THE RELATIONSHIP OF OBJECTIVE AND SUBJECTIVE FAMILY INCOME ADEQUACY T0 SELECTED MEASURES OF PERCEIVED LIFE QUALITY Dissertation for the Degree of Ph. D. , MICHIGAN STATE UNIVERSITY :. NORLEEN MARION ACKERMAN ' 197T I ‘d- l v-HFCIS I “ .I. I ‘ cfla‘xk‘h-d - . o th' E}: :" -'. .2 ‘2 Q '- '..- lb . II" -.‘,.1“ - {A I. ._- I”: T» I n' - ' - 3 ‘h I . t V I‘ A. 'l iii A '-.',‘ ”at, E . 3 I {1' .5 / 7%; ué “fit/7.9:) ’42}: <§ 0’ LIIr T ; }/ MICIIIJJ _ data”: University U5.- 6: "I M.“ '.- r1" 1"" - L .- ..r~.,. .--_’___ . ’ . ’ [v.7 . .Iu'!‘ ‘ "i A. ‘1‘;: :g-Ifi}, PLACE If! HEI’U TO AVOID I 'ATED,‘ DATE ' Wynnj MSU I.N'|AM""‘“‘" ‘ ' 4' \ THE PF'ALT FAMI'T )Wi\f The: "anal tc seLv \ \ research _‘. §bitween 1(6'”? .rnI ‘ I.u-,z' “'LJ‘ h _ l Hf 1w vflc 11-” ‘ ,_JIA- ”IT . - Ar“ .1 L a4” :I I : I "VAL who d-€iercfi 'L 7 1 "m adeguufiy, lot 1 mt guhgectivc oéogn;:v “y ixfch wnt» stbcut to cantextunl ilhacctive finally 'rnome adeqxacy v.0 In ilifli‘P 5y the leOpc-donc~head§ vi family antes. “.g '. I .~ .n ‘- I,‘.’; .‘n ,IfiIquacv was In chome~needs ratio coulntoa E ABSTRACT THE RELATIONSHIP OF OBJECTIVE AND SUBJECTIVE FAMILY INCOME ADEQUACY TO SELECTED MEASURES OF PERCEIVED LIFE QUALITY BY Norleen Marion Ackerman The study was designed to relate family income adequacy to selected measures of life quality. The three research questions were (1) What was the relation- ship between income adequacy and selected life quality measures? (2) What was the relationship between the congruency of income adequacy measures and selected life quality variables? and (3) Did groups who differed in level of objective adequacy, level of subjective adequacy, and in congruency differ with respect to contextual variables? Subjective family income adequacy was an assess- ment made by the respondent-heads of family units. Objective adequacy was an income-needs ratio computed by dividing family income by a consumption standard which was adjusted for family composition and geographic area of residence. The consumption standard, and the equivalence scales used to adjust it, were developed by the Bureau Norleen Marion Ackerman of Labor Statistics and were based on family income needs for a moderate level of living. Congruency was a measure of the degree of agreement and direction of disagreement between objective and subjective family income adequacy. The life quality measures were satis- faction with perceived family income, level of consump- tion, and overall life quality. The data used were collected in the fall of 1974 by the Survey Research Center of the University of Michigan. The data are identified as the Economic Incentives, Values and Well-Being Project, Part IV. The 1,046 interviewed respondents were husband or wife or one adult heads of family units or individuals living alone. They were drawn from a multistage area probability sample of private households in the contiguous United States. The results were generalizable to the family types and geographic area specified. The major statistical procedures implemented to test the hypothesized relationships were two-factor analysis of variance, one-way analysis of variance, and multiple classification analysis. Parametric and non- parametric statistics were used with contextual variables to describe groups of respondents which differed in levels of income adequacy and congruency. Results of the analysis indicated that satis- faction with family income, level of consumption, and Norleen Marion Ackerman overall life quality increased as family income increased. More of the variance in satisfaction was explained by subjective adequacy levels than by objective adequacy levels. Subjective and objective adequacy together, how— ever, explained more of the variance than did either measure individually. Results indicated that family income adequacy, objectively and subjectively measured, was more strongly related to those satisfaction variables most directly involving money resources: family income and level of consumption. Family income adequacy was less strongly associated with overall life quality, a more global measure where the effect of income adequacy was more diffuse. For each of the three life quality variables, satisfaction scores were similar for the congruency groups whose subjective adequacy level was as high as, or higher than, their objective adequacy level. However, for those congruency groups whose subjective adequacy level was lower than their objective adequacy, satis— faction was significantly lower on all three life quality measures. While congruency groups did differ in satis- faction, the group differences explained very little of the variation in satisfaction with family income, level of consumption, and overall life quality. Norleen Marion Ackerman Variation in contextual measures as family income adequacy increased were, in most instances, similar for objective and subjective adequacy. As income adequacy increased, the number of children and the age of the oldest child decreased, indicating reduced demands on income. As income adequacy increased, family money income increased, as did those characteristics associated in previous research with higher incomes: being male, being white, having more education, and, for male respon— dents, having a white collar occupation. Objective and subjective adequacy groups did differ, however, in the occupational trends of female respondents. As objective adequacy increased, fewer females were housewives; as subjective adequacy increased, more females were house- wives. The higher subjective adequacy where the respon- dent was a housewife may be due to the economic value of additional household production by the wife, or might be due to the value placed on leisure time. Neither of these were included in computing the objective adequacy ratio used here. The congruency group with the lowest satisfaction scores had assessed their subjective adequacy as being at least two levels lower than was their objective adequacy level. They might be termed better off than they think they are. The group had the youngest respondent-heads and the youngest age of oldest child, indicating younger Norleen Marion Ackerman families. They had the second highest income and edu- cation of the five congruency groups. They scored highest on the material wishes index, were experiencing deteriorating financial situations, and had not achieved their consumption goals. These characteristics, and the general economic situation of families in the fall of 1974, indicate that the theory of relative deprivation might explain their dissatisfaction. The conclusions reached on the basis of this research are that (l) the objective adequacy ratio developed and used here is a useful method for standard- izing income adequacy by family size, (2) income adequacy needs to be measured both objectively and subjectively in order to explain variation in perceived life quality measures, and (3) incongruency between objective and sub- jective adequacy can identify those among the higher levels of objective income adequacy who are more dis- satisfied with selected life quality measures. THE RELATIONSHIP OF OBJECTIVE AND SUBJECTIVE FAMILY INCOME ADEQUACY TO SELECTED MEASURES OF PERCEIVED LIFE QUALITY BY Norleen Marion Ackerman A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Family Ecology 1977 Copyright by NORLEEN MARION ACKERMAN 1977 ACKNOWLEDGMENTS Sincere appreciation is extended to my guidance committee: Dr. Beatrice Paolucci, Dr. Warren Samuels, Dr. Dennis Keefe, and Dr. Jane Oyer. I am especially grateful to Dr. Paolucci, the chairman of the committee; her advice, encouragement, and direction have greatly contributed to my professional development and to the completion of this study. Dr. Samuels provided insight- ful questions and suggestions throughout the project. Dr. Keefe furnished helpful ideas and information. Dr. Oyer's comments and questions were valuable in inter- preting study results. Other faculty members provided guidance and support. Dr. Peter Gladhart gave prompt replies to data analysis questions and reviewed parts of the dissertation. Dr. Margaret Bubloz's interest and knowledge of life quality research were helpful.‘ Dr. Linda Nelson's careful review and questions on the study findings were valuable. The data used here were collected by Burkhard Strumpel, Gerald Gurin, and Richard Curtin and made ii available through the Inter-University Consortium for Political and Social Research and the Michigan State University Political Science Department. Neither the collectors, the Consortium, nor the Political Science Department bear any responsibility for the analysis nor the interpretation presented here. Appreciation is expressed, however, to Dr. Strumpel and his colleagues, for early release of the data, and to Ann Robinson and Harriet Dhannak for facilitating the transfer of the data. Assistance in study design and statistical analy- sis was provided by Dr. Mary Andrews of the Institute for Family and Child Studies and by Suwatana Sookpokakit of the Office of Research Consultation. Their help is greatly appreciated. The College of Human Ecology and the Family Ecology Department at Michigan State University made available an assistantship and a research grant. The Cooperative Extension Service at the Ohio State University granted a temporary research assignment and an extended study leave. College and extension faculty made adjust- ments in their duties or programs to enable me to complete my doctoral program. All of these are deeply appreciated. Finally, I appreciate the encouragement and sup- port of fellow graduate students, family, and friends throughout this project. And I express sincere appreci- ation to my parents for their support throughout my edu- cational career. A TABLE OF CONTENTS Chapter I. INTRODUCTION . . . . . . . . . Statement of the Problem . . Conceptual Framework. . . Research Objectives Hypotheses . . . Assumptions. . . Definitions. . . II. REVIEW OF LITERATURE . . . . . . Basic Concepts. . . . . . . . Objective Measures of Economic Well- Being I I I I I I I I I I Relationships Between Measures of Objective Income. . . . . . Subjective Measures of Economic Well- Being . Congruency Between Objective and Sub- jective Measures . . . . . Perceived Life Satisfactions . . . III. METHODOLOGY . . . . . . . . . Data Source. . . Procedures for Sampling and Data Col- lection . . Selection of the Study Sample. . Description of the Study Sample . . Description of Variables . . . Income Adequacy Variables . . . Satisfaction Variables . . . Contextual Variables, Demographic. Contextual Variables, Economic Per- ceptions . . . . . . . . iv Page \DmU‘ltb 12 15 15 19 24 Chapter Analyses of Data . . . . . . . . Statistical Models. . . . . . . Computer Programs . . . . . . . IV. RESULTS OF DATA ANALYSES . . . . . . Income Adequacy and Satisfaction. . . Satisfaction with Perceived Family Income . . . . . . . . . . Satisfaction with Perceived Level of Consumption . . . . . . . Satisfaction with Perceived Overall Life Quality . . . . . . . . Strength of association . . . . Congruency and Satisfaction . . . . Contextual Variables. . . . . . . Subjective Adequacy Groups . . . . Demographic characteristics . . . Economic perceptions . . . . . Objective Adequacy Groups . . . . Demographic characteristics . . . Economic perceptions . . . . . Congruency Groups . . . . . . . Demographic characteristics . . . Economic perceptions . . . . . Summary . . . . . . . . . . . V. OVERVIEW, DISCUSSION, AND IMPLICATIONS . Overview of the Study . . . . . Discussion of Findings . . . . . Limitations of the Study . . . . Implications for Further Research . Implications for Educational Programs Page 63 64 68 70 70 78 81 83 86 90 96 97 97 103 105 105 109 112 112 117 119 123 123 125 135 136 138 e 1 3 3 g 4. 4 5 a 1 .I- 1 P I I I I c I I I I " “if SCALES BUREAU OF LABOR STATISTICS EQUIVALENCE CODING VARIATIONS B. i. A '31L-I; ‘4 .Imennnix a. ‘l 'ssLscwsb BIBLIOGRAPHY . I o 1 I". ha; 5 , V 1’ I-fAPEENDICES I)“ -. .._.,. - -‘fl _ -wv---v_._ .— .- LIST OF TABLES Table Page 1. Marital status and family type . . . . . . 47 2. Total family income before taxes in 1973. . . 48 3. Characteristics of the respondent . . . . . 49 4. Subjective family income adequacy distri— bution . . . . . . . . . . . . . 51 5. Objective family income adequacy distribution . 53 6. Objective adequacy ratios and total family income distribution by percentiles of the study sample . . . . . . . . . . . 54 7. Congruency group frequency distribution . . . 57 8. Descriptive statistics for satisfaction variables . . . . . . . . . . . . 59 9. Two—way analysis of variance for mean dif- ferences in satisfaction for objective and subjective income adequacy. . . . . . . 72 10. One-way analysis of variance for mean dif- ference in satisfaction with overall life quality for objective adequacy . . . . . 75 11. Means, standard deviations, and post hoc con- trasts of effects of subjective adequacy on satisfaction variables . . . . . . . . 77 12. Means, standard deviations, and post hoc con- trasts of effects of objective adequacy on satisfaction variables . . . . . . . . 79 13. Probability of difference in mean scores on overall life quality for adjoining objective adequacy groups . . . . . . . . . . 85 O-——v v..-” - _.._ .. v Table 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. Strength of association between income ade- quacy and satisfaction . . . . . . . One-way analysis of variance for effects of congruency on satisfaction variables . . Mean scores, standard deviations, and post hoc contrasts of effects of congruency on satisfaction variables . . . . . . Subjective adequacy group means, standard deviations, and statistics for contextual variables which are demographic charac- teristics . . . . . . . . . . . Post hoc contrasts of the difference between subjective adequacy groups on demographic characteristics . . . . . . . . . Modes, percentages, and statistics on con- textual variables for subjective adequacy groups . . . . . . . . . . . . Modes, percentages, and statistics on eco- nomic perception variables for the sub- jective adequacy groups. . . . . . . Objective adequacy group mean scores, standard deviations, and statistics on contextual variables. . . . . . . Post hoc contrasts of the difference between objective adequacy groups on demographic characteristics‘ . . . . . . . . . Modes, percentages, and statistics on con- textual variables for objective adequacy groups 0 I C I I I O I O O C 9 Modes, percentages, and statistics on eco— nomic perceptions objective adequacy groups 0 I O I O I O I O O O I Congruency group means, standard deviations, and statistics for contextual variables . Post hoc contrasts of the difference between congruency groups on demographic char- acteristics. . . . . . . . . . . Page 87 92 94 99 100 102 104 106 108 110 111 113 115 v-..“ ‘~.~—v—aV—— _. _. Table 27. 28. 29. 30. 31. 32. 33. Modes, percentages, and statistics on con— textual variables for congruency groups . . Modes, percentages, and statistics on eco- nomic perception variables for congruency groups 0 I I I O I O I C I C O . Summary of findings by hypothesis with sig- nificance level and strength of associ- ation . . . . . . . . . . . . . Revised scale of equivalent income for urban families of different size, age, and com- position . . . . . . . . . . . . Equivalence scale of comparative costs by city and region for the total budget for a four- person family at a moderate level of living, Autumn, 1973 . . . . . . . . . . . The identification of consumption achievement values from variables V97 and V98 . . . . The identification of perceived financial progress, recent, from variables V54 and V 7 I I I I I I I I I I I I I I ix Page 120 141 142 150 151 --. LIST OF FIGURES Figure Page 1. Model of the relationship of levels of objec- tive and subjective family income adequacy, and their congruency, to selected measures of life quality . . . . . . . . . . 10 2. Congruency groups identified by their levels of objective and subjective family income adequacy . . . . . . . . . . . . 56 3. The identification of congruency groups by their levels of objective and subjective family income adequacy. . . . . . . . 145 -v—.__Hv--._ v-.. ,-—.—'. - CHAPTER I INTRODUCTION One of the concerns of a nation is the level of living or the well-being of the families and individuals who are its citizens. Well-being has been measured for many years by economic indicators, such as family income and per capita income; the total of goods and services produced, as measured by the Gross National Product; and changes in prices, as measured by the Consumer Price Index. These have been supplemented by social indicators, such as counts of medical personnel and facilities avail— able, indexes of crimes reported, and acres of recreation lands. These economic and social indicators are quanti- tative, objective measures. They do not measure the quality of the environment in which people live, nor of the goods and services they use, or people's satisfactions with them. This was emphasized by Angus Campbell in 1971 when he said: Since World War II we have seen average family income rise dramatically and the number of families below the poverty line reduced by more than half. . . . But ~O-O‘--- ‘~ ' -' during those same years, we have seen a prodigious increase in the crime rate, . . . a spreading epi— demic of the use of drugs. . . . Civil disobedience and civil disorder are an everyday occurrence. It would take a brave man, indeed, to argue that the quality of life in these United States has been improving. . . . (pp. 2-3) Increasingly, the View of the better life as one in which people accumulate more and more things has been questioned. Today, the concern is for quality of life rather than for additional material possessions. While objective indicators of the quality of life have been developed and used, subjective or perceptual indicators are also important. Regarding subjective indicators, Strumpel (1975b) states: It is one of the often-stated purposes of the social indicators movement to measure output rather than input, well—being rather than command over material resources, personal health rather than the number of hospital beds, . . . Individual citizens, to the advocates of subjective or perceptual indicators, . . . are the final judges of societal output and welfare. One way of obtaining their verdict is to monitor people's satisfaction or happiness with their circumstances, their expectations of the future. . . . (Strumpel, 1975b, p. 2) Just as the nation has studied economic indicators over time, economists, family economists, and others have studied the expenditures of families (U.S. Dept. of Agri- culture, 1935; Morgan, 1958). These studies have documented how families do spend their money and have served as the basis for determining family income requirements to meet certain expenditure patterns and certain levels of living. However, such studies do not tell us how the command over resources, and the resultant attainable level of living relates to perceived overall life quality and its compo- nent parts, or domains. These studies have focused on the resource base, the input to well-being, not on the output of well-being or the satisfaction derived from it. Satisfaction with one's level of living, also termed quality of life, is more than a matter of the con- ditions of the physical, economic, and social setting. It is also a matter of how these are judged by oneself and others. The standards applied to the life situation, as well as the situation itself, influence the assessment of life quality. As Campbell and Converse (1972) state: We have become deeply impressed at the degree to which subjective states can "pull apart“ from what might be deducted on the basis of our current ways of understanding objective situations. . . . man . . . can display bitter discontentment with objec- tive situations that by retrospective standards are overflowing with abundance. Both the fact of dislocations between the objective and subjective, as well as the social importance of such dislo- cations has been . . . illustrated in recent decades by the revolution of rising expectations. Discontentment with objective conditions has appeared to be increasing over exactly the same period that those conditions have at most points and by most criteria been improving. . . . (PP- 8-9) We know little about the subjective assessments families make of their economic situations, or about how these assessments compare with objective assessments of family well—being. Are they similar? If not, how does the incongruency influence satisfaction with perceived overall life quality, and with those domains more -5..- - a--__—_V. . _ A ~.._—. directly related to economic factors, particularly family income and the family's level of consumption? Family economists work with individuals and fami- lies, providing information and counseling in the manage— ment of family resources to obtain family goals which enhance the level of living and quality of life of the family. As they do so, family economists need to be aware of both the objective and subjective assessment of family resources, their degree of congruency, and their relationship to satisfaction with the goals attained through resource use. Statement of the Problem Perceived overall life quality is viewed as the satisfaction derived from one's level of living. The level of living is achieved through the use of resources to attain goals. Hence, goals defined, implemented, and achieved, i.e., managerial activity, lead to the desired end, satisfaction with perceived overall life quality. The ability to achieve goals is determined, in part, by the availability of family money income. The adequacy of that income can be viewed both objectively, in relation to budget standards, and subjectively, in the perceptions of family members. The primary purpose of this research is to deter- mine the relationship between income adequacy and satisfaction with perceived overall life quality and two of its component parts, perceived family income and per— ceived level of consumption. Income adequacy is measured objectively and subjectively, and the two measures are combined in a measure of congruency. The secondary purpose of this research is to describe the members of groups which differ in level of subjective adequacy, level of objective adequacy, and degree of congruency. Groups are differentiated with respect to the demographic characteristics of the families, and the demographic characteristics and economic percep— tions of the respondent-heads. Conceptual Framework A managerial conceptual framework is used in this study. The respondents are family heads who are either the husband or wife or the one adult in the family unit. The focus is on the family's money resources and their relationship to the satisfaction derived from goal achievement. The respondent's perception of the ade- quacy of the family's money income and perception of personal satisfactions with selected life domains is elicited. It is recognized that these personal per- ceptions are likely to be influenced by the respondent's situational context as a member of the family unit. Money income is viewed as the resource to be managed. Satisfactions are viewed as the evaluation of v a... I l 6/. s ,., the outputs of the use of money income. Families manage their money income to achieve the goals from which satis— faction is derived. The resource input to the managerial process, while not the only input, is important to the success of the process. As Gross, et al., (1973) observe: Without resources, no amount of motivation or appli- cation of the managerial process will result in the achievement of goals. 0n the other hand, large stocks of resources will contribute nothing to satisfaction unless individuals or families are motivated and capable of managing their use. (p. 175) Availability of resources has been assumed to be related to goal satisfaction: Recognizing, developing, allocating and using human and material resources in ways that achieve one's goals contributes substantially to satisfaction with living. Each activity is an important compo- nent. And research data now point to a possible correlation between how effectively families deal with one or more of these components and their general quality of living. (Schlater, 1970, p. 48) Adequacy of money income can be judged in relation to a standard developed outside the family unit, such as a standard budget. Resource adequacy can also be judged by a family member, in relation to that individual's perception of family needs and wants. Gross, et al., (1973) state that “the adequacy or inadequacy of resources can be assessed only in relation to goals" (p. 174). Speaking of economic resources in particular, Deacon and Firebaugh (1975) state: " . . . the gap between economic goals and accomplishments affects the degree to CI-w'—-vv-- ._.-'__V___Il 'v._..___ 'v which economic well—being is perceived as adequate" (p. 242). To better understand family management, it is necessary to understand the factors affecting satis- faction with goal achievement. It appears that both objective and subjective factors must be examined. The standards specified and used influence the amount of resources used and provide a basis for evalu- ating the level of goal achievement. The degree to which standards are met influences the level of satisfaction derived from goal achievement. As Paolucci, et al., state: Standards "measured" in qualitative terms . . . tend to be subjective. . . . Examples . . . include "quality of life." . . . Qualitative standards describe what the family considers to be essential; family members will exert effort to secure them and feel dissatisfied and uncomfortable if they are not attained. (Paolucci, et al., 1977, p. 133) While standards are applied to individual goals, they are also applied to complexes of goals. Gross, et al., (1973) observe that "Standards of living differ from separate standards in that they consist of clusters of standards woven into organic wholes, each part related to every other part" (p. 128). Just as there are clusters of standards, there are also clusters of goals: In actual life . . . there is . . . a "tangled web" of goals--a goal complex. In this chain all pur- poses or ends are considered goals up to the final one or ones, which are probably values. These later ends are increasingly vague—-a characteristic more common to values than to goals. (Gross, et al., 1973, p. 126) In a hierarchy of goals, short-term or specific goals form the steps which build toward more long-term or more general goals, until one reaches final goals. The sum of present goal achievement forms the content or level of living. The level of living, together with one's aspirations and expectations of future changes in one's level of living, form the quality of life. The evaluation of one's quality of life is referred to in this study as satisfaction with perceived overall life quality. Such satisfaction could be referred to as the ultimate goal of the managerial process, or the terminal value which motivates behavior. Research Objectives The objectives of the research are: 1. To determine whether there is a relationship between family income adequacy, objectively and subjectively measured, and satisfaction with family income, level of consumption, and overall life quality. 2. To determine whether there is a relationship between the congruency of objective and subjec- tive family income adequacy and satisfaction with perceived family income, level of consumption, and overall life quality. m—' -“w To determine whether there are differences between the levels of objective and subjective family income adequacy, and congruency groups, with respect to contextual variables. The model of the relationships studied are shown in Figure 1. Hypotheses The hypotheses formulated for this study are stated below in the form of expected findings: General Research Question 1: Is there a relationship between income adequacy and selected life satisfaction variables? There is a difference in satisfaction with per— ceived family income for those who differ in objective and subjective family income adequacy. There is a difference in satisfaction with per- ceived level of consumption for those who differ in objective and subjective family income adequacy. There is a difference in satisfaction with per— ceived overall life quality for those who differ in objective and subjective family income adequacy. As the level of subjective family income adequacy increases, there is an increase in satisfaction with perceived: (a) family income, (b) level of consumption, (c) overall life quality. As the level of objective family income adequacy increases, there is an increase in satisfaction with perceived: (a) family income, (b) level of consumption, (c) overall life quality. 10 .muwamsv ovaa mo mmusmmwfi Umuowamm o» .mocmsumcoo menu can .Momsvopm mfioocw hHHEmM m>Huownn5m paw 0>Huomnno mo me>oH mo mfiSmcowumawH can No H0002 .H .mwm meowumouumm mofiumflnouomumno UHEocoom oazmmumofioo mmDOMU m0 mZOHBmHMUmmn whaflmsummouwm new haamo Ifinmmuwoswp Hmmmwp mocwsnmdoo can homswopm mEoocw mo maw>oa CH Hmmmwp 0:3 muons ow 30m aufiamso mung Hamum>01 monogamGOUI coaumasmsoo xomsvmv< mo Hm>mq- w>nuowmnou NEMMMH I homsuwcm H. mu w>uuoonnsmu “cues coauommmfiumm whomsvam wEoocH mo mam>0a cw mmocwuwmwwc an owcflmamxm on ado Momaomnd mzoozH MBHA¢DO mqu monummwfi muwamsw oufla wouomamm MqHz¢h cw :oHumaum> onu mo gone 30: Wi— *‘~‘v ‘A-..’ v __ 11 General Research Question 2: Is there a relationship between the congruency of income adequacy measures and selected life satis- faction variables? H6: There is a difference in satisfaction with per— ceived family income for those who differ in the congruency of their objective and subjective family income adequacy. There is a difference in satisfaction with per— ceived level of consumption for those who differ in the congruency of their objective and sub- jective family income adequacy. H - There is a difference in satisfaction with per— ceived overall life quality for those who differ in the congruency of their objective and sub- jective family income adequacy. H ' The congruent group will have the highest mean scores on satisfaction with family income, level of consumption, and overall life quality. General Research Question 3: Do those groups who differ in level of objective adequacy, level of subjective adequacy, and in con- gruency differ with respect to contextual variables? H There is a difference among (a) subjective family income adequacy groups, (b) objective family income adequacy groups, and (c) con— gruency groups with respect to: 10’ (i) demographic characteristics: -number of adults, age of respondent-head, number of children, age of oldest child, and total family income of the family unit; housing status and urbanization of resi- dence —sex, racial, or ethnic group, education and occupation of the respondent (ii) economic perceptions: -current concerns, material wishes, con- sumption achievement, and recent and intergenerational financial progress .-v-.—-vv are 2 12 Assumptions The assumptions upon which this study are based Respondent-heads have accurately reported their total family income for the most recent calendar year, 1973. The total family income is allocated to the family unit living in the respondent-head's household and is not also providing for others living outside the household, e.g., not providing alimony or child support to another family unit, nor providing support for someone in a nursing home. The Bureau of Labor Statistics standard budgets developed for urban wage earners and clerical workers may be applied to all families. The age of the female spouse may be substituted for the age of the male spouse when using the Bureau of Labor Statistics equivalence tables. for differences in family composition. Respondent-heads can assess their subjective family income adequacy, and their satisfaction with perceived family income, level of con- sumption, and overall life quality. 13 Definitions Overall Life Quality.--A person's subjective evaluation of the satisfaction, acceptance, or pleasure derived from his present level of living, related to felt needs and wants. Level of Consumption.-—Encompasses the food, fuel, and other nondurable goods used up, the services of houses, automobiles, clothing, and other durable and semidurable goods utilized; the services of human beings and public goods and services used by the individual or the family during a given period of time (Davis, 1945). Standard of Consumption.--Encompasses the same entities as level of consumption, but is a normative concept describing what a person or family realistically desires, and strives to achieve (Davis, 1945). Level of Living.--Includes all goods, services, and conditions consumed or experienced in living: those goods and services acquired in the marketplace plus nonmarket goods, services, and conditions such as public facilities, social status, and location of residence (Davis, 1945). vv C—VV‘ 14 Standard of Living.--Includes the same entities as level of living, but is a normative concept describing what a person realistically desires and strives to achieve (Davis, 1945). Money Income.—-The total income in money before taxes which a family receives in a one-year time period. Real Income.--The total family income from money income, home production, services derived from goods owned, and services derived from public goods. Objective Family Income Adequacy (Objective Adequacy).—-The assessment of budget makers as to the adequacy of the total money income of the family unit. It is based on expenditure studies and scientifically determined needs and is adjusted for family composition and geographic region of residence. Also referred to as economic well—being. Subjective Family Income Adequacy (Subjective Adequacy).--The personal judgment of the individual as to whether family income is adequate. It is assumed to be based on one's knowledge of family income and family goals, or family needs and wants. -- .. CHAPTER II REVIEW OF LITERATURE Research and related literature are reviewed in five major categories: basic concepts, objective measures of economic well-being, subjective measures of economic well-being, congruency between objective and subjective measures, and satisfaction with selected measures of per- ceived life quality. Basic Concepts Two types of basic concepts are discussed: income concepts and standards and levels of consumption and liv- ing. Any measure of economic well-being has, as one of its components, a measure of income. Three major con- cepts related to income appear in the literature: money income, real income, and psychic income. Money income is "the flow of purchasing power that comes into the control of an individual or family in a given period of time" (Fitzsimmons & Williams, 1973, p. 6). Money income includes the wages and salaries of family members, capital income such as rent, interest, dividends, and royalties; business and farm income not allocated to 15 z16 ) \y/ labor; and transfer income. Hence, it includes market transactions where the goods or services have been given a dollar value. Money is an important medium of exchange, yet there are other means of acquiring goods and services. In the evaluation of the well-being of families, one is interested in the total income available to families, whether in the form of money, in the provision of ser- vices, or goods provided in kind. This can be measured by real income, which has been defined as: . . . a flow of commodities and services available for the satisfaction of human wants and needs over a given period of time. . . . The concept of real income is a recognition of the contribution to the family made by resources other than money and includes the use of family time, energy and abili- ties along with community resources. (Gross, et al., 1973, p. 487) Gross, et al., indicate a preference for the use of real income as a measure, stating: " . . . real income is a more accurate basis than is money income for comparing the welfare of different families at a given point in time . . .“ (p. 487). The third income concept is that of psychic income. It is: " . . . the enjoyment or satisfaction derived by people from use of their real income in a period of time" (Fitzsimmons & Williams, 1973, p. 261). The authors express a preference for measures of psychic income, saying: "Psychic income might be the most l7 important form, since people are primarily interested in satisfaction realized rather than in the goods them- selves" (p. 261). While it might be concluded that the hierarchy of preferred concepts is from psychic down to real income, then to money income, actual use of the concepts in research is the reverse. As Fitzsimmons and Williams (1973) state: " . . . because enjoyment and satisfaction are difficult to observe and evaluate, money and real income are more often used as a measure of the family's well-being" (p. 261). Money income is more commonly used than real income in research because it is a more familiar concept and the information about it is less expensive to obtain and to develop for use in analysis. Of the three income concepts, money income and real income are objective measures of income; psychic income is a subjective measure, not of income itself, but of the satisfaction derived from the use of income. A 1945 article by Davis provides the benchmark commonly used by family and consumer economists to develop conceptual definitions which distinguish between consumption and living, and between level and standard. The distinctions Davis draws are: Consumption means the commodities, their uses, and EEEVIEE§_ESnsumed; livin includes consumption and much more: working conditions, cushions against major and minor shocks, freedoms of various kinds, and what I have called "atmosphere." The level of 18 consumption or living, . . . is that actually experienced, enjoyed or suffered by the individual or group; the standard of consumption or living is the level that is urgently desired and striven for, . . . (pp. 2—3). Davis sometimes uses the word "content" rather than level, feeling that content better suggests compo- sition and quality as well as size. With these terms, he develops the following definitions: Actual consumption or consumption level is . . . a sort of aggregate of the food, quI, and other nondurable goods used up; the services of houses, automobiles, clothing and other durable and semi— durable goods utilized; and the services of human beings used, by an individual or group, in a given period of time. (pp. 3-4) The consumption standard . . . is the consumption level that is earnestly desired and eagerly striven for, in respect of quantities, qualities, and pro- portions of the various goods consumed or wanted for consumption. (p. 6) The content of living is a reality experienced by an individuaI-br group. It is made up of a complex combination of consumption, working conditions, possessions, freedoms, and "atmosphere," and the balance or harmony among them, in relation to needs and felt wants. (p. 7) The standard of living . . . is the . . . content of living . . . Which an individual or group earnestly seeks and strives to attain, to maintain if attained, to preserve if threatened, and to regain if lost. . . . it is no less a reality than the experienced content of living. (p. With changes in wording and updating of the terms used, these definitions are very much in use today (Schlater, 1970; Hafstrom & Dunsing, 1973; Rudd & Kline, 1976). 19 Objective Measures of Economic Well-Being Approaches to the objective measurement of eco- nomic well—being include the measurement of command over resources, the measurement of resources used or available for use, and the combination of these first two measures in the development of income-consumption needs ratios. The first approach measures income or financial resources. The concepts of money income and real income, defined above, are examples of this approach. Real income includes nonmoney transfers, services from durable goods, and nonmoney income from unpaid work, of which unpaid housework is likely to be the largest component, as well as the components of money income. A second approach to the objective measurement of economic well-being is that of measuring consumption. Two examples of this are the level of expenditures and the money value of consumption. The level of expenditure approach sums the purchases of the family unit during a given time period. It excludes goods and services obtained without the use of money and the value of services obtained from durable goods purchased during earlier time periods. It includes the total price of durable goods purchased during the current time period even though the service life of those goods extends through future time periods. This method of accounting for durable goods results in overestimating the '—— A—.~ 20 consumption of families during their establishment years when the investment in durable goods is larger, and underestimates the consumption of the elderly who con— tinue to use a flow of services from durable goods pur- chased in earlier time periods. The money value of consumption approach quantifies the total value of goods and services consumed during a given time period regardless of whether expenditures of money are involved. It excludes current expenditures for durable goods (including only their service value during the present time period). It also excludes expenditures for nonconsumption items, such as life insurance and those gifts and contributions which go to persons outside the immediate family. Money income and the level of expenditures are simple measures of those income and consumption items, respectively, which have market prices. Real income and the money value of consumption are more comprehensive measures, attempting to include all items of current income or consumption, regardless of whether financial transactions are involved. The third approach to the objective measurement of economic well—being is more holistic. It relates income to a consumption standard which is usually adjusted for differences in family composition. These measures have been termed "well-offness ratios," ._..—. __ 21 "income-needs ratios,“ and "welfare ratios" (Morgan & Smith, 1969; Morgan, et al., 1974; Strumpel, 1976). The basic format of a welfare ratio equation is an income component in the numerator and a consumption standard in the denominator. The income component may be total money income or total real income, measured in varying degrees of completeness. The denominator, or the consumption component, however, is somewhat different from consumption itself. It is a consumption standard which has been adjusted for differences in family com- position. Three examples of denominators, using different consumption standards, are reviewed here. First, a welfare ratio which considers the number of family members and makes some adjustment for economies of scale is used by Strumpel (1976), Curtin (1976), and Yuchtman (1976). It is computed by: Welfare = Total yearly money income of the family unit Ratio $I,400 + ($766 x nfifiBer of family members) The denominator of this equation recognizes a basic consumption package, indicated by the $1,400, required by one family member which also is shared with other family members. It adds a given amount, $700, for the additional individual needs of each family member. A second, and more complex, set of consumption standards adjusts for the age of the head, age of the oldest child, number of adults, and number of children. 22 These are the standard budgets developed by the Bureau of Labor Statistics (BLS) of the U.S. Department of Labor. Budgets have been developed at lower, moderate, and higher levels of living. The dollar value of the standard budgets, for a selected family of four in the fall of 1973, is $8,181 for the lower level, $12,626 for the moderate level, and $18,201 for the higher level of living (U.S. Department of Labor, undated). Equivalence scales are provided for adjusting either the consumption budget, or the family income needed to achieve that budget, for families of differing composition (U.S. Department of Labor, 1968). Another set of equivalence scales is available to adjust the BLS budgets for geo— graphic area of residence (U.S. Department of Labor, undated). The BLS budgets are based on the dollar value of selected market baskets of consumption items. These market basket items are derived primarily through a Survey of Consumer Expenditures and scientifically identified consumption needs, where such needs have been determined. The Survey of Consumer Expenditures is conducted about every ten years, with new market baskets of consumption items identified following the survey and new standard budgets developed. The cost of the budgets are updated annually with the Consumer Price Index. 23 Any one of the BLS budgets can be used in the computation of a welfare ratio, in an equation such as: Welfare _ Total yearly family income Ratio - BLS Standard BLS equivalence number Budget x for family composition W Such an equation adjusts the BLS standard budget, which is developed for a selected family of four, for the composition of the particular family for whom the ratio is being computed. The larger the resulting welfare ratio, the better off the family is economically. The consumption needs standard developed by Orshansky, which is the official United States poverty threshold, can also be used to compute a welfare ratio. That standard starts with the United States Department of Agriculture food requirements, adding up the indi- vidual requirements of family members and then adjusting for economies of scale in feeding peOple. Additional economies of scale are allowed for by using three times the food requirement as a total needs estimate, except for families of one or two, where a multiplier larger than three is applied. An additional adjustment is made for farm families to allow for their greater nonmoney income (Orshansky, 1965a, 1965b). Morgan, et al., (1974) in their work, used a welfare ratio which was a modifi- cation of the methodology of Orshansky. 24 The determination of consumption needs poses some special problems. For example, although nutritional needs have been scientifically identified, those needs can be met through an array of possible foods which vary greatly in price. "Needs" standards thus are normative; they are selected by the developer of the standard. In the instance of the BLS budgets, such judgments of needs are made with reference to the food consumption patterns of families as those patterns are identified through the Survey of Consumer Expenditure data. Relationships Between Measures of Objective Income Research studies have been conducted which compare money income and the money value of consumption, and which report correlations between income measures and welfare ratios. Rudd and Kline (1976) compare the measures of money income, defined as total family income after taxes, and the money value of consumption for rural families in the United States. They conclude that: Findings suggest that for rural families money value of consumption is linearly related to income, that the variance of money value of consumption is pro- portional to income and that money value of con- sumption is more equally distributed than income. . . . A comparison of how the two measures . . . rank rural families by consumption status indicates that income is most likely to rank families in- accurately at very low and very high levels of income. (p. 1) 25 However, these findings apply only to rural families. Rural families are more likely to engage in more hours of home production and, thus, to consume more goods and services without paying for them in the market. Evidence of this is reported from a 1964 study which found that one factor differentiating the number of hours of home production by family heads and their spouses is urban or rural residence. Those who live in rural areas report 861 hours of home production per family per year (exclud- ing housework), while those not living in rural areas report only 292 hours of home production per year (Morgan, et al., 1966). Thus, for a sample of all families in the United States, the difference between real income and money income is likely to be less than in the sample analyzed by Rudd and Kline. A study by Morgan and Smith (1969) considers the strength of association between components within a simple and a more complex welfare ratio. The simpler welfare ratio is computed by dividing family income by family consumption needs. The more complex ratio multi- plies a net real income figure (money income plus non- money income minus costs of earning an income) by leisure time, then divides the product by family consumption needs. Morgan and Smith conclude that: . . . a more complete measure of economic well- offness is still heavily dominated by real income and real income is dominated by money income. But 26 this is no reason for complacence. . . . there are differences, and, more important, the domination will almost certainly turn out to be less at the bottom of the income distribution. (p. 455) Morgan and Smith (1969) also examine the relation- ships between eleven different measures of economic status; some of these are income measures, others are ratios of economic well-being. The simplest of the measures is total money income of the family for one year. The correlation between money income and net real income, as measured here, is .99, indicating a very strong association between the two measures. The cor- relations between total money income and two different welfare ratios is somewhat less, but still quite strong. Those correlations with total money income are .85 for the dollar income/needs ratio and .84 for the net real income/needs ratio. The correlations of a food only ratio, computed by dividing food consumption by food needs, is lower with income measures, .34 with money income, and .35 with net real income, than with economic welfare ratios, .52 with the money income/needs ratio, and .54 with the net real income/needs ratio. Regarding the correlations of all eleven measures, Morgan and Smith observe that: . . . they [the correlations] become progressively smaller as our measures become more complex, taking account first of family structures, then of dif- ferences in leisure, and finally in differences in what the family must pay for housing. One indi- cation of the validity of the adjustment for family 27 size is that the food adequacy measure . . . cor- relates more highly with the ratios of income to needs than it does with any of the absolute income measures. (PP. 456-57) Subjective Measures of Economic WelIeBeIfig The economic welfare of individuals or families has tended to be measured in terms of the objective presence or absence of financial resources. This approach ignores the psychological perspective of the person. As Cantril has stated (1965): Everyone--whether of high or low status, . . . has subjective standards which guide behavior and define satisfactions. . . . The problem is to learn what these standards are in a person's own terms and not judge them by our own standards. (p. 21) Factors other than current command over resources are a part of the person's subjective judgment of the adequacy of his income. Macdonald (1963) provides an illustration of one factor: A childless young couple with $3,000 a year is not poor in the way an elderly couple might be with the same income. The young couple's sta- tistical poverty may be a temporary inconvenience; if the husband is a graduate student or a skilled worker, there are prospects of later affluence or at least comfort. But the old couple can look forward only to diminishing earnings and increas- ing medical expenses. (PP. 91-92) While numerous budget studies have been done, and a number of needs standards developed, less study has been made of people's subjective judgments of income and its adequacy (U.S. Department of Agriculture, 1935; Morgan, 1958). 28 The perceived income adequacy of homemakers under the age of sixty-five, with one or more children living at home, is reported for randomly sampled selected areas in thirteen states. Much higher percentages of a sample of black families in east central Texas report their incomes are not adequate than do rural migrants in California or than do rural families and urban low-income area families sampled in the midwest. The highest fre- quencies of incomes perceived as being adequate occurs in the four rural small places family samples in the mid- west (Iowa Agriculture and Home Economics Experiment Station, 1974). Rainwater (1974) has researched the subjective dimension of the level of living, or "the assessment people make of how well or poorly people are enabled to live at a given income" (p. 110). In a 1971 survey using a quota sample of Bostonians, respondents were given qualitative, descriptive labels of the level of living of specified families and asked to name the dollar amount of income the family would need to live at that level. An example of the questions is: Mr. and Mrs. Smith have two children and are gen- erally considered to have.a comfortable living, not particularly high and not particularly low. What is the lowest income they could have and still be considered to have a comfortable level of living? (p. 95) Similar questions were asked for levels of living labeled as "poor," "get along," ”prosperous and substantial," 29 and "rich." These levels of living were chosen by the researcher to designate points along a continuum from a very low level of living to a very high level of living. Analysis of the dollar amounts of income respondents named as being needed for the qualitatively described levels of living reveals that: Taking comfort as 100 percent, we find that poverty requires an income slightly less than half of the comfortable level, while getting along connotes a living level requiring about two-thirds of the income required to be comfortable. For the pros- perous, substantial level, something over a quarter again as much is necessary, and being rich requires over twice as much money. (p. 111) Additional analysis indicates that the respondents' background characteristics and their implicit scale of dollars seldom accounts for more than a few percent of the variance in the dollar amounts of income named. When the respondents are separated by social class, they all name similar dollar amounts of income for the lower levels of living. The working class, however, see the lower limit of richness to be about four times the upper limit of poverty while, for the upper middle class, richness starts at an income that is six times the poverty line. However, there is a great deal of vari- ation in dollar amounts stated within the classes of respondents. Two criteria in the phrasing of the questions seem to influence the dollar amount of income named by the respondents. The specified level of living is 30 clearly the most important source of variation. Yet, when the level of living is controlled, there is a strong relation between family size and the dollar amount of income needed by respondents. Significant interaction between living level and family size results in greater dollar increments for additions to family size at higher levels of living. In the judgment of the Boston sample, couples with five children need about 50 percent more income than couples with no children in order to be equally well-off economically. Rainwater (1974) also analyzes time series data from the Gallup Polls. This provides information on people's conceptions of the yearly income needs for two levels of living, a "get along" level and a higher level described as "health and comfort." The question regard- ing the "get along" level was used almost yearly from 1946 to 1969. It asked: "What is the smallest amount of money a family of four (husband, wife, and two children) needs to get along in this community?" (p. 52). The mean amount reported by respondents is $2,226 in 1946 and $6,225 in 1969. Measured in constant dollars, there is about a 50 percent increase in the amount of goods and services Americans felt were necessary for a family to get along, over the 23-year period. The incomes stated range from 46 to 58 percent of per family disposable 31 income. The incomes stated average 106 percent of the take-home pay of a worker in private nonagricultural industry. Factors which influence the amount specified are the respondent's family income, the amount the family spent on food, the respondent's educational level, and the size of the community. Correlations between these factors and the get-along income range from .21 to .30. Factors which have minimal effect are race, sex, number of children, and, after controlling for community size, region of the United States. Analysis of the Gallup Poll question which spe- cifies a somewhat higher level of living, described as "health and comfort," shows a similar 50 percent increase in the constant dollar amount of income reported as needed over the 25-year period from 1939 to 1963. The income amount stated ranges from about two-thirds to a little over three-fourths of per family disposable income. Congruency Between Objective and Subjective Measures There is some evidence that people view their income adequacy or economic well-being from both an objective and a subjective perspective. A limited amount of research examines the relationship between these objective and subjective measures of income adequacy. 32 Rainwater (1974) compares the Gallup Poll question on the amount of income needed to live at a level of "health and comfort" with the Ornati budget for a minimum comfort level of living. He concludes that "there is quite a high degree of agreement between budget makers and the public" in their judgments of the income needed for this level of living (p. 52). Comparisons are also made between the subjective judgments on the Boston sample and several objective budget standards with respect to the increments of income needed by larger families, if they are to maintain the same level of living as family size increases. As the number of children in the family increases, Rainwater determined that all of the objectively computed equiva- lence scales of increasing income needs increase much more rapidly than does the scale developed from the sub- jective judgments of the Boston sample. The difference is greatest when the Boston sample's responses are com- pared with the equivalence scales of the Bureau of Labor Statistics. The motivation for, and the dilemma of, the search for congruity between objective and subjective measurements is pointed out by Campbell, Converse, and Rodgers (1976): Whenever one begins to explore unfamiliar terrain with unfamiliar measuring instruments . . . the nearest approach to a guide is common sense expectation. . . . most of the strong expectations 33 derive in one way or another from the simple proposition that "the better people's situations are, the more satisfied they should be with them." If we are to go on to ask what we mean by "better," we typically mean "objectively better." . . . we are moving into something of an impasse, if we carry this reasoning to its logical extreme. If the only data we were to trust were those in which reports of satisfaction were congruent with objective satisfaction, however defined, then we would have served little purpose in investigating subjective perceptions to begin with. Yet if this kind of subjective investigation has any particular inspir- ation at all, it depends on the observation, rather well documented, that subjective assessments do depart from objective situations in meaningful and important ways, about which all too little is known. Our dilemma is thereby sharpened. we become most suspicious of bias or measurement inadequacy when subjective assessments come into conflict with objective situations, although such discrepancies taken substantively are almost the principal reason for the conduct of the study. The only route out of the dilemma is to come to grips with the unexpected result with all the analytic tools and side information we can bring to bear, in order to learn what part of the finding, if any, stems from distortions that can be readily understood, and what part seems to reflect something of substantive interest in the real world. (p. 115) When a person's subjective assessment of a situ— ation differs from an objective assessment of it, there is incongruity, inconsistency, or dissonance between the two assessments. According to Festinger's theory of cognitive dissonance, the individual is motivated to overcome the dissonance. This can be done by (l) chang- ing one of the elements in the dissonant relationship, (2) adding new cognitive elements that are consistent with one of the already existing cognitions, and/or (3) by decreasing the importance of the cognitive elements involved in the dissonant relationship. Thus, 34 the existence of dissonance within a person's cognitions should be observable in a person's actions (Festinger, 1957). A special case of cognitive dissonance is recog- nized as relative deprivation. The conditions for a feeling of relative deprivation are (1) an expectation that is felt to be legitimate or deserved and that is expected to be fulfilled within a given time period and (2) a high perceived probability, which increases rather suddenly, that the expectation will not be fulfilled. Here, the dissonance occurs between the expectation which is thought to be deserved and its nonoccurrence; the dissonance felt is that of injustice or inequity (Morrison, 1971). In sum, the fact that subjective and objective assessments may be inconsistent is the principal reason for the use of subjective assessments. An individual may not necessarily be aware of the objective assessment. In fact, if it disagrees with his subjective assessment, he may try to ignore it to avoid the incurrence of dis- sonance . Perceived Life Satisfactions Two major studies conducted by the Survey Research Center of the University of Michigan in the early 19708 focused on peOple's perceptions of life satisfactions or quality of life (Campbell, et al., 1976; Andrews & Withey, 35 1974). A central theme of both studies was the interplay between the satisfactions people derive from different aspects of their lives and how these aspects combine to form their overall evaluations of perceived overall life quality. Both studies indicate that a person's overall satisfaction with life is based largely on the satis- factions that flow from a variety of different domains. Andrews and Withey report that the four domains which contribute most to perceived overall life quality are satisfaction with (l) yourself, (2) your family life, (3) the amount of fun in your life, and (4) a money index developed from satisfactions with family income and level of consumption. The remaining domains of the twelve Andrews and Withey identify are your health, your job, goods and services, your house or apartment, things to do with your family, time to do things, spare time activities, and an index of satisfaction with national government. Andrews and Withey (1974) observe that the twelve domains do as well in explaining satisfaction with over- all life quality as do a much larger list. They indicate that information concerning income, sex, race, age, family life cycle stage, and education add nothing more to understanding why some people are content with their lives and others are not. 36 Nyblad (1975) concludes that there is no great variance in perceived life quality when age, sex, area of residence, or income vary. She finds that social factors contribute most to one's overall life quality, economic factors are second, and environmental factors last. Both Campbell, et al., and Andrews and Withey state that people reserve their ratings of greatest satisfaction for those parts of their lives that are the most personal and intimate. The more removed a domain is from the person, the more dissatisfied they are likely to be with that domain. In the Campbell, et al., study, respondents are most satisfied with their marriage, second most satisfied with their family life, and least satisfied with savings and investments. Satisfaction is also low in the domain of level of consumption, which Campbell, et al., refer to as standard of living (1976). Andrews and Withey report that persons of low social and economic status, urban residents, and blacks find life less satisfying than do their counterparts. Young people are less pleased with their lives than are older people. In general, there is little difference in the satisfactions expressed by men and women (ISR Newsletter, 1974). Campbell, et al., recognize the importance of income as a resource, stating (1976): 37 . . . when all resources and resource satisfactions are gathered together . . . the preeminence of satis— faction with one's standard of living as a predictor of such feelings is dramatic, and another income related satisfaction, that with savings, makes a substantial contribution as well. (p. 380) The Campbell, et al., study provides an example of the importance of income as a resource. When asked "Up to now, have you been able to satisfy most of your ambitions in life or have you had to settle for less than you had hoped for?" almost 60 percent of the sample responded positively. However, the variation in responses is "more tightly associated with the classic resources of income and education than any of our other global measures of life satisfaction or felt well-being" (pp. 380-81). The 40 percent of the population who answered the question negatively were asked the further question "What are the main things that (have) stood in your way?" About one-half of those responding mentioned purely financial limitations; some of the other replies at least imply financial constraints (p. 381). Because of the importance of variation in one's financial situation for feelings of well-being, as well as the greater amount of expressed dissatisfaction in this domain, much more detailed studies have been ini- tiated, Campbell, et al., state. They refer to Strumpel as "a central and up-to-date reference" (p. n381). Strumpel (1975a) observes that the "most straight- forward subjective correlate of an individual's objective 38 income position is his satisfaction with income" (p. 78). Yet, a more complete decomposition of subjective economic welfare would include, besides income satisfaction, satis- faction with one's level of consumption or the extent to which the present income is seen as providing for a com- fortable life, perceived equity of monetary rewards from the job and income expectations (Strumpel, 1975a). The concept of satisfaction with level of consumption, according to Strumpel, serves "as a proxy for the degree to which present family income satisfies consumption needs or aspirations; also for the experience of financial constraint, the difficulty of making ends meet, or the felt denial of present wants in the sphere of consumption or income allocation" (p. 78). Strumpel (1975a) notes that satisfaction stands for acceptance rather than approval: Some people, more easily than others, accept unde- sirable and undesired situations and environments. As a yardstick against which to measure reality they employ a barely acceptable minimum of "tolera- bility." Others use a "fair," "good," or even "ideal" state of affairs as anchoring points for the evaluation of their reality. There is reason to assume that the use of different standards is not randomly distributed over persons or domains. It has been noted, for instance, that under- privileged groups (as long as there is little hope for betterment) tend to be complacent, even to express much satisfaction with their condition, while the perceived possibility of change has been observed to arouse aspirations. (p. 79) Strumpel feels that if an individual's standards for evaluating his condition are tainted by the perception 39 of its persistence, satisfaction scores by domain appear in a new perspective. Marriage is likely to be considered rather permanent; one's job, or at least the kind of job one can hope to occupy, is somewhat fixed. In contrast, people's income and level of consumption change; vari- ations in income, prices, and material needs occur con- stantly and are often difficult to anticipate (Strumpel, 1975a). In addition, the social acceptability of dissatis— faction differs from domain to domain. Dissatisfaction with income and level of consumption is accepted in a culture stressing mobility, opportunities, and individual progress. This is less true of the job, and not at all true of marriage (Strumpel, 1975a). Strumpel has found a strong relationship between satisfaction with income, reports of past increases in income, and a sense of well-being. The relationship is constant, even after the effects of income level are taken into account. The distribution of satisfaction across the income distribution suggests that the societal average serves as the reference point for lower-income Americans; their dissatisfaction is highly correlated with the distance from the mean (Strumpel, 1975a). The members of households with somewhat higher incomes, those who are comfortable to fairly affluent but not wealthy, do not become much more satisfied with further 40 increments of income, according to Katona (1964). Strumpel (1976) also reports that satisfaction with one's level of consumption responds to reality; actual income correlates positively with measured satisfaction with level of consumption. In sum, a number of measures of objective income adequacy have been developed and used in research. Few measures of subjective adequacy have been empirically tested; little has been researched in the area of con- gruency. In recent years, work has been done on satis- faction with overall life quality, with analysis of those domains which contribute to satisfaction with overall life quality. Economic domains are a contributor to life quality, with the strength of their relationship appearing to be less than the social domains but more than the environmental domains. CHAPTER III METHODOLOGY This study is undertaken to determine the relation- ship of family income adequacy to selected measures of satisfaction with perceived life quality and to specific contextual variables. Data collected in a national study conducted by the Survey Research Center (SRC) of the Uni- versity of Michigan are used to test the research hypothe- ses. This chapter describes the data source, sampling and data collection procedures used in the SRC study, selection and description of the study sample, development of specific variables, and data analysis procedures. Data Source The data used in this study were collected from households during September-November 1974, as a part of the Fall Omnibus Study of the Survey Research Center of the University of Michigan. These data are identified as SSA-3512, Part IV, "Economic Incentives, Values, and Subjective Well-being” by Burkhard Strumpel, et a1. 41 42 (Inter-University Consortium for Political and Social Research, 1976-77, pp. 119-20). Information from the Bureau of Labor Statistics (BLS) United States Department of Labor is used in the development of the objective family income adequacy measure. Procedgpes for Sampling and Data Collection The SRC interview population is a cross-sectional sample of American adults, eighteen years of age or older, who live in private households in the United States, excluding Alaska and Hawaii. Transients and individuals living in institutions, dormitories, large rooming houses, or on military bases are not included. The households are selected through a multi-stage area probability sampling technique. Probability selection is enforced at all stages of sampling. The interviewer has no freedom of choice among housing units or among household members within a sample dwelling. The final SRC sample consists of 1,519 individuals. Data are collected by personal interview with the selected respondent in each household. Trained SRC inter- viewers use the interview schedule prepared for the study. That schedule contains fixed questions with either fixed answers or open-ended answers. Additional items are gathered by interviewer observation, e.g., race and sex of the respondent. 43 Data are coded, transformed, and then stored on magnetic tape by the SRC staff. A copy of the magnetic tape was made available to the researcher by the Inter- University Consortium for Political and Social Research, Ann Arbor, through the Michigan State University Depart- ment of Political Science, a Consortium member. Selection of the Study Sample A sample of 1,046 of the 1,519 respondents to the fall 1974 SRC survey of Strumpel, et al., are used for this study. The criteria for selecting the study sample are (1) complete data on life satisfaction measures and subjective family income adequacy and (2) availability of specific information required to compute the objective family income adequacy measure. The first criterion reduces the sample by less than 100 persons. The deletions for not providing on- the-scale responses to the 3 satisfaction variables include 44 on perceived overall life quality, 20 on perceived level of consumption, and 22 on perceived family income. An additional 6 respondents who did not provide information on subjective family income adequacy are‘deleted. The second criterion, presence of all information necessary to compute objective income adequacy measures, requires survey data on age of family head, age of oldest child, number of parents in the household, number of 44 children under 18 in the household, and total family income before taxes in 1973. This criterion also requires BLS income equivalence numbers to adjust the BLS standard budget for differences in family composition. The deletions by the second criterion include 159 respondents who did not provide information on their total family income and 331 who could not be assigned the most specific type of BLS family income adjustment equivalence number. This most specific type of equivalence number is based on the age of the family head, age of oldest child, and presence of husband and wife, or one parent, or one adult living alone (see Appendix A). The BLS age of male family head classification for equivalence numbers is reinterpreted in this study to be the male or female respondent-head. Still, 71 adult children and 8 respon- dent-heads are deleted because the age of the head is not known. Seven families are deleted because they are unusual age-of-parent, age-of-child combinations for which there are no specific BLS family income adjustment equivalence numbers. While equivalence numbers are available for family units which include adult children living at home, such children, when not the survey respondent, cannot be identified in the SRC sample data. Therefore, all 237 cases with more adults in the family unit than just the respondent, and a second adult if the respondent is married, are deleted. 45 Eight additional respondents are deleted because the comparability of their objective family income ade- quacy measures with those of other respondents is ques- tioned. These are married respondents whose spouses do not live at home. Maintenance of two separate residences is likely to require more income to achieve the same moderate level of living. There is also the possibility that the respondent may have interpreted the question regarding total family income as being either the income of the family unit at that address or of the family unit including the absent spouse. The sum of individual deletions is 676, while only 473 cases are deleted from the SRC sample in the formation of the study sample. Thus, 203 cases are deleted for multiple reasons. Possible examples of respondents deleted for multiple reasons are (l) the respondent who fails to give on-the-scale responses to two of the life satisfaction measures and (2) the respondent who is a parent of the family unit head, who does not report the family unit's total income in 1973. Description of the Study Sample The sample selected for this study contains 1,046 respondents. The major characteristic on which the study sample differs from the SRC sample is that of marital status. There are 6.7 percent more married respondents in the study sample than in the SRC sample and 5.0 percent 46 less in the never married category. Other percentages-- separated, divorced, and widowed-—are quite similar in both samples, as shown in Table 1. The differences appear to be related to the second deletion criterion-- that respondents who cannot be assigned specific BLS family income adjustment equivalence numbers are excluded from the study sample. The never married respondent- heads may more frequently share their family units with other adults, such as a nonrelated adult or an elderly parent who moves into their home. If this is the case, such respondents are deleted from the study sample. Comparing response categories, a lesser dif— ference between the SRC sample and the study sample is that 2.5 percent less of the study sample are one—person households and 2.5 percent more are married couples with children under the age of 18 living at home, as shown in Table 1. This difference is consistent with and related to the difference in marital status between the study sample and the SRC sample, but is less pronounced. In all other characteristics compared, the study sample is very similar to the SRC sample. Comparisons of family income are shown in Table 2; comparisons of the respondent's sex, age, racial or ethnic group, edu- cation, and occupation are reported in Table 3. Since the samples are so similar, the results of the study are generalizable to those family units in private households 47 TABLE l.--Marita1 status and family type Characteristics of the Study Sample SRC Sample Respondent-Head N % N % Marital Status 1046 100.0 1519 100.0 Married 767 73.3 1101 66.6 Separated 34 3.3 50 3.3 Divorced 57 5.4 87 5.7 Widowed 98 9.4 160 10.5 Never married 90 8.6 206 13.6 Not ascertained - - 5 .3 Family Type 1046 100.0 1519 100.0a One person 215 20.5 284 23.0 One adult, one or more children 64 6.1 75 6.1 Married couple, no children at home 342 32.7 405 32.8 Married couple, children age 17 or younger living at home 425 40.6 471 38.1 Other family typesb - - 284 - types." aPercentages exclude the category "other family b Family units which include other adults in addition to the respondent-head and, if the respondent is married, the respondent's spouse. TABLE 2.--Total family income before taxes in 1973 48 1973 Family Study Sample SRC Sample Inc°me N % cum. N % cum. 9:9 Under $2,000 46 4.4 4.4 57 4.2 4.2 $ 2,000- 2,999 51 4.9 9.3 72 5.3 9.5 3,000- 3,999 59 5.6 14.9 73 5.4 14.9 4,000- 4,999 34 3.3 18.2 50 3.7 18.5 5,000- 5,999 44 4.2 22.4 62 4.6 23.1 6,000- 7,499 78 7.5 29.8 94 6.9 30.0 7,500- 8,999 78 7.5 37.3 94 6.9 36.9 9,000- 9,999 63 6.0 43.3 73 5.4 42.3 10,000-10,999 79 7.6 50.9 98 7.2 49.5 11,000-12,499 71 6.8 57.5 94 6.9 56.4 12,500-14,999 109 10.4 68.1 148 10.9 67. 15,000-17,499 95 9.1 77.2 123 9.0 76.3 17,500-19,999 71 6.8 83.9 84 6.2 82.5 20,000-22,499 54 5.2 89.1 76 5.6 88.1 22,500-24,999 33 3.2 92.3 45 3.3 91.4 25,999-29,999 24 2.3 94.6 37 2.7 94.1 30,000-34,999 27 2.6 97.1 36 2.6 96.8 35,000 or more 30 2.9 100.0 44 3.2 100. Don't know - - 40 - Not ascertained - - 119 - TOTAL 1046 100.0 1519 100.0 aPercentages exclude the categories "don't know" and "not ascertained." 49 TABLE 3.--Characteristics of the respondent Study Characteristic Sample SRC Sample N % N % Sex Male 463 44.3 654 43.1 Female 583 55.7 865 56.9 Age 18 to 34 418 40.9 582 38.3 35 to 54 333 31.8 488 32.1 55 to 64 151 14.4 221 14.5 65 to 94 144 13.8 219 14.4 Racial or Ethnic Group White 914 87.4 1298 85.5 Black 94 9.0 159 10.5 Chicano, Puerto Rican, Mexican- or Spanish-American 23 2.2 33 2.2 American Indian 6 .6 6 .4 Oriental 6 .6 8 .5 Other 3 .3 7 .5 Not ascertained - - 8 .5 Education Up to 8 grades 176 16.8 263 17.3 9 to 11 grades 125 12.0 202 13.3 High school diploma 355 33.9 504 33.2 Some college, no degree 209 20.0 312 20.4 Bachelor's level degree 132 12.6 166 10.9 Advanced degree 49 4.7 62 4.1 Occupation Unemployed, retired, disabled 171 16.3 256 16.9 Housewife, widow 240 22.9 331 21.8 Laborers and service workers 107 10.2 174 11.5 Operatives 86 8.2 130 8.6 Craftsmen and foremen 75 7.2 115 7.6 Clerical and salesworkers 136 13.0 204 13.4 Managers 113 10.8 156 10.3 Professionals 118 11.3 150 9.9 Not ascertained - - 3 .2 NOTE: 1519 for the SRC sample. N = 1046 for the study sample and N 50 in the United States, excluding Alaska and Hawaii, which include no adults other than the husband and wife or the one-adult head. Description of Variables Income Adequacy Variables The independent variables in this study are sub- jective family income adequacy, objective family income adequacy, and the congruency or agreement between objec- tive and subjective income adequacy. Subjective family income adequacy is developed from the question "Do you feel that your total (family) income is enough for you (and your family) to live as comfortably as you would like at this time?" The four response options are "very comfortably," "comfortably," "not too comfortably," and "not at all comfortably." Thus, while the study analyzes subjective income "ade- quacy,” the scale by which respondents judged their sub- jective income speaks of "comfort." Does the word "comfort" lead respondents to indi- cate a lower level of income adequacy than might a less luxurious sounding average income term? It does not seem to do so, judging by study data. The modal value in the study is "comfortably," the response of 58.6 per- cent of the sample. The next largest category is "not too comfortable," 25.6 percent of the sample. Hence, 51 a "comfort" based scale of measurement does not appear to have depressed the distribution. The distribution of the study sample by subjective family income adequacy is given in Table 4. TABLE 4.-—Subjective family income adequacy distribution ..... I“::::.2§:§::CY Number Percentage 3232;32:32 SA-la Not at all adequate 62 5.9 5.9 SA-Z Less than adequate 268 25.6 31.5 SA-3 Adequate 613 58.6 90.2 SA—4 Very adequate 103 9.8 100.0 TOTAL 1046 100.0 aSA is subjective family income adequacy. The second income adequacy variable is objective family income adequacy. Unlike subjective family income adequacy, it is not based on the respondent's perception. It is a computed welfare ratio which represents the relation of family income to the income that family unit would need to live at the moderate level of living envisioned by the designers of the BLS standard budget. The objective adequacy ratio was developed through a series of steps. First, each respondent is assigned a BLS family composition equivalence number which adjusts the BLS standard budget at the moderate level, $12,626, 52 for differences in income needs based on differences in the number and ages of people in the family unit. For example, one person under the age of 35 and living alone is assigned an equivalence number of 37, indicating that a person needs 37 percent of the income required by a family of four, with a head age 35-54 and the oldest child age 6-15, to achieve an equivalent level of living. Secondly, each respondent is assigned a BLS geo- graphic equivalence number to adjust the moderate level standard budget for regional differences in income needs based upon climate and life styles. For example, a respondent living in Boston is assigned a geographic equivalence number of 118, indicating that that family needs an income of 118 percent of the 0.8. average standard budget income level to attain the same moderate level of living. The objective family income adequacy variable is then calculated by the following equation: Objective Adequacy = __ Total family income in 1973 Ratio Standard Adjustment Adjustment for budget; X for family X geographic moderate composition region level 100 100 The resulting ratios are grouped into six ordinal categories of objective family income adequacy. These categories, their descriptions, and frequencies are reported in Table 5. Just under 30 percent of the 53 sample have objective adequacy ratios of less than 100, indicating that they do not have a sufficient family income in 1973 to attain the level of living represented by the BLS standard budget at a moderate level of living. The middle groups in the sample, those with ratios of 100 to 299, are 68.6 percent of the total. They have from one to just under three times the income needed to attain the level of living represented by the BLS moderate level standard budget. TABLE 5.--Objective family income adequacy distribution Objective Cumulative Group Adequacy Number Percentage Ratio Percentage OA-la 16-49b 79 7.6 7.6 OA-2 50-99 233 22.3 29.8 OA-3 100-149 247 23.6 53.4 OA-4 150-199 180 17.2 70.7 OA-5 200-299 186 17.8 88.4 OA-6 300 or more 121 11.6 100.0 TOTAL 1046 100.0 aOA is objective family income adequacy. bThe ratio is obtained by dividing total family income by family needs adjusted for family composition and geographic region, The distributions of objective family income adequacy ratios and of total family income in 1973 are listed by percentiles in Table 6. A caution in reading the table is that it should not be assumed that the 54 family which is at the 50th percentile of the objective income adequacy is the same family which is at the 50th percentile of the family income distribution. TABLE 6.--Objective adequacy ratios and total family income distribution by percentiles of the study sample . Objective Total Family Percentile Adequacy Ratio Income in 1973 5th 43 Under $2,000 10th 57 $ 3,000-$3,999 15th 71 4,000- 4,999 20th 81 4,000- 4,999 25th 91 6,000- 7,499 30th 100 7,500- 8,999 35th 111 7,500- 8,999 40th 122 9,000- 9,999 45th 131 10,000-10,999 50th 143 l0,000-10,999 55th 151 11,000-12,499 60th 164 12,500-14,999 65th 179 12,500-14,999 70th 198 15,000-17,499 75th 217 15,000-17,499 80th 241 l7,500-l9,999 85th 272 20,000-22,499 90th 319 22,500-24,999 95th 405 30,000-34,999 100th 988 35,000 or more The objective adequacy ratio is a measure of family income divided by family income needs. A large family with an income at the 50th percentile would have an objective adequacy ratio which is less than that of a one-person family unit which is at the 50th percentile of the income distribution. Table 6 indicates that the median objective ade- quacy ratio for the study sample is 143, or 143 percent of 55 the income needed to live at the moderate level of living represented by the budget standard. It also shows that the median income of the study sample family units is $10,000 to $10,999. Seventy percent of the families in the study sample have incomes large enough to live at or above the BLS moderate level standard budget. The third income adequacy variable, congruency, is a measure of the degree of agreement and direction of disagreement between a respondent's objective and subjec- tive family income adequacy. Congruency groups are identified and coded by the matrix of objective and sub- jective adequacy groups shown on page 56. As Figure 2 illustrates, OA-3, OA-4, and OA-S are treated alike in developing the congruency groups. This is because post hoc contrasts during the early stages of data analysis show these three groups do not differ significantly from each other in their mean scores on the satisfaction variables. The congruent group in the congruency variable is C-3. C-2 and C-4 are one degree away from being con- gruent. C-1 and C-5 are two degrees away from being con- gruent. C-1 and C-2 differ from the congruent group in that their subjective adequacy is higher than their objective adequacy. They could be termed worse off than they think they are. C-4 and C-5 differ from the congruent group in that their objective adequacy is 56 momswmcm msoosw MHHEMM 0>Huommnsm can m>Huommoo mo mam>ma Mamba an omwmwucocfl mmsoum mocmsnmcou .N .mfim muo «no Nuo «no H:o Hno mumsemem sum> elem euo muo muo muo «no Huo mumswmue mumm muo «no VIU vlu mlu «no mumsvoom ems» mmmq Nuem muo muo mno mao euo mac mumeeoem Ham #m #02 Hldm mmmloom mmmloom mmaloma mvalooa mmlom melma owumm oaumm oaumm oaumm owumm oaumm oldo mldo ¢I€O mldo NIdO Aldo homsvmca mfioocH m>wuomnno mo Ho>mq an mmsouu Momsvmcd mEoocH m>euommnsm «0 Hm>mq we mmsouo 57 higher than their subjective adequacy. They could be called better off than they think they are. The frequency distribution of the congruency groups is reported in Table 7. Over half of the study sample, 54.0 percent, are in the congruent category. Their objective and subjective family income adequacy levels are in agreement. The remaining respondents are about equally distributed between those who differ from the congruent group, C-3, in that their subjective income adequacy is greater than their objective income adequacy, and those that differ in the opposite direction. TABLE 7.--Congruency group frequency distribution Group Description Number Percentage C-l SA>>0Aa 29 2.9 C-2 SA > CA 202 19. 3 C-3 SA==OA 565 54.0 C-4 SA< 0A 219 20.9 C—S SA<ma cosmoamwcmwm n .Eooooum mo mmmummo mwoam niooo.v moo.m- oiooo.v Hmm.o- crooo.v Hme.e- Aauaaaocoouoo omumaucumnm moonwaum> omaoom NNm. Nmm.m hco.a vam.m mvm.H mnm.v wmh.H whN.v GOHflMH>wO ILIIILIII CM 0: ouaocoom oiooo.c amH.au oxooo.c mme.au olooo.o sea.on impeaaoooouoo oaommucuuum moosmwum> omaoom coo. o~a.o cam. mam.m mom.H Nmm.¢ mNm.H Hmm.m GoduMa>mo oumosmum com: oxooo.o. mom.eH oiooo.o oom.oau oiooo.o ooo.au impedanmoouoc oauumumumdw. mmUGMHHM> UmHoom boa. mmm.m OHH.H omh.v mmm.H mmm.m mam.H hhd.~ soauma>mo oumcsmum GM”: vndm can mtflm mldm can Nldm mldm 6am Alfim omummupsoo mmdouu wumswmom mu0> cucm muooomoc muam mumsqmom can» moon Nudm mumswmom Ham um #02 Hlflm moosmummmwo mdouo muaaodo moan Hamnm>o po>fimoumm cues scauommmaumm cowumadmsou mo Hm>mq ow>wmoumm spas coauomwmflumm mfioosH mflwfimm om>wmoumm snag sowuommmflumm mammamsd mo mama moanmwum> coauommmHuMm so Momsqovm 0>Huommnsm mo muommmm mo mummnusoo Don “mom can .msowumfl>mp pumpsmum .mGMOZII.HH mamdfi 78 family income, level of consumption, and overall life quality increases as level of subjective income adequacy increases. In regard to objective income adequacy, H5, the results are less clearcut. Therefore, the relationship of objective adequacy to each of the satisfaction var- iables will be handled individually. Satisfagtion with Perceived Family_Income For perceived family income, satisfaction mean scores increase as the objective adequacy ratio increases. However, for groups OA-3, OA-4, and OA-S, the difference in mean scores on satisfaction with family income is less than for the groups with lower objective adequacy ratios, or for those with the highest objective adequacy ratios, as shown in Table 12. This lesser difference between objective adequacy group mean scores on satisfaction with family income becomes more apparent when the groups are contrasted and t-tests of significance are applied. Contrasts between groups with objective adequacy ratios of 100 to 299 have a lower level of significance or are not significant. The comparison of OA-3 and OA-4 is significant at the .040 level; the comparison of OA-4 and OA—S is not sig- nificant. Contrasts between groups with lower adequacy ratios, OA-l with OA-2 and OA-2 with OA-3, are 79 ammo. .I A“ V .Eopomnm mo mmmummc ovoa "Hm>ma oosmoamwsmflmo n .msouo mnu mo muwnfime mo owumu momsqmom msooca m>wuomnno map mum .mvan ma Hl coaoom goodneum> omaoom QOUGMflum> pmaoom mmo.H ham.m mam. mab.m mHo.H hoa.m HNH mmmnoom mndo mmo.a mom.m moo.a Hom.m mm~.H mmm.¢ mma mmmuoom mndo mva.a Ham.m hoo.a mmm.m 5mm.a mmm.v omH mmanoma undo Noa.a mmm.m mvo.a NmH.m mmv.H mam.v new moanooa mudo Ham.a mmm.m ham.a mv>.v vmv.a omm.m mmm mmuom mudo mom.a Hum.o omm.a Hoa.¢ How.a oma.m an cmolwa H140 amwwmmwmm aams amwwmmwmm aaas ammwmmwmm aama maoaauaaoaa cacao wuaamso mqu soflumadmsoo mo QEoocH haflemm mommo mam» and Hamum>o cm>Hmoumm Hm>oq cm>wmoumm om>flmoumm mo mm wwma Suez cowuomMmHumm sues coauommmwumm saw: cowuommmwumm Hmnfisz moanmwum> sowuommmflumm so momswwom m>auomnno mo muommmm mo mummuusoo no: umom can .mGOfiumw>mo oumocmum .msmmzln.ma mamas 80 significantly different at the .000 level of probability. The contrast between the two groups with the highest adequacy ratios, OA-S and OA-6, is also significant at the .000 level of probability. Thus, the effect of objective income adequacy on satisfaction with perceived family income is greatest at the extremes of the distribution. Very high and quite low objective adequacy ratios make a significant dif- ference in satisfaction. Differences in satisfaction mean scores within the middle objective adequacy groups are consistent in direction but are smaller in magnitude and are sometimes not significant. Because mean scores on satisfaction with family income are less significant for OA-3 and OA-4, and not significant for OA—4 when contrasted with OA-S, these three groups are combined for two additional contrasts. The contrasts of OA-2 with OA—3,4,S and of OA-3,4,5 with OA-6 are significantly different at the .000 level of probability. This produces four groups which are sig- nificantly different at the .000 level of probability. These groups are OA-l, OA-2, OA-3,4,5, and OA-6. In summary, with regard to HSa' the results of analysis show that satisfaction with perceived family income increases as the level of subjective adequacy increases. However, the difference in satisfaction mean scores is significant for only four of the five compared 81 pairs of objective income adequacy groups. Difference in satisfaction is slight for those with objective adequacy ratios of 100 to 299. By combining three of the objective adequacy groups which have quite similar satisfaction mean scores, four levels of objective family income adequacy which differ significantly in satisfaction with family income mean scores are identified. Satisfaction with Perceived Level of Consumption For perceived level of consumption, satisfaction mean scores increase as objective adequacy ratios increase with one exception. Groups OA-4 and OA—5 had satisfaction mean scores which are essentially the same. The results of contrasting adjacent objective adequacy groups by their mean scores on satisfaction with level of consumption show a pattern quite like the con- trasts of the same groups by their mean scores on satis- faction with family income. Differences in means scores on satisfaction with level of consumption are significant at the .000 level for those groups with low objective adequacy ratios, OA-l with OA-2 and OA-2 with OA-3. Those groups with the highest objective adequacy ratios, OA-S and OA-6, are significantly different at the .002 level. Contrasts involving the middle groups on objective income adequacy do not reach the .05 alpha level selected by this study. The contrast of OA-3 with OA-4 has a 82 probability of .191 and the contrast of OA-4 with OA-S has a probability of .822. Again, the effect of objective income adequacy on a satisfaction variable is greatest at the extremes of the objective adequacy distribution. Very high and quite low objective adequacy groups, when contrasted with their adjacent objective adequacy groups, do differ significantly in satisfaction with perceived level of consumption. Two additional contrasts are implemented to com- pare the combined group, OA-3,4,5 with objective adequacy groups OA-2 and OA-6. These contrasts are significant at the .000 level. This produces four groups which are significantly different at the .000 level of probability. These groups are OA-l, OA-Z, OA-3,4,5, and OA-6. In summary, with regard to H5b’ the results of analysis show that satisfaction with perceived level of consumption increases as objective income adequacy increases, with one exception. Groups OA-4 and OA-S are very much alike in satisfaction mean scores. The difference in satisfaction mean scores is significant for three of the five compared pairs of adjacent objective adequacy groups. Differences in satisfaction are not significant for those with objective adequacy ratios of 100 to 299. By combining three objective adequacy groups which have similar means scores on satisfaction, four 83 levels of objective family income adequacy which differ significantly in satisfaction with level of consumption are identified. Satisfaction with Perceived Overall Life Quality For perceived overall life quality, combining six objective adequacy groups into three levels of ade- quacy provides all the differentiation of income adequacy needed to explain variation in satisfaction with overall life quality. Those with the lowest objective adequacy ratios have the lowest mean scores on satisfaction with overall life quality. Those in groups OA-Z, OA-3, and OA-4 have similar mean scores on satisfaction with life quality; and those with the highest objective adequacy ratios, OA-5 and OA-6, have similar satisfaction mean scores. Clustered into these three levels of objective adequacy, OA-l, OA-2,3,4, and OA-5,6, satisfaction with perceived overall life quality increases as objective adequacy increases. However, within the clusters OA-2,3,4 and OA-5,6, the pattern of increasing satis- faction with increasing objective income adequacy ratios is not present. For consistency with the analysis of the two other satisfaction variables, the same set of contrasts is implemented here. As before, the two groups with the lowest objective income adequacy ratios, groups OA-l and 84 OA-2, differ at the .000 level of probability. The one other contrast which is significant is that between groups OA-4 and OA-S, significant at the .015 level. The con- trast between the two highest groups, OA-S and OA-6, has a probability of .879. Thus, the dividing line in dif- ferentiating moderate from high satisfaction scores by objective adequacy groups has shifted down to the dif- ference between groups OA-4 and OA-5. For the other two satisfaction variables, family income and level of con- sumption, that dividing line between middle and higher objective income adequacy has been between groups OA-S and OA-6. The dividing line between low and middle objective income adequacy ratios, with regard to their satisfaction with overall life quality, has also shifted downward. OA-2 is significantly different from OA-3 for the other two satisfaction variables, family income and level of consumption. Here the dividing line between low and middle objective adequacy groups in their satisfaction with life quality is between OA-l and OA-2. The mean scores on satisfaction with life quality are quite similar for groups OA-2 and OA-3, being 5.223 and 5.389, respec- tively. Contrasts of OA-2 with OA-3 and OA—3 with OA-4 show that these paired groups do not differ significantly from each other. 85 As with the other satisfaction variables, two additional contrasts are performed. For consistency, groups OA-3,4,5 are clustered and then contrasted with OA-2 and OA-6. The contrast of OA-2 with OA-3,4,5 is significantly different at the .018 level; the contrast of OA-3,4,5 with OA-6 is not significant. Thus, the set of four groups identified here are significantly dif- ferent in two of three cases. The four groups and their probabilities are shown in Table 13. TABLE 13.--Probability of difference in mean scores on overall life quality for adjoining objective adequacy groups . . Pooled Variance Objective _ . . Group Adequacy Ratios (Pgogzgiiityfa OA-l 16-49 OA-2 50-99 -3.640 (.000)b OA-3,4,5 100-299 -2.363 (.018)b OA-6 300-988 -1.300 (.194) aT-statistic and probability indicated for con- -trasts of that group with the one immediate preceding it in the table. bSignificance level: p S .05. In summary, with regard to HSC' the results of analysis showed that satisfaction with perceived overall life quality increases as objective income adequacy ratios increase if the objective adequacy groups are 86 clustered into low, medium, and high adequacy ratio cate- gories rather than left as six more differentiated cate- gories. There is a significant difference in satisfaction with overall life quality between those with adequacy ratios which are low, OA-l, and those with moderate ade- quacy ratios, starting with group OA-2. There is also a significant difference in mean scores on satisfaction with overall life quality between those with moderate adequacy ratios and those with higher adequacy ratios, the dividing point being between OA-4 and OA-S. Strength of association. Having identified sub- jective and objective adequacy groups which differ sig- nificantly in mean scores on satisfaction variables, the next research question is: How much of the variation in satisfaction mean scores is explained by groups who differ in objective and subjective family income adequacy? Table 14 provides a summary of the contributions of objective and subjective adequacy to explaining the variance in satisfaction variables. The individual contribution of each factor is shown by eta square. The contribution of subjective adequacy to explaining differences in mean scores on satisfaction, as determined by eta square, is .35 for family income, .32 for level of consumption, and .11 for overall life quality. Thus, 35 percent, 32 percent, and 11 percent 87 TABLE l4.--Strength of association between income adequacy and satisfaction Satisfaction with Perceived: Family Level of Overall Income Consumption Life Quality Subjective Income Adequacy Etaa b .59 .47 .33 Eta Square .35 .22 .11 Betac .54 .40 .31 Objective Income Adequacy Eta .35 .34 .20 Eta Square .12 .12 .04 Beta .12 .18 .10 Multiple R .601 .499 .340 Multiple R Square .361 .249 .120 aEta is an unadjusted score for an individual factor. b Eta Square is the equivalent of multiple R Square for an individual factor. cBeta is an eta score adjusted for the contri- bution of other factors. 88 of the variation in satisfaction scores is explained by subjective adequacy. The strength of association, or amount of variance explained, decreases as one moves from satisfaction variables more closely related with economic matters, family income and level of consumption, to the satisfaction variable which encompasses a wider range of life experiences and is therefore more diffusely related to economic matters, perceived overall life quality. The contribution of objective adequacy to explain- ing differences in mean scores on satisfaction, as deter- mined by eta square, is .12 for family income, .12 for level of consumption, and .04 for overall life quality. As before, the strength of association is greater for the two satisfaction variables more closely related to economic matters than it is for satisfaction with overall life quality. Note, however, that less of the variance in satisfaction mean scores is explained by objective adequacy than is explained by subjective adequacy. For example, in Table 14, the eta square statistic for satis- faction with family income was .35 for subjective adequacy and .12 for objective adequacy. A comparison of the beta and eta scores of the two income adequacy measures shows the relative effect of each of the other. Beta scores for subjective adequacy are somewhat smaller than eta scores, representing the adjustment for the other factor, objective adequacy. 89 The adjustment is from .59 to .54 for family income; from .47 to .40 for level of consumption, and from .33 to .31 for overall life quality. In contrast, the beta scores for objective income adequacy are more greatly reduced from the eta scores, representing the adjustment for the other factor, subjective adequacy. Thus, the unique contribution of objective adequacy to variation in satisfaction is less. This is shown by the changes from eta to beta of .35 to .12 for family income, of .34 to .18 for level of consumption, and of .20 to .10 for overall life quality. However, the magnitude of change from eta to beta scores is somewhat different for level of consumption than for the other two satisfaction measures. The adjust- ment from eta to beta is proportionately largest for level of consumption among the three sets of subjective adequacy statistics. And the adjustment from eta to beta is proportionately smallest for level of consumption among the three sets of objective adequacy statistics. It appears that the contribution of objective adequacy has greater uniqueness in its contribution to level of Consumption than in its contribution to the two other Satisfaction measures. Likewise, subjective adequacy has less uniqueness in its contribution to level of cOnsumption than in its contribution to the other two Satisfaction measures . 90 The final set of statistics, multiple R and multiple R square, provide a summary measure of the total contribution of objective and subjective adequacy, taken together, to mean scores on the satisfaction lneasures. Again, the size of the contribution of income adequacy measures to satisfaction scores decreases as <3ne moves from those most directly related to income to -those less directly related to income. The multiple R square decreases from .36 for family income to .25 for level of consumption to .12 for overall life quality. Congruengy and Satisfaction One-way analysis of variance is used to test the :relationship between congruency and each of the three asatisfaction variables. The congruency groups are :identified by the degree of agreement and direction of (disagreement of their objective and subjective adequacy :scores. Group C-3 is the congruent group, its members identify themselves with a subjective adequacy group “filich is equal to the group in which they are placed by tflaeir family's objective adequacy level. Groups C-1 and C-2 place themselves in higher subjective groups t"ham their placement on the objective adequacy measure- Imentn They might be called worse off than they think they are. Groups C—4 and C-5 place themselves in sub- jective adequacy groups which are lower than their 91 objective adequacy levels. They could be called better off than they think they are. The hypotheses to be tested are: H6: There is a difference in satisfaction with per- ceived family income for those who differ in the congruency of their objective and subjective family income adequacy. H7: There is a difference in satisfaction with per- ceived level of consumption for those who differ in the congruency of their objective and sub- jective family income adequacy. H8: There is a difference in satisfaction with per- ceived overall life quality for those who differ in the congruency of their objective and sub- jective family income adequacy. The results of one-way analysis of variance, as reported in Table 15, show that at least one congruency group differs from the others in the mean scores on satis- faction with family income, level of consumption, and overall life quality. Thus, H6' H7, and H8 are all sup- ported. Congruency group differences are significant at the .000 level of family income and at the .001 level for level of consumption and overall life quality. Since there are significant differences between congruency groups on satisfaction mean scores, the next hypothesis deals with which groups differ: H9: The congruent group will have the highest mean scores on satisfaction with family income, level of consumption, and overall life quality. 92 TABLE 15.--One-way analysis of variance for effects of congruency on satisfaction variables Source of Variation Sum of Mean df Squares Squares F-Statistic (Probability) A. Satisfaction with Perceived Family Income Between Groups Within Groups Total 160.958 4 2080.175 1041 1.998 2241.133 1045 40.240 B. Satisfaction with Level of Consumption Between Groups Within Groups Total 27.223 4 6.806 1487.913 1041 1.429 1515.136 1045 C. Satisfaction with Overall Life Quality Between Groups Within Groups Total 33.624 4 8.406 1439.095 1041 1.382 1472.719 1045 20.137 (.000)a 4.762 (.001)a 6.081 (.001)a aSignificance level: p E .05. 93 Mean scores on the satisfaction variables and post hoc contrasts are implemented to test the hypothesis. The results are reported in Table 16. H9 is rejected. The congruent group, C-3, had the highest mean score on only one satisfaction variable, level of consumption. That mean score of 5.197 was not significantly different from the mean scores of 5.193 for C-2 and 5.172 for C-l on satisfaction with level of con- sumption. Differences in satisfaction mean scores for congruency groups C-1, C-2, and C-3 were not significant for any of the three satisfaction variables. Differences between the congruent group, C-3, and the incongruent groups who are described as better off than they think they are, C-4 and C-5, however, are significant at levels of .002 and .000. The congruent group has higher satisfaction scores than do groups C-4 and C-5 on all three satisfaction variables. The findings show that when subjective adequacy is lower than objective adequacy, satisfaction mean scores are lower. The reverse situation, lower objective adequacy than subjective adequacy, does not result in lower satisfaction mean scores. In fact, those with lower objective than subjective income adequacy have mean scores on all three satisfaction variables which do not differ significantly from the mean scores of the congruent group, C-3. Thus, subjective judgments of .mo. w m we Hm>ma cosmowmwsmwmo .Soommum mo momnmmo Heoan .smsu mmma nose osm .smsu mmma umsszOm .ou Hmsvm .smsu Hmummum umssmEOm .swsu Hoummum nose "mum mlu ou HIU How msmflm one .mosmsnmsoo me 0 use momsqmom m>wuomnno me so .momsvmom m>wu00nn5m ma 4mm 94 oooo. mma.v oaoo. mov.m oooo. mam.b m.¢uo osw m.m.ano omoo. nmo.m ooao. 5mm.~ oooo. mvm.m «no use m.~.HIU oomo. mmo.~ mod. Hmm.a oooo. ¢m>.m mlu osm v10 oooo. mmm.m omoo. mmo.m oooo. mmm.m muo osm muU oaoo. mmH.m oaoo. ma~.m oooo. mom.m v10 use mlo ham. mew. mum. mmo.| mNH. mam.al mnu osm ~10 Hen. Ham. mam. moa.l mom. mmo.a mlu osm H10 Amoaaanonouoc AsuaHmmmnmuoc isomHaommoumc oflumwumumlmp owummumumnm owumwumumHMWx omummuusoo mmsouo nmosmasm> omaoom smosmflum> omaoom noosmflum> omaoom mmn.a mvm.e HH¢.H mam.v HN>.H nom.~ Hm s0vv4m ammo mmm.a mHH.m mm~.H omm.e mom.a wmm.m mam s0v4m “elo mma.a mav.m ema.a hma.m th.a Nmm.v mom so 0 4m “muo th.H omv.m mm~.a mmH.m hav.H mnm.¢ mom «oAsm ”mlo mmm. va.m eoH.H NhH.m eoH.H mmm.v mm msoAAsm ”H10 ammwmmmmm aams amwwmmmmm saws ammwwwwwm saws aaoaasauoaa aaouo muflamso owes sowumssmsoo mo msoosH mHHEmm mommo mmsouw Hamum>o om>wmoumm Hm>mn om>wmonom ©m>floosmm mo mosmsumsou suHB sOfiuommmHumm sues sofluommmwuwm suflS sowuomwwwumm Hwnfisz moanmflum> sowuommmqum so mosmsum Isoo mo mpommmm mo mummuusoo Don umom use .msOHHMfl>mw pumosmum .mmnoom somZII.mH mqmde 95 income adequacy have a depressing effect on satisfaction scores when subjective adequacy is lower than objective adequacy. Lower objective adequacy than subjective ade- quacy, however, does not have a significant effect on satisfaction means scores. As a followéup, three additional contrasts were run. For those who are better off than they think they are, C-4 and C-5, the degree of better-offness is sig- nificantly different at the .000 level for family income, at the .036 level for overall life quality, and is not significant for level of consumption. The three congruency groups which do not differ significantly in satisfaction mean scores, C-1, C-2, and C-3, are clustered and contrasted with C-4 and then with C—4,5. For all three satisfaction variables the cluster of C-l,2,3 is significantly different from C-4 and C-4,5. These contrasts provide additional support to the findings above that those who have lower subjective than objective adequacy scores differ from the remaining three congruency groups and that their satisfaction mean scores are sig- nificantly lower. In summary, congruency, in itself, does not appear to be the criterion on which groups differ. Only when income adequacy variables differ in the direction of lower subjective adequacy than objective adequacy do mean scores on the three satisfaction variables drop. 96 Subjective judgments of income adequacy are important in that when one feels one's income adequacy is lower than the BLS-based objective adequacy ratio defines it to be, one's satisfaction with family income, level of consumption, and overall life quality is lower. While congruency groups C-4 and C-5 do differ significantly from C-l,2,3, the amount of variation in satisfaction mean scores is very small. The eta square statistic, which measures strength of association, is .0052 for family income, .0003 for level of consumption, and .0005 for overall life quality. This is very close to zero explanatory power. Thus, congruency groups provide an example of a variable which is significant but is not important. The importance is in the indi- vidual variables, objective adequacy and subjective adequacy, not in the degree of their congruency or the direction of their incongruency. Contextual Variables One-way analysis of variance and nonparametric models are used to relate objective and subjective family income adequacy and congruency to contextual variables. The purpose of the analysis is to describe the membership of the objective and subjective adequacy and the con- gruency groups. For example, how do respondents in dif- ferent subjective adequacy groups differ in education, number of children, and current concerns? 97 The contextual variables are of two types, demo- graphic characteristics and economic perception. The hypothesis tested is: H10: There is a difference among (a) subjective family income adequacy groups, (b) objective family income adequacy groups, and (c) con- gruency groups with respect to: (i) demographic characteristics: -number of adults, age of respondent-head, number of children, age of oldest child, and total family income of the family unit; housing status and urbanization of resi- dence -sex, racial, or ethnic group, education and occupation of the respondent (ii) economic perceptions: -current concerns, material wishes, con- sumption achievement, and recent and intergenerational financial progress For each of the income adequacy variables, results of the analysis are reported first for demographic char- acteristics, then for economic perceptions. The demo- graphic variables might be termed objective variables in that they are observable facts. The perception of the person is not an integral part of such measures as chronological age and housing status. Those contextual variables which are called economic perceptions, on the other hand, are the respondent's perceptions regarding selected economic matters. Subjective Adequacy Groups Demographic characteristics. Each subjective adequacy group differs at a probability level of less 98 than .001 from all three other subjective adequacy groups on satisfaction mean scores on perceived family income, level of consumption, and overall life quality (Table 11). The subjective adequacy groups also differ in a number of their demographic characteristics and in their eco- nomic perceptions. Five demographic characteristics of the respon— dent's family units are entered as dependent variables in one-way analysis of variance for the factor, subjective family income adequacy. All five characteristics are significant, as reported in Table 17. The strength of association as measured for the sample by eta square is highest for family income and number of children, at .123 and .121 respectively. Lower strengths of associ- ation occur for age of oldest child, .036, and age of the respondent head, .023. The remaining characteristic, number of adults in the family unit, has a very low strength of association, .008. The four subjective adequacy groups differ in that, as subjective adequacy increases, (1) number of children and the age of the oldest child decreases, (2) the age of the respondent-head and total family income increases, and (3) the number of adults in the family unit increases slightly. Those subjective adequacy groups which are significantly different at the .05 level, as determined by post hoc contrasts, are identified in Table 18. .mo. w m "Hm>wH cosmoflmwsmflmo .mosmwum> mo mammamsm mmzlwso Eoum oaumflumumlm ms» mo upwawnmnoumn .homsvmcm mEoosfl m>wuomnn9m ma 4mm 99 mmv.oso aaa.ao eam.om oem.eo .>0o .oom oooo. .coso msoocH aas.asa moo.msm moo.am moo.am :80: mag. Noam season so.ss Ho.os aa.ms em.ms .>0o comm .oum oooo. .ooum uncoocOAmmm oa.mo om.~e No.88 o~.om com: mNo. meow so was as. me. we. we. .>0o .oum demo. .nouo mosses ao.a oa.a mo.a oo.a c802 moo. «cum mo umnsoz om.m mo.o Nm.m om.m .>oo oaacu .oum oooo. .nosm ommoso mo.m Hm.e ma.e mm.o cums omo. «mum mo mos ma. mm.a oe.a om.s .>0o .oum oooo. n.noso cmuoaaco as. mm. oH.H ~a.a s80: HNH. «mom so Honsoz mos m 2 mac n 2 New w 2 mu m z mums mos mums mos mums 00¢ moanmwum> >um> mumsqmcm sass mmmq Has us uoz moaumwumum Hoopxmusou anew mncm «lam causm mOflumHHmuomumso owsmmnmosmo mum scans moanmflum> Hmsuxou Isoo How moaumflumum psm .msOfluMfl>mp pumoswum .msmmE msoum womsvmom m>wuomnnsmll.ha flames 100 .mo. w m ”Hm>mH cosmoamwsmwmm onoo.v maaoo.v Aooa.v mfiamo.v mamoo.v moo.m| mmm.m| hem.a nom.~ hoa.m ensm w mlam mfiooo.v Aaea.v mamao.v Ammm.v mavao.v mvm.m| mnv.a mmm.~| mmm. mev.m mnsm a Numm Aonm.v maooo.v Aavm.v mamv0.v mamoo.v voa.al mmv.m1 omv.l Noo.m who.m Nusm w diam Aummulav “ammunav Aummulev summulev Aummulev mossesm> mosmfinm> mossesm> oosmwum> wosmaum> cmaoom omaoom omaoom pwaoom omaoom omummnusou OEOOGH Umwm UHHSU mQSOHU maflemm Ipsoosommmm Onwwsos umuoao Mommawso HMHOH. MO mm¢ M 32 MO $04 N g2 mofiumwsmuomnmso oasmmumofimp so museum womsqmom m>wuomn95m smwzumn mosmHmMMHU on» mo mummuusoo oos umom||.ma mamas 101 Nonparametric statistics, the chi square and Cramer's V, are used to test the relationship of the remaining demographic characteristics to subjective adequacy. The results are reported in Table 19. The characteristics of race, education, occupation of male and female respondents, urbanization of residence and housing status are all significant. One characteristic, the sex of the respondent, had a probability of .112 on the chi square statistic. The strongest association between demographic characteristics and subjective adequacy, as measured by Cramer's V, is that with the occupation of male respon- dents. The association is .254 on a scale where zero is no association and 1.00 is a perfect association. The next highest strengths of association are race, at .182, and housing status at .174. 'Strengths of associ- ation in the .100 to .144 range are shown by urbanization of residence, occupation of female respondents, and edu- cation, as shown in Table 19. The four subjective adequacy groups differ in that (l) the two highest subjective adequacy groups are racially more white, (2) home ownership increases as subjective adequacy increases, (3) with the exception of SA-2, more males are in white collar occupations and more females are housewives as subjective adequacy rises, (4) educational level rises as subjective adequacy rises, l()2 a cue: moons sH omosuuusdwpms ms» m>ond .ouoa so ooo.om mo soflumHsdom omosu mum mmsouo oao omosu mun mmsouo (m How wouuomwu nmomusooummo .homsqopm o>wuoofinsm me am a .mo. w m "Hm>oa mossOMWMsofimm H.ww «pH. > m.uuaouu m.ea o.os m.am m.oe ocssom o u we museum Ho msso coco. oussqm wso msemsox ooa. > m.umsmuo mosopfimom m.~v m.ae m.~v pm.ov m.- AN u up no Hausa ammo. oumsvm «so sowusnwssnuo Ammm u zv ONH. > m.uoemuu musmpsommum H.mv o.~v «.mm m.ov ~.Hv AN 8 up causes no awesomso: smoo. mumsvm «so sOHummsooo Ammv I zv vmm. > m.uuemuu musuosommmm v.mm v.mv m.w~ o.m~ m.~v ma u up can: no umaaoo muss: mooo. mumsvu H50 sOwuwmsooo m.mm eve. > u.nwsduo m.~m o.av m.v~ on.om msonwo ma 0 up usovsommmm Hoosom soar Mooo. cannon “so no sofiusosom Nae. > m.uufiuuu m.ao m.am s.ms ¢.hh v.sm w u up usopsommom cuss: cooo. mucosa sso so once one. > m.uusmuu m.mv o.vm ~.aw h.mm h.mm m u up usopsommum mamsmm NHH. mumsvm «no mo xom moa w 2 man u mom N mm W was: and mum: on< «ya: out muo> couscous sass mums Add as uoz omuusooumm manoaum> vuau00nnsu How moanuwuo> Husuxousoo so mowuwHuuuu pad .uomausoouum .noooznl.ma Manse 103 with the exception of group 2, and (5) urbanization of residence rises as subjective adequacy rises, with the exception of SA-Z. Economic perceptions. Differences in economic perceptions variables by subjective adequacy groups are all statistically significant. The strength of associ- ation ranges from .257 down to .148, as illustrated in Table 20. The strongest association is between subjective adequacy and consumption achievement. If the respondents have met their past consumption goals, or expect to do so in the future, they are in a higher subjective ade- quacy group. Material wishes and current concerns have strengths of association with subjective adequacy of .207 and .198, respectively. Those with fewer material wishes, and those whose first ranked current concern is not financial security, subjectively judge their family income to be more adequate. . The strength of association between financial progress and subjective adequacy is .176 for recent financial progress and .148 for intergenerational financial progress. The four subjective adequacy groups differ in economic perceptions in that, as subjective adequacy increases (1) more respondents have consumption goals they have achieved, (2) the respondents have a lower 104 .uswnm oo>mwsou H0\osm omuommxm sues mmosullsmwome as» o>onm woos» mum mmsoum so MOM oouuomou momnusoouomo .ON GMSU ONCE HO mmhoom xmwflfl mwflmfl3 awesmums sues omosulnsmwooe ms» m>onm «mos» mum mmsosm sm sou couscous mommusoouomo .mossvoom o>auoonnsm no «mo .mo. w m "Ho>oa cosmowmwsmwmu Hmsowumuosom ovH. > m.HQEflHU IHGHQH o.oa ~.oo m.oa ~.mm m.ma o u no .uuouooum «mo souuom mooo. wussvm «so Huwosuswm ova. > m.n0&muo usooom «.mN m.mm m.sm m.om m.sc «a n no .uooumoum d0flquOHH09OD mcoo. Gunfivm H‘U HMfiOGflflHh a.mm emu. > u.uuscuo ~.~c H.H¢ m.o~ om.ea o0>0sno< m u so us0s0>0ano¢ uoz sooo. osmsvm wso soaumssmsoo o.~m aoN. > u.uusuuo o.oa c.a~ e.ec oo.Hm xmocH nus: «a n no mucus: om n ma uooo. cannon “no seasons: «.mm was. > o.u0suuo o.om m.oe ~.eo o.ss musesomm ma u no usuoosoo Hsdossswm dooo. cannon Hso usouuso mumswoos mumswmos mumsvoss ommusoouom _ mu0> couscous cons mums Add as uoz one muooouuo uosuusucum HMMMMMMMWm eusm mnam ~u¢m nausm moo: osqscm unsoum homsvoou o>wu Iowans» as» How moasmaum> soaumoouos owsosooo so uoquudusuu ass .uomsusooumm .umoozfl.o~ mamas 105 material wishes index score, (3) fewer respondents report financial security to be their first ranked current con- cern, (4) fewer respondents judge their recent financial progress to be negative, and (5) more respondents state they are better off than their parents had been at the same age (intergenerational financial progress). Objective Adequacy Groups Demographic characteristics. Many of the objec- tive income adequacy groups differ in their mean scores on satisfaction with family income, level of consumption, and overall life quality (Table 12). The objective adequacy groups also differ in a number of demographic characteristics and economic perceptions. Five of the demographic characteristics are entered as dependent variables in one-way analysis of variance with the factor, objective adequacy. All five are significant at or below the .003 level, as shown in Table 21. The strength of association, as measured by eta square, is by far the greatest for family income. It has an eta square of .563. Such a result is to be expected because family income is the numerator in the equation by which objective adequacy ratios were com- puted. Objective adequacy is the ratio of family income to the denominator, family needs, with needs determined 106 .mosmflum> mo mammamsm hasnmso Eosm owumwumumuh was NO muwawnsnoum .mo. w Q "Ho>oa mosMOflmwsmwmo a .homsvoos m>wuoonnsm ma sow mma.so Nao.oo moa.mo coo.eo ooe.mo see.ao .>0o .pum oooo. .nonm «EoosH me~.e~o moo.asm ems.mso omm.oso ome.oo eao.mo c802 mom. Noam season o~.os ~o.ms mm.ms He.as aa.ms oo.os .>0o mace .osm omoo. .noum -ocooso mom es.se oa.~e ao.se om.~c eo.me ao.se :80: use. Noam so was as. as. ea. He. cs. om. .>oo .oom oooo. .nouo mosses ~o.s aa.s om.s ma.s oa.s om.s can: aao. «new so soossz aa.m oe.m ea.m am.m om.c sm.o .>0o cssso .oum oooo. .aoud uuooso ee.a Hm.m mm.e a~.m so.m oo.o sou: omo. menu so cos as. ea. oo.a a~.H em.s om.s .>0o . .oum oooo. n.noud cuuosaso as. mo. no. s~.H om.s so.s com: Nos. Noun mo noossz Has u 2 cos n 2 cos n 2 sea u z mmm u 2 as n z moss oom ammuooa oasusms oms-oos mauom menoa mos as u mosoosso> oeumm owumm oaumm oaumm owumm owumm .u .u um Hmsuxousoo onso muco e- assuxousoo so mowumwumum osm .msowumw>oo osmosmum .mosoom smmE msoum momsvoom m>wuoonnoal.aw mqmsa 107 by the moderate level standard budget, adjusted for family composition and geographic region of the United States. The other demographic variables tested by analy- sis of variance are the four characteristics used to determine the family unit's BLS family composition adjustment number. In strength of association with objective adequacy, number of children is highest, with an eta square of .102: age of oldest child was next, with an eta square of .058. The association of objective adequacy with variables involving adults in the family unit is lower, being .029 for number of adults and .017 for age of the respondent-head. The six objective income adequacy groups differ in that, as objective adequacy increases, (1) family income increases, and (2) number of children and the age of the oldest child decreases. Those objective adequacy group differences which are significant at the .05 level, as determined by post hoc contrasts, are identified in Table 22. Contrasts are computed for objective adequacy groups OA-l, OA-2, OA-3,4,S, and OA-6, which are identified in earlier contrasts with satisfaction variables as being significantly different in eight of nine contrasts (Tables 11, 12, 13, and 16). Nonparametric statistics, the chi square and Cramer's V, are used to test the relationship of the 108 .momsvmpm mEoosfl waflfism o>fluommno ma 400 .Eoommum mo moonmmo ovoa a .mo. w m ”Ho>oa mossowmwsmwmm aioao.o alsoa.o aiaao.o alooo.o aiaoa.o ovo.mH| omm.m| vmm.m hHH.m ~Hm.m misc a m.v.mu.H owldo a Also “umouuev Aumoulao Rummunav humouusv naumoulav musMHum> mosmwum> mosMflum> mosmwss> mosswsm> ooaoom ooaoom ooaoos poaoom ooaoos omummsusoo mmsoso oEoosH comm oaflso Mawsmm lusoosommom omwwsos umooao MOMWHHso sauce mo mos m assz so was u nesz mowumwuouomumso owsmmumosop so masonm momsvmom m>euomnao somzums mosoHOMMHo may no mumsnusoo oos umomnu.mm mqmsa 109 remaining demographic characteristics to objective family income adequacy. All characteristics tested are significant at the .000 level, as reported in Table 23. Four characteristics have strengths of association of .21 to .27, as measured by Cramer's V. These are racial or ethnic group, education, occupation of male respon- ‘f’ '3'“? dents, and housing status. Strengths of association between .10 and .19 are found for sex of respondent, occupation of female respondents, and urbanization of 4 residence. The six objective family income adequacy groups differ in that, as objective income adequacy increases, (1) fewer respondents are female, (2) more respondents have more than a high school education, (3) more male respondents have white collar occupations and fewer female respondents are housewives, and (4) more respon- dents are white, with the exception of group OA-S. Economic perceptions. Differences in economic perceptions by objective adequacy groups are all sig- nificant at the .000 level, as reported in Table 24. The strengths of association are highest for two per- ceptions, material wishes and consumption achievement, at .148 and .163 respectively, as measured by Cramer's V. For most objective adequacy groups, material wishes index scores decrease and the number who have achieved their 1510 .ouoa Ho ooo.om mo sowumasmom m sues moons sw omosullsmwowe osu o>onw woos» mum pmuuomou mommasuouomo .uBonwo Hoosom smfls m smsu soauoosow whoa saw: omosUIIsm«ooE msu o>ons omosu mum pouuomou nommusoouomo .mo. w o “scans coccosuscosmo .uoosuoos o>wuoonao nu son a.oo mom. > u.uossuo o.oo H.ms m.~a H.oa e.sm o.a~ ossaom as m n no uduoum no usso nooo. oumsvm wso mswmsom mod. > u.sosuuo oosoowmum m.wm o.om m.mm H.ov m.vm om.we m.s~ on I up no Hausa naoo. ouosvn Hso sowuouwsdnua .mmm I 2. had. > u.uosduo nusoosommus v.mm «.mm v.em o.ne o.ov H.av ~.av ma I no mausom mo mo>wsomsom nooo. ounsvu «so scauossooo snow I zo new. > n.uosduo nusoosomuum o.mm m.~m m.mv H.Hv 5.H~ m.n m.~e an I no was: no Huaaoo cues: nooc. onusvu «so scauumsooo a.mm oa~. > n.uosduo usoosomuum m.mm o.hm ~.N¢ s.om a.a~ 0v.aa neodmfio ON I up no Hoocom nose nooo. mucosa ago cosponsom mew. > u.s08duo b.wm s.mw m.~m m.mm s.nm o.~w c.5m m I up usoosommus oases nooo. mucosa «no mo doom emu. > m.uusduo s.mn «.mm s.am s.om v.vm H.hm n.mm m I up osmosomous . masses nooo. oumsvu Hso mo xom +oom mawuoow amanoma avanooa «atom meted 0 onto mldo Vida mice «:40 waldo o nusoouom o ansdus> osu whooouuo noapmauuum assuxousoo hsomouso moo: onEdm sw msouo mo ommusoouos “moo: oamfism nmsoum 5onsuoos 0>Huoonno How moanmwun> ausuxousoo so uowuuwuuuu vs: .uomousoouom .novozll.mw Manda 111 .sHmm om>oflsom H0\osm omuommxo sues mmosunnsmwoms ms» w>osm mmosu mum mmsosm so you sophomou mommasmouomo .oN sass macs mo mouoom amps“ mmsmws Hawsmums spas mmosunnsmwoms ms» o>osm mmosu mum masoum so How couscous mommusmouomo .mo. 1 m "Ho>0H cosmowmwsmwm v a .aomsvmom o>wuooflso ma «on HmsoHumuosmm mNH. > m.uusouo nuwusH m.mm o.mn v.em o.am c.wh s.mm m.mh OH I up .mmmumoum mmo Houuom sooo. oussvm wso Hmwosmswh see. > m.uwsouo oedema m.Hv m.ov s.mm H.ve o.mm m.om m.ae oN u no .mmmumoum sowusuoHuouoo sooo. ohmsvu “so Hmwossswh h.mm mos. > m.uufimuo m.mm H.me v.mm o.mm «.mm om.- oo>owsos ow u no usmso>owsos uoz sooo. oumsvm Mso sowumasusoo m.~m med. > m.HoEMHo o.ma H.m~ h.m~ m.s~ v.ov ov.em xmosH sma3 cm I no mosmwz canes nooo. mucosa sso assumes: «.mm de. > m.HmEsHU o.me m.oe ~.~m m.om «.mm H.mc muauooom m~ u we usuoosoo Hmwosmswm sooo. oussvm wso usouuso +oom mamuooa masuoms aesuoos omuom aeuos mocucoouom coauooouoo owumm osumm oflums odysm owumm owumm osw muommumo mowumwumum owsosoom ouso mnco «use muse «use usuco woos oaosom . mssoum homsvoom o>auomnso usofiummouom oafiosoom so mowumwumum osm .ummmusoouom .mooozI|.v~ wands 112 level of consumption goals increases as objective ade- quacy increases. The strength of association with objec- tive adequacy, as measured by Cramer's V, is .121 for current concerns, .091 for the family unit's recent financial progress, and .128 for intergenerational financial progress. Congruency Groups Demographic characteristics. Five demographic characteristics of respondent's family units are entered as dependent variables in one—way analysis of variance for the factor, congruency. Three characteristics, family income, age of oldest child, and number of children, are significant at or below the .004 level, as shown in Table 25. Age of the respondent-head is significant at the .043 level. Number of adults is not significantly different by congruency group. The strength of association between family income and congruency is .123, as measured by the eta square statistic. The association between congruency and child-related variables is .015 for both number of children and age of oldest child. Age of the respondent- head has a very low strength of association, .010. The five congruency groups differ in that (1) family income increases as one moves from C-l to C-5, (2) age of oldest child is lowest for the most incon- gruent groups, C-1 and C-5, and highest for the somewhat .mo. w m ”Ho>oa cosmowmwsmwmo .mosmwum> mo mamaamsm wmslmso Eosm owumflumumlm us» no wufiaasmsonm 113 n .momsowom o>fluomnso ma so .momsomos o>wuomnssm ma 4m .hososumsoo ma on aos.am osm.mo ess.am moo.oo Hms.~m .>0o .oum oaoo. .aouo meoosH moo.mso mma.msm mma.mam ~me.mm oom.mo com: mus. Nous assess ~m.ea oo.oa ~m.oa ms.as mm.o~ .>0o one: .oum omeo. .noum -ucoosoouum os.mm ~c.~e o~.ce mm.me ou.ae s60: oHo. Noam so was me. ow. no. me. om. .>0o .oum oNH. .nouo usages oo.a os.a os.H ~5.H mm.H sous moo. ~uum mo soassz es.m mH.m oa.o sm.o ma.o .>0o oaaso .oum omoo. .nouo uaooso oo.m oo.m es.e mm.m ~m.m s80: mac. aces mo mos oo.H mm.a -.H em.a ma.~ .>0o . .oum oeoo. n.nosm couoaaso oo.s no. so. om.a «H.H sows mso. None so umnssz an n z mHN u z mom I z now u z ow n z «ovvsm sovsm so u am «casm sexasm mosumsuuum Hmmwmwwmww muo «no muo mno menu moasmwum> Hmsuxousoo Mom mowumflumum psm .msowumw>op oumossum .msmmfi msoum aosossmsooll.mm mqmse 114 incongruent groups, C-2 and C-4, (3) number of children is highest for groups C-1 and C-2 whose incongruency is in the direction of higher subjective adequacy, and (4) respondent-heads are younger in age for the most incon- gruent groups, C-1 and C—5, with the youngest mean age found for group C-S. Selected post hoc contrasts, test- - 4 r- a”. if ing the significance of congruency group differences, are reported in Table 26. The age of the respondent head and the total family income each have three of four contrasts - which are significant at or above the .013 level. Chi square and Cramer's V statistics are used to test the relationship of the remaining demographic char- acteristics to the congruency variable. Five of the seven characteristics are significant at or above the level of .027, as reported in Table 27. The occupation of male respondents and the sex of the respondent, with chi squares of .126 and .127, respectively, are not significant. The strength of association, as measured by Cramer's V, is in the .098 to .120 range for all five variables which are significant. These variables are occupation of female respondents, housing status, urbani- zation of residence, and the race and the educational level of the respondent. The five congruency groups differ in that (1) 'C-1 and C-2, which are incongruent in the direction of 115 TABLE 26.--Post hoc contrasts of the difference between congruency groups on demographic characteristics Congruence Groups Pooled . . Contrasted Variance T-statistic Number of Children b ! c-3 and C-1, 2a 1.975 .048 ;] C—3 and C-5 -.262 .794 i, C-1 and C-2 -.607 .544 i; C-4 and C-5 -.667 .505 L: Age of Oldest Child C-3 and C-2,4 -.560 .576 C-3 and C-l,5 -l.835 .067 C-1 and C-5 .338 .735 C-1,5 and C-2,4 -l.544 .123 Age of Respondent-Head C-3 and C-1 -.772 .440 c—3 and c-5 2.900 .004b C-1 and c-5 2.482 .013b c-1,2,3,4 and c-5 2.482 .013b Total Family Income b C-1 and C-2 -3.250 .001 c-2 and c-3 -7.528 .000b c-3 and c-4 -4.352 .000b C-3 and C-5 -1.384 .167 aC is congruency, the congruency groups are described in Figure 2. bSignificance level: p 5 .05. 1216 .OHOE HO ooo.om mo so«um«smom m su«3 woman s« omosullsm«oms osu o>osd omosu mum sophomou momuusoososo .neo«m«o «oosom su«s m smsu so«usosoo ouoe su«3 omosullss«oms osu m>osu omosu mus ouuuomou oommusoouomo .aososuusoo n« o s .mo. w m »«o>o« oosdu«u«so«mo H.mo FNH. > u.uossuo w.vm ~.wm m.mw a.«o o.«m ms«»sm n« m I up usumum so msso sooo. ounsou «so ms«msom m.- moH. > u.uosduo mosoo«uus o.m~ m.we s.~e «.om om.om stmIsos «N I up no «muss msNO. oussou «so so«um~«s¢sso snow I zv o~«. > u.uofiuuo ousoososmms s.m« s.sm w.ov o.om w.«n «.«v «N I no oHQEGm mo mo>«3onsos uv«o. ouuson «so so«uumsooo snow I zo h««. > u.umsduo nusoosomnms m.mm m.sv m.ev «.vm o.o« m.~v cm I up can: no uu««oo ou«s3 o~«. ousson «so so«uomsouo m.mm mo«. > n.uusduo usoososuos m.«¢ «.mv m.mm m.m~ ov.vm nso«m«o ma I up «0 soosom nose amoo. mucosa and sosucooom moo. > a.smsdso a.sm o.om m.oo m.sm m.~o e.ao c u so usoocomuom ou«sz nova. oussou «so no cuss omo. > m.sussuo h.mm «.Nm m.wm «.mm m.mo h.mm v I up usoosomuos o«msmm sNH. «assoc «so no son «wwwsm wmwsm «cmuqam Mmmmm «wmwam oomusoouom osnusuc> osm auoooumo no«uu«umum «usuxousoo huomouso moo: oHQEdm s« msouo mo omnusoouom “moo: o«m&mw assoum aososumsoo Hon uo«s¢«u¢> «usuxdusoo so mo«uu«uuuu osu .uumsusoouos .no@0:uu.s~ wands 117 higher subjective adequacy, contain fewer white respon- dents and fewer persons with more than a high school edu- cation, (2) C-1 contains fewer home owners, and (3) C-5 contains fewer respondents who live in areas with a popu- lation of 50,000 or more and more female respondents who were not housewives. .1! Economic perceptions. Differences in economic perceptions by congruency group are significant at or above the .022 level for three of the five variables tested, as shown in Table 28. Material wishes and inter- generational financial progress are not significant. The strength of association for the three perceptions which are significant are similar, .108 for consumption achievement, .106 for recent family financial progress, and .091 for current concerns. The five congruency groups differ in economic perceptions in that (1) C-1 and C-2 respondents (a) less frequently name financial security as their first ranked current concern, and (b) more frequently have achieved their consumption goals, and (2) C—5 respondents (a) more frequently rank financial security as their current concern, (b) more frequently report deterioration with regard to recent family financial progress, and (c) less frequently have achieved their consumption goals. ‘2! I.I.II,III..( . . H .s«mm oo>o«som H0\osm oouommxo su«3 omosunlsm«oue ms» o>oss omosu mum mmsoum o sow oouuomou momsusoouomo .o~ smsu whoa wo mouoom soos« mmsm«3 «m«uoumfi su«3 omosulusm«oma osu o>osm omosu mum mmsouo o uOM sophomou mommusoouomo 118 .mo. w m u«o>m« oosso«m«sm«ms .mososuosoo m« on «sso«umuosom vho. > m.umfiuuo thousH m.mh m I up .muoumosm o.«s s.ms m.ms s.«m «no umuumm mm«. oumsom «so «m«osms«m wo«. > u.umasuo usooum m.«v m« I no .mmoumoss H.om m.so m.ov s.om so«umuo«u0u0o nooo. mucosa «so «m«occs«s s.mm moH. > m.umsmuo oo>m«sos m« I up usoso>o«sos o.m« m.m~ >.mm m.ov uoz sooo. osmsom «so so«usfismsoo m.~m one. > m.uossuo songs sm«z cs I no umsu«z «.mv m.mm «.mw «.mm ow I m« «Hm. «Mason «so «m«uounz «.mm «mo. > m.uosmuo >u«usomm ow I up msuoosoo e.ns m.om c.~m o.he «m«osms«m swmo. oumsom «so usouuso Hm I z mam I z mom I z mom I 2 mm I z umcusmoumd o~nu«uo> 40vv¢m «cvsm so I am soxsm «0AAam one muoomumo uo«um«uoum so«um00u0o mno vuo mlo «no m«no "woos m«msmm o«sosoom mdsosm aosossmsoo How mo«su«us> so«umoouom o«sosooo so mo«um«usum osm .wommusmouom .moUQZII.m~ msmsa 119 Summagy The results of the data analysis can be sum- marized as follows (see Table 29): 1. Relationships between objective and subjective family income adequacy and satisfaction variables: (a) The effects of objective and subjective ade- (b) (C) quacy, while somewhat correlated, are addi- tive. Subjective adequacy explain more of the difference in satisfaction scores than does objective adequacy. Toqether, however, they explain more of the variation in satis- faction scores than either does alone. Objective and subjective adequacy are more related to the satisfaction variables dealing with items which have a more direct economic base, family income and level of consumption, than to the satisfaction scores for overall life quality, in which the economic base is more diluted by other factors. Satisfaction scores increase as subjective adequacy increases, and as objective adequacy increases. The differences in satisfaction mean scores are greatest for objective ade- quacy at the extremes of the objective adequacy distribution. 120 sou was so suHHHoooouo one .«« mmHsoB s« pouuooou who hose N .umfiuom m«su ou mo>«omfiosu osoH uos oo o .> m.HoEuso counooou su«3 moHsd«Hu> omosu sou cocoon duo oouuooou su«3 mmHsm«ss> ones» How o«um«uuuouo osu mo hu«««smsouo s osm .mm .vm How mos«os«w uses s .m« vsn .NH so. NoH. mH. ooo. mH. ooo. .coo .uuouoouo HusoccsHo HH. ooo. so. ooo. oH. ooo. .0ou .ououoouo Hu«osocso HH. ooo. mH. ooo. mN. ooo. usosm>0«soo co«uosoucoo so. HHN. mH. ooo. HN. ooo. mosuss Hc«u0uuz mo. NNo. NH. ooo. oN. ooo. usuoocoo ucmuuso mH. ooo. sN. ooo. sH. ooo. usuuuu oc«u=om oH. sHo. HH. Hoo. oH. moo. sosuuu«ccnuo NH. «Ho. oH. ooo. NH. ooo. uoHcsom .so«ucooooo NH. mNH. mH. ooo. mN. ooo. mmHus .coHumooooo HH. moo. NN. ooo. «H. ooo. co«ucosom oH. ooo. mN. ooo. oH. ooo. doom oo. sNH. mH. ooo. oo. NHH. xom NH. Hoo. mm. ooo. NH. ooo. osoosH sH«scs Ho. moo. No. moo. No. ooo. ooocuusoosoooou mo cos Ho. mNH. mo. ooo. Ho. ooo. masses «0 nonsoz No. moo. mo. ooo. so. ooo. oHsso uuooHo «0 mos No. ooo. oH. ooo. NH. ooo. smuoHHoo no sonsoz “nOHss«Hn> «ssuxmusoo o«m o«m o«: O» cous«ou mo«sd«um> homsooom osoosH Hoo. Hoo. so. Nmo. mm HH. Hoo. mm su«Hcoo 0o«H ««nso>o ooo. Hoo. NH. Hoo. Ne NN. Hoo. Ne coHuosdmsoo no «o>os moo. ooo. NH. Hoo. H: mm. Hoo. Hm osoocH sH«suo “su«3 so«uomwm«umn ou wouMHou mo«s¢«us> hossvowm ofioosH > m.uoEmuo «mum so > m.HoEsso «mum so > m.umEmso Noam so so oss sm soozuom aososumsoo moosooos oEoosH sHHsoo o>auoosoo howsooos oEoosH sHseao 0>Huoososm m«uosuoows mso mo msooo so«um«00mmw mo sumsosum osm «o>os oosmo«m«sm«m su«3 nm«mosuooas ms mms«os«u mo assessmnu.m~ msmse 121 2. The relationship between congruency and satis- faction variables: (a) Those congruency groups who are better off than they think they are, C-4 and C-5, have lower mean scores on all three satisfaction variables. The strength of association between congruency and satisfaction variables, however, is very slight. 3. The description of those who are in the different groups of each of the income adequacy variables: (a) Subjective adequacy groups differ in that, as subjective adequacy increases, family . income increases, number of children and the age of the oldest child decreases, home ownership increases, more respondents are white, more males are in white collar occu- pations and more females are housewives, material wishes decrease, fewer list financial security as their first ranked concern, and more respondents have consumption goals which they have achieved. (b) Objective adequacy groups differ in that, as objective adequacy increases, family income increases, the number of children and the age of the oldest child decreases, home ownership increases, more respondents are (C) 122 white, more males are in white collar occu- pations and fewer females are housewives, material wishes decrease, and more respondents had consumption goals which they have achieved, with the exception of group OA-6 on consumption goals. Congruency groups differ in that, as the group number increases from C-l to C-5 (increasing objective adequacy and decreasing subjective adequacy, with congruency at C-3), family income increases up to C-4 and then decreases, the number with financial security as the top ranked current concern increases, and the number whose consumption goals have been achieved decreases. In addition, the age of the respondent-head is lower in C-1 and lowest in C-5, the number of white respondents is lowest in C-1 and C-2, and home ownership is lowest for C-1 and second lowest for C-S. CHAPTER V OVERVIEW, DISCUSSION, AND IMPLICATIONS The study is summarized and findings discussed in relation to related research studies and theories. Implications of the findings are related to research and educational programs. Overview of the Study Satisfaction is viewed as an evaluation of goal achievement, which results from the use of resources. The satisfactions analyzed are those of the respondents, while income adequacy is that of the family units with whom the respondents share resources. The primary focus of the research is on the relationship between family income adequacy, objectively and subjectively measured, and satisfaction with family income, level of consumption, and overall life quality. The degree of congruency between objective and subjective adequacy is also tested in relation to the three satis- faction variables. The secondary focus of the research is on the description of objective and subjective family income adequacy by contextual variables. 123 124 The data used in this research were collected in the fall of 1974 by the Survey Research Center of the University of Michigan and are identified as the Economic Incentives, Values, and Well-being Project, Part IV. A study sample of 1046 respondents who are the husband or wife or one-adult family head is drawn from the larger multistage area probability sample of private households in the contiguous United States. Satisfaction and subjective family income adequacy variables from the larger study are used here. Infor- mation from the Bureau of Labor Statistics is used to develop a measure of objective family income adequacy. A measure of congruency is developed from the interrela- tionship of objective and subjective family income ade- quacy. Family related contextual variables used in the research include the number of adults, number of children, age of oldest child, age of the respondent-head, and the total income of the family unit, housing status and urbanization of residence. Respondent-related contextual variables used are demographic characteristics: sex, race, occupation and education; and economic perceptions: current concerns, material wishes, consumption standard achievement, and recent financial progress. 125 Both parametric and nonparametric statistical models are used to test the hypothesized relationships and differences among variables. Discussion of Findingg The discussion of the findings is organized around the three research questions. Research Question 1: Is there a relationship between income adequacy and selected life satisfaction variables? The income adequacy of the family unit, whether assessed by the respondent or judged by the standards of budget makers, explains variation in each of the three satisfaction measures. Subjective adequacy accounts for more variance in satisfaction than does objective adequacy. This would appear to be a logical outcome. The respondent, as an individual within the family unit, can consider the adequacy of the family's income in relation to the particular goals of the family unit. Satisfaction is an assessment of goal attainment in relation to needs and wants. Hence, a measure of income which is assumed to assess the adequacy of income in relation to needs and wants should be more strongly related to satisfaction than one which does not consider the particular family's goals. 126 Each of the three satisfaction variables is found to increase as income adequacy increases. Greater income adequacy provides more resources for goal achievement. Thus, the family might be expected to derive greater satisfaction from goal achievement when more resources are available to facilitate goal achievement. Past research reports have stated that income, unadjusted for family size, is correlated with life satisfaction. Hayes and Stinnett (1971) found that life satisfaction among husbands and wives in their middle years is significantly higher among those with higher income. From this, they suggest that life satisfaction may be dependent to a large extent on the earning capacity of the family and its resulting level of living. Hafstrom and Dunsing (1973) relate both subjective family income adequacy and income before taxes, unadjusted for family size, to a variable which is comparable with satisfaction with overall life quality. For a sample of "typical" homemakers, they found subjective adequacy to have the highest correlation with life satisfaction: income before taxes has a lesser but still moderate correlation with life satisfaction. The findings in this research are consistent with those of Hafstrom and Dunsing. In both cases, subjective adequacy predicts more of the variation in life satisfaction than did objective adequacy. However, the objective adequacy -—-- - «II-w 127 variable used in this research is adjusted for family composition and geographic area of residence. The objective measure used by Hafstrom and Dunsing is unadjusted family income. Using data from a 1972 national sample of house- holds, Strumpel relates satisfaction with family income to an income measure adjusted for number of family meme bers. Within an overall pattern of increasing satis- faction with income as objective income increases, he found that satisfaction increases most steeply for the lowest 15 percent and the highest 10 percent of the sampling distribution. Between the extremes there is a long, gentle increase in satisfaction with family income, except for a dip at the thirtieth percentile (Strumpel, 1975a). The sample used in the study reported here also shows steeper increases in satisfaction with family income at the extremes of the distribution, but no dip. Satisfaction increases with each increase in objective adequacy across the six categories. The dip in Strumpel's data may be due to sampling error, since dividing the sample into twenty categories greatly reduces the number of cases in each category, as opposed to the six categories used here. The effects of objective and subjective income adequacy, while having some commonality, are additive in the research reported here. The two measures together - -' ".M- .7451?! 128 explain more of the variation in satisfaction than does either alone. This finding gives evidence that the two kinds of indicators should be used as compliments to each other. Similar results regarding objective and subjective measures of a phenomenon are found in a study by Newman (1975) of the desire to change household resi- dences. Among short-term residents, subjective indi- cators explain more of the desire to move than do objec- tive indicators. Among long-term residents, the two types of measures are about equal in explaining the desire to move. Together, the objective and subjective indicators explain more of the desire to move than do either type of indicator alone. In sum, this research, and that of Newman and of Hafstrom and Dunsing, demonstrate that subjective measures are stronger predictors than are objective measures. This research, as well as Newman's, gives evidence that the combined measures are better predictors than either type used alone. Income adequacy, both objectively and subjectively measured, explains more of the variation in the satis- faction with family income and level of consumption than of the variation in overall life quality. A possible reason is that family income and level of consumption are domains of satisfaction in which goal achievement is likely to be facilitated or constrained by economic ' .- ’7 “my 129 resources. On the other hand, overall life quality is a much more generalized concept; all domains of one's life contribute to overall life quality. While income adequacy is significantly related to life quality, it is one factor among many. As a result its contributions are more diluted by the many other components of life quality. Research Question 2: Is there a relationship between the congruency of income adequacy measures and selected satisfaction variables? For each of the satisfaction variables, congruency groups C-4 and C-5, those who are better off than they think they are, differ from the remaining three congruency groups. Groups C-4 and C-5 are least satisfied with family income, level of consumption, and overall life quality. Those in groups C-1 and C-2, who are worse off than they think they are, have satisfaction scores which are not statistically distinguishable from the scores of the congruent group, C-3. While C-S has the youngest family heads and the youngest age of oldest child, indicating younger family units, group C-l is second lowest in both of the cate- gories. In other ways, the two most incongruent groups are quite different. Group C-l contains more blacks, more female respondents, and is second lowest in — . - 7:94:31” 130 percentage of respondents with education beyond high school. In other research, these factors have been associated with lower levels of income (Schiller, 1973). Group C-l does have the lowest income level in this sample. These respondents may have recognized their traditionally more limited economic opportunities and '7 new lowered their goals to a level which better coincided with their incomes. This would narrow the gap between what is and what is aspired to. More modest goals are more attainable, so satisfaction with goal attainment may have increased. In contrast with group C-l, group C-S, which has the youngest family units in the sample, is second highest in education, has the most male respondents, the most white respondents, and the fewest female respondents who are housewives. These are factors associated, in other research studies, with higher incomes. And C-S does have the second highest income level of the con- gruency groups. The economic perception variables are also help- ful in understanding group C-S's lower satisfaction scores. Their consumption goal achievement is lowest, their material wishes index scores (indicating stronger wishes for more consumption items) are highest, they are the group most concerned with financial security and have the fewest respondents who report recent 131 financial progress. Their aspiration for more things and their failure to meet aspired to consumption goals would indicate unfulfilled aspirations. The gap between their goals and their present situation is likely to be wider than for the other congruency groups. Thus, when evaluating where they are in relation to where they would like to be, this wider gap could lead to lower satisfaction. These suggestions of the adjustment in aspirations to one's economic situation as an explanation of the dif- ference in satisfaction levels between groups C-1 and C-5 are based on Kurt Lewin's theoretical model of aspirations (Lewin, et al., 1944). Briefly, the model states that aspirations are not static, they tend to grow with achievement and decline with failure. Aspirations are influenced by the performance of family, friends, and other reference groups. Aspirations are reality oriented; it is common for them to be slightly higher or lower than the level of accomplishment. From empirical studies, Katona (1975) reports that high levels of aspiration do occur following gratif- ication of numerous wants for many people. People who do not desire any special expenditures are mostly old or poor. Katona states: The relatively few desires expressed by low-income people appear to indicate that even our wishes and desires are reality tested: . . . Newer 132 articles or products used by relatively few people . . . are desired primarily by upper- income people who are well supplied with standard goods. (Katona, 1975, p. 157) While the theory of aspirations is a possible explanation of the levels of satisfaction of groups C-1 and C-5, the theory of relative deprivation can also be applied to group C-S. This group is younger, racially contains more whites, contains more males, and contains more respondents with more education than the other con- gruency groups. In comparison with others, and by past experience, a group with these characteristics might realistically expect economic advancement. However, at the time of this study in the fall of 1974, the country was experiencing the second year of double digit inflation and salary and wage increases were not keeping pace with price increases. Families were experiencing lower constant dollar income and higher prices on such basic commodities as food, clothing, and gasoline (Ackerman & Bowers, 1974). In an economy where people had come to expect a much more gradual inflation, outpaced by wage and salary increases, the economic conditions of 1973-74 are a rather sudden change, and are in opposition to what might have been previously expected. It is a rather sudden blockage to fulfillment of expectations which had previously appeared to be realistic. Thus, dissonance is created between the unfulfilled expectation which is thought to 133 be deserved and the inequity or injustice felt from its nonfulfillment. Thus, group C-5 may be experiencing relative deprivation. Group C-l, having lowered its expectations in keeping with the lesser achievements of its reference groups of people who are also younger, but more racially black, more female, and less educated, does not have the same higher expectations as group C-5. Therefore, their expectations cannot be blocked to an equal degree by the inflation of 1973-1974. Research Question 3: Do those groups who differ in level of objective adequacy, level of subjective adequacy, and con- gruency group differ with respect to contextual variables? For the most part, objective and subjective adequacy groups have differences in contextual variables which would appear to agree with logical expectations. As income adequacy rises, as measured both objectively and subjectively, income rises, home ownership increases, more males are in white collar occupations, the number of children and the age of the oldest child decreases. In one demographic characteristic the two kinds of measures present opposite results. If one contrasts the lower two categories of subjective adequacy with the upper two, one finds that as subjective adequacy rises, more female respondents are housewives. If one 134 contrasts the lower three objective adequacy groups with the upper three, one finds that fewer female respondents are housewives. While the pattern is not perfectly linear across all categories of either measure, this difference does exist. The nonhousewives are either employed or are retired or disabled; most are employed. Such employment should add to family unit income and increase its objectively determined income adequacy. The fact that subjective adequacy shows the reverse pattern might indicate the home production of the house- wife adds appreciably to the nonmarket income of the family unit, a factor which is not incorporated into the objective adequacy measure. Or the higher subjective adequacy of groups which include higher percentages of housewives may indicate the economic value of the greater leisure time of the housewife. Since percentage of females employed and satisfaction with overall life quality are known to vary with marital status, it might be useful to extend the analysis, controlling for marital status. Among congruency groups, C-S, those who are economically better off than they think they are, has the lowest average age for the respondent-head. It is the group which judges its income adequacy to be much lower than that adequacy is by the standards of budget makers. SuCh a difference might be the result of greater 135 aspirations, as discussed previously, possessed by this group. This group does have the highest score on the material wishes index, an indication of higher aspir- ations. This finding relates to that of Crosby. She found, among a random sample of homemakers, that the respondent's perception of difference between expected and present levels of living had an inverse relationship with the respondent's age (Crosby, 1970). Limitations of the Study The general purposes of the study were accom- plished. However, the advantages of a large and general- izable national sample of households acquired at little cost in time and money to the researcher were in part balanced by the need to accept and adapt measures used by others who did not have quite the same research goals. One limitation has to do with the information available on the family's economic situation. Family income is measured by one question and grouped into eighteen categories. The data set does not include information on the level of debts or assets, other than home ownership, of the respondents. Thus, the objective adequacy measure used here is not as detailed as it might be. It could not consider longer term aspects of the family's financial situation. The two most incongruent groups have the youngest ages of respondent-heads in the sample and the youngest children. Since younger families 136 are known from other research to have the highest per- centage of credit use and the highest percentage of debt, it is possible that data on debts and assets might pro- vide further explanation of the results found here. The research design used here did not control on third variables to explore their effects as suppressor or distorter variables (Rosenberg, 1968). Such methods could be used to further understand the relationships analyzed here. The study does not consider possible demands on family income from outside the family unit living in that one household. If there are financial allocations for child support or alimony from a previous marriage or aged parents with financial needs to which the family unit contributes, these are not recognized in the compu- tation of the objective adequacy ratio. If a researcher designed their own questionnaire and collected their own data, these possibilities could have been included. Implications for Further Research The objective adequacy measure used here is com- pared with a measure of subjective adequacy in its ability to predict satisfaction with life quality and two of its domains. The objective adequacy measure might be further explored. How does it compare with a simple money income of the family measure in explaining satisfaction? How much of the variation in objective adequacy is due to 137 the income component of the ratio, and to the family composition component of the ratio? In this research, the contextual variables are used to describe those who differ in congruency and those who differ in income adequacy, objectively and subjectively measured. Another research question would be to use the contextual variables to predict the respondent's income adequacy and congruency. The demographic variables of the family unit, age of respondent-head, age of oldest child, number of children, and number of adults could be developed into a family life cycle variable. With that variable, dif- ferences in satisfaction and in income adequacy could be related to the stages of the life cycle. A future study could be designed to determine differences in the level and standard of consumption of respondents and relate those differences to objective and subjective income adequacy. This would provide a measure of the gap between levels and standards, the magnitude of one's aspirations. The findings could lend support or refutation to the proposition that subjective adequacy is a judgment of the similarity or difference between present and desired level of consumption. A future study might incorporate questions to obtain data on personality and self-concept of the respondent. They could be used to analyze and interpret 138 differences in satisfaction measures and subjective adequacy and congruency measures. The research reported here analyzed data from only one family member, a respondent-head. To understand better the family as an interacting unit of individuals, data are needed from husbands, wives, and children. Their perspectives may differ, yet they all draw from and share the same pool of family resources, including family income. Implications for Educational Programs The findings here seem to indicate that educators working with families cannot assume that objective measures provide a complete picture of family income adequacy. Even more important is how the family views its income adequacy in relation to family goals. Explor- atory discussions with the family will be needed to determine their goals, their felt needs and wants, before the educator can work with the family in assessing the ability of family resources to meet those goals. Based on the aspiration theory, Katona's findings, and the increasing material wishes found here for those who were incongruent in that they are better off than they think they are, the educator needs to be alert to the influence of past goal achievement on the current level of aspirations. If past goals have not been reached, aspirations may be lowered, if past goals have 139 been attained, aspirations may increase. The educator will need to explore with the family whether the changes in aspirations fit, in degree, with the ability of family resources to facilitate reaching the new aspir- ations. For example, if increased economic well-being is a goal which has been achieved when the family head moves from unemployment to full-time employment, is the aspiration to buy several new household durables on credit greater than is warranted by the increase in family income? The educator might be able to aid the family in judging the stability of future income as well as its current sudden increase before the family allows their newly formed higher aspirations to lead to imple- menting goals to buy several durable goods on credit at one time. In effect, the educator would be asking the family whether their increased income is as great as their increased aspirations. One of the areas of concern to an educator is how to increase the overall well-being of families who have low incomes and little prospect in the short term for an increase in the adequacy of their incomes. Every effort should be made to assist these families in manag- ing their incomes. Their economic situation, however, limits what can be accomplished. One of the insights gained from this research, as well as other research, is that income adequacy has less effect on satisfaction 140 with overall life quality than on satisfaction with family income and with level of consumption. Thus, it might be well to focus educational programs on those domains of life satisfaction which are less constrained by income adequacy, such as family relations. This can enhance life satisfaction for the families and allow them to function at the best level possible, while both the families and professionals continue to work toward economic policies which will improve their economic well-being. APPENDICES APPENDIX A BUREAU OF LABOR STATISTICS EQUIVALENCE SCALES APPENDIX A BUREAU OF LABOR STATISTICS EQUIVALENCE SCALES TABLE 30.--Revised scale of equivalent incomé'for urban families of different size, age, and composition (4-person family-husband, age 35-54, wife, 2 children, older 6-15 = 100) Age of Head Size and Type of Family Under 35 35—54 55-64 One person 37 38 33 Two persons: Husband and wife 50 61 60 One parent and child 40 59 62 Three persons: Husband, wife, child under 6 62 69 -- Husband, wife, child 6-15 62 83 89 Husband, wife, child 16-17 -- 92 89 Husband, wife, child 18 or over -- 83 86 One parent, 2 children 68 77 84 Four persons: Husband, wife, 2 children, (older under 6) 71 79 -- Husband, wife, 2 children, (older 6-15) 76 100 105 Husband, wife, 2 children, (older 16-17) -- 114 126 Husband, wife, 2 children, (older 18 or over) -- 96 110 One parent, 3 children 88 97 -- Five persons: Husband, wife, 3 children, (oldest under 6) 85 95 —- Husband, wife, 3 children, (oldest 6-15) 94 115 119 Husband, wife, 3 children, (oldest 16-17) -- 128 138 Husband, wife, 3 children, (oldest 18 or over) -- 118 124 One parent, 4 children 108 117 -- Six persons or more: Husband, wife, 4 children or more, (oldest under 6) 98 -- -- Husband, wife, 4 children or more, (oldest 6-15) 107 130 139 Husband, wife, 4 children or more, (oldest 16—17) -- 145 —- Hueband, wife, 4 children or more, (oldest 18 or over) -- 149 -- One parent, 5 children or more 124 137 -- 1 The scale values shown here are percentages to be applied to the total cost of a budget (excluding State and local income taxes and disability pay- ments) for the base family (4 persons--hueband, age 35-54, wife, 2 children, older child 6-15 years) to estimate the total income required to provide the same level of living for urban families of different size, age, and composition. In addition to the cost of goods and services for family consumption the total budget costs include gifts and contributions, life insurance, occupational expenses, employee contributions for social security, and Federal income taxes. Estimates of personal taxes paid to State and local governments and of payments for disability insurance may be added in those urban areas where applicable. SOURCE: U.S. Department of Labor, 1968, p. 14. 141 142 TABLE 31.—-Equivalence scale of comparative costs by city and region for the total budget for a four-person family at a moderate level of living, Autumn, 1973 Index of Index of Compar- Compar- Area ative Area ative Costs Costs Urban United Lancaster 98 States 100 St. Louis 98 Kansas City 99 MetrOpolitan areas 102 Baltimore 99 South 93 Los Angeles 99 West 99 North Central 101 Green Bay 99 Northeast 110 Cedar Rapids 100 Seattle 100 Nonmetropolitan Portland 101 areas 90 Indianapolis 101 South 85 West 90 Cleveland 101 North Central 93 Detroit 101 Northeast 98 Champaign-Urbana 103 Minneapolis- Austin 87 St. Paul 103 Houston 90 Philadelphia 103 Orlando 90 Baton Rouge 90 Washington 103 Dallas 90 Milwaukee 105 Chicago 105 Nashville 92 Buffalo 105 Atlanta 93 San Francisco 106 Dayton 93 Bakersfield 93 Hartford 109 Wichita 94 New York 114 Boston 118 Denver 96 Honolulu 118 Durham 96 Anchorage 131 Cincinnati 96 San Diego 97 Pittsburgh 97 SOURCE: U.S. Department of Labor, undated, p. 19. APPENDIX B CODING VARIATIONS NOTE: I. * II. data. APPENDIX B CODING VARIATIONS The format in which the following information is presented is: Study Variable: Tape variable number; interview sChedule IOcation. Wording of question or identifi- cation of interview schedule source of the data. Values used Values used in this study_ on source tape Description Any additional information. SATISFACTION MEASURES SATISFACTION WITH PERCEIVED FAMILY INCOME: V87; Blc. (How do you feel about) . . . The income you (and your family) have? 1. l. Terrible 2. 2. Unhappy 3. 3. Mostly dissatisfied 4. 4. Mixed 5. 5. Mostly satisfied 6. 6. Pleased 7. 7. Delighted 8,9,0. No feelings at all; Never thought about it; NA; DK. SATISFACTION WITH PERCEIVED LEVEL OF CONSUMPTION: V88; Bld. (How do you feel about . . .) . . . Your standard of living--the things you have like housing, car, furniture, recreation, and the like? Same code as V87, above. SATISFACTION WITH PERCEIVED OVERALL LIFE QUALITY: V8534§Ifi. How do you feel about your lifE as a whole? Same code as V87, above. INCOME ADEQUACY MEASURES SUBJECTIVE FAMILY INCOME ADEQUACY: V96, B4. DO you feel that your total (familinncome is enough for you (and your family) to live as comfortably as you would like at this time? Would you say very com- fortably, comfortably, not too comfortably, or not at all comfortably? *Deleted from the study sample because of missing 143 144 SA-l 5. Not at all comfortably SA-2 4. Not too comfortably SA-3 2. Comfortably SA-4 1. Very comfortably * 8,9 DK,NAa OBJECTIVE FAMILY INCOME ADEQUACY: A computed variable, developed as ISIlows: 1. Objective Adequacy = Each respondent was assigned a Bureau of Labor Statis- tics equivalence scale number which adjusts the BLS standard budget for a moderate level of living for differences in family composition. The variables used, which are reported in greater detail under "Con- textual Variables, Demographic," on the pages which follow, were: V21: Age of oldest child V22: Number of children in the family unit V45: Number of adults in the family unit V48: Age of respondent In addition, V27, marital status was used to delete from the sample those family units with two adults in which the second adult was not the spouse of the respondent. Each respondent was assigned a Bureau of Labor Statis- tics equivalence scale number which adjusts the BLS standard budget for a moderate level of living for differences in geographic location of residence. The variable used was V4; Probability sampling unit (PSU) code. Values of 001 to 996 identify specific PSU's in the sample. Selected examples of the code are: 131-132 Detroit City 133-134 Detroit suburbs 365-366 Huston, Tex. 697-698 Sheboygan, Wisc. The family income variable (see Contextual Variables, Demographic, which follows) was recoded to dollar amounts. The objective income adequacy measure was then com- puted, using the following formula: Total family_income in 1973 Stafidard' Family Geographié Budget, Composition Location Moderate Equivalence Equivalence Level Number Number 160’ 100% Ratio * Deleted from the study sample because of missing data. aDK is don't know; NA is not ascertained. 145 Computed ratios were then grouped into 6 values of an objective family income adequacy measure. OA-l. O.A. ratio of 16-49 OA-2. O.A. ratio of 50-99 OA-3. O.A. ratio of 100-149 OA-4. O.A. ratio of 150-199 OA-S. O.A. ratio of 200-299 OA-6. O.A. ratio of 300 or more. CONGRUENCY between objective and subjective family income adequacy: A built variable, developed as follows: Objective family income adequacy was reduced from the 6 values identified in the list above to 4 values by combining values 3, 4, and 5. This choice was made based on post hoc contrasts following one-way analysis of variance of the differences in satisfaction mean scores for the 6 levels of objective family income adequacy. Values 3, 4, and 5 did not differ sig- nificantly in satisfaction mean scores. Subjective family income adequacy was used in its same form. A congruency matrix was developed and the cells were coded as follows: Levels of Levels of Objective Adequacy Subjective OA-l OA-2 OA-3,4,5 OA-6 Adequacy Ratio Ratio Ratio Ratio 16-49 50—99 100-299 300 plus SA-l: Not at C-3 l C-4 l C-5 C-5 all adequate , SA—Z: Not too C-2 C-3 l C—4 C-5 adequate [— SA-3: Adequate C-1 C-2 C-3 C-4 SA-4: Very C-1 C-1 C-2 C-3 adequate Fig. 3. The identification of congruency groups by their levels of objective and subjective family income adequacy. 146 4. The congruency groups are: C-l. SA>>OA: Subjective adequacy much greater than objective adequacy. C-2. SA>'OA: Subjective adequacy somewhat greater than subjective adequacy C-3. SA==OA: Subjective adequacy equal to objective adequacy. C-4. SA<