedness, income level, purchase of several major budget items, and employment of wife. 15 l6 Shaffer's (1964) findings in analyzing Survey of Consum- er Finance data for income and expenditure patterns related to the life cycle point up possible trouble spots: the need for awareness of the implications for financial planning when the wife leaves the labor force after working; heavy expenditures for durable goods in the young, married, child- less stage; indications that many families have risky finan- cial positions because of poor money allocation, extensive use of mortgage and consumer credit, inadequate preparation for possible current income reduction and life insurance proqrams inconsistent with needs, especially with younger children. Analysis of data in the Kahl and Davis study (1955) showed the best single index of socioeconomic status to be an occupational scale. In addition, they said "observation suggests that the core of status is a culturally defined, group-shared style of life" whose resultant values and mode of living influence expenditure patterns (p. 322). One study by Mathews and Slocum (1969) found that mem- bership in a social class influences patterns of credit card usage: "The lower classes tend to use their credit cards for installment financing to a greater extent than upper classes" (p. 72). (The other classification for card use was conve- nience.) In a later study (1970), in which they compare social class and income as indicators of consumer credit be- havior, they believed their findings showed that income is also a useful variable to use for understanding credit l7 behavior. In addition, the data indicate "that social class does not significantly differentiate credit behavior within all income categories; however, social class does appear to be a valid segmentation variable in the upper income categories" (p. 71). In a study of 100 financially overextended families in the Detroit metropolitan area, the sample (not random) had been solicited from professional counseling sources. Huber (1965) statistically compared the problem group with the general population of metrOpolitan Detroit (based on the 1960 census) in many areas--including education, income, occupa- tions, number of working wives, age of head of household and number of years married--for possible clues to the problem family's indebtedness. He found much similarity between the two groups; one exception was that the income of black families in the problem group were a little below white families in the study and considerably better than in the local community as a whole. He believed their income level identified them more with the "installment credit problems of 'our' middle-class population as a whole than those peculiar to the low-income Negroes like Caplovitz's The Poor Pay More" (p. 18). Further analysis of the overextended families of the study showed that in the employment area, 50 percent of the families reported declining incomes, a third of which were considered drastic drops, a third reported rising incomes and the balance fluctuating ones. This would seem to indicate 18 that many families face changing income and for heavily debt- committed families the budgetary readjustments are hard to make. Age-wise, over a third of the study's heads of family were 34 year olds or less compared to a fourth in the metro- politan area. Huber states, "It is to be expected, however, that a study of overextended families would tend to consist of younger families in View of the peculiar high-cost demands associated with starting a home and a family" (p. 20). Another finding pointed up the crucial early years of marriage; the couples of the study designated the early years of marriage as the time when their financial problems began, 29 percent indicating money problems from the begin- ning of marriage. Other reasons given for debt problems were current overbuying, overuse of "easy credit" especially for high cost durable items due not only to impulsiveness and impressionability but also to lack of discriminating consumer information, no planning ahead for future needs, and neither savings nor insurance protection programs for emergencies. Herrmann (1966) in a survey of studies of families in bankruptcy (Brosky, 1965; Dolphin, 1965; Herrmann, 1965; Myers, 1961) found that certain characteristics may have pre- disposed bankrupts to financial difficulties. Among these were: (1) youth and lack of financial experience, (2) the heavy expenses for families in the first stages of the family life cycle, (3) income declines which interfered with debt- carrying capacity, (4) unanticipated expenses such as major illness, accident, and other unexpected misfortunes-~debts 19 from medical services constituting one of the major catego- ries of debt in all four studies, and (5) total debt loads which had grown to unmanageable size--this accumulation (usually a gradual process) appearing to be the principal cause of the financial distress. The majority of the bankrupts in these studies were found to be married men in their early 303 or younger, work- ing in blue—collar occupations, whose incomes were typically lower than those of other families in their community. The Survey Research Center completed a four-year panel study (1967-1970) entitled "Consumer Durables and Installment Debt: A Study of American Households" (1973). This survey parallels the annual "Survey of Consumer Finances" (SRC) for those years in that a national cross-section of primary family units were interviewed and the same questionnaires were used for data-gathering. One major difference, however, was that for the panel study, the interviews were repeated each year with as many of the original families from the first interview as could be located or responded. As indicated by the study's title, installment debt is one of the two major dependent variables on which the analysis of this volume focuses. The thrust of its direction is in looking for aggregate trends. The SRC used an estimate of average outstanding installment balances in its analysis. This estimate was obtained by summing the total outstanding installment debt for each of the four interview years and dividing by four. A four year average annual income (grouped 20 into income classes) was frequently used as a major variable. Its findings on the pattern of average levels of out- standing debt across income groups were: For families with an income less than $10,000, installment debt balances grow with income. Throughout this range installment debt is a constant prOportion (about 12 percent) of income, except at the very bottom of the income distribution. At incomes above $10,000 but below $20,000 a year, out- standing installment debt is almost a constant amount, and hence declines rapidly as a proportion of income. At very high levels of income, $20,000 or more, installment debt balances begin to decline, not only relative to income but also in absolute amount (p. 153). A look is taken at demographic correlates of income including stages of the family life cycle. It was shown that the use of installment debt is strongly influenced by family life cycle and its components. But overall they con- clude that controlling for these does not change the basic relationship of installment debt to income. Attitudes toward installment debt were explored. For this a debt attitude index was devised to reflect attitudes toward using debt. Not only do attitudes strongly affect installment debt use, but the variable change in attitudes among panel families had a statistically significant impact on debt use. They also assessed the impact of family charac- teristics including family life cycle and found that "atti- tudes of families at different stages of the family life cycle differ considerably" (p. 132). "Even after taking account of other factors which in- fluence installment debt use, the impact of attitudes on out- standing balances is striking" (p. 134). 21 Ryan's study (1968) sought "to ascertain and measure factors associated with excessive installment debt burdens and to identify economic and demoqraphic characteristics of the excessively indebted." Debtors were classified according to debt burden using the ratio of installment debt payments to income, income level and liquid asset holdings. Those whose liquid assets exceeded debt by at least $200 were not considered in trouble. The remainder were considered exces- sively indebted and classified as in some trouble (ST) or as in deep trouble (DT), the latter being a subgroup of the former. Those in the DT subgroup were debtors with 40 per- cent debt payments to income ratio, and those with 20-39 per- cent debt ratio but with diSposable income under $4000. The ST group consisted of debtors with 20-39 percent debt ratios with disposable income of $4000 or more, and those with 10—19 percent ratios with income less than $6000. Thus "40 percent of the debtors were classified in some trouble with respect to installment debt" (9. 64). Some of Ryan's findings were: laborers, service workers, unemployed (whether of short or long duration) and the retired were above average in the likelihood of experi— encing debt trouble; and, the greatest prOportions of debtors, of varying degrees of seriousness, were among the unmarried, the poor, and those under 25 years of age, or 65 years or older. Her life cycle analysis showed that under 45 years old, those "married with no children" had the highest percent of 22 overburdened debtors, closely followed by the "unmarried, no children" stage (which was less than 5 percent of the sample). The highest percentages were found among those over 45 years old, then the "ummarried," closely followed by "married, no children, head retired." In their chapter on "Installment Credit in Perspective," Moore and Klein (1967) review and summarize data on borrower characteristics from which they conclude: The use of installment debt has increased sharply in all income groups since the mid-thirties, but more so in the middle and upper income range than in the lower groups (p.20). . . . and, . . . there has been a substantial increase in recent decades in the incidence of both installment and non— installment debt among all income, occupation, and age groups. The over-all increase is indeed more striking than the redistribution among income, age, or occupation groups (p. 31). They also note that installment debt, automobile paper, the latter's largest component, and total consumer debt, "have all grown at a faster rate than either personal income or total sales of durables" (p. 4). Mandell's report on Credit Card Use in the United States (1972) contains some interesting findings relevant to the debt function of credit card use and the resultant effect on the debt pattern. Data for this study, collected through the Survey Research Center, found four "major" determinants of credit card use—-level of family income, education of the family head, age of the family head and family life cycle stage, and Size and location of the community. His data on occupational 23 credit card use shows the higher occupational status catego— ries of "professional, technical and kindred workers," and "managers, officials" having a higher percent of use (80 percent and 72 percent respectively), while the "laborers and service" category with 36 percent and "Operatives" with 42 percent are at the low end. Regarding the family life cycle findings, Mandell reports: . . families who have the greatest need and make the greatest use of credit cards, both in terms of expenditures and credit features, are young families with children at home (p. 13) . . . and, . . They realize that their needs are greatest at this age and life cycle situation and so they borrow against their expected higher future levels of in- come. Therefore, these families are not only the greatest users of credit cards, but also the greatest users of the credit aspect of the cards (p. 17). In analyzing those who use the credit/debt feature of their credit cards, Mandell notes that there are two possibka methods of repayment: 1) "pay as soon as you can;" and 2) pay the debt off a little at a time, "analagous to in- stallment debt" (p. 90). Of those families using the credit feature, "slightly more than half treat their debt as an in- stallment type of loan" (p. 90). The first repayment method is most likely to be used by older families, those with higher income, and the more highly educated. Those most likely to use the second repayment method are younger families with heads under 35 years of age, and those earning less than $10,000 per year. In view of the large number of credit card users who use their cards as a credit instrument, he 24 hypothesizes that "some credit card users are substituting credit card debt for the more traditional installment debt paper" (p. 95). The data show that an increasing prOportion of the po- pulation with credit card debt is at the middle and higher income levels. Mandell concludes, from part of the data analysis: . . . the addition of credit card debt to other types of non-automobile and non-mortgage debt does tend to change the patterns of debt distribution in the population (p. 99). He notes, however, that the sizeable change occurs among the wealthier families. This may partly be offset by heavier use of the "pay as soon as you can" repayment method. Brown's (1969) research defined social class as a "population segment identified by occupational class with income held constant" (p. 128), and tested for social class behavior differences in broadly defined family investment categories using a stratified sample by income class. The category "human capital" included education and medical care while the "consumer durables" category included household durables and automobile. (The third category, "financial equities and real property," included personal insurance and net change in assets and liabilities.) She found the following differences among social classes: White collar classes invest more in education, me- dical care and insurance than blue collar families of the same income [while] . . . Unskilled workers . . show a preference for household durables (p. 133). 25 No relation was found between social class and spending on autos. She concluded that the results indicate a "need to consider social as well as economic explanations for the financial activities of families" (p. 136). The marketing study by Martineau (1958), considered a classic in its field, was conducted in Metropolitan Chicago and tested the relationship of social-class membership to spending behavior, including purchasing patterns and Spend- save aspirations. Social class was operationalized using the part of Warner's Index of Status Characteristics based on occupation, source of income, and housing type with the weighted scores converted to social class level. (The full ISC includes dwelling area as a factor, but provides for a weight adjustment where missing.) His study showed "a social-class system operative in a metrOpolitan area" and "class membership is an important determinant of the individual's economic behavior" (p. 125). In addition, he states, There is certainly a rough correlation between income and social class. But social class is a much richer dimension of meaning. There are so many facets of behavior which are explicable only on a basis of social class dynamics (p. 125). Wasson (1969) presents some analysis he made of U.S. Department of Labor, Bureau of Labor Statistics data (BLS, "1960-61 Survey of Consumer Expenditures," and "1964 Supplemental" study) which show that occupational classes cut across income groups and "Occupational class, not income . . 26 determines the proportion of spending allocated to some . . important categories" (p. 235). He contends that this data analysis reinforces Martineau's early 1958 study of the value of social class over income class for market segmentation and concludes: . . . market segmentation is influenced strongly by a complex of cultural influences, of which occu- pation and the other elements of social class are important components (p. 238). He admits that occupation is not the only cultural factor, but believes the BLS data "demonstrates the need" to look at "occupation first, and only then at income level" in marketing studies (p. 238). CHAPTER III METHODOLOGY Data Source The source of data for this study is a four-year panel study (1967-1970), undertaken in survey form, by the Survey Research Center, Institute for Social Research, at the University of Michigan, Ann Arbor, Michigan. This study, entitled "Consumer Durables and Installment Debt: A Study of American Households," paralleled the annual "Survey of Consumer Finances" conducted by the Survey Research Center for those years in that a national cross-section of primary family units were interviewed and the same questionnaires were used for data gathering. One major difference, however, was that for the panel study, the interviews were repeated each year with as many of the original families from the first interview as could be located or who responded. Another dif- ference was that families with heads aged 60 or older were excluded from the panel in the initial interview. While the SRC's analysis of the panel data is related to total installment debt, the data base is broader in scope and allows for other analyses. Data are available to classify both installment and noninstallment debt by the purpose for which the debt was incurred. The SRC study, 27 28 therefore, provides an opportunity to obtain empirical data to analyze both types of debt as well as total family debt. This study will utilize the data from one year of the panel study. The plan is to look for patterns, not trends or averages; this can be done using one year's data. The year 1968 (designated Wave II, i.e., second year, by SRC) has been chosen for analysis for two reasons: 1. one of the independent variables, socioeconomic status, is most accurately coded for 1967 and 1968 because only in those years was it obtained by direct question; 2. the year 1968 was chosen over 1967 because use of the later year permits correction or resolution of incon— sistencies or ambiguities that may arise in the earlier interview when data on the same families are available for more than one interview. Procedures for Sampling and Data Collection The SRC samples represent cross-sections of the main- land United States pOpulation, excluding Alaska, living in private households. Excluded are transients, residents of institutions, and persons living on military bases. House- hold refers to dwelling unit, the basic unit for sampling. "The method known as multistage area probability sampling is used to select a sample of dwelling units representative of the nation" (Katona, et al., 1969:235). 29 The SRC has had extensive experience in researching family economic behavior as well as many years experience in survey methods. It maintains a nationwide staff of inter- viewers who are selected and trained by traveling super- visors. "The interviewers are instructed in the careful and uniform use of the fixed-question Open-answer technique. .. . Many questions are answered in the respondent's own words, which the interviewers record verbatim (or as nearly verbatim as possible). Nondirective probes are used to clarify the answers received" (Katona, et al., 1969: 236). In early 1967, the SRC initially interviewed a national cross-section of 2,604 primary family units whose heads were under age 60. The interviews were repeated each year, with the final interview early in 1970. Each year's interview was conducted during the first quarter of the year. The four annual interviews were held at intervals of approximately twelve months with those families who could be located or who had responded. On the fourth round, a panel of 1,436 families remained.* The questionnaires for each of the waves of interview— ing are very similar but do have some differences.** The time frame of the data was as follows: *Data from this SRC Panel Study are available in the form of a merged four year family tape. Hence, any year's analysis would be for those families that completed all four years of interviewing. **The questionnaire for Wave II (1968 year) is repro- duced in the 1968 Survey_of Consumer Finances, (1969) Chap— ter 15, pp. 245-279. 30 l. The annual or disposable income was for the previous year (e.g., for Wave II, the 1968 year, the disposable income was for the year 1967). 2. The amounts and types of debt were at the time of interview. The data collection, preparation, editing, coding and re—editing were carefully supervised and coordinated by experienced professionals throughout the various survey Operations until the final documentation and archiving of the data in 1971 (Hendriks, et al., l973:vii). Study Design and Operational Definitions The purpose of this study is to look for family debt patterns as related to family life cycle stages and socio- economic status levels by analyzing the components of install- ment and noninstallment debt. Nominal definitions for design of this study are: Family Unit - "All persons living in the same dwelling unit who are related by blood, marriage, or adoption." A single person unrelated to the other occupants in the dwelling unit or living alone is a separate family unit. Debt Patterns include debt incurrence and type of debt. Debt incurrence is debt classified (in five categories) by purpose for which incurred: 1. Additions and Repairs 2. Car 3. Durables 31 4. Other debt (e.g., other major transactions, travel) 5. Medical and Dental Type of debt is classified as installment or noninstall- ment. Independent Variables The independent qualitative variables are stages of the family life cycle (FLC) and socioeconomic status levels (SES). Stages of the Family Life Cycle The complete Survey Research Center classification con- tains ten life cycle stages in its panel study. Following review of a frequency distribution of extracted raw data, it was found that the two "retired head" stages had a very small number of family units. Therefore, these two stages were. combined with the equivalent stages with "head in labor force" and the number of stages was reduced to eight. Further analysis of the extracted data following the initial change showed the number of families at each stage of the FLC was not large enough for meaningful analysis, particularly at the early stages and the "any age, single with children" stage even if adjustments of the socioeconomic levels were made. The SRC Life Cycle classification was modified and Operationalized in this research study as follows: 32 Under Age 45 Stage I - Unmarried, no children and Married, 2 or more adults, no children Stage II - Married, 2 or more adults, youngest child under 6 Stage III - Married, 2 or more adults, youngest child 6 or over Age 45 or over Stage IV - Married, 2 or more adults, children at home Stage V - Married, 2 or more adults, no children* and Unmarried, no children *(The term "no children" means no children under 18 living at home.) These changes, primarily collapsing to fewer levels, were made in the most homogeneous manner thought possible. For the same reason, the stage "any age, unmarried, has children," the most heterogeneous and one of the smallest groups (65 families), was eliminated from the final opera- tional structure. Socioeconomic Status Levels The SRC panel study classifies families by socioeconomic status using the Duncan socioeconomic index ratings. Dr. 0. Dudley Duncan used the percent 'Excellent' or 'Good' ratings for the ninety titles of the 1947 NORC [National Opinion Research Council] prestige study as the criterion for the derivation of a regression equation expressing the criterion as a linear function of income and education. He based the two regression weights on age-specific education and income patterns and sub—classified some large occupation groups by industry to get more precise SES ratings for detailed occupational 33 titles from the 1950 census data for the civilian male population* (Scheffler et al., ed., 1971:5). SRC coding is available for three Duncan score ranges: 1) from 00-96; 2) decile scale; and 3) scores bracketed into 19 categories of five-score ranges. The original intent was to use the Duncan Decile Scale which has a range of 0—9. It divides the entire experienced labor force of 1950 (the popu- lation Duncan used to derive his socioeconomic scores) into tenths according to the socioeconomic scores of their occur pations. The tenth of the labor force who worked in occupa— tions with the highest socioeconomic scores were assigned the decile scale score of 9. The tenth with the lowest scores were assigned a score of 0. A raw data frequency breakdown showed a small number of family units at several of the SES levels, and the ten levels were collapsed to five. A further analysis of the data, extracted in tabular matrix form, showed the number of families per cell using five SES levels was not sufficient for meaningful analysis. The low end of the SES range had the smallest frequencies, so the ten levels of the Duncan Decile scale were regrouped and operationalized for this study as follows: *The background and development of the Duncan SES index is contained in Occupation and Social Status, by Albert J. Reiss, Jr., Chapters VI-VII, pages 109-161, and Appendix B, Table Bl, pages 263-275, by Otis Dudley Duncan. 34 Low level - The range from 0 through 3, Middle level - The range from 4 through 6, High level — The range from 7 through 9. This provided a 5 by 3 matrix with 15 cells. The table below (Table 3.1) shows the distribution of the study sample population for each of these cells. Table 3.1. Frequency Distribution: Total Sample POpulation by Family Life Cycle Stage and Socioeconomic Status Levels (N=1252) Family Life Cycle Stages Socioeconomic Status Levels Row I II III IV V Total Low 22 84 45 76 76 303 Mid 11 102 69 58 68 308 High 64 180 116 142 139 641 Column Total 97 366 230 276 283 1252 Dependent Variables The dependent variables are formed by classifying the components and totals of installment and noninstallment debt. The following SRC definitions utilized for this study delineate the scope of the data:* *These have been taken or summarized primarily from Consumer Durables and Installment Debt and supplemented from working papers provided the author by SRC. {35 Installment Debt - A11 private, non—mortgage debt of the family unit (or household) which is subject to two or more regular payments regardless of timing. Thirty-day charge accounts and transactions in which the purchaser promised to pay within thirty days are not included, but revolving charges and open-ended accounts owed to a store on which a set amount is supposed to be paid regularly are included. Total Outstanding Installment Debt - The dollar amount of outstanding debt owed by the family on the day it was interviewed. Non-Installment Debt - Private, non—mortgage debt that is owed to a financial or commercial institution which is not subject to more than one payment (even though, despite the original agreement, the debtor may make more than one payment) and all debt owed to individuals or non-financial or non-commercial institutions (such as a hospital or school). Total Outstanding Non-Installment Debt - The dollar amount of debt owed at the time of the interview. Debt Exclusion - Debt incurred for groceries, utility bills and taxes; debt incurred for business and investment purposes. Annual Income - Total family income after the deduc- tion of estimated federal income tax liabilities, hereinafter called Disposable Income. All debt is outstanding debt (as different from monthly or annual debt) and hereafter the word "outstanding" is 36 assumed without being repeated for each debt variable. Thus the dependent variables are: Installment debt for each of the five categories of debt incurrence and their total -- Additions and Repairs installment debt Car installment debt Durables installment debt Other installment debt Medical and Dental installment debt Total installment debt Noninstallment debt for each of the five debt incurrence categories and their total —— Additions and Repairs noninstallment debt Car noninstallment debt Durables noninstallment debt Other noninstallment debt Medical and Dental noninstallment debt Total noninstallment debt Total incurrence debt (installment and noninstallment) for each category -- Total Additions and Repairs debt Total Car debt Total Durables debt Total Other debt Total Medical and Dental debt Total Debt - (the grand total) is the sum of total installment debt and total noninstallment debt. 37 The debt dependent variables are operationalized by calculating a ratio of outstanding dollar amount of debt to the family's disposable income arriving at the "degree of debt," the term used for this debt measure. This gives the percent of debt for each of the debt variables. The reason for using this debt-to-income ratio as the dependent debt measure is to assess the comparative impact of the degree of debt by adjusting out the income differences. Procedures for Analyzing the Data The data were analyzed using the Michigan State Univer- sity computer facilities. The various descriptive and statis- tical analyses were carried out primarily utilizing the SPSS: Statistical Package for the Social Sciences (Nie et al., 1975) programs, and were run on a Control Data Corporation (CDC) 6500 computer. Following the preliminary analysis discussed in the Operational definitions above (page 31), the sample popula- tion was separated into two groups: those with debt, and those without debt. A frequency distribution was obtained for each debt variable giving the number of families (N) in each cell, the number of families in each cell with no debt (N') and the percent of families in each cell that had no debt (i.e., without debt) N'/N. The with debt group was statistically analyzed, where possible, with the two-way analysis of variance test using the degree of debt measure (debt to income ratio). The "F" 38 ratio was the basis for testing for significant differences and the .05 probability level was used for significance and for acceptance or rejection of the hypotheses. The addi- tional descriptive statistics obtained were the cell, row, and column means, and the grand mean. Utilizing the statistical breakdown, a table of means was compiled for each variable which showed a significant interaction or main effect. Where there was an interaction, the means were graphed to look at the interactive effect. From the tables, a comparison of marginal means was made, and for the significant variables an examination could then be made of the cellular differences. A post hoc analysis was made for those variables whose main effect "F" ratios were significant at the .05 level or better to determine where the significant differences were occurring. An analysis of variance statistical test was made for each possible contrast of levels for the significant independent variable holding the other independent variable constant. This method is not applicable to a two—way inter- action; hence the analysis was computed for main effects only. The original intent was to use the analysis of variance statistical method for each of the dependent debt variables to look for possible interaction between life cycle stages and socioeconomic status levels, and if there was none, to look at the main effects and determine where any of the significant differences were. 39 There was a wide range, sometimes sizeable, of the N or number of families in the with debt suprpulation for each debt variable. In some cases the number was extremely small. This imposed a severe restriction on completing the two—way analysis of variance (ANOVA) statistical test across all levels and stages of the independent variables. For the test, there must be at least two (2) per cell or five (5) per level to get a meaningful analysis. Therefore this statis- tical test had to be limited to those debt variables that met this criteria—-at least at most levels. Thus the group of noninstallment debt incurrence variables had to be entirely eliminated from this method of analysis. In addition, not all of the ANOVA tests on the remaining debt variables were able to include all stages of the family life cycle but do include all levels of socioeconomic status. All stages of the family life cycle were included in the ANOVA statistical test with the following exceptions: 1. Missing FLC stage I only are Addition and Repairs installment debt Durables installment debt Total A and R installment debt Total Durables installment debt 2. Missing FLC stages I and V are Medical and Dental installment debt and Total Medical and Dental installment debt. CHAPTER VI FINDINGS AND DISCUSSION Frequency Distribution Since the number of families having debt for the dif- ferent incurrence and type categories varied considerably and the incidence of no debt was predominant, it was believed that a frequency distribution of the total sample population for each debt dependent variable using the no debt classification would be meaningful. This series of tables gives: 1. The number of families (N) of the total sample popu- lation who are cross-classified in each cell, i.e., in a particular family life cycle (FLC) stage at a given socio- economic status (SES) level. For example, 22 of the sample families are FLC stage I and at the low SES level (I/L); 2. The number of those families in a cell having no debt (N') for each named debt variable; and, 3. The ratio of the number of families in the cell with no debt (N') to the number of sample families in that cell (N) expressed as a percent. 40 41 The number and percent for the families with debt can be obtained by subtraction.* The individual frequency distribution tables can be found in Appendix A, Tables A.l - A.18. The frequency distribution tables are summarized by debt groups. The pattern of the individual incurrence debt variables (additions and repairs, car, durables, other and medical and dental) for the installment, noninstallment, and total debt groups are compared to the relevant total. The results are shown in Tables 4.1 - 4.3. In addition, the totals for type of debt are summarized in Table 4.4. For each of these incurrences, the FLC stage and SES level where both the highest and lowest percent of families with no debt occurs is shown in order that patterns, if any, can readily be discerned. In the installment incurrence group, a pattern can be seen among families in the highest percent of no debt and in the lowest percent of no debt for both family life cycle (column totals) and socioeconomic status (row totals). *For example, the number of families (N') having no car installment debt was 749, or 59.8 percent of the total N (1252). Therefore, 503 or 40.2 percent had car installment debt. 42 Table 4.1. Installment Debt: Percent of Families With No Debt (Tables A.l - A.6) # Installment FLC Stages SES LGVGIS Variable Highest Lowest Highest Lowest Additions and Repairs I IV L M Car V I/II* L M Durables V II H L Other V II L M Medical and Dental V III L M Total Installment V II L M *The difference between the stages is only 0.4 percent. #L-low; M-middle; H-high A pattern emerges where the percent of no debt for the FLC is highest at stage V and lowest at stage II; for the SES, it is highest at the low level and lowest at the middle level. Except for the lowest percent related to FLC, the patterns are strong. Exceptions to this pattern occur for several of the incurrence variables. Because of the strong general pattern shown by the highest and lowest percent of no debt families in relation to FLC and SES among the individual installment incurrences it is not unexpected that the percent of families with no total installment debt follows a similar pattern. 43 The frequency distribution of no debt for noninstallment debt incurrences also has a pattern-—but it is not as strong as in installment debt incurrences: Table 4.2. Noninstallment debt: Percent of Families With No Debt (Tables A.7 - A.12) Noninstallment FLC Stages SES Levels Variable Highest Lowest Highest Lowest Additions and Repairs V/I** II M H Car' I III/II** M. H Durables I/V* II H/L*** M Other V I L H Medical and Dental V III M H Total Noninstallment V III L/M* H *Both have the same percent **The difference is 0.6 percent ***The difference is 0.1 percent The most consistency is found at the lowest percent of no debt for SES. This is at the high level with one excep- tion-—durables, where it is at the middle level. There are tendencies in other areas. The highest percent of no debt most frequently occurs at the middle level of SES, but with some exceptions. For FLC, the lowest percent of no debt is generally at stage III or stage II except for other; and the highest percent of no debt tends to be found mostly at stage V. 44 Total noninstallment debt normally follows the majority pattern but reflects the diversity of the individual non— installment incurrences. The total incurrence debt group is fairly consistent: Table 4.3. Total Incurrence Debt: Percent of Families With No Debt (Tables A.13 - A.18) Total FLC Stages SES Levels Variable Highest Lowest Highest Lowest Additions and Repairs I/V* DWGII** L M Car V II L M Durables V II H L Other V II L M Medical and Dental V III L H/M** Total Debt V II L M *Difference is 0.5 percent **Difference is 0.4 percent The most consistent pattern is found in the highest percent of no debt. For FLC, it is at stage V, except for additions and repairs. The lowest percent of no debt for FLC stages has considerable variation, but stage II predomi- nates. For SES, the highest percent of no debt is at the low level, with durables the only exception-—at the high level 45 Table 4.4. Type of Debt Group: Percent of Families With No Debt (Tables A.6, A.12, A.18) Type of FLC Stages SES Levels# Debt Variable Highest Lowest Highest Lowest Total Installment V II L M Total Noninstallment V III L/M* H Total Debt V II L M *Both have the same percent #L-low; M-middle; H-high Debt in the Sample Population For an overview of the with debt population, the means (percent of debt-to-income) and its percent in the total sample population are tabulated. The results are given in Table 4.5. From this, comparisons can be made within debt groups and between groups for similar debt variables. In the installment incurrence group, the highest degree of debt (means percent) is for car debt, with other debt being the second highest. The lowest degree of debt is for durables debt. The noninstallment incurrence debt group could not be analyzed with the two-way analysis of variance because of the very small sample size. Tables of means were also not included because the very small percent of the population could easily distort the means and was not considered typical of the general population. The grand means are included in 46 Table 4.5. All Debt Variables by Percent of Total Population and Grand Mean Degree of Debt* for the With Debt Population Installment Noninstallment Debt Debt Total Debt Variable Mean N Mean N Mean N Additions and Repairs 11.2 8.5 9.7 1.4 11.3 9.5 Car 15.0 40.2 30.8 3.0 16.5 42.3 Durables 4.6 23.5 16.5 1.1 5.2 24.5 Other 12.9 27.7 19.2 9.3 16.0 33.5 Medical and Dental 6.4 4.2 16.6 5.1 11.7 8.7 Total 18.5 65.3 21.5 18.1 22.8 70.2 *expressed as a percent the table only as a bridge and as a possible indication of the change and direction from the installment incurrence means to the total incurrence means. For the total incurrence debt group, the highest degree of debt is for car debt, almost matched by other debt. The smallest degree of debt is for durables. Though the relative placement for the totals are the same as for the installment incurrence group, the means percent or degree of debt is higher. Looking at the totals of the type of debt (Table 4.5) installment debt was owed by 65.3 percent of the sample popu- lation having a mean debt to income ratio of 18.5 percent, 47 and is used much more frequently than noninstallment debt which was owed by 18.1 percent of the families. However, the debt ratio was higher at 21.5 percent. Some kind of debt, that is, total debt was owed by 70.2 percent of the families; the debt ratio being 22.8 percent. This would seem to indi— cate that Some families are assuming both installment and non- installment debt which is compatible with observed experience. Tests of Hypotheses The results of the two-way analysis of variance for the three groups of variables--installment incurrences, total incurrences, and totals for type of debt--are presented in Tables 4.6 - 4.8. Hypotheses were previously stated in a directional form. The hypotheses are here restated in a non-directional null form to facilitate acceptance or rejection. They are: I.. There will be no difference in the degree of debt for each of the five debt incurrence categories, that is, additions and repairs, car, durables, other, and medical and dental, both for the installment incurrence debt group and the total incurrence debt group A. between stages of the family life cycle, B. between levels of socioeconomic status, and C. no interaction between family life cycle and socioeconomic status. II. There will be no difference in the degree of debt between the three types of total debt--installment, noninstall- ment and (grand) total 48 A. for stages of the family life cycle, B. for levels of socioeconomic status, and C. no interaction between family life cycle and socioeconomic status. Debt Incurrence Categories Installment Examination of the installment debt incurrence group (Table 4.6) reveals that only durables installment debt had a significant main effect and this was for socioeconomic status, which had an F ratio of 5.659 with 2 degrees of freedom at the .004 level of probability. Based on these results, Hypothesis I for the install- ment incurrence group is accepted or rejected as follows: A. Between stages of the family life cycle, the statistics show no significant differences from this main effect. Therefore, the null hypothesis is accepted. B. Between levels of socioeconomic status, as indicated in Table 4.6,there was a significant main effect for durables installment debt. There were1x>significant differences for ad— ditions and repairs, car, other,cn medical and dental install- ment debt. Therefore, for durables installment debt the null hypothesis is rejected. For additions and repairs, car, other, and medical and dental installment debt the null hypothesis is accepted. C. There was no interaction between the effects of family life cycle and socioeconomic status. Therefore, the null hypothesis of no interaction is accepted. 49 Table 4.6. Two-Way Analysis of Variance of Family Life Cycle and Socioeconomic Status for Installment Debt Incurrences Installment N# Soumce of Digrees F Level of Variable (25) Variation Free cbm Ratio Probability Addition & 106 Repairs (8.5) Main effect of SES 2 2.061 .131 (No stage I) of ED: 3 1.042 .379 Interaction SES by FLC 6 .717 .999 Residual 94 Car 503 (40.2) Main effect of SES 2 2.501 .081 of FLC 4 2.020 .089 Interaction SES by FLC 8 1.417 .186 Residual 488 Durables 298 (23.8) Main effect of SES 2 5.659 .004* (No stage I) of FLC 3 2.255 .081 Interaction SES by FLC 6 1.266 .272 Residual 286 Other 347 (27.7) Main effect of SES 2 1.084 .340 of FLC 4 .816 .999 Interaction SES by FLC 8 1.248 .269 Residual 332 Medical & 53 Dental (4.2) main effect of SES 2 2.015 .144 (No stage I, of FLC 2 .130 .999 V) Interaction SES by FLC 4 .082 .999 Residual 44 *Significant at < .05 level #Nmber of families with each debt %-Percent of population with each debt 50 9:221 The total incurrence group results (Table 4.7) shows only total car debt had a significant main effect for socio- economic status with an F ratio of 4.317 with 2 degrees of freedom at the .014 level of probability. Total other debt, part of this group, was the only debt variable tested which shows a significant two-way interaction between family life cycle and socioeconomic status with an F ratio of 2.264 with 8 degrees of freedom at' the .022 level of probability. The means percents (degree of debt) were plotted on a graph (Figure 4.1) to look at the interactive effect. The lines are obviously not parallel and do cross each other repeatedly. The high level of SES is quite stable. But low and middle levels of SES cross several times. Low SES level has a very high degree of debt at FLC stage I and a high degree of debt at stage IV. Middle SES level has a high degree of debt at FLC stage V and a moderately high degree of debt at stage II. In each case the high points for low level SES come one FLC stage before a rise in degree of debt for middle SES. 51 Table 4.7. Two-Way Analysis of Variance of Family Life Cycle and Socioeconomic Status for Total Incurrence Debt 'Ibtal N Source of Defies F Level of variable (%) variation F 3 Ratio Probability Additions & 119 Repairs (9.5) Main effect of SES 2 1.947 .146 (No stage I) of FLC 3 .836 .999 Interaction SES by FLC 6 .859 .999 Residual 107 Car 529 (42.3) Main effect of SES 2 4.317 .014* of FLC 3 .123 .999 Interaction SES by FLC 8 1.001 .435 Residual 514 Durables 307 (24.5) Main effect of SES 2 1.831 .160 (No stage I) of FLC 3 1.069 .363 Interaction SES by FLC 6 .978 .999 Residual 295 Other 420 (33.5) Main effect of SES 2 1.898 .149 of FLC 4 1.137 .338 Interaction SES by FLC 8 2.264 .022* Residual 405 Medical & 109 Dental (8.7) Main effect of SES 2 2.303 .103 (No stage I, of FLC 2 .195 .999 V0 Interaction SES by FLC 4 1.140 .342 Residual 100 *Significant at < .05 level 52 MEAN DEGREE OF DEBT Z 60 A 50 M 40 q 30 1 Middle SES ’7 ‘0 I Low SES Il"' " ’ "I". ‘| ’ ' IIIHIIIIIIIIII“““ 10 " High SES — I ‘,o 0 ‘g 0 l l l 1 l 7 T T T T— I II III IV V FAMILY LIFE CYCLE STAGES Figure 4.1 Total Other Debt: Mean Degree of Debt Interaction Between FLC & SES 53 Based on these results, Hypothesis I for the total incurrence group is accepted or rejected as follows: A. Between stages of the family life cycle, the statistics show no significant differences from this main effect. Therefore, the null hypothesis is accepted. B. Between levels of socioeconomic status, as indi- cated above from Table 4.7, there was a significant main effect for total car debt. There were no significant main effects for additions and repairs, durables, other or medical and dental. Other showed a significant interaction. Therefore, for total car debt the null hypothesis is rejected. For additions and repairs, durables, other, and medical and dental the null hypothesis of no difference for main effect is accepted. C. An interaction between the effects of family life cycle and socioeconomic status was found, as noted above, and shown in Table 4.7, for total other debt. No other inter— action was found. Therefore, the null hypothesis of no inter- action for total other debt is rejected. For additions and repairs, car, durables, and medical and dental, the null hypothesis of no interaction is accepted. Type of Debt Category Looking at the type of debt group (Table 4.8) reveals a significant main effect for socioeconomic status for each one in the group. The results of the significant main effect of socioeconomic status, all with 2 degrees of freedom were 54 Table 4.8 Two-Way Analysis of Variance of Family Life Cycle and Socioeconomic Status for Types of Debt Type N Source Degzies F Level of‘ variable (%) variation Freedom. Ratio Probability final IhstiUnent 818 (65.3) Mahleffixt: of SES 2 4.235 .015* of FLC 4 1.585 .175 InUflectRXI SES by FLC 8 1.712 .091 Residual 803 Tbbilfibn- insbiUhent 226 (18.1) Mahleffian: of SES 2 4.477 .012* of FLC 4 .242 .999 Infinactkn SES by FLC 8 1.245 .274 Residual 211 Tbtal 879 (HLZ) Mahlefflafi: of SES 2 5.557 .004* of FLC 4 .729 .999 lhtenxxion SES by FLC 8 1.564 .131 Residual 864 *Significant at < .05 level 55 as follows: total installment debt had an F ratio of 4.235; total noninstallment debt had an F ratio of 4.477; and total debt (grand total) had an F ratio of 5.557 at the levels of probability of .015, .012, and .004 respectively. Thus there was a significant effect of socioeconomic status on all three types of debt. The test results applicable to Hypothesis II for the type of debt are accepted or rejected as follows: A. Between stages of the family life cycle, the statistics show no significant differences from this main effect. Therefore, the null hypothesis of no effect is accepted. B. Between levels of socioeconomic status, there was a significant main effect for total installment debt, total noninstallment debt, and total debt. Therefore, the null hypothesis of no effect is rejected for all debt variables. C. There was no interaction between the effects of family life cycle and socioeconomic status. Therefore, the null hypothesis of no interaction is accepted. The post hoc analysis, to look for where the signifi— cant differences were occurring, was completed for those debt variables whose two-way analysis of variance tests were statistically significant as noted above. To summarize, all significant main effects were from the effect of socioeconomic status. An ANOVA was therefore computed separately for each possible contrast of socio- economic status levels with the total family life cycle 56 range. (The same exception applies as in the complete two-way analysis of variance computation; that is, install- ment durables debt eliminated FLC stage I.) The effect of this was to analyze for differences between each SES contrast holding the effect of the stages of the family life cycle constant. The results are presented in Table 4.9. The significant contrast levels of socioeconomic status are related to comparable directional hypotheses (see Chapter I, p. 12). There were no significant differences in the contrast between the low and middle levels of socioeconomic status. The results for the individual variables are: For durables installment debt, it was hypothesized that a linear pattern will emerge, with the largest ratio at the low level and the smallest at the high level.* Signifi- cant contrasts were found between low and high levels at the .001 level of probability; and between middle and high levels at the .017 level of probability. The hypothesis is supported. For total car debt, it was hypothesized that an inver- ted uneven V pattern will emerge, with an intermediate ratio at the low level, the largest ratio at the middle and the smallest ratio at high. A significant contrast was found *Comparison of means for significant main effects can be found in the next section. The level for the largest and smallest ratios (percents) is discussed. 57 Table 4.9. Post Hoc Analysis of Variance Contrast of All Socioeconomic Levels for Significant Debt Variables Degrees Variable Contrast of F. Level of Levels Ratio Probability Freedom Durables SES Between: Ingggtlment Low & Middle 1 .681 .999 Middle & High 1 5.740 .017* Low & High 1 11.234 .001* Total Car SES Between: Debt . Low & Middle 1 3.052 .078 Middle & High 1 .003 .999 Low & High 1 5.784 .016* Total In- SES Between: Sggtlment Low & Middle 1 .658 .999 Middle & High 1 9.528 .003* Low & High 1 3.205 .070 Total Non- SES Between: mgitaDébt Low & Middle 1 2.036 .153 Middle & High 1 2.116 .144 Low & High 1 8.265 .005* Total Debt SES Between: Low & Middle 1 .680 .999 Middle & High 1 6.547 .010* Low & High 1 9.825 .002* *Significant at < .05 level 58 between the low and high levels at the .016 level of probabi- lity. The hypothesis was not supported. For the totals of type of debt--total installment, total noninstallment, and total debt-~it was hypothesized that a linear pattern will emerge, with the largest ratio at the low level and the smallest ratio at the high level. For total installment debt, a significant contrast was found between the middle and high levels of socioeconomic status at the .003 level of probability. The hypothesis was partly supported. For total noninstallment debt, a significant contrast was found between the low and high levels of SES at the .005 level of probability. The hypothesis was supported. For total debt, significant contrasts were found between low and high levels of SES at the .002 level of probability; and between middle and high levels of SES at the .01 level of probability. The first significant contrast supports the hypothesis. Comparison of Means for Significant MaIn Effects Using the descriptive statistics, a matrix table of means (percent of debt to income mean) was formed for each variable tested by the two-way analysis of variance in order to facilitate a comparison of means. These are shown in Tables 4.10 - 4.15.* *Table 4.12 is for total other debt and should be viewed in conjunction with Figure 4.1. 59 In examining those debt variables with significant main effects we note the following: 1. For installment durables debt (Table 4.10), the degree of debt is highest for low SES and decreases linearly through the middle and high levels of SES. Between the middle and high SES levels, the difference in degree of debt is 1.2 percent while between the low and high SES levels, the degree of debt difference is 1.8 percent. Both differences were statistically significant on the post hoc analysis as noted above. 2. For total car debt (Table 4.11) the pattern is different. The low level of SES is much higher than middle or high SES, the latter being very close in degree of debt. Although low and middle levels differ more in their marginal means than low and high SES levels, in the latter case, the contrasts are greater across all stages of the family life cycle. This may be the reason why the low and high SES con- trast differences were statistically significant while the low and middle SES contrasts were not, on the post hoc analysis. 3. For total installment debt (Table 4.13), the degree of debt is highest for middle SES level, followed by low level, with the high SES level having a much lower degree of debt. The large difference in the mean between middle and high SES levels corroborates the post hoc statistical signi- ficance noted above. 60 A visual examination shows that the high and low means are not consistent across the stages of the family life cycle. FLC stage I has the highest degree of debt and FLC stage III has the lowest degree of debt. However, the highest degree of debt for the low SES level is at FLC stage IV, while the lowest degree of debt for high SES is also at stage IV. Neither trend, however, is sufficient to be statistically significant. 4. For total noninstallment debt (Table 4.14), the degree of debt follows a pattern similar to installment durables debt (1 above) in that it is linear with the highest at low SES level and lowest at high SES level. However, the differences are greater. The difference in the degree of debt between low and high SES levels is over 25 percent, while the difference in the degree of debt between the middle and high levels of SES is 7.5 percent. Thus, the post hoc statistically significant difference for the contrast between low and high SES can be descriptively seen. 5. For the (grand) total debt, (Table 4.15), the degree of debt is again in decreasing order from the highest mean percent at the low SES level through the middle SES level to the smallest mean at the high SES level. However, the comparative size of the spread is not the same as total noninstallment debt (4 above). The difference in the degree of debt between low SES and high SES is 9.3 percent, while the difference in the degree of debt between middle and high 61 SES levels is 5.2 percent. In the post hoc analysis, both of these contrasts were statistically Significant, but the first contrast (low/high) had a lower significance level. The tables of means for the debt variables that were not statistically significant are presented in Appendix B. 62 Table 4.10. Durables Installment Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies)* Socioeconomic Family Life Cycle Stages Row Total Status Levels I II 'III IV V % (N) Low 5.1 5.9 4.4 4.2 9.7 5.5 (5) (39) (ll) (31) (10) (96) Mid 1.6 5.2 5.3 4.8 4.2 4.9 (l) (32) (21) (17) (ll) (82) High 5.9 4.1 3.5 2.3 4.0 3.7 (11) (56) (23) (30) (17) (137) Column % 5.4 4.9 4.4 3.6 5.6 4.6 Total (N) (17) (127) (55) (78) (38) (315) *Numbers in parentheses are cell frequencies; ratios are in percent. Table 4.11. Total Car Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies)* Socioeconomic Family Life Cycle Stages Row Total Status Levels I II III- IV V % (N) Low 18.7 28.8 13.9 27.6 15.2 23.3 (9) (42) (21) (26) (13) (111) Mid 18.3 13.9 14.9 15.5 13.1 14.5 (5) (55) (33) (26) (22) (141) High 16.2 12.6 15.0 13.4 19.6 14.8 (33) (83) (58) (61) (42) (277) Column % 16.9 16.8 14.7 17.1 17.0 16.5 Total (N) (47) (180) (112) (113) (77) (529) *Numbers in parentheses are cell frequencies; ratios are in percent. 63 Table 4.12. Total Other Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies)* Socioeconomic Family Life Cycle Stages Row Total Status Levels I II III IV v % (N) Low 50.2 11.8 9.7 33.5 15.6 19.8 (4) (32) (10) (21) (12) (79) Mid 9.2 22.9 8.6 11.1 32.7 18.2 (3) (55) (30) (25) (16) (129) High 16.1 12.9 12.6 12.1 15.5 13.3 (20) (78) (46) (39) (29) (212) Column % 20.3 16.0 10.9 17.1 20.3 16.0 Total (N) (27) (165) (86) (85) (57) (420) *Numbers in parentheses are cell frequencies; ratios are in percent. Table 4.13. Total Installment Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies)* . . Family Life Cycle Stages Row Total SOCioeconomlc Status Levels I II III IV V % (N) Low 21.0 19.8 14.6 25.7 14.6 19.8 (12) (68) (30) (48) (28) (186) Mid 31.8 25.0 17.1 19.2 18.6 21.2 (7) (84) (52) (43) (37) (223) High 21.6 15.9 15.4 13.6 20.3 16.5 (41) (131) (87) (90) (60) (409) Column % 22.6 19.5 15.8 18.1 18.5 18.5 Total (N) (60) (283) (169) (181) (125) (818) *Numbers in parentheses are cell frequencies; ratios are in percent. 64 Table 4.14. Total Noninstallment Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies)* Socioeconomic Family Life Cycle Stages Row Total Status Levels I II III IV V % (N) Low 38.8 67.4 19.7 33.5 22.1 39.8 (4) (15) (12) C13) (5) (49) Mid 2.4 23.9 19.5 8.8 49.4 21.9 (2) (23) (ll) (9) (5) (50) High 8.7 9.8 23.0 14.1 13.0 14.4 (10) (40) (31) (31) (15) (127) Column % 15.4 25.0 21.6 18.0 22.1 21.5 Total (N) (16) (78) (54) (53) (25) (226) *Numbers in parentheses in percent. are cell frequencies; ratios are Table 4.15. Total Debt: Mean Ratios of Debt by FLC and SES (Cell Means and Frequencies)* Socioeconomic Family Life Cycle Stages Row Total Status Levels I II III . IV V % (N) 33.9 33.2 20.4 34.1 16.2 28.6 (12) (71) (33) (49) (32) (197) Mid 32.5 29.1 19.4 20.5 24.0 24.5 (7) (91) (57) (44) (39) (238) High 23.1 17.4 22.6 16.6 20.4 19.3 (42) C142) (91) (100) (69) (444) Column % 26.3 24.6 21.2 21.9 20.5 22.8 Total (N) (61) (304) (181) (193) (140) (879) *Numbers in parentheses are cell frequencies; ratios are in percent. CHAPTER V SUMMARY, CONCLUSIONS AND IMPLICATIONS Summary The purpose of this study was to analyze the variations in the ratios of debt to income utilizing the components of installment and noninstallment debt to determine if debt patterns exist in relation to stages of the family life cycle and socioeconomic status levels; and to interpret the findings in respect to implications for family financial management. Debt patterns were defined to include 1) incurrence and 2) type. 1. Debt incurrence is debt classified by purpose for which it was incurred, that is, additions and repairs, car, durables, other, and medical and dental. 2. Type of debt includes installment, noninstallment and total. Family was defined as all persons living in the same dwelling who are related by blood, marriage, or adoption. A single person unrelated to the other occupants in the dwelling unit or living alone is a separate family unit. Installment debt was defined as private, non-mortgage debt subject to two (2) or more regular payments. 65 66 Noninstallment debt is generally private, non-mortgage debt not subject to more than one payment. Degree of debt is the term referring to the dependent debt to income ratio measure (expressed as a percent). Income refers to disposable income which is the total family income after the deduction of estimated federal income taxes. The source of data for this research study was one year (1968: Wave II) of a four—year panel study on consumer dura- bles and installment debt conducted by the Survey Research Center, University of Michigan covering the years 1967—1970. The data were made available on a four-year merged family tape. The SRC life cycle classification was modified for this study and Operationalized into five (5) stages of the family life cycle. The Duncan Socioeconomic Status Decile scale was modified and operationalized into low, middle and high levels of socioeconomic status. The total sample population numbered 1252 families (primary family units). The debt components were classified by debt incurrence, by debt type, and their respective totals. The dependent debt variables thus formed were: installment, noninstall- ment and total incurrence variables for each of the five purposes; totals for installment debt, noninstallment debt, and the (grand) total debt. 67 The debt measure was the ratio of outstanding dollar amount of debt to the family's disposable income expressed as a percent. Degree of debt is the term used for this debt measure. The five noninstallment debt incurrence variables had to be eliminated from the major statistical analysis test because of the very small number of families who had debt in this category. The number across the levels and stages of the independent variables did not meet the minimum criteria for a meaningful analysis. The statistical method used for analysis was the two- way analysis of variance (ANOVA) test. The level of signi- ficance was at the .05 probability level. Where there were significant differences from socioeconomic status (main effect) as a result of the two-way ANOVA, a post hoc analysis was made to determine where the significant differences were occurring,that is, at or between which levels of SES. The method used for this was an ANOVA for each possible con- trast of levels for the significant variable (SES), holding the other independent variable constant (FLC). Conclusions The results of the two-way analysis of variance data analysis showed a significant effect of socioeconomic status levels on the following debt variables: durables install- ment debt, total car debt, total installment debt, total noninstallment debt, and the (grand) total debt. The post 68 hoc analysis showed that none of the significant differences were between low and middle levels of SES; but significant differences occurred between low and high SES levels, between middle and high SES levels, or both—~depending on the individual debt variable. The basic direction of differences was linear with the largest debt ratio at the low SES level and the smallest ratio at the high SES level. Exceptions were: total installment debt where middle SES level had the largest ratio; and total car debt where middle and high SES levels were almost the same for the smallest ratio. There was only one significant interaction effect of family life cycle and socioeconomic status. This occurred for total other debt. There were no significant effects (main effects) from stages of the family life cycle. There are several possible reasons why this was contrary to past studies (Shaffer, etc.). The use of credit has become more widespread and its differences in relation to the stages of the family life cycle may have become diffused. The change may, in part, also reflect the expanded use of credit cards. One other possibility is that the study sample pOpulation is under- represented for young families where important differences are likely to occur in relation to later stages of the FLC. The SRC panel study did not replace lost families after the initial interview. Both because of faster changes in 69 status at this age (single to married, or no children to with children) and because younger families are generally more mobile, the lost or dropped families are likely to occur disprOportionately at the younger stages. The empirical results of this study indicate that credit/debt patterns are influenced by socioeconomic status level. While not all significant differences were as generally expected--that is, degree of debt and level of socioeconomic status would be inversely related--the heaviest debt burdens were found most likely to occur at the low level of SES. Although stages of the family life cycle did not show statistically significant patterns of degree of debt for most debt categories, a pattern similar to what would be expected on the basis of previous findings was reflected in the frequency distribution. In most debt categories, the largest percent of families with debt was at FLC stage II (young, married, youngest child under 6 years of age) while the smallest percent was at stage V (older, married with no children at home or single). Stages of the family life cycle and socioeconomic status levels showed the only statistically significant interactive effect with respect to total other debt. This may be of special importance and may indicate a changing pattern of installment credit/debt from the traditional, more formal, secured-type of installment loan to expanded use of 7O newer types of credit instruments for a wide variety of purposes. Implications from This Study For Education and Counseling Analysis of the empirical data indicates that the SES level has an effect on the degree of debt both for types of debt and for some purposes for which debt is incurred. Therefore, the concept of socioeconomic status has relevancy in understanding the use of consumer credit/debt. Incorpora- tion of this concept into the learning process may be useful in family financial management in general and in the use of consumer credit in particular. Socioeconomic status can also provide an additional tool to use in a "preventative approach" to assist families in avoiding overextended debt or other financial problems. The family needs information to get maximum benefit from the allocation and use of its credit resource and to understand factors that influence its family financial planning; its SES level can help place the family in the proper universe for behavior comparison. The significant interaction between FLC and SES on total other debt, coupled with the showing in the literature that the FLC concept is related to the management of money resources establishes a further relationship--that is, a family's stage of the life cycle affects the degree of debt incurred for total other debt. Therefore, FLC continues to 71 be a viable concept to be considered in a family's use of credit. The information gained from this study on the relative degree of debt for the various purposes for which debt is incurred can be used by teachers in connection with consumer credit. It can provide a basis for more accurately antici- pating financial problems as well as encouraging improved decision-making. For Research There are several possible areas of research from which to obtain additional empirical data on family debt patterns (see suggestions for further research below). Any new know- ledge gained would contribute to helping families prevent, ease, or alleviate financial or economic problems. Such information would also have important implications for family financial management in the allocation and use of economic resources-—short, intermediate, and long term. Suggestions for Further Research Additional analysis of data for other years would show if patterns are consistent over time or vary with changing economic conditions,for example, in times of recession or inflation. A replication of the study design and analysis could be made for another year of the four-year panel debt study. In addition, if comparable data can be obtained for years of substantially changing economic conditions, such 72 information could provide insights as to how families adjust their debt habits to external economic conditions; and what problems, if any, arise in relation to boom, recession, inflation, or a combination of recession and inflation. The longitudinal approach would also uncover any pattern changes due to debt dynamics. Another approach would be to pursue alternative ways of analyzing noninstallment debt per se as well as in relation to installment debt. One method of analyzing the noninstallment debt components would be a follow-up study of this thesis using the same study design but changing the statistical method of analysis. The Chi Square statistical test could use the frequencies observed as the dependent debt measure. Since the frequency of no debt is so high, using the no debt frequency as the measure would probably give a more meaningful and precise pattern. Any statistical significances from the Chi Square test could then be analyzed post hoc using an intercellular contrast with which to deter- mine where differences between the frequencies observed and frequencies expected were occurring. An additional follow-up study based on this thesis design could be a comparison of the degree of debt for out- standing versus annual installment debt components. The "rule of thumb" used in education and counseling is: above twenty percent of income for consumer credit (without mortgage) is the point of overextension. Is this annual or 73 outstanding debt? Is there a difference in the relative use of the components? Further analysis might be made of the concept of socioeconomic status in relation to consumer credit/debt not only as an effect in itself but also in relation to the family life cycle--especially with a full range of the family life cycle stages with both young families and retired stages fully represented in the research study. A possible alternative method to test the concept of socioeconomic status would be to treat it as a quantitative variable, at least as grouped intervals that are continuous. Study of the other debt category could be expanded. Not only has the aggregate amount of dollar debt increased, with larger percentages of families using debt, but it is also being used for more purposes. Therefore, a study of the other category broken down into more homogeneous cate- gories may be fruitful. A possible breakdown for other personal debt could be: 1. travel, recreation and hobbies, adult education expenses; 2. major changes in family structure: marriage, divorce, funeral expenses; 3. unexpected expenses from lawsuits or fines, unemployment, other personal crises; and, 4. all other debts which are private and discretionary. 74 An investigation of the growth of other debt in rela- tion to the expanded use of credit cards may be worth- while. APPENDICES APPENDIX A FREQUENCY DISTRIBUTION OF INDIVIDUAL DEBT VARIABLES 75 oz 0“. .unmo usonpaz Hamv.mnu ca mmaaaemm mo umnssc ms» ma .2 mo oaumu may m2 w .2 .HHmo may :a mmfiHfiEMM mo MODES: may ma 2 "muoz« m.mmmm¢smmm~2 m.«ammammmmm m.omflmmaama~ a.mmnvmafiom~ o.mm” «admom m.~m "Hmhm a::.stsoo . p p _ w h p w (p m n J m.mmmasmnavo ~.~anaoamama m.mmnmm ”~42 v.mm"~o ”82H m.mmu oogoma «.mv" Hmvm _:s: >— _ L— b _ _ — L _ L L m . ~.mfi mhanmom H.mmua¢ "mm m.ommmm “mm o.mm"hm "mo o.m¢" aqnmoa o.mm "a“HH “:2 p m _) >- p w P — p H )m t) n.mwmamaflmom ~.vmm¢o “on v.mm"mm Lon o.mm"m~"mv w.mmu mvnvm H.mm "mamm so; _ n _ H _ H b L F e _ h o\o —Z Z o\o ”Z Z o\o 12 Z o\o .Z Z o\o .Z Z o\o ,Z Z m_m>m.._ 2=2m .58 26m > >_ .= _. _ 28888 momma 225 23 >=Emu .o_oo.w «undo oz spas mwwHflEmm "coausnfluumflo wocmsqmum pawn ucwsaamumcH umo .~.< manna .Boamn muoc wmma . . . . 4 a . s L . . I . . . . n H 314$ .32 6 mm ” momummm o mm H 3mm: 2. mm" womnomm a «flammnomm mém "mm"; .28 c5200 0 . o h A o w _ o L P u w _ 4 H Hmuvmmuavw a may mmmmma o mm" mmfimva m 4” mmnmafi m.~mrn©H6mH m.mm ”Heads :9: c 4 _ . h by .11.). u . )._. . :le I) p C _ 4» _ . m mmnmnmmmom m cm" mmumo m 0mm omhm .mmuvm x31 ~.Hmuma "NOH m.am mm HHH “:2 o L 4 .Ju l (l w o p M-) _ p L h) H mmnmmmnmom mm " mm "on m mm n 34.: 9mm" 3 .3 ESL“: "SW Boodmmam 2,3 _ p H _ b h _ __ _ _ H H 0\o ,2 Z o\o ”Z Z 0.». ,Z Z o\o "Z 2 oo ,2 2 0\o PZ Z m_m>®4 952m .SOF 30m > >_ :_ __ . o_Eocoom 89% 996 8.4 3.8mm. -o.oom *unmo oz spas mwaaasmm "cowusnfluumflo wocmoqmum unmo usmEHHmumcH whammmm paw msoflufiood .H.¢ OHQMB 76 .2 on .2 «0 oflumu may ma .23 39:? :8 9L... 5 8:23 no 885a 9: 3 m .2 .Hawo on» cw mmflaflEMM mo Hogans ms» ma 2 "muoz« .fimmoa “33 Hémwmmwmmmm 4.8L 3&2 ISmeminomm mémnmmmmmom 223:3 .28 55.8 n L L L _ w L L L) w w L q 9211:. TE Nému Sdmfl 98m 34% N ST: ms: m SLLZJOL: o mighty :9: 4 L L L. L 223312; v.3" am “mm 9mm" mm “mm 4.3": "S 93"? "NS SNELL L: 22 L L H m h _ L — P b L L) m.mhLmMN “mom N.mmuwlhm 1.05 $.55.“ mm «on 0.0m" mm "mv b.mmnwm “vm m.om "ONnNN 26.. L. n L n L n L L L P L L 0\o F 2 2 0\o p Z 2 0\o L 2 2 0\o F 2 2 0\0 F2 2 0\o L 2 Z m_Q>0J mama .80... 26c > >_ L: __ L 2695an $0me 996 23 >=Emu .oLoom «upon 02 apex mmHHHEmm «cowusnflnumflo moswswmum unmo acmeHHmumcH umnuo .v.< manna .BOHwQ wuos wwm« . . a . a . . L 4 . . a . L . . ._ m whuhmmummma o mmhmemmmmm h annmmauwnm H whumhauomm m mm“ amfloom m.mmuom1m E::.Et:60 a b L o L o w L o L L. .1 L t a w wnfleomnavo m LmLNNHLmML a mammaaumsa N omuma “SSH m.mm LQNHEmL m.mm”mmao £9: . . u . L t 1.): EL} ....” )7 --ilL, L I L L!) IrlrJJLIIi . q manmm~_mom m mmnam ”we L.OLLLV _mm o.mw"mv “as m.mm “on woa m.om_oaaL L22 w L 1 1. .4 4 _r , L -. L L L .6 L_ L. m.m£ Lawnmom m.mmnoo “on m mmnmv Lea o.manvm “m4 o.mm “ma Fm m.MMLLHaN 205 L L L L L h L L L L H h 0\0 —Z 2 Q0 ’2 Z Oxo L2 2 o.\o L2 2 06 L2 2 00 L2 2 m.®>®4 mBmL .28. 30m > >_ :. LL _ ULEocoomm momma $96 83 >__Em.L -oLoom apnea oz nuflz mmHHHEmm "coflusnfluumflo hocwskum undo usmEHHmumcH mmanmuso .m Jam OHQMB 77 .366 ”393:. :89 mfi 5. $32me we “when: 93 3 .Hamo man ca mmflawfimw mo Hones: on» ma 2 .z o» .2 Lo caumu wnu ma w z . "muoz« L.Lamvmv mamNL m.mmmmmammmm «.vmmma MoLm m.o~m Hmmomm L.-w mammwm L.mmLLmLLa .mszzcsoo . b L _ o L o L w o (L L o L « Nommmmm “ave m.mmnmh mama mom“ mm ”N: 0 mm" mmuwaa N hm" mvnoma m mm “mmmvw :91 . _ L L _ _ L L L L L w.LN"mm mmam m.m¢LLm "mm m.m~mmL mam m.v~" hanmm o.LL" manmoa ¢.omL¢ NHL .22 P p L( w L P L L — L w m.mmmngunmom «.mmmmv Lon m.wmum~ LLL m.mmL manmv o.mL" mauvm m.vaOLL~N .2: L n L h L n L L L t L L 0\o L 2 2 0\o L 2 Z O\o L 2 Z O\o L. 2 Z 0\0 L 2 2 0\o L 2 Z w.®>01_ 932w LEOL 26m > >. L: L. L OLEocoom momme 295 2: >283 .oLQom «unmo oz snag mwflHHEmm "coflusnfluumflo wocmskum undo usmEHHmumcH Hmuoe .m.¢ OHQMB .30Hmn wuo: wmm« . . l . L 4 . L (4 . . L . _ L . )4 L m «Ammammmmma a Lannmmmmm o mmnmommmnm m Ham oaflomm o vauvvmmmm m amnmmxa a::.staao o.mmnmooLLvo L.LmhmmawamL m.mmLomahmaL v.HmL wonmLL v.4mLOLHBmL m.om"~mwo £2: 0 L LII):L..)IL (I..-) _ I) a:_ . 4 -.-)(iP I'Li) 11);) ((#5)) . 1L- _ . m.mmhmmmmwom m.mmmao ”mo .m.em mm "mm Lm.Hm" mommo «.mmuvm WOL m.om_0L:L oeL L1 0 l; 4. ALT .r L L L :lklallllrl:))lL ,‘LILL , L.mmnommnmom a.maLmL LSL v.LvaL ASL L.La" Henmv N.mmnom Wm m.mmno~a~ sou L L L L L L L L L L P. L Ox»... L2 z 0\0 L2 2 M 9.0 L2 2 Q0 L2 2 06 L2 2 0w L2 2 m_®>®1_ mDLmL _mLO._. 30m > >_ :_ : _ OLEOCOOQW mommLm 930 23 >26? -oLoom Lunwo oz cuflz mwflaflfimm “:ofluqnfluumflo mocwswmum unmo unmEHHmumsH Hmucmo paw Hmofipmz . m.¢ OHQMB 78 .unmp usonqu HHmo may :H mmHHHsmL Lo amass: may mH .2 on .2 Lo oHumu may mH w .2 .HHmo may CH mmHHHEmw mo Hogans wcu mH z "muoz* .mevHNHmmmmH N.»mwmnmwmmm H.memo~wo»~ H.mmnH-mom~ H.mmnvmmmoom o.mmnmmmnm _SOLCE=60 . L L _ L. L L m a L I H To? omen HE 4.831.332 $21312; 91.2816: $21312: b.8146 “3 as: m _ L M _ _ L L L L IL . .Ha“ oomLmom H.Hmumm "mm m.mmLHm “mm o.mmumm “as H.Hm"mm "NOH m.om_0HLHH .22 )1? 4w I4_ I».- H L P — b L L I o.»mn ¢m~_mom ».mmumb .mn H.mmLmH LS» m.nmuqe "me «.mmnom Lam o.o0Hmmnmm 303 L h L H L h L IL L L LII L o\o .2 Z o\o L 2 Z Ax. L 2 Z o\o L 2 Z 9.... L 2 Z pxo L 2 Z w _ m>m1_ mBmLm Lm::.30m > >_ _= : L oztocoom mmomLm 905 m3 LLLEmuL .o_oom «unmo oz :qu mmHHHEmm "COHHDQHHumHQ wocmsqmum pawn usmEHHmumchoz Hmu .m.< mHnt .3OHQQ QHOC GQW¥ m. mamm~H_~mmH 8. mm _Nm~_mmmLm. ma.mHN_SHN m.mmm mmmomm H.mm mammaom o.mm wwmrm azzLetsoo L L m w H L _ m. mmLCmm .va m. mm mmH HHH a. La mmL_NHL 4 Ha" MHHLHL m.Hm LmLHamH H.00Huvmvm as: v II .-.-I M . _ . III- ) ILL- I.» -L L L J ,L I L IL I I . II L I H) -) 4. ma. mom_mcm o. OOHmo .mo u. OOHmm am 0. mm. monmo Lp.mm "HoflNCHLiOOHLHHHH 22 IIoIIILrI) .9 III. . IoIII I) _ L IIIIH _ I L IIILIII III I IL. I L) I. I.» L H manmmNLMOH o oeHmH L L L» mm_ .mL Lea .0. con mv_mv m.La “m Hm m.mm ammmmwlzou L L L L L L L L L h L I. .z z p. .z 2 L o. .z z a. .z z .0 .z z o. .z 2 38:; mag :20» Box > >_ L: __ _ oLEocoomm mmomLm $96 33 LLLEmL -oCom Hunmm oz cqu mmHHHEmm .COHuanuumHo mocwskum undo ucmEHHmumchoz mHHmmmm paw mcoHqup< .h.¢ OHQMB 79 .2 on .unmo usonuLg LLmo mnu :L mmLLLEmm Lo amass: may mL .HHmo may cH mmHHHEww mo HmnEd: may mH .2 mo oLumu mnu mL m z . z "wuoz« L.ommeLLLNmNL m.meLoNmeN m. mmL LqN_oLN o. omL LON. .OMN m.mmLmNmmem o. Lm me_Lm ax: 5:200 » L L _ w w L.mmLLLm +Lvm m.mmLomLmeL o. mmLmNLLN¢L m. mmme LmLL q.mmLLoLLomL m. hm Lovam gmLL . _ L L L L I L L m.LmmeN Lmom m.mame me v.Lame me L.m¢Lom Lao m.Lmme LNOL a. omLOL_LL .32 I w) ILI I“ F L L — L.NmLLwN Lmom L.meNL LmL m. om.L mo LoL m.mmLNv Lmv o.meaL Lqm q. mmLmLLNN 304 L L L L L L L L L L L L oxo L2 2 o\o L2 2 L 0\o L2 2 o\o L2 2 o\o L2 2 o\o L2 2 m_®>m.L 9:9m :30» 30m > >_ L: __ _ 2:588 $95 986 23 2::an -oLoom «pawn oz SUHZ mmHHHEmm ":oHuanuumHQ mocmsvmum pawn pameLLmumchoz uwzuo .OL.¢ anms .Bonn muo: mmma m mmmmNLLNmNLL9LzfimmmN_mmNLm. mm .mLN LwLNLm. mm LmNN _LomN m.nm Lmewmm c.00LLLme a::.stnao o 4 L L L o w w r L v m@ hmw_Hvo o. 00H. mmH mmHo. OOH NvHLNVH m mm. vHH LoHH m mm LthBmH 0.00Hva&m :91 LMWIMMWII.QMWIIII .--I-.. rI-II.».II.L . . . IL -L .I-.L I.. -II..L IIII» .: I L LIIII. 0 LOL 0. o0L_ mo .mo Lm mm.Lm _mm m. Imm we Loo LL.vm Lem NOL o.o0LLLLLL .32 . .. .. I. I. II III L II -L I -I- . . III)- .L II-. ,. .-. .LII. L ,I I IL-IILIJ. m mJ LomLmom o ooLLoL LoL a. 00L 0L LLL .m.LmLLv Lmv Lm.mmL mmLLm Lo.OOL_NNWN 204 L L . L L . L L L L L L o. o .L o .0 L2 2 \0 L2 2 L 0 L2 2 90 L2 2 00 L2 Z 06 L2 2 mL®>®J so 0 , 3.5m. L L. 2, m > 2 LL. __ L ULEocoom mmmmLm mLo>O 33 37cm; -OLUOm 9 unmo oz SUHB mmVHHHEL..nL unmo ucmEHHmumCHCOZ moHnmusa "coLusnLuumLa mocmsvmum . m ..LH wHQmLH. .2 cu pawn usocuflz HHmo may cH mmHHHEmm mo Hogans map mH .2 we oHumu an» mH w z . 80 .HHmo msu :H mmHHHEmM mo Hones: ms» mH z "muoz« m. HmoNOH.~m~H N.H¢4mmmgmmm m.om4m-4mhm m.mhmthmom~ h.mhummmummm m. mm 4Hm.hm E::.Et360 _ _ 0 _ _ L‘ w + m. om mewam «.mmuvammMH «.mhuHHHmmvH m.mb4mm 4mHH m.hhuovH"omH v. vm mvmuvm .tr: _ L L . L _ L L L. H m.m& www.mom m.~m"mm "mm m.qmumv "mm H.vm"mm "am m.nnumh "NCH m.Hw"m "Ha .32 b F .p . H _ L L L L L f m.mm «mmflmom v.mm4Hb “mu m.~mnmo “on m.m54mm nmv H.Nm4mo 4cm m.Hm4mH4~m 304 L. 4 4 .L L 4 L L P L r L £0 ,2 Z o\o '2 Z o\o ,2 Z o\o .Z 2 Q0 F2 2 o\o L2 2 m_m>o.._ 82% .98 26m > >_ =_ = _ o_Eocoom $35 226 2... >=Emu .o_oom «pawn oz :uH3 mmHHHEmm "COHuanuumHo mocmswmum unmo ucmEHHmumchoz Hmuoa .NH.< mHQme .30Hmn muoa mmm* m. 333.32 3; 403 ”SN m. mm. .momi rim mémm ofiomm ~.mm 4va 4031.3 #93 .28 25.8 L _ L L w w v. we moo Hvo o.mm “ananH h. mm MMH WNVH o.vm4 mongH n.Hm “meBmH m.mm 4Hm&m :9: H u w w L -LII_ 4-|+-;¢,; ¢;:+’J .;+ le m.mm_mmm4mom m;mm _hw ”we wo.wm_wm .mmiLm.Hm4 mwumw H. mm .mm NCH 0. OOH HHHH .22 IL L4 ‘ o.-!|F|l||| -i‘ rill“-..il.l|rli L .-- L F .ltlllrrrll I, IILII. o.mm4mmmnmom jHUQOH Joni no 4: no“ u 4mm v vmu mm4mv m.mm 4mm Fm o.QOH4-NNWN 304 L L L. L L L L L L L L L_ L .3 ,z z a» 2 z 4 ca ,2 z 0a .z 2 ca 72 z 90 ,z 2 29,3 So 30 £22m _ L m > >_ :_ __ _ o_Eocoom mmmmum 99,0 83 >:Em.¢ -oBom «pawn oz Sufi: mmHHHEmm ":oHuanuumHo mocmswmum unmo ucmEHHmumchoz Hmucmo flaw HMUHUwz .HH.< wHQMB 81 .2 on .2 mo oflumu may ma m . .unmn usonufl3 Hamo wnu cL mmLHLemm mo Logan: man mH .z .HHmo may :H mmHHHEmw mo Hanan: mnu mH z "muOZu . L 4 4 J 4 H mON4mmN H.mm4 mmfith m.Hm4mHH4OMN m.om4 mmflwwm m. Hm. “romim LEO... :E:_oo . L L L L w T H n.5m4mmh 4 E 93 43 F3 9543 4m: 93:3 ES mam" 342: «13ng no.1 L _ _ NmNHF.~L H w.wm4vmm -—-—--—p———--4 L _ L L L L L _ 3 4mm «.34 mm 43 «.343 "mm 31 $42: méflo 4: 2.2 _ L L L L . P L L L w 843 9343 4.: mamzm 43 0.34 ~13. fimmnmjmm 26.. N. vm 4pm..” 4mom me v.mm 4NmH 4mom m.Nm L L L L L L L L L L L o\o L2 2 o\o L 2 Z o\o L 2 Z o\o L 2 Z o\o L2 2 o\o L 2 Z w _ m>04 2.65 .20» 30m > >_ =_ = _ OLEocoom .080m 385 296 83 2.8an «name 02 nqu mmHHHEmm "coHuanuumHa wocmsvmum pawn umo Hmuoe .vH.< mHnme .30Hmn muo: mmm« o. om4LNHH4 NmNH m. mm. .va mmmN o. 2? SN 4whN v. hm. 4HON .omm v.om 4 Hmmmomm mimm .HHm 45m .mLOL c2560 L. w L _ m. mm4vnm4 Lew 0. mm .mmH 4LML m. mm NNH4~¢L m. ~m_mm _wHL 0.0m“ menomH m.mm onLL 50;. . 4 Lil. -... 4|- 4, 1!: . .-!__‘\‘ 4 . : L 1 L. .. IIL.IIL;I:;71IALLIJ_IJ . 9. mm. mnN4mom m om4mm 4mm .N. mmém 4mm .m.Hm4mm 4mm N.om 4 Na 4NOH w.Hm a .HH U.S. Ll . 4! -.lni +1.4 .11: :L-IIIL-I;L..-iiiT - :L L.!‘|Lr-l:ia L. 4 .7 w. Nm4 omN4mom H mm4mh 43. .m. -mmmmo 4.0.5- 4m.mm4Nv .mc m.om 4 on Gm m.mm .HN .NN 30 L L L L L L 4 , . 4 L 4 4 L. 4 e. .z z ... .z z 4 .... ,2 z 0.. .z z 0.. .z z ... .z z 233 mam ESP 301 > >_ 4: : _ OLEocoLomew mmomLm $30 2.1. BELLE -oBom «name 02 nqu mmHHHEmm "coHuanuumHQ xocwswmum unmo muHmmmm can mcoHuHUU< Hmuoe .MH.< anme 82 . .z o» .2 Lo oLumu mg» mL L . .unmo uaoguL3 LLou on» cL mmLLLamu mo “magma may mL .2 .Hawo us» ca mmLHHEMM mo Hwnfidc may ma 2 umuoz.‘ mNNnmNN N.$4LNLBLN YNXLEONN mém4SN4$m N.NL42.43 2:25:28 _ p L L b L1 [w SLFS @3212; Wow"? 43L LéflNoioL: 9343.43 :9: L _ L L L L L w 4 Nm 43 934mm 43 L134 mm 43 H.342. 4N3 WNLBJ“: Es. 93426 43392 L maimNL‘ 4L3 fimL. » _ 134218». m.£. L —--y———L—L-L{ m L¢ 4* m L _ L L P L w b QNEVNN 48m N.L.m 43. 4: LIN: mflfi 92.4 mm 43 9343 43 984342 33 L L L L L L L L L L L L .\. .z z .\. .z 2 q. .z z .\. .z z .\. .z z .\. .z z 223 egsm 30* 30m > >_ L: = L oLEocoom .oBom 803m 205 2.... 2.8mm «yawn 02 Spas mmeHEmm "cofiusnfluumwa hoqmskum Lama umnuo Nance .mN.¢ magma .30Hmn muo: mmmLL Liv“ .LL .... ..: M N¢ 4NmNH mm 4mvm4mmN L. HhmmmH4WNN m wthbH4OMN N. no 4 mmflomm m mm 4 om4ha LSOL. c5300 3: 4Com 4:... WE ..NNLFNL WNENSNZLVNES E: m.$41NNN4.omL N.NN 434$ :91 . w 4 4 _r i 4 4 ‘llLisrL 1147!, tILi‘LiliLLfyian 144'; . m K _LNN 42: Law Rm .3 RAF? .3, H.345 4S 5:: 3 4N3 mg; .33 22 o W fi ill HI M . wLI a P L L L L H h . m 3 EN 48m m mm 43 42. N $43 43 0.2.3. 6L. v.34 3 .5 WE LNLNN so _ _ _ L L L L L L L L L L 0\0 L2 2 oxw L2 2 o\o L2 2 o\o L2 2 9.0 L2 2 80 L2 2 m_®>m.4 mam .mLOP 30m > >_ =_ : . OLEocoLomLLw $96 225 .2... >:EmL .oaom «pawn oz cuflz mmflaflsmm "coflusnauumflo mocmsvmum unmo mwNamuso Lagos .mN.< magma .z o» .2 mo oflumu mnu ma w .uan usozuflz Hamo may aw mmflHHEmm mo “$285: mnu.mfl .z .Hawo wnu :H mmHHflEMM mo HmnEsc ms» ma z “muoz« d 4 4 4 4 d u d 4 H «.mflmnm “33 m.om _ 2&2: H5113 "m2 m4...“ 3. “on.“ ma: Snomm H.233; .98 5:28 . . H _ _ a w . L4 w V p a Y >.onmnmanaqo v.om“ chmmma o.m~" ~¢.~¢H o.Hm mmuoaa H.Hm" mmsmfl v.vaNNWm 52: M _ _ H u " F J— % L #-ll-!l.~i|l1.|+ “.mwnon "mom m.~¢" mmumm H.v~_ ¢H_mm q.nn «Humo m.oau HHWCH v.mmflv ma .22 b w # H h _ p b _ w M w ,o.mmuwoafimom m.nm" evflmh m.mm“ FNJon n.mfl «Humv m.mau mawm m.mq”oawm 204 _ h p H _ n p L _ _ _ b «a .z z «a ,z z «a ,z z «a .2 z «a _z 2 so ,2 z 296.. mama Each 301 > >_ =- = _ a o_Eocoom 885 230 2] >228“. .ofiom «name oz nuwz mmfiHflEmm "cofiusnwuumwn hocmsvmum “58 25.09 61¢ 383. 83 .30Hmn muo: mmmn . 4 fi . ,. i a . 4, 4 q mmhemfiummm; mm “Emummm. h $.mmmmilo.~mnomiomm v.3 » , 4 p v P w b l. 93 Em 43. 5mm 52 m3 «..me Rim: 55.2mm be: 5.3 l+ (lAl‘l L _ ._ ,, _ ._ H magnum gem A.3 .3 3 A.3 3 ,mm 9.6.me 3 ~43 om NS mg; 3.: 92 <>——-—< ommuoom um.Hm «mm Km .98 c5200 > p r F 32:: ,.m.~m mm 3 :9: 1‘ 44I4l> wiomhmmm “2: 7mm .3 .5 .539 mm 3 waivm a; 3.3.. E «m .98 .omfmmLEj » p _ » > p 4 _ L. J” p , ¢ fL 00 ,z z 00 ,z z 00 ,z z , no .2 z W at .z z . co ,2 z ~;m>m4 L .r y. ,r 3:30 ESP 30E > >_ :_ : _ 9:31.615 mwGfium 37:79. 3...; 3:59.; -ZCOM- «ammo oz :uflz mmflaflsmm "coflusnfluumfla xvcmswwum pawn Hmucwo 0cm HMOHUwZ Hmuoe .ha.¢ mHQME APPENDIX B MEAN RATIOS OF INDIVIDUAL VARIABLES NOT SIGNIFICANT Table 8.1. dditions and Repairs: to FLC and SES 84 Installment Debt, Mean Ratios of Debt (Cell Means and Frequencies)* Socioeconomic Status Levels Low Mid l;1,x1h Jnlumn Tatal *Numbers Tai>le El.2. Family Life Cycle Stages I II III IV V 17.3 3.0 13.2 11.6 -- (7) (3) (8) (3) 1.4 27.7 12.9 14.0 8.0 (2) (9) (5) (8) (9) 13.8 7.5 8.3 8.2 7.9 (l) (13) (18) (17) (6) 8 8 16.2 8.6 10.8 8.6 (u, (3) (29) (26) (33) (18) Row Total % (N) 12.9 (21) 15.2 (33) 8.4 (57) 11.2 (111) in parentheses are cell frequencies; Car Installment Debt: (Cell Means and Frequencies)* Mean Ratios of Debt to FLC and SES ratios are in percent. Socioeconomic Status Levels LOW 11 i (:11 Column Total *Numbers in parentheses Family Life Cycle Stages I II III IV V 18.7 6.2 14.0 20.7 15.8 (9) (39) (20) (24) (12) :2.) 14.2 14.8 16.0 13.6 (4; (53) (32) (25) (21) 16 2 12.3 13.5 13.2 19.6 (3“ (80) (54) (59) (38) % 17.2 13.8 14.0 15.5 17.2 ) (172) (106) (108) (71) (N) (46 Row Total % (N) 17.0 (104) 14.8 (135) 14.3 (264) 15.0 (503) are cell frequencies; ratios are in percent. (35 *Numbers in parenthrscs are cell frequencies; Table 3.3. Other installment Debt: Mean Ratios of Debt to FLC and SES (Coll Means and Frequencies)‘ Sucineconomic Family Life Cycle Stages Row Total Statas Levels I II III IV V % (N) “0w 25.4 2.2 9.8 29.1 9.5 16.3 (2) (28) (9) (17) (9) (65) '11 {.1 18.3 7.0 9.8 20.3 13.9 ‘" .n (47) (28) (21) (14) (113) ”1(h 14.7 10.9 8.9 8.2 15.5 11.0 1 (16) (as) (38) (27) (22) (169) . 14.8 13.6 8.3 14.2 15.8 12.9 r l (N) V?l) (141) (75) (65) (45) (347) *humbtxi 1D arintheses are cell frequencies; ratios are in percent. Table 8.4. ficiical and Dental Installment Debt: Mean Ratios of Debt to FLC and SES (Cell Means and Frequencies)* Sc _.gconomic Family Life Cycle Stages Row Total [WW 3.7 4.8 2.9 3.9 1.2 3.6 (2) (4) (4) (2) (1) (13) ! _ , “ 133.3 8.4 7.4 8.8 1.1 12.8 i" 1) (8) (6) (3) (1) (19) ,) 3.6 4.3 4.0 2.5 3." 3.8 ‘ ‘” , (2) (10) (10) (6) (4) (32) L Column 7 23.7 6. 4.8 4.5 2.9 6.4 Total (in ( ) (22) (20) (11) (6) (64) ratios are in percent. gurioecowcmic Status LOVDIS 236 Total Additions and Repairs Debt: Mean Ratios of Debt to the and 568 (Cell Means and Frequencies)* Family Life Cycle Stages Row Total I IT III IV V % (N) 5.3 26.6 3.0 13.2 11.6 13.9 (I) (S) (3) (8) (3) (23) 1.4 25.2 11.5 14.0 8.0 14.5 (2) (10; (6) (8) (9) (35) 13.8 6.7 11.5 7.6 6.9 8.7 (3) (17) (20) (20) (7) (67) 8.2 15.2 10.6 10.3 8.2 11.3 ("‘ (6) (35) (29) (36) (19) (125) II Ir nth-5?s are cell frequencies; ratios are in percent. Tital Durables Debt: Mean Ratios of (Cell Means and Frequencies)* ST)? 1‘3t"‘-t".(r“"‘)7 ..‘ Status 10V” Columi Tetal "T‘§‘.1rTI'lI-‘7 {”0 ‘1) Debt to FLC and SES Family Life Cycle Stages is I II III IV v 3.1 5.9 4.6 4.2 9.7 (5) (39) (11) (31) (10) .~ 8.3 5.4 4.9 4.- 13 (lb) (22) (17) (ll) 5.9 4.0 6. 2.3 4.0 (11) (38) (2 ) (30) (17) . 5.4 5.7 5.7 3.6 5.6 (N) (17) (133) (58) (78) (38) in parentheses Row Total % (N) 5.6 (96) 6.3 (87) 4.2 (141) 5.2 (324) are cell frequencies; ratios are in percent. 87 Table 8.7. Tara] Medical and Dental Debt: Mean Ratios of Debt to FLC and SES (Cell Means and Frequencies)* Socioeconnmic Family Life Cycle Stages Row Total “M“ I15 I II III Iv v 3. (N) \w 3.7 37.3 20.5 3.4 1.2 22.2 " (2) (in) (11) (4) (1) (28) "4 103.8 6.1 13.9 7.2 2.9 12.2 " (l) (12) (12) (4) (2) (31) H - 1.8 3.1 12.7 10.1 3.8 7.0 .11.] “3) (:4) (17) (15) (6) (67) Fglunn o 13.0 11.3 15.2 8.4 3.3 11.7 .* al (N) (8‘ (46) (40) (23) (9) (126) ‘1. {—4 5.4 ’ I Ev "‘ers in parentheses are cell frequencies; ratios are in percent. REFERENCES REFERENCES Bigelow, Howard F. "Toward a Theory of Family Finance." Journal of Home Economics, Vol. 32, April 1931, pp. 325—332. Bonde, Ruth L. "Means and Ends." Consumer Credit in Family Financial Management. Proceedings of—a Na- tional Workshop at University of Wisconsin, Oct. 9-12, 1967, Washington, D.C., American Home Economics Association, 1967, pp. 144-152. Board of Governors. Federal Reserve System. Federal Reserve Bulletin, Vol. 61, No. 5, A 45 (May 1975) and Vol. 58, No. 10, A 56 (October 1972), Washington, D.C., United States Government Printing Office. Brown, Kathleen H. "Social Class as an Independent Variable in Family Economics Research." Journal of Consumer Affairs, Winter 1969, pp. 127-136. Duvall, Evelyn Millis. Family Development. Philadelphia: J. B. Lippincott Co., 1967. Gayle, Mary A. "Relationship of Family Life Cycle Defini- tions to Family Financial Activities." Unpublished Ph.D. dissertation, Purdue University, 1966. Gross, Irma H. and E. W. Crandall. Management for Modern Families. New York: Appleton-Century-Crofts, 1963. Hendricks, Gary, and Kenwood C. Youmans, with Janet Keller. Consumer Durables and Installment Debt: A Study of American Households. Ann Arbor,Michigan,Survey Research Center, Institute for Social Research, University of Michigan, 1973. Herrmann, Robert 0. "Families in Bankruptcy--A Survey of Recent Studies." Journal of Marriage and the Family, Vol. 28, 1966, pp. 324-330. Huber, Milton J. "A Profile of the Overextended Credit Family." Consumer Finance News, Vol. 49, May 1965, pp. 6, 7, 18-20, 22, 24-25. 88 89 "A Study of Financially Overextended Families." Detroit, Michigan, Merrill-Palmer Institute, 1965. Katona, George; William Dunkelberg; Jay Schmiedeskamp; and Frank Stafford. 1968 Survey_of Consumer Finances. Ann Arbor, Institute for Social Research, 1969. Kahl, Joseph A. and James A. Davis. "A Comparison of Indexes of Socio-Economic Status." American Sociolo- gical Review, Vol. 20, 1955, pp. 317-325. Lansing, John B. and Leslie Kish. "Family Life Cycle as an Independent Variable." American Sociological Review, Vol. 22, October 1957, pp. 512-519. Lansing, John B. and James N. Morgan. "Consumer Finances Over the Life Cycle." Consumer Behavior, Vol. II, Lin- coln Clark Ed) New York, New York. UnIversity Press, 1955, pp. 36-53. Lawson, Edwin D. and Walter E. Boek. "Correlations of Indexes of Families' Socio-Economic Status." Social Forces, Vol. 39, December 1960, pp. 149-152. Mandell, Lewis. Credit Card Use in the United States. Ann Arbor, MichIgan: Institute for Social Research, University of Michigan, 1972. Martineau, Pierre. "Social Classes and Spending Behavior." Journal of Marketing, Vol. 23, October 1958, pp. 121— 130. Mathews, H. Lee and John W. Slocum, Jr. "Social Class and Commercial Bank Credit Card Usage." Journal of Marketing, Vol. 33, January 1969, pp. 71-78. Moore, Geoffrey H. and Philip A. Klein. The Quality of Consumer Installment Debt (Studies in Consumer Install- ment Financing, No. 13). New York, National Bureau of Economic Research, 1967. Morse, Richard L. D. "Historical Development of Consumer Credit.” Consumer Credit in Family Financial Management. Proceedings of a National Workshop at University of Wisconsin, Oct. 9-12, 1967, Washington, D.C., American Home Economics Association, 1967, pp. 20-26. Nie, Norman H.; C. Hadlai Hill; Jean G. Jenkins; Karin Steinbrenner; Dale H. Bent. SPSS: Statistical Package for the Social Sciences. New York: McGraw Hill, Inc., 1975, (2nd ed.). 90 Reiss, Albert J. Jr.; Otis Dudley Duncan; Paul K. Hatt; and Cecil C. North. Occupation and Social Status. New York: The Free Press, 1962. Ryan, Mary Ellen. "An Analysis of Economic and Demographic Characteristics of Consumers Associated with Excessive Installment Debt." Master's Thesis, University of Minnesota, 1968, (Published by Federal Reserve Bank of Minneapolis). Scheffler, Joan; Richard Rice; and Judith Kaplan (eds.). Occupation Reference File Workbook. Ann Arbor, Survey Research Center, Institute for Social Research, University of Michigan, 1971. Schlater, Jean (dir.). National Goals and Guidelines for Research in Home Economics. Study sponsoredIBy Association of Administrators of Home Economics, East Lansing, Michigan, October, 1970. Shaffer, Carol Warwick, "Income and Expenditure Patterns Related to the Life Cycle." Unpublished Ph.D. disserw tation, Michigan State University, 1964. Sheldon, Eleanor B. (ed.). Family Economic Behavior. Philadelphia: J. B. Lippincott Co.,SI973} Slocum, John W. and H. Lee Mathews. "Social Class and Income as Indicators of Consumer Credit Behavior." Journal of Marketing, Vol. 34, April, 1970, pp. 69-74. Wasson, Chester R. "Is it Time to Quit Thinking of Income Classes." Consumer Behavior in Theory and in Action. Stewart H. Britt (ed.). New York: John Wiley and Sons, Inc., 1970, pp. 234-238. Fist. of Fan-33y {String-y )-.'....£_Lge;1 Shiva Carversity East Lansing, Michigan "111111111 1111 1111115