m..- f E. .5 .5, ‘ IIIIIIIIIIIIIIZIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII ' aye o)5' ”JOLM‘I . 310454 0251 :I M35533”? " 542.: s» gflfie i University I k A 1—K This is to certify that the dissertation entitled CONSUMER SATISFACTION: ATTITUDES AMONG RETIREES presented by Dawn Thorndike Pysarchik has been accepted towards fulfillment of the requirements for Ph . D . degree in Family Ecology ajor professor Date August 10, 1982 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 A D -5 r . ~un-w Hr" RETURNING MATERIALS: MSU Place in book drop to LIBRARIES remove this checkout from —_ your record. FINES will be charged if book is returned after the date I . stamped below. 1-week" 11“..-- 15 ‘1 “8’2 “999 a": ‘ 1) _ fivI/iIJ', , ._‘_ 390686.11? $355; .1“ 5 J? ! 3 I990 U 1 - 5 JUNEEJBQW’ 5P 122 " :00 “33”.WWW m, 271993? - FEBO71993 "O 9174 1 i8; " ‘\ a: “I 251 5., 75\ CONSUMER SATISFACTION: ATTITUDES AMONG RETIREES By Dawn Thorndike Pysarchik A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Human Ecology 1982 ABSTRACT CONSUMER SATISFACTION: ATTITUDES AMONG RETIREES By Dawn Thorndike Rysarchik The purpose of the study was to assess the attitudes of Michigan retirees at three stages of retirement (early, age 60 to 64; middle, age 65 to 69; and later, age 70 and above) toward selected marketplace practices, consumer dissatisfaction, and complaint reso- lution. The research was cross-sectional in design and used mail survey methodology. Attitudes were measured by Likert-type statements and closed and free-response questions. The study was restricted to retirees who had been retired for at least one year and were 60 years of age and older. The survey was sent to 866 retirees from six Michigan cities and towns during March and April 1982. One hundred ninety-six usable questionnaires were analyzed in the study. While the hypotheses were not found to be statistically sig- nificant, descriptive analysis indicated some differences in attitudes between groups. The younger retirees reported a greater tendency toward price consciousness, comparison shopping, and the use of news- paper advertising for product information. Retirees in the middle Dawn Thorndike Pysarchik stage of retirement most frequently ranked new products and brands as inferior in quality to older established products. Retirees in the later stage of retirement reported the greatest frequency in paying cash for consumer purchases, preparing shopping lists, expe- riencing health and transportation problems hampering shopping activities, and needing shopping assistance by sales personnel. All three age groups reported more product problems than service problems. Specifically, product problems related to food and clothing and service problems concerning general repair were the most frequently reported. The most frequently cited reason for dissatis- faction was the quality of the product or service. Blame for consumer problems was most often attributed to the retailer, followed by the manufacturer. The complaint action taken most frequently by all age groups was to complain to the retailer. The majority of retirees who reported a complaint concerning a consumer problem were not satisfied with the resolution of the complaint. ACKNOWLEDGMENTS The author would like to extend her deepest appreciation to her doctoral committee chairperson, Dr. Marilyn Nagy, for her support and effort throughout the research study and graduate program. Sin- cere gratitude for their contributions is also extended to her guidance committee: Dr. Jane Dyer, Dr. Barbara Stowe, and Dr. Donald Taylor. The author wishes to thank friends and colleagues for the encouragement and support given throughout her doctoral program. She is especially grateful to Dr. Rick Rollenhagen, Mrs. Necia Black, and Dr. Brenda Witter for their assistance with the statistical analysis of the data; Mrs. Becky Hard for proofreading the disserta- tion; and Mrs. Sue Cooley for the final typing of the dissertation. A very special acknowledgment is given to her husband, Morry, and son, Ryan, for their endless patience, understanding, and love, which made it possible for the author to complete the research study and doctoral program. 11 TABLE OF CONTENTS' Page LIST OF TABLES ......................... vi Chapter I. INTRODUCTION ...................... l Statement of the Problem ............... 2 Objectives of the Study ................ 2 Profile of the Senior Citizen Market ......... 3 Growth Patterns of the Elderly Population ....... 4 Retirees: One Market Segment or Many? ......... 9 Buying Power of the Elderly .............. l4 Retirement Trends ................... l6 Significance of the Study ............... l7 Conceptual and Theoretical Framework ......... l9 Theoretical Definitions ................ 2l Operational Definitions ................ 21 Assumptions of the Study ............... 22 Overview of the Research Study ............ 23 11. REVIEW OF LITERATURE .................. 24 Life Satisfaction Among Retirees ........... 24 The Economic Security of Retirees ........... 28 The Effects of Inflation on the Economic Security of the Elderly ................... 3l The Impact of Information Processing on the Market- place Behavior of Elderly Consumers ......... 36 Consumption Patterns of the Elderly .......... 42 Consumer Satisfaction/Dissatisfaction Among the Elderly ....................... 46 Attribution Theory ................. 53 Complaint Behavior of the Elderly ........... 56 Government Regulations and Consumer Satisfaction/ Dissatisfaction .................. 57 Social Programs and the Elderly ............ 60 Consumer Education .................. 6l Implications and Summary ............... 63 iii Page III. RESEARCH DESIGN AND METHODOLOGY ............. 65 Research Design .................... 65 Hypotheses ...................... 66 Group l ....................... 66 Group 2 ....................... 67 Group 3 ....................... 67 Sample Selection ................... 67 Sample Criteria .................... 68 Sample Characteristics ................ 7l Sample Characteristics by Stage of Retirement . . . . 80 Development of the Survey Instrument ......... 85 Pretest of the Survey Instrument ........... 87 Pretest Data Analysis ................. 9O Factor Analysis ................... 9O Reliability Testing ................. 92 Cronbach's Alpha Test of Internal Consistency . . . . 94 Methodology ...................... 95 Data Collection .................... 97 Data Preparation .................. 98 Data Analysis ..................... 99 Crosstabulation and Chi-Square Analysis ....... 99 Analysis of Variance and Multivariate Analysis of Variance .................... lOl Hypothesis Testing ................. 103 IV. DATA ANALYSIS AND FINDINGS ............... lO4 Attitudes of Retirees Toward the Marketplace at Different Stages of Retirement ........... 105 Group l Hypotheses Testing ............. 109 Other Findings--Attitudes of Retirees Toward the Marketplace .................... ll7 Consumer Problems of Retirees at Different Stages of Retirement .................... l2l Group 2 Hypotheses Testing ............. l2l Other Findings--Consumer Problems of Retirees . . . . l3O Complaint Behavior Among Retirees at Different Stages of Retirement .................... 131 Group 3 Hypotheses Testing ............. l32 Other Findings--Complaint Behavior Among Retirees . . l37 V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ........ l4O Summary of Research Methodology and Data Analysis . . . 140 Summary of Findings .................. l42 Attitudes of Retirees Toward the Marketplace . . . . l43 Consumer Problems of Retirees ............ l44 Complaint Behavior Among Retirees .......... l46 iv Limitations of the Study ............... l46 Conclusions ...................... 149 Recommendations .................... l53 Recommendations for Advertisers ........... l53 Recommendations for Retailers ............ l54 Recommendations for Manufacturers .......... l54 Recommendations for Public Policy Makers ...... l56 Recommendations for Educators ............ l56 Recommendations for Future Research ......... 157 APPENDICES ........................... l6O A. QUESTIONNAIRE ...................... l6l B. VARIMAX ROTATED FACTOR MATRIX: PRETEST DATA ....... l68 C. RELIABILITY COEFFICIENTS--COEFFICIENT ALPHA: PRETEST DATA ..................... l7l REFERENCE LIST ......................... l73 LIST OF TABLES U.S. Population Changes of Persons Age 65 and Above . . . . Michigan Population Projections for the Aged From l980 to the Year 2000 .................... Changes in Aged Dependency Ratio From 1979 to the Year 2035 ........................ Life Expectancy of Residents of the State of Michigan, l90l- l977 ...... . .............. . . . Market Segmentation Among Elderly Consumers ........ Comparison of Frequency in Satisfaction/Dissatisfaction of Products and Services Among the Elderly and General Population (Best & Andreasen Study) ....... Most Frequently Reported Consumer Problems Among the Elderly (Best & Andreasen Study) ............ Comparison of the Most Frequently Reported Reasons for Consumer Dissatisfaction Among the Elderly and General Population (TARP Study) ............. Comparison of Age of Retiree Sample to Michigan and U.S. Population Age 60 and Above ............... Comparison of Retiree Sample Age 60 and Above to Michigan and U.S. Population Age 60 and Above by Sex ....... Comparison of Retiree Sample Age 60 and Above to Michigan and U.S. Population Age 65 and Above by Ethnic Background ....................... Comparison of Pre-retirement Occupation of Retiree Sample Age 60 and Above to Michigan and U.S. Population Comparison of Retirement Income of Retiree Sample and U.S. Population Age 60 and Above ............ Comparison of Educational Levels of Retiree Sample Age 60 and Above to U.S. Population Age 65 and Above ..... vi b-bwww .10 .11 .12 .10 .11 .12 .13 Comparison of Marital Status of Retiree Sample Age 60 and Above to U.S. Population Age 65 and Above ...... Comparison of Retiree Sample Age 60 and Above to Michigan Population Age 60 and Above by County .......... Comparison of Retirees' Residence Location by Stage of Retirement ....................... Comparison of Retirement Income by Stage of Retirement Comparison of Retirees' Gender by Stage of Retirement . . . Comparison of Education by Stage of Retirement ...... Attitude Scale Groupings of Survey Instrument Items . . . . Reliability Coefficients for Attitude Scale Groupings: Actual Test Data .................... Multivariate Analysis of Variance of Attitudes Toward the Marketplace by Stage of Retirement ......... Comparison of Retirees' Attitudes Toward the Marketplace Comparison of Attitudes Toward Advertising as a True Picture of Products and Stage of Retirement ....... Comparison of Attitudes Toward Television as a Source of New Product Information and Stage of Retirement ..... Comparison of Attitudes Toward Shopping Only for Need and Stage of Retirement ................. Comparison of Frequency of Product Problems by Retirees' Stage of Retirement ................... Comparison of Frequency of Service Problems by Retirees' Stage of Retirement ................... Comparison of Frequency of Product Problems by Years in Retirement ...................... Comparison of Frequency of Service Problems by Years in Retirement ...................... Frequency of Type of Product or Service Problem ...... Complaint Actions Resulting From Consumer Problems vii Page 78 79 81 82 83 84 106 108 111 113 118 119 120 123 125 128 129 131 133 Page 4.14 Attribution of Blame in Consumer Problems ......... T34 4.15 Summary of Analysis of Variance in Complaint Action by Stage of Retirement and Attribution of Blame ...... l35 4.l6 Summary of Absolute and Relative Frequencies and Means for Attribution of Blame in Product or Service Problem . l36 4.l7 Actions Taken to Resolve Consumer Problem ......... l38 4.l8 Level of Satisfaction With Action Taken to Correct Problem ......................... l39 4.l9 Explanation of Why No Action Was Taken to Resolve Consumer Problem .................... l39 viii CHAPTER I INTRODUCTION Today's marketplace could generally be characterized as a highly complex, competitive, and technically oriented marketing system. Recognizing that the formal education and marketplace training of elderly consumers occurred in the past when the products, services, and market system were considerably less complex than they are today, it is not surprising that the elderly often feel alien— ated (Burton & Hennon, l981). The sophisticated and impersonal nature of the marketplace has rendered the aged consumer less able to cope effectively with the economic consequences of a dynamic market system. An analysis of consumer satisfaction among retirees should focus upon the relative position of the elderly in today's economy. Retirees face physiological, sociological, psychological, and economic adjustments that affect their sovereignty in the marketplace. Spe- cifically, the ability of the elderly consumer to participate effec- tively in the marketplace is abated through diminished coping strategies, physical mobility, health, ability to process market information, income, and psychosociological adjustments. Statement of the Problem Previous research has identified the consumer attitudes and problems of the elderly by employing a temporal definition of "elderly" as those persons aged 65 and above. Specifically, no attempt has been made to identify the consumer attitudes and problems of retired persons at different stages of retirement. Due to the potential for reduced ability to cope in the marketplace by consumers in advanced stages of retirement, there is a need for research data concerning the specific concerns of retirees. The purpose of the study was to analyze the attitudes of Michigan retirees at different stages of retirement (early, age 60-64; middle, age 65-69; and later, age 70 and above) in relation to marketplace practices, the nature and extent of consumer problems, and the method of problem resolution. Objectives of the Study Due to the limited empirical research on the attitudes of retirees regarding consumer satisfaction/dissatisfaction and complaint resolution, this research study was undertaken to fulfill several objectives. l. Determine the attitudes of Michigan retirees at different stages of retirement (early, age 60-64; middle, age 65-69; and later, age 70 and above) regarding selected market practices: a. external information sources (advertising, family, friends, sales personnel) b. shopping assistance (sales personnel, family, friends) c. shopping habits (brands, shopping lists, price/quality relationship, comparison shopping, new products and coupons) d. the effects of transportation and health on market activity e. enjoyment of shopping f. cash and credit purchases 2. Determine the nature and extent of consumer problems identified by Michigan retirees at different stages of retirement (early, middle, and later): a. product problems/complaints b. service problems/complaints 3. Determine the relationship of demographic and situational characteristics to methods of consumer complaint resolution used by Michigan retirees at different stages of retirement (early, middle, and later). Profile of the Senior Citizen Market Traditionally, a stereotypic approach has been applied to that marketing which has been directed at the elderly, both from an income and consumer characteristics perspective (Allan, l98l). In The Economics of Aging, Schulz (1980) outlined some of the results of this situation. He stated that the research and statistics reported categorize the elderly as one homogeneous group aged 65 and above. There is no such thing as the collective aged; the aged are as diverse as the population itself, and this is just as true for economic status as it is for other areas. . . . A second major problem in analyzing data on the aged is that it is often not appropriate to group people together who differ in age by as much as fifteen or twenty years. (pp. 18, 20) Allan (1981) reinforced the heterogeneity of the aged through analy- sis of statistics which connoted that "families headed by a person between age 65 and 71 are better off financially than with a head over age 72" (p. 15). Allan also found in her research that elderly persons having a better financial position are those who are married rather than single and white rather than minority. Growth Patterns of the Elderly Population Review of the demographic characteristics of the elderly reveals that the buying power of this group is significant. Trends reSponsible for the importance of the senior/retiree market today are the (1) increase in the percentage of the population aged 65 years and older, (2) decrease in the number of children per family in the general population, (3) increase in the rate of growth in the senior market, (4) increase in longevity, (5) increase in early retirement, and (6) increase in financial resources (Bernhardt & Kinnear, 1976). Future growth trends in the United States will be occurring within the elderly segments of the population due to the declining birth rate and increasing life expectancy (State Plan on Aging, 1981-83, p. 7). This phenomenon is illustrated by a comparison of persons aged 65 and above from the years 1900 to 2000 (projected) in Table 1.1. Specifically, the growth patterns of the population aged 65 'and above have been and are projected to be dramatic. During the period from 1920 to 1950, the growth rate of this segment averaged 36 percent. From 1950 to 1960, the population over age 65 increased by another 35 percent (Siegel, 1976). During the 1960 to 1970 decade, the 20 percent increase in this group accounted for a greater population expansion than the 13 percent increase in the total popu- lation (Schulz, 1980, p. 5). Further analysis indicates that although the next three decades will evidence slowed growth in the 65-and-over population, a sharp increase in this age cohort is anticipated in the first part of the let century (Schulz, 1980). Table l.1.--U.S. Population Changes of Persons Age 65 and Above. Year 2000 1900 1980 (projected) Number of Persons Age 65 3 million 25 million 31 million and Above Percentage of Total 0 Population 4% 11% 19% Source: Population Reference Bureau, 1980. Projected growth trends also indicate an aging of the elderly segment. During the period from 1977 to 2000, the age group 65-74 will increase by about 20 percent; however, the group aged 80 and above will increase by 67 percent (Sheppard, 1978). The median age in the United States is another measure of the increasing age of the total population. The median age was just under 28 in 1970; in the year 2000 it will be 35, and in the year 2020 it will be 37 (Schulz, 1980, p. 8; Business Week, 1978, p. 64). Male/female population trends for the state of Michigan support the projection of Sheppard (1978), who stated that the aged female sector will continue to grow faster than the aged male sector. In 1980 there were 2.7 percent more females than males in Michigan. It is projected that this will increase to 3.6 percent more females by the year 2000 (Table 1.2). Table 1.2.--Michigan Population Projections for the Aged From 1980 to the Year 2000. Ages 1980 2000 (projected) Male Female Male Female (figures listed in thousands) 60-64 198 222 215 211 65-69 148 169 163 179 70 and Over 253 373 359 564 Median Age 28.19 29.51 34.18 36.16 % Total Michigan Population 8.7% 11.4% 10.2% 13.8% Age 65 and Above Source: Michigan Department of Management and Budget, Population Projections for Michigan to the Year 2000 (Lansing: Infor- mation Systems Division, Office of the Budget, 1978), able 3. Another important measure of population shifts is the aged 1 Although the aged dependency ratio represents dependency ratio. only an approximation of those dependent (or no longer in the work force) on the working population, it is frequently used since other measures of future labor force participation are not available (Schulz, 1980). Table 1.3 illustrates the changes in the actual (1979) and projected (2035) aged dependency ratio. From the table it is possible to see the rather dramatic increase in the aged dependency ratio from 1979 to the year 2035. Table 1.3.--Changes in Aged Dependency Ratio From 1979 to the Year 2035. 1979 2035 Aged Dependency Ratio .193 .360 Source: Richard S. Foster, "Demographic Trends and Social Security" (Kansas City, Missouri: Society of Actuaries Regional Meet- ing on Pensions, May 24-25, 1979). The changing dependency ratio has particular impact on the future of the social security system. The Annual Report of the Board of Trustees of Old-age, Survivors, and Disability Insurance Program (OASDI) (1979) projected the following: 1. In addition to the aged dependency ratio increasing from .193 in 1979 to .360 in the year 2035, the number of bene- ficiaries (including those under retirement age) per 100 covered (those paying in) (presently 31.3) will rise to 50 By 2035. 1Aged dependency ratio is computed by dividing the population of aged dependents or nonlabor force (those age 65 and older) by the worker population age 20 to 64. 2. Expenditures as a percentage of taxable payroll are projected to rise from 10.4 percent in 1979 to approximately 17 percent in 2035 due to the changing ratio of beneficiaries to covered workers. The political consequences of these demographic changes must also be analyzed. Robert Hudson (1978), a political scientist, enumerated three issues resulting from fiscal strains as the aged dependency ratio increases. 1. Conflict over new policies and funding for the elderly. It has been projected that public support through transfer pro- grams for aged dependents is three times that for children (Clark & Spengler, 1978). This projection is made on the basis that exclusive of education, parents typically provide the financial support of their children. 2. The accountability for agencies serving the aged will be actively scrutinized. 3. The political power of the elderly will be tested. In view of the fact that four out of five persons aged 65 or older vote, thei; aggregate voting power is significant (Sargent, 1982, p. 3 . While the United States is undergoing significant shifts in the age distribution of the pOpulation, the population as a whole is also living longer. Table 1.4 illustrates the trend toward increased longevity. Table 1.4.--Life Expectancy of Residents of the State of Michigan, 1901-1977. Year 1901 1977 Male Female Male Female Life Expectancy at Birth 53.4 55.1 69.3 76.3 Source: Health Department Statistics, State Plan on Aging, 1981-83 (Lansing: Office of Services to the Aging, 1980), p. 7. The cumulative effect of increasing life expectancy and a decline in the birth rate results in an older population that is xgrowing in both number and proportion at a faster rate than the popu- lation as a whole (Rones, 1978, p. 3). Retirees: One Market Segment or Many? The demographic changes that currently are in process rein- force the aggregate significance of the elderly in the marketplace. The paradox, however, is that marketers are just now beginning to recognize the unique consumer needs of elderly persons. Some retail- ers are conducting programs on physical fitness, investments, and art in their stores. Additionally, they are directing advertising efforts at elderly consumers and investigating their specific transportation problems (Greenfield, 1981, p. 33). Likewise, some manufacturers are making strides to market to older customers. Wilson Sporting Goods has developed a line of golf clubs to assist the physically disadvantaged elderly golfer, Levi Strauss has designed a line of clothing for the "more mature" figure, and food companies have produced a range of salt-free foods in response to the dietary problems of the elderly (Allan, 1981, p. 13). Concomitantly, transportation companies are offering special incentive programs for the older traveler who can travel at off times and seasons. Greyhound has participated in a special promotion entitled "the Senior Class" initiated by the state of Tennessee, Office of Tourism. This program offered a vacation promotional package coordinated with discounts for seniors at restaurants, hotels, 10 and tourist attractions (Allan, 1981, p. 73). Other service busi- nesses such as bankers, insurance, and brokers are recognizing the aggregate economic significance of the elderly through special investment programs and discounts on banking services (Allan, 1981, p. 14). A review of marketing literature indicated that the only criterion used for market segmentation of the elderly consumer group is age. The minimal amount of market segmentation research that has been completed commonly uses the temporal definition of the elderly as persons over 65. The existent studies have not typically analyzed the heterogeneity of the elderly as a function of differential aging rates due to psychological, biological, and social factors (Phillips & Sternthal, 1977). The result of this mode of segmentation is that the over-65 market is approached as a single homogeneous market by most marketers (Towle & Martin, 1976). Why has this sector of the population been viewed as one large group of consumers for so long? As little research has been conducted to define the needs of elderly market segments, Block (1974) suggested that marketers regard the aged as desiring similar products and ser- vices to the general adult population. Klippel and Sweeney (1974) indicated that the uncertain level of purchasing power and indis- tinctive market characteristics contribute to businesses' lack of interest in the seniors as separate market segments. Additionally, Reinecki (1964) stipulated that although the elderly are becoming an important market force due to their potential purchasing power, "the 11 'older market' will remain for some time to come primarily a market for staples" (p. 63). Ben Enis (1977) enumerated four criteria that he suggested must exist in order for business to justify a market segment: (1) identification, (2) accessibility, (3) responsiveness, and (4) significance of the current and potential buyers. Analysis of the elderly using these four evaluative criteria indicates that some of the criteria could readily be met. Other criteria, however, require further study before conclusions can be drawn regarding mar- ket segmentation. In review of the "identification" factor, Enis (1977) sug- gested that marketers do not find it difficult to identify retirement- age population segments in view of the growth in retirement communities. Additionally, more of the aged are living in their own homes, inde- pendent of their children, than ever before. Research has indicated the following household characteristics for individuals aged 65 and above: 1. More than 50 percent of the households consist of mar- ried couples. 2. Ten percent of the households contain one parent, usually the mother, plus one child. 3. Forty-four percent of the households have a single person living alone or with an unrelated person. 4. Eighty percent of the households own their own home, and 80 percent of those who own do not have a mortgage. 12 5. Less than 5 percent of the population reside in an institution. 6. Only 5 percent of this age cohort have health problems that restrict them to their homes (Allan, 1981, p. 14; Flint, 1980, p. 52; Linden, 1976, p. 51). Recognizing these lifestyle characteristics, Gelb (1978) suggested that it would be possible to access elderly persons by gearing segmentation strategies to them. While newspapers represent the primary media information source for the aged as a whole, further market segmentation combined with information processing analysis might suggest other appropriate methods of accessing elderly persons. To assess the "responsiveness" criterion of market segmenta- tion, some information is available; however, further investigation is also required. Lawther (1978) found that social integration and social activity affect consumption behavior in the elderly. Further, Phillips and Sternthal (1977) stated that "little is known about older persons' differential sensitivity to marketing variables which is basic to development of strategies to influence their consumption behavior" (p. 445). Research has indicated, however, that elderly consumers are typically among the last to adopt a new product or concept (Phillips & Sternthal, 1977). The fourth component, significance of the aged marketing sector, can be indirectly approached through an analysis of buying power and physical and life-style characteristics related to age. Buying power analysis is based on studies of population trends and income per person. A study of physical and life-style characteristics 13 of the aged must identify attitudes of the elderly regarding "age- oriented products, promotional appeals, pricing or distribution strategies--a 'pro-age' mind-set" (Gelb, 1978, p. 42). When consider- ing these two variables, one must look at their interrelationship: what is the "proportion of the elderly reporting buying behavior distinct to their age group" and whether or not this age group can afford the extra cost of marketing specifically for them (Gelb, 1978, p. 44). In an attempt to further define market segments existing within the elderly market, a national study of 209 elderly consumers (extracted from the data base of the 1978 Target Group Index) was undertaken. The objectives of the study were to analyze psychographic and buying-style characteristics of the 65-and—over age group (Towle & Martin, 1976). The findings of the study identified six natural seg- ments in the senior market and matched psychographic characteristics with each segment (1975, p. 467). From Table 1.5 it can be determined that the two most signifi- cant segments are the "conspicuous consumer" group (34.6 percent), which is characterized by changing brands and purchasing products of which their friends approve, and the "saver/planner" group (25.1 percent), which purchases unknown brands for economic reasons and plans ahead for expensive purchases. These findings, in addition to the fact that only 8.4 percent of the elderly respondents identified themselves as "brand loyalists," have implications with regard to further analysis of the consumption behavior of elderly adults. Table 1.5.--Market Segmentation Among Elderly Consumers. 14 Segment Psychographic Description EldgtIyeMgiket Saver/Planner Frank, candid, self- 25 1% (buys unknown brands) assured, confident ' Brand Loyalist Brave, courageous (does not buy for reserved, conventional, 8.4% approval of friends) insecure, not stubborn Information Seeker . . , (persuasible) K1nd, s1ncere 10.1% Economy Shopper Not brave, not dominating, (not brand loyal) not egocentric, not frank, 10.6% candid, funny, witty Laggard Not witty, not kind, not 11 27 (not persuasible) reservitive, liberal ' ° Conspicuous Consumer Stubborn, egotistical, 34 67 dominating ‘ ° Total 100.0% Source: Jeffrey G. Towle and Claude R. Martin, Jr., "The Elderly Consumer: One Segment or Many?" In Advances in Consumer Research, vol. 3, ed. BeVerlee B. Anderson (Cincinnati: Assoc1ation for Consumer Research, 1975), p. 467. Buying Power of the Elderly Carole Allan (1981), a marketing consultant who specializes in studying the buying behavior of older people, stated that the level of spending by older consumers is significant even though most marketers have not been particularly responsive to this group's pecuniary needs. She hypothesized that the levels of expenditures might even be greater if a more purposeful attempt was made to iden- tify the consumer demands of the elderly. 15 It is time for marketers to change their traditional views of the elderly. The process of aging is a continuum--there is no set chronological age at which one suddenly becomes old, loses interest in lifetime pursuits, or stops being open to new ideas. The major difference between interests at various stages of the life cycle is often in how ends are achieved. (p. 17) Within the United States there are 28 million households headed by an individual aged 55 and above who together are responsible for over 25 percent of all consumer spending or $160 billion (Bureau of Labor Statistics, 1981, p. 16). Further analysis indicates that of all consumer segments, the 55 to 64 age group is the prime consumer segment based on per capita spending. Focusing on the 65-and-over age cohort reveals a situation antithetical to what most would suspect. This group accounts for consumer expenditures in excess of $68 billion even though average household spending is less than that of other age groups (Allan, 1981, p. 16). Research has shown, however, that per capita outlays for the 65-and-above group exceed the average in the following product and service categories: vacation travel, women's hair care, specific types of women's apparel, home care services, cigars and pipe tobacco, alcoholic beverages, gifts and contributions, mobile homes, newspaper subscriptions, and food (Allan, 1981, p. 17). It has also been illustrated that those persons aged 65 and over in the higher income brackets are among the "prime customers" responsible for above average consumption of fine merchandise and services such as furs, jewelry, retirement homes, stocks, cruises, and domestic travel (Allan, 1981, p. 17). 16 Retirement Trends In the past, employers' policies regarding mandatory retire— ment have been characterized as varied and changeable as they have responded to various governmental, economic, and societal forces. Given the projected population shifts toward the elderly segment, retirement policies will likely be liberalized in the future to encourage elderly workers to remain in the declining work force (Flint, 1980). Compulsory retirement policies are found most commonly in those industries that have a high incidence of pension coverage. Thus, industries such as manufacturing, public administration, trans- portation, communication, and public utilities, have often forced the departure of older workers from their jobs at a speci- fied age, usually 65. On the other hand, trade, services and finance, insurance and real estate tend to be less likely to have either pension coverage or compulsory retirement. These are all industries with a high proportion of older workers. Even when mandatory retirement provisions are present, the majority of workers retire voluntarily. (Rones, 1978, p. 9) The impact of federal legislation on mandatory retirement remains questionable as current statistics indicate that workers are voluntarily retiring at earlier ages. Early "legislation protected workers up to age 65" (Rones, 1978, p. 9). Amendments to the Age Discrimination in Employment Act were passed in 1978, which sheltered employees from forced retirement based on reasons of age only, until age 70. The trend toward both voluntary and induced early retirement is of significance today. Flint (1980) profiled changes in retire- ment age over the past 20 years. Twenty years ago 5% of men 50 to 64 had dropped out of the work force, now it is 10%; 20 years ago 8% of those 55 to 59 had bailed out from work; last year it was 18%; among those 60 to 64 the dropout rate climbed from 19% to 38%. (p. 55) 17 There is some evidence to suggest that recent levels of inflation may be slowing this trend; however, it is expected that early retirement will remain a viable consideration for many workers in the future. What are the factors affecting an individual's decision to retire early? The Social Security Retirement History Study (1978, pp. 2-3) submitted that health is the most frequently reported reason for early retirement. "However, the health factor may be exaggerated because pressures from a work-oriented society may induce older people to cite it rather than a desire for increased leisure" (Rones, 1978, p. 4). Adequate retirement income has also been iden- tified as an important determinant in the early retirement decision. Quinn (1975) stated that the interaction between health and retirement income must also be considered. Those older workers recognizing a health problem are more likely to retire early if they view their proposed retirement income as adequate. Other determinants contribut- ing to the retirement decision are job satisfaction, number of depend- ents, and local unemployment (Rones, 1978, p. 5). Significance of the Study This study attempted to more definitively identify consumer attitudes, problems, and methods of complaint resolution among Michi- gan retirees by temporally categorizing them into stages of retirement. In this manner it was possible to attain a more detailed analysis of the particular consumer concerns of retirees during different stages of retirement. To date, no other study has attempted to analyze retirees in this manner. 18 Individuals at varying stages of retirement possess a wide range of values, attitudes, and consumption patterns. Given the aggregate impact of this sector on society and the economy, it becomes crucial that persons making decisions on behalf of retirees have a sensitivity to the salient consumer concerns of these age groups. The results of the study could be of importance to govern- mental agencies at all levels, consumer advocates, educators, and concerned marketers. Increased knowledge of the marketplace attitudes and problems of retirees will result in the development of more effec- tive public policy, consumer education programs, and marketing prac- tices. The impact of the retired consumer in the marketplace will continue to increase in significance in future years (Binstock, 1975; Dodge, 1962). The need for assistance through consumer education for the elderly is an omnipresent one-~a need that will increase as the elderly population increases. Any assistance that can be provided to the elderly consumer will increase his/her efficiency and effec- tiveness in purchase decisions. The State Plan on Aging, 1981-1983 stated that in the future a greater percentage of older persons will be living in rural Michi- gan communities. Geographical movement of retirees to outlying areas will present challenges of client accessibility and need awareness to policymakers and agencies serving the elderly. As this study sur- veyed retirees from both urban and rural locations in the state of Michigan, the findings from this study could provide a base of 19 information to assist agencies in better serving both urban and rural retirees in the future. Conceptual and Theoretical Framework This study draws from and contributes to several conceptual frameworks: decision-making theory, attribution theory, and the human ecological systems theory. Decision making is "a basic activity of the family organization. It is concerned with integrating values, goals, standards and resources in such a fashion that action results" (Paolucci et al., 1977, p. 129). Decision-making theory is particu- larly useful when analyzing the consumer complaint behavior of retirees. This framework provides a way of understanding and managing complaint behavior. As individuals and families have limited resources that they must manage, successful resolution of consumer problems con- tributes to family financial well-being. Decision making is necessary when individuals and families recognize that a consumer problem exists. The decision regarding the appropriate resolution action is particularly important to retirees living on relatively fixed income. Some individuals make the deci- sion to aggressively pursue the resolution of problems; others decide to take a more passive approach. Whatever the resulting action, exe- cution requires management skills. This study looks at the levels of complaint action taken by retired individuals and families and thus contributes to knowledge about family decision making. Attribution theory defines the concept of locus of control as the internality or externality of perceived blame for a consumer 20 problem. Valle and Wallendorf (1977) found that the type of action taken by a dissatisfied consumer tended to be consistent with the psychological distance of the attribution from the consumer. The distance ranged from internal attribution or “oneself" to increas- ingly external attributions (people known by the consumer, the retailer or retail representatives, the manufacturer, and the "larger social system") (Krishnan & Valle, 1979, p. 446). Further analysis indicated that an individual's perception of whether the nature of the problem was stable (does not vary over time) or unstable (varies over time) contributes to their cognition of locus of control (Kirshnan & Valle, 1979). Attribution theory supports the human ecological framework. Through application of this theoretical framework, a more comprehen- sive analysis of the interface between consumers and other social systems is effected. A study investigating the consumer satisfaction of retirees cannot overlook the interdependence of factors affecting the ulti- mate consumption decision. Consumer decision making is a complex interchange of behavioral, economic, and marketing factors (Sproles et al., 1977). In this regard, the application of the human ecologi- cal perspective to the study of consumer satisfaction provides a meaningful conceptual framework. The interdisciplinary nature of this approach results in a global analysis of the consumer-commercial interface with other systems. As the results of consumption decisions affect the well-being of individuals and families, the importance of consumer satisfaction 21 in quality of life is eminent. The notion of quality of life has been an underlying goal in human ecological studies. This study will con- tribute to the improvement of consumer competence and family resource use (Goal IV of the National Goals and Guidelines for Research in Home Economics, 1970, pp. 42-50). Specifically, the research find- ings will add to an understanding of consumer service needs (Goal IV-l); consumer choice making and behavior (Goal IV-2); consumers and the marketing system (Goal IV-3); resource development, allocation, and use (Goal IV-6); and levels of living (Goal IV-7). Theoretical Definitions Attitudes toward marketing--Measures designed to appraise consumers' attitudes toward marketing practices and performance with regard to promotion, distribution, price, and product (Pruden & Longman, 1972, p. 59). Gerontologyr-"A branch of knowledge dealing with aging and the problems of the aged" (Webster's Seventh New Collegiate Dic- tionary, 1967, p. 351). Level of livingf-A measure of lifestyle that reflects how an individual presently lives. The concept is not restricted exclusively to consumption (Davis, 1945). Standard of liviggf-A measure of lifestyle that reflects an individual's wants and endeavors for attainment (Davis, 1945). Operational Definitions Retiree--A person who no longer holds a full-time job with an employer and receives retirement income of some kind. Widows 22 who never worked outside the home but receive retirement income based on their husbands' work experiences are considered to be retirees. Stages of retirement-~Early stage of retirement, ages 60 to 64; middle stage of retirement, ages 65 to 69; later stage of retirement, ages 70 and above. Length of time in retirement--Group l: 1 to 2 years; Group 2: 3 to 5 years; Group 3: 6 to 10 years; Group 4: 11 to 20 years; Group 5: 21 years or more in retirement. Assumptions of the Study All research must make some assumptions. This researcher made three assumptions in the design and execution of this research. 1. Attitudes can be accurately measured using a Likert-type scale. This assumption refers to the reliability and validity of the research instrument. It assumes that the responses given are, in fact, the responses of the retiree and not those of the retiree's children and/or friends who may have assisted the retiree in complet- ing the survey. 2. Census data reported for individuals age 60 and above or 65 and above are representative of the retired population. As Census data do not provide demographic data for retirees as a separate group, the researcher must make the assumption that the data available from the Census Bureau represent the characteristics of retirees. 3. Mail survey research methodology is an appropriate and accurate mode of data collection from the elderly. While nonresponse 23 is high in mail survey research methodology, comparisons of the resulting sample to population statistics provide a check on the generalizability. The researcher assumed that nonresponse would not negate the value of the study through biased results. Overview of the Research Study Chapter II presents a review of research and literature germane to this study. Discussion of attribution theory development as it relates to consumer studies is also presented within Chapter II. The limitations of the study are discussed at the end of this chapter. Chapter III contains an explanation of the research design and methodology used in this study. Specifically, the sample selec- tion and criteria, procedures for the development of the survey instrument, pretest methodology, data collection procedures, and data analysis used in hypothesis testing appear in this chapter. The analysis of the data and findings are discussed in Chapter IV. Discussion and presentation of both the desCriptive and inferential statistical analysis appear in this chapter. Chapter V contains a summary of the research methodology and data analysis; and a discussion of the limitations, conclusions and recommendations for marketers, educators, public policy makers and future research. The Appendices and Bibliography follow Chapter V in the dissertation. CHAPTER II REVIEW OF LITERATURE We have witnessed the evolution of society from one that respected the wisdom of the elderly to one that often views old age as "a symbol of uselessness, and old persons become a social burden" (Dodge, 1962, p. 143). It has been hypothesized that this atti- tudinal change emanated from the transition of the United States as an agrarian to an industrial economy. This attitudinal transfor- mation is further supported by the emphasis on the youth population of today. In view of these societal changes, marketers must begin to realize the crucial role they play. Marketing "has an opportunity to play a significant role in the drive for social justice which is replacing the drive for security or affluence among many members of our society" (Holton, 1967). Technological advances, the pressure of economic competition, and changing social and consumer expectations have combined to make the task of responsible marketing to the elderly market segments very difficult (Barnhill, 1972). Life Satisfaction Among Retirees Gerontology. as an area of research inquiry, gained impetus with the passage of the Older Americans Act in 1965. The Act has been amended, but the purpose remains the same-- 24 25 To enable older Americans to enjoy living in security and with dignity after years of contribution to the economy and society. The intent of this legislation as stated in Section 101 of the Older Americans Act, is to assist older people in securing equal opportunity to the following services and opportunities. 1. An adequate income upon retirement in accordance with the American standard of living. 2. Access to the best physical and mental health which science can make available and without regard to economic status. 3. Suitable housing which reflects Special needs and prefer- ences at an affordable cost. 4. Full rehabilitation services for those who require institu- tionalization. Employment opportunities free of age discrimination. Retirement in health, honor and dignity. Meaningful activity within the widest range of civic, cul- tural and recreational opportunity. 8. Efficient and coordinated community services which are readily available when needed. 9. Immediate benefit from proven research knowledge which can sustain and improve health and happiness. 10. Freedom, independence and the free exercise of individual initiative in planning and managing their own lives. (Michigan Office of Services to the Aging, 1980, p. 2) Review of the ten objectives reveals that all, to a greater or lesser degree, impact on the level of consumer satisfaction or dis- satisfaction among the elderly. Consequently, the import of the consumer component in gerontological research cannot be overlooked when attempting to measure the overall life satisfaction of elderly Americans. Clark Leavitt (1976) discussed the underlying relationship between "satisfaction" and quality of life. He suggested that it is not unnatural for humans to endeavor toa "positive state." The literature has termed this state: "gratification, pleasure, posi- tive hedonic tone, happiness, joy, enlightenment, self-actualization, 26 nirvana" (Leavitt, 1976, p. 252). Davis (1945) cautioned, however, that researchers must clearly identify how they are measuring "quality of life." Davis noted that an assessment of "standard of living" reflects an individual's wants and endeavors for attainment. How- ever, an appraisal of "level of living" reflects how an individual presently lives, which is not restricted exclusively to consumption (Davis, 1945). Early measures of quality of life were centered on macro- economic or general welfare concerns of the United States (Leavitt, 1976). With the passage of time, the importance of social, cultural, and environmental goals prompted researchers to apply a broader defi- nition to "quality of life" studies (Gross, 1969). Traditional measurement techniques for quality of life have been predominantly centered on objective indicators; however, subjective indicators are now being employed as measurement tools (Leavitt, 1976). Objective measures have clustered around six general areas: (1) health and safety; (2) education, skills, and living standards; (3) income; (4) economic equality; (5) human habitat; and (6) art, science, and free time (Terleckyj, 1972). Several criticisms of using an objective approach to the measurement of a complex concept have surfaced. First, when measuring these areas, secondary data have been used. It has been suggested that there are existing social, bureaucratic, and political biases attached to the compilation of these secondary data (Leavitt, 1976). To use existing programmatically or'institutionally biased data to measure quality of life typically results in a measure that simply profiles the economic distribution of 27 resources (James, 1972). Second, methodological approaches have been used to measure each of these six areas separately. Measures are not taken to determine where individuals are in their attempt to reach these six goals. Third, "while objective data can provide certain insights, only perceptual and behavioral measures are capable of fully defining the relevant variables and generating the desired assessments" (Leavitt, 1976, p. 253). Leavitt also believed there is a need to determine the level and rate at which society is progressing toward these goals. Subjective measures of the quality of life have taken the form of attitudinal and belief assessment (Leavitt, 1976). Although this methodological approach has been applied in a problem-specific manner, collectively the studies have contributed to a greater understanding in quality of life assessments. A criticism of the general applica- tion of subjective indicators is that the resulting measure does not weight all factors relative to importance for the individual. Com- paratively little research has addressed the weighting concern. Studies that have made an attempt to resolve this problem have been conducted by Gitter and Knoche (1971) and Andrews and Withey (1974). Campbell et a1. (1976) submitted that there was a relation- ship between general 1ife satisfaction and the concept of quality of life. These researchers have identified items that contribute to an individual's perception of quality of life. Using Campbell's (1976) quality of life indicators, a follow-up study was conducted by Bearden et:al. (1979) to determine the most significant antecedents of quality (If life as viewed by elderly respondents. Elderly respondents 28 identified family life satisfaction, health, and financial security as the significant determinants of quality of life. Bearden, Gustafson, and Mason (1979) used a path-analytic technique to predict and explain the causal relationship of health, financial status, consumer alienation, level of living, and consumer satisfaction to overall life satisfaction of the elderly. The find- ings indicated that health, alienation, and level of living satis- faction were related to general life satisfaction in the elderly. Other studies have corroborated the findings of the Bearden et a1. studies. A research investigation conducted by Barfield and Morgan (1978) indicated that an individual's satisfaction with retire- ment was based on two factors, specifically, good health and a favor- able financial position. Another study conducted in Chicago revealed that "an adequate income is the single most important determinant of the well—being of persons at any stage of the life cycle, but is par- ticularly critical for elderly" (Gerontologist, 1976, p. 28). A third factor-index study revealed that social and physical isolation as well as economic achievement were the most salient factors affect- ing the elderly's feelings of life satisfaction (Abdel-Ghany, 1977). The Economic Security of Retirees In his book, The Economic Status of the Retired Aged in 1980: Simulation Projections, Schulz (1968, p. 2) cautioned that due to the relatively dependent economic status of retirees, it is crucial that society in general address two socioeconomic concerns regarding retirement: 29 1. What is an "adequate" retirement income? 2. What should be the respective roles of individuals, private industry, and governments in planning and providing income for retirement? The Michigan Region VI Task Force on Economic Security for the White House Conference on Aging (1981) developed a definition of economic security applicable to the elderly. Economic security of the elderly is not a dollar amount to be assigned for the population. Rather it is a way of life which varies for each individual with the following component factor: ability to live at or near a pre-retirement standard of living with income above the poverty level. (p. 2) The task force specifically identified five areas that together con- tribute to the economic security of the aged. 1. Social security payments form the foundation of retirement income for the retiree; it accounts for the major source of retirement income within the United States. However, the solvency of this program is currently in question and is projected to continue to experience fiscal difficulties in the future. Public and private pensions make a supplementary contribution to retirement income but only 40%-50% of retirees are covered by these pensions. Supplementary Security Income/Income Maintenance Programs comprise a means-tested source of income for certain disad— vantaged persons such as divorced women, displaced homemakers and immigrants. The entitlement of these individuals to social security benefits qualifies them for such low benefits that it would be difficult for them to live. These programs supplement social security through such sources as $51 bene- fits and food stamps. Taxation, especialky property/school taxes, encumbers indi- viduals living on fixed incomes. The cumulative effect of additional taxes such as income and sales tax places addi- tional financial stress on the retiree. Pre-retirement financial planning should be aimed at the young and middle-aged individual to assure sound financial planning for retirement. The task force felt that governmen- tal policy changes were necessary to facilitate pre-retirement planning. 3O Retirees have identified social security, disability payments, and private pensions as the three major retirement income sources (Rones, 1978, p. 5). However, income derived from these sources does not preclude the fact that many retirees seek part—time employment to supplement their income. In view of this, another trend is emerg- ing. During the past decade the percentage of men aged 60 and over engaged in part-time employment has increased from 19 to 23 percent (Rones, 1978, p. 5). Major factors accounting for involvement in part-time employment are health, income, and part-time employment opportunities. The economic well-being of aged Americans is as varied as other demographic characteristics. The following statistics exemplify the diversity in income sources for senior citizens age 65 and above. 1. Less than one in seven aged persons is considered "poor" as delineated by official government poverty indices. In the aggregate, 21 million persons over 65 are not in poverty. 2. In 1980, estimates revealed that 3.3 million households aged 65 and above had incomes of $15,000 or more. By 1990 this number is projected to increase to 5 million. 3. In 1978, the average income for the total 16 million house- holds headed by an individual aged 65 and above was $10,291, which was 3/5 of the national average income or $17,730. 4. In 1978 the aggregate household income was $162 billion for all persons aged 65 and above. 5. In 1978 the yearly per capita income for elderly persons over 65 was approximately $500 less than for the general population; higher than the age groups 35 to 44 and under age 25; and slightly less than that of those aged 25 to 34. 6. Nearly 70 percent own their own home and approximately 80 per- cent of those who own do not have a mortgage. (Allan, 1981, pp. 14-15) Dodge (1962) made some additional points that provide further insight into the economic situation of the elderly. 31 With a high rate of mortgage-free home ownership and reduced family size, older persons do not have the same financial obliga- tions as younger people. Although the negative effect of infla- tion must be considered in any analysis of older persons' financialconditions, the sources of income may counterbalance increased living costs. Retirement plans, income in kind, understatement of assets, favorable tax exemptions, and govern- mental plans and aids may indicate a heretofore unrecognized dis- posable income availability. (p. 155) ' The Effects of Inflation on the Economic Security of the Elderly It has been frequently verified that when asked to rank the social and economic problems of greatest concern to American consumers of all ages, inflation is listed as the greatest concern. A spiraling effect defines the relationship of inflation, prices, and income. Inflation can affect both the purchasing power and yields of finan- cial assets. Economic well-being, or the level of living, "basically reflects the level and pattern of expenditures, which depend impor- tantly on income and relative prices" (Torda, 1972, p. 5). Inflation is one salient factor affecting income and spending; likewise, spend— ing patterns influence job opportunities and the rate of inflation. Consumers have criticized the 19705 and early 19805 as a period of continually increasing prices. Rising prices are not a phenomenon indigenous only to the inflationary periods of the 19705 and 19805 as price increases have characterized most of the post-World War II period (Torda, 1972). According to Torda (1972), the elderly are frequently identi- fied as the segment of society that is most adversely affected by inflation due to their low and fixed incomes. 32 Inflation is widely regarded as "unfair" in its effects on income distribution. All too little is known about the distri- butional effects of inflation, and much more research needs to be done in this area. One group which obviously is hurt by inflation is older persons living on fixed retirement income. (p. 4) An analysis of the disproportionate effects of inflation on various components of the economy provides insight into the pecuniary problems of the elderly living in an inflationary economy. The net result of a highly inflationary economic period is the determination of "winners" and "losers." Consumers who possess assets that have the potential to rise in price faster than inflation are winners. Those consumers who have undertaken considerable debt will eventually become winners since future repayment will be with less valuable dollars. Conversely, the losers are those who own fixed price assets and/or have sources of income that are not adjusted for inflation. The concept of winners and losers through inflation leads economists to the basic conclusion "that inflation per se will not change the total real value of output and income, but will impact primarily on the distribution of that real output among the popula- tion" (Juster, 1978-79, p. 16). "Thus, the major function of infla- tion is defined in terms of shares of pie, where there is a largely random and hence undesirable element of chance in the distribution of rewards from economic activity" (Juster, 1978-79, p. 16). Employer policies encouraging early retirement have contributed to the economic insecurity of some retirees. The impact of inflation on an individual who retires early is manifested in three ways. First, it may leave him with a pension that steadily declines in value. 33 Second, he foregoes any future salary increased upon which his pension is calculated. Third, he will be living for a longer period of time on pension income, which typically does not keep pace with inflation. While social security is indexed for cost-of-living adjustments, those who receive social security payments at the earliest possible age for aworker, age 62, will receive an ongoing actuarial reduction in their social security pension (Main, 1974). Even with special economic inducements from the employer for early retirement, in the face of inflation, a retiree may find that the retirement income that he assumed would be ample will now support only a modest lifestyle. It has been suggested that the feasibility of early retirement is questionable unless one can meet "the rule of 85" as proposed by Union Mutual Life Insurance Company (1974, p. 51). The rule advises that if one's age plus length of work service add up to 85, it may be financially feasible for one to retire. However, one must also examine other factors that contribute to the early retirement decisions before resolving the issue. Factors such as health, interest in present job, other financial responsibilities, age of spouse, and leisure-time interests affect the decision to retire. Rao and Kiser (1977) stated that it is crucial for marketers to have a clear understanding of consumer inflation psychology and adaptive behavior in today's economy. To effect mutually beneficial marketplace adaptations in inflationary times, it is further neces- sary "to identify specific forms of adaptive behavior both in terms of present actions and future intentions" (Rao & Kiser, 1977, p. 13). In their study, Rao and Kiser found that the effects of inflation 34 have changed general consumer behavior patterns and that consumers hedge against inflation through: 1. more comparative shopping 2. greater care and maintenance of the products they use 3. spending more time on shopping activities Additionally, the data concluded that during high inflation consumers are not willing to lower their level of living by purchasing lower quality products or foregoing their favorite brands. The researchers emphasized the importance of this information for marketers to prevent wasteful misallocation of resources in the marketplace. Plonk and Pulley (1977) did a study of the financial management practices of retired couples with higher-than-average financial resources and were able to draw some conclusions on purchasing behavior changes due to inflation. When a rise in the cost of expenditures occurred, the couples took action to adjust their income and expenditures. They changed the composition of their assets to obtain a higher return, took part-time employment and substituted time and skills for money expenditures. They curtailed expenditures by selling automobiles, and reduced expenditures for entertainment, clothing, food and travel. (p. 257) As the sample used in the study had financial resources higher than the national average for the elderly, it was not surprising to note that 80 percent of the 50 couples reported use of credit cards. 'Traditionally, the elderly have not been significant users of credit. This practice may indicate a coping strategy in periods of rising prices. Schulz (1980) agreed with the traditional argument regarding 'the negative effect of inflation on a retiree's income; however, he 35 pointed out that the effects may be less than supposed. Today the retiree's major source of income, social security benefits, auto- matically increases with inflation as measured by the Bureau of Labor Statistics Consumer Price Index (CPI). Likewise, Supplemental Security Income (SSI) and food stamps are also adjusted to the infla- tion rate. Although private pensions are not generally indexed to inflation, the trend toward periodic increases in pension benefits is gaining favor by private business. Separate surveys have reported 80 percent of employers raised pensions to the already retired over the past five years. AT & T raised retirees' pensions 16 percent on December 1, 1979. General Motors plans to increase the pensions of past retirees 30 percent over the next three years (Flint, 1980, p. 52). Flint (1980) suggested additional factors that have not been widely acknowledged when considering the well-being of retirees in inflationary periods. He stated that consideration should be given to the fact that the elderly are not affected by spiraling home costs as 80 percent own their homes and most receive some break on increasing property taxes. Rising medical costs are borne in large part by Medicare and Medicaid and not the retiree. Various discounts are given to the elderly by bus companies, banks, and retailers, which reduce the effects of inflation. Various types of in-kind transfers such as food stamps and energy assistance help the lower income retiree in coping with inflation. 36 The Impact of Information Processing on the Marketplace Behavior of Elderly Consumers Due to the nature and extent of information to which indi- viduals are exposed, consumer information processing is becoming a key area of inquiry within the broader arena of consumer research (Wilkie, 1978). "Consumer information processing refers to sequences of mental activities employed by consumers in consumption contexts. These activities can at times be related to learning, evaluation, or deci- sion processes; in each instance primary emphasis is on cognitive or thinking dimension" (Wilkie, 1978, p. 223). "To summarize, information processing is a multi-stage activity consisting of search, receipt, evaluation and integration of information into an individual's problem solving process" (Cravens, 1970, p. 656). McGuire (1976) further delineated the stages suggested by Cravens. McGuire identified eight sequential stages through which an individual goes to process information affecting a consumption deci- sion: (1) exposure to information, (2) perception of the information presented, (3) comprehension of what is perceived, (4) agreement with what is presented, (5) retention of what is accepted, (6) information search and retrieval, (7) deciding among the available options, and (8) acting on the basis of the decision. Information processing by the elderly is affected by several concomitant social, psychological, and physical changes (Phillips & Sternthal, 1977). Social changes resulting from role attrition and loss of central function affect the type and quantity of information exposure. Such role shifts include changes from employee to retiree, 37 married to widowed, and parent with child responsibilities to parent in the "empty nest" (Cumming & Henry, 1961). While these social trans- formations evolve over one's lifetime, it has been stated that retire- ment acts as a catalyst to hasten the attenuation in social activity (Phillips & Sternthal, 1977). Rosow (1970) suggested that the post-75 age period is characterized by a severe reduction in social contact. Psychological adaptations in the elderly accompany role reduc- tion manifesting in narcissism (Cumming 8 McCaffrey, 1961). This phe- nomenon also causes a reduction in social participation of the aged. There is an attrition in energy levels "available to the ego for responding to, or maintaining former levels of involvement in the out- side world" (Rosen & Neugarten, 1960, p. 65). Other psychological changes result in diminution of emotional interactions and self- assertiveness (Rosen & Neugarten, 1960). Additionally, reduction in bodily functions contributes to the social and psychological changes previously discussed (Blitstein, 1971). Several research studies have indicated that a relationship exists between the reduction in social participation of the elderly and their increased reliance on the mass media (Phillips & Sternthal, 1977). Findings of a longitudinal study by Graney and Graney (1974) indicated that the elderly view more television and increased their reading to compensate for a reduced activity level. Schramm (1969) reported an increase in television and newspaper exposure in those individuals 60 years of age and older. Regarding media selection, Schramm (1969) found that elderly individuals (approximately 60 years and over) selected media on the 38 basis of information value as compared to younger adults who make their choices based on entertainment qualities. Similarly, Steiner's research found that elderly adults (over the age of 60) identified newspapers while younger adults recognized television as the most important media source (1963). Information sources affecting decision making appear to change with advancing age. There is evidence to suggest that extended family members form an essential component of the interpersonal contacts of persons aged 65 and older (Shanas & Hauser, 1974). Payne (1960) found that following retirement, elderly persons often experience a reduc- tion in the confidence of their own decision-making abilities. Conse- quently, the extended family becomes an important link in the decision-making practices of elderly adults (Payne, 1960). Rosow (1970) found a direct relationship between an elderly person's reference group and physical proximity. Specifically, the elderly have a greater propensity to rely on peers as a reference group when they have a similar marital, social, sex, and age classi- fication and when they are in close physical proximity. Rosow observed this relationship when conducting research in a residential environment where more than 50 percent of the living units were occu- pied by elderly residents. Due to the similarity in background and physical proximity, there was a significant increase in interaction between those aged 65 and above. Another consideration affecting the information processing efficiency of the elderly is that of one's ability to learn at an 39 advanced age. Phillips and Sternthal (1977) identified three arenas that mediate the learning ability of the elderly: 1. The nature of the stimulus 2. The type of situation in which information is pre- sented 3. The personal characteristics of information recipi- ents (p. 447) In analyzing the first factor, Botwinick (1973) reported that "pacing," the rate at which the information stimulus is presented, has a significant effect on the older adult's ability to learn. In a situation where the older learner can receive information at a self- determined,ratherthan externally determined pace, his learning ability approaches that of the younger learner. However, when functioning in an externally paced information setting, his learning proficiency falls below that of the younger learner. Birren (1969) hypothesized that this diminution in learning occurs due to a slowdown in the capacity of the central nervous system to process information. Con- sequently, when given adequate time to process information, the older learner does not experience a decrease in learning ability. A review of studies related to the second arena enumerated by Phillips and Sternthal (1977) identified situational conditions that impact on learning efficiency among the elderly. Factors such as the relevancy of the stimuli and the information load affect information processing. As research has demonstrated that "the difficulty of dis- criminating between relevant and irrelevant information increases 4D with age, elderly persons must be selective about the information they process (Rabbitt, 1965, p. 233). This contention was given support through an experimental study that investigated two age groups, a younger group with a mean age of 19 and an older group with a mean age of 67. When unrelated stimuli were presented to the two groups, the older group exhibited a reduced ability to perform the information processing task (Rabbitt, 1965). As contemporary society places a heavy value on being informed, a situation resulting in information overload may occur. To investigate this phenomenon, Jacoby et al. (1974) analyzed the effect of information load on product brand purchase decisions. From the findings of the study, the researchers identified the following conclusions. . .While consumers do feel more satisfied and less confused, they actually make poorer purchase decisions with more infor- mation. . . . (p. 66) . Increasing package information load tends to produce: (1) dysfunctional consequences in terms of the consumer 'sability to select that brand which was best for him and (2) beneficial effects upon the consumer's degree of satisfaction, certainty, and confusion regarding his selection. In other words, our subjects felt better with more information but actually made poorer purchase decisions. (p. 67) The personal characteristics of older subjects may help to compensate for deficiencies in learning abilities. It has been demonstrated that the reduced speed of the elderly in processing information can be offset by their reliance on their lifelong experi- ence (Birren, 1969). It is believed that accumulated life experiences assist individuals in processing information more quickly. The assumption is that with increased experience the adult forms broader concepts (an organized unit of information 4l rather than an isolated fact). Thus, while young persons process more bits of information per unit of time, an older adult processes fewer but larger chunks of information per unit of time. (Birren, 1969, p. 25) Consequently, in decision-making situations where an elderly adult has had previous experience, no decrease in processing efficiency was evidenced (Szafran, l968). One's proficiency in information processing can be affected by one's propensity to be influenced by persuasion, conformity, and suggestibility (Phillips & Sternthal, 1977). Research studies have implied that in certain instances there appeared to be an age rela- tionship to these three forms of influence. Studies have revealed that elderly adults demonstrated no greater susceptibility to these types of influence than younger adults except in situations where they viewed themselves as lacking the capability to make a decision (Nahemow, l963). Additionally, social isolation of the elderly appeared to increase their susceptibility to influence by society. Phillips and Sternthal (l977) suggested that a knowledge of information processing by the elderly has implications for communi- cations, production, and other marketing mix variables. l. If the communication criterion is information exposure, emphasis should be placed on reaching the elderly's extended family. . . . Attention should also be given to elderly con- sumers per se, particularly if the market is characterized by residential concentration of the aged. . . . In such situa- tions, the elderly's exposure to and reliance on their age peers as sources of information are likely to be substantial. 2. Newspapers are the most appropriate medium if the objective is to have elderly individuals learn new information. . . . Bernhardt and Kinnear (l976) found that a majority of the elderly respondents in their study read a daily newspaper and 70 percent read a Sunday newspaper. They also determined that the elderly preferred newspapers and broadcast media 42 for news information. Mason and Smith (l973) determined that in addition to newspapers, personal observation is a primary source for product information among the elderly. Of particular note, in the dissemination of information to the elderly, is that distracting materials should be avoided. 3. Television viewing by the elderly is heaviest during the following time periods. before 9. 00 a. m. (25 percent of the elderly). from 9:00 a.m. to 6:00 p. m. (33 percent of the elderly), and between 6:00 p.m. and 7:00 p.m. (66 percent of the elderly to watch the news). Seniors listen to AM rather than FM radio, in their homes, during the 6:00 a.m. and 9:00 a.m. period (Bernhardt & Kinnear, 1976). 4. If the strategist's aim is to influence elderly individuals, strategies based upon self-perception theory are likely to be more effective than the traditionally used persuasion approaches. 5. Older adults are often among the last to adopt a new product, service or idea. . . . One such strategy entails minimizing the complexity that the elderly perceive a new product to have. 6. .-. Because of their lower fixed incomes, the elderly are likely to be more price sensitive than other segments of the population. 7. Because of the reduced mobility that accompanies aging, the elderly are less able than younger adults to adapt to scat- tered urban shopping facilities and to engage in extensive search and shopping behavior (Anon., American Psychologist, 1974). . . . Products and services must be brought closer to the aged individual by either offering delivery services, increasing the density of retail outlets, or using door-to- door personal selling strategies. 8. If personal selling strategies are used, special problems in inter-personal interaction with the elderly should be recog- nized. The aged consumer may be unable to process what seems to be ordinary sales conversation because it resembles an externally paced task. (pp. 450-51) Consumption Patterns of the ElderLy Despite the recognized significance of the senior market, relatively little research has been conducted on the consumption activities of older adults. Bernhardt and Kinnear (l976) regarded the existing studies as limited to "implications of the increasing 43 number of individuals age 65 and over, their share of aggregate expen- ditures, and the type and sizes of products purchased by this group" (p. 449). To effect a more comprehensive understanding of the unique needs of this group, one must be cognizant of underlying factors affecting consumption behavior. A fundamental principle is that aged consumers represent a composite of consumption patterns and values that have been formed over many years. Furthermore, many aged per— sons are reluctant to change these purchase patterns. Consequently, purchase decisions in old age reflect those attitudes developed over a lifetime (Barnhill, 1972). The findings of a study conducted by Reinecki (l964) found that age itself had little influence on the expenditures of elderly consumer units (with the exception of medical expenditures). The amount of income and the size of the unit were, however, the two factors that accounted for the variation in buying patterns. These data lend support to the notion that the elderly are, in fact, a truly heterogeneous market. Bernhardt and Kinnear (1976) studied 3,435 adults age l8 and older in Atlanta, Georgia, and identified some consumption behavior characteristics among elderly shoppers. l. The elderly typically shop at higher priced department stores rather than discount stores even if it means they must travel further to the downtown shopping district. 2. Significantly fewer elderly consumers possess retail credit cards (one in six) and gasoline credit cards (fewer than one in four) compared to younger consumers. 3. Services used most frequently by seniors are long distance telephone calling (65%), taxi cabs (32%), mass trans1t (43%), and laundromats (23%). (pp. 451-52) 44 A study conducted by Mason and Smith (1973) provided addi- tional insight into the elderly's shopping patterns. The data from this study indicated that most elderly people preferred to shop in the morning hours, made their food purchases most frequently from chain supermarkets, and made their clothing and household furnishing purchases in the central business district. Other research indicated that "retirees allocated more of their budget than non-retirees for current consumption on food at home, shelter (both rental and home ownership), fuel and utilities, public transportation and medical care" (Torda, l972, p. 7). Compared to other age cohorts, retirees spend less on food away from home, apparel and upkeep, private trans- portation, personal care, reading and recreation, and other goods and services in general. In many instances, discussions centered on elderly consumers are often focused on the financially disadvantaged position of the elderly in the consumer role. In poverty areas, "shopping alterna- tives are limited to a small number of retailers, the majority of which are small scale high priced operations" (Barnhill, 1972), p. 79). Consequently, poor individuals pay from 7-2l percent more for goods and services (Sturdivant, l972). The added limitation of decreased mobility results in relatively few poor elderly consumers engaging in comparative shopping. As a consequence, aged low-income consumers have a tendency to fall prey to high-cost credit, repossessions, and in some instances court actions (Barnhill, l972). 45 In an attempt to assess the retail needs and wants of elderly consumers, Lambert (1979) conducted personal interviews with indi- viduals in two age groups, aged 55 to 64 and 65 and above. The broad nature of the open-ended questions permitted the elderly respondents to reveal a wide variety of concerns and needs related to their retail experiences. Content analysis of the interviews indicated that generally, the elderly are not a homogeneous market. Specifically, the salient concerns of the elderly shopper could be clustered into five areas. l. Oiscounts--0ue to the unstructured time schedules of the elderly, larger discounts than the conventional lO% senior discounts may be offered to elderly consumers who shop during slack shopping hours of the store. 2. Personnel training--Older consumers are attracted to retail operations which offer them courteous and helpful service. Therefore, training of personnel should stress how to approach and assist elderly consumers with regard to their unique con— sumer needs. 3. Special assistance, facilities and services--Among other things, elderly shoppers often have difficulty in locating products, making product/service decisions, reaching products on high shelves, checking out, and carrying purchases to their cars. Retailers who provide assistance and special service in these areas will gain increased patronage on the part of the older consumer. 4. Transportation-—0ften times older consumers have difficulty simply getting to the marketplace. A transportation service, whose cost is shared by several retailers and/or mall manage- ment, which picks up the elderly shopper at their place of residence would be of great assistance to the elderly. It was suggested that costs could be curtailed by providing the ser- vice only on selected days and times of the day; necessitating only a part-time driver. 5. Interactive communication with older consumers--The respond- ents inferred that communication between retailer and the elderly needed to be improved in order to provide a sense of importance and purpose for this sector of society. It was suggested that in addition to traditional market research, forums might be held to explore market opportunities as well as possible social interaction. (pp. 35-37) 46 When the suggestions generated by these elderly respondents were reviewed by a sample of retailers, varying levels of interest were indicated with regard to practicality and feasibility of implementa- tion. Consumer Satisfaction/Dissatisfaction Among the Elderly Russo (l979) identified the area of consumer satisfaction/ dissatisfaction (CS/D) as having originated with the l965-75 era of the consumerism movement. He contended that it had an empirical base in measures of social indicators and "some traditional business- oriented analysis of complaint behavior" (p. 453). Recognizing the historical beginning of 05/0, Krishnan and Valle (l979) stressed the significance of investigating consumer complaint behavior from a marketing management and social welfare viewpoint. The advantages to the marketer included a source of feedback regarding the acceptability of products and services and an opportu- nity to develop and/or revise products, services, and policies to comply with the needs and wants of the consumer. John Goodman (l979, p. 47) co-director of a study of complaint resolution completed by the Technical Assistance Research Programs, Inc. (TARP) for the Office of Consumer Affairs, further defined the benefits of effective complaint resolution for business as (l) maintenance of market share, (2) low-cost market research data base, (3) lower warranty and ser- vice costs, (4) improved employee productivity, and (5) decreased government involvement in corporate affairs. 47 There are social welfare benefits to be derived from the identification of appropriate corrective measures for disadvantaged consumers and the development of effective consumer protection pro- grams and legislation. Haines (1979) discussed an additional perspective for the study of consumer complaint behavior. The longer run effects of consumer dissatisfaction, can some- thing be done about it? . . . Research on consumer satisfaction ought also to end in helping consumers, as well as companies, governments, and quasi-governmental organizations, some of which impinge on our lives without having any market to discipline their behavior. (p. 451) Sproles and Geistfeld (1978) stated that in order for a consumer satisfaction/dissatisfaction (CS/D) study to be meaningful, it must be conducted in relationship to a specific consumer product or service area. The researchers posed specific questions in rela- tion to their study in clothing and textiles which suggested rele- vancy to any CS/D study. 1. What is the magnitude of consumers' general satisfaction and dissatisfaction with the overall product category (of cloth- ing and textiles)? . . . prices, . . . levels of quality, . . styles/assortments available, . . . types of stores retailing the products, sizes available, . . . performance of fibers, fabrics, construction. . . . To what extent are consumers satisfied with specific products (i.e. jeans, shirts, suits); . . . classifications (i.e. categories of men's, women's and children's wear)? How satisfied/dissatisfied are consumers huith clothing and textiles) as compared to other major product and service categories? What product characteristics (of the clothing-textile system) in its actual end use are most likely to result in consumer satisfaction or dissatisfaction over the “wear life" of the product? 48 5. Who are the dissatisfied consumers? Do they have a unique prof11e 1n terms of general orientations toward consuming (cloth1ng). lifestyle, and demographic characteristics? 6. What types of (clothing and textile) complaints are actually registered with retailers and/or manufacturers? To what extent does the complaining behavior of consumers represent actual levels of satisfaction/dissatisfaction in the market? 7. How do governmental regulations (of clothing and textiles) affect CS/D? To what extent is federal regulation more effec- tive than voluntary self-regulation by industry in providing ultimately for consumer satisfaction? 8. Should the (clothing and textile) market (or any sub-market) be considered to be performing efficiently (or in the consumer's or society's interest) if consumers are satisfied? What are the implications for policy makers? (p. 383) The literature reveals many issues related to the study of CS/D in products and services. Problems of definition, conceptuali- zation, theorization, and measurement are some commonly cited key research concerns (Sproles & Geistfeld, 1978). The study of consumer discontent or dissatisfaction is a relatively new area of study in market research. In recent years the concepts of consumerism, consumer protection, and the complex com- ponents of consumer dissatisfaction have been researched using both theoretical and empirical approaches (Lundstrom & Lamont, 1976). At the same time, however, questions have arisen regarding the relevancy of the research and the conceptual measure as the thrust has been on "protecting the consumer . . . without always knowing how he wants to be protected" (Advertising Age, 1973, p. 10). Along with this concern is the lack of a standardized instrument to analyze the intricate construct of consumer dissatisfaction (Lundstrom & Lamont, 1976). 49 Russo (1979) raised a question about the validity of the data base used in CS/D studies. In the critique he discussed the follow- ing issues: 1. The concerns of using survey methodology due to nonrepresenta- tive returns, interviewer oriented bias, the "weak level of the responses." 2. The sampling of only those individuals who have complained about a product or service. This instance would occur when a study was being conducted on the data from the consumer complaint files of a company or governmental agency. 3. The nonrepresentative nature of sampling from persons who reiterate only the "extreme experience." This situation may reflect exaggeration of the situation due to memory biases interjected when the occurrence is reconstructed by the con- sumer. (p. 453) Measurement of the consumer dissatisfaction construct has raised some questions by researchers in the past. Various approaches to the measurement of consumer dissatisfaction have been developed. Attitude scaling methodology developed to detect the intensity of consumer dissatisfaction with the marketing policies of business has been employed by Barksdale and Darden (1972), Greyser and Diamond (1974), Hustad and Pessemier (1973), Miller (1974), and Lundstrom and Lamont (1976), among others. The assumption is that consumer satisfaction/dissatisfaction attitudes can be measured on a scaled continuum indicating the extent and direction of the attitudes (Lundstrom & Lamont, 1976). Through analysis of various scaling techniques used by other researchers, Lundstrom and Lamont (1976) have defined consumer dis- satisfaction to include the following arenas: l. The product strategies of business including product design, quality, safety, performance, packaging and proliferation. 50 2. Business communication and information in the areas of per- formance data, warranties and disclaimers, personal selling, and advertising practices. 3. The impersonal nature of business and retail institutions in the areas of self-serv1ce reta1ling, sales personnel and cus- tomer service and merchandising practices. 4. Socioeconomic and political forces including an inflationary economy and'concern over pricing practices, government regu- lation of business and marketing, and the involvement of business in current social trends that are contrary to busi- ness practices. (p. 374) Krishnan and Valle (1979) emphasized the need for empirical support of a theoretical framework to conceptualize CS/D behavior. They described the research in this area as being atheoretical and descriptive. They argued that the literature abounds in studies that merely analyzed various types and incidence of complaint behavior generated from the records of consumer protection agencies and com- panies. Russo (1979) supported this critical analysis with three con- siderations for theoretical approaches. First, he specified that theory building in the consumer satisfaction/dissatisfaction area must scrutinize: 1. the type of theory that is appropriate, explaining output or process 2. the breadth of phenomena to be explained, a general theory to include all goods and services 3. the necessity and usefulness of such a theory The second consideration referred to the practice of over- theorizing without sufficient empirical support. The third caution raised the question of research validity due to substantial reliance 51 on survey techniques and social psychological experimentation pro- cedures. In 1976 Best and Andreasen conducted a national CS/D study for the Center for Responsive Law. The objectives of the research were "to portray the general level of satisfaction with purchases of a number of different types of products and services, to identify prob- lems encountered by a cross section of the population and to determine the actions consumers take when they encounter problems" (Bernhardt, 1981, p. 110). The study analyzed the types of problems encountered with 26 prodUCts and 8 services. Although the results were published by Best and Andreasen for the general population, the data were not analyzed regarding the specific problems of the elderly. Bernhardt (1981) conducted further analysis on the previously collected data and found differences in satisfaction/dissatisfaction of the elderly and the total population. Generally, the elderly reported somewhat greater dissatisfaction with "all purchases," "all services," and "all products" than the total population (p. 111) (Table 2.1). Specifically, the elderly population indicated dissatisfaction with a similar set of products and services as the general population; however, the rank ordering of the products and services was different. It should be noted that the results of this study did not necessarily indicate consistency with other studies. However, this study looked at many products/services and not just one product category as did other studies (Wall, Dickey, & Talarzyk, 1977) (Table 2.2). 52 Table 2.1.--Comparison of Frequency in Satisfaction/Dissatisfaction of Products and Services Among the Elderly and General Popu- lation (Best & Andreasen Study). Percent Satisfaction and Dissatisfaction Total Population Elderlya Satisfied Dissatisfied Satisfied Dissatisfied All Products and Services 87.8% 12.2% 83.7% 16.3% All Services 85.1% 14.9% 81.2% 18.8% All Products 88.8% 11.2% 84.6% 15.4% N = 28,574 N = 4.304 Source: Kenneth L. Bernhardt, "Consumer Problems and Complaint Actions of Older Americans: A National View," Journal of Retailing 57,3 (1981). aAge 65 and above group. Table 2.2.--Most Frequently Reported Consumer Problems Among the Elderly6 (Best 8 Andreasen Study). Frequency of Problem Product/Service Problem % N Appliance Repair 40.8 65 Mail Order 36.8 105 Grocery Items 33.2 404 Car 32.5 81 Eyeglasses 31.5 173 Car Parking 31.2 76 Source: Kenneth L. Bernhardt, "Consumer Problems and Complaint Actions of Older Americans: A National View," Journal of Retailing 57,3 (1981): 113. aAge 65 and above group. 53 A second national project conducted in 1976 by the Technical Assistance Research Program (TARP) for the U.S. Office of Consumer Affairs had similar objectives to the Best and Andreasen study. This study included all products and services and analyzed "who experienced problems and the types of problems" (Bernhardt, 1981, p. 110). It also investigated the method of complaint resolution most frequently used by consumers who took action. Just as with the Best and Andreasen data, Bernhardt conducted further analysis on the findings of the study. Of the 22 possible products and service-related problems that were asked about, the rank ordering for the type of problem was the same for both the general population and the elderly (Table 2.3). Recognizing that the elderly typically had lower incomes and probably made fewer purchases in total, the relationship of the number of problems to the number of products and services purchased is probably higher than is actually represented by these data. Attribution Theory Attribution theory, developed by Harold Kelley (1972), has been applied to research in various disciplines. Attribution theory is "concerned with the perceived causes of events" (Krishnan & Valle, 1979, p. 445). Weiner et a1. (1972) further developed the theory to describe the relationship between possible attributions and achieve- ment situations. Krishnan and Valle (1979) stated that "a success can be attributed either to something about the actor (internally) or to something about the environment or situation (externally)" (p. 445). 54 Table 2.3.--Comparison of the Most Frequently Reported Reasons for Consumer Dissatisfaction Among the Elderly and General Population (TARP Study). . . General Elderlya Reason for Dissat1sfaction Population Population Having Problem Having Problem Store Didn't Have Advertised Product 8.1% 4.5% Unsatisfactory Performance/ Unsatisfactory Repair 6.6% 3.2% Unsatisfactory Service 5.1% 2.4% N = 2,513 N = 494 Source: Kenneth L. Bernhardt, "Consumer Problems and Complaint Actions of Older Americans: A National View," Journal of Retailing, 57,3 (1981): 116. aAge 65 and above group. In addition, the performance can be attributed to something that does not vary over time (stable) or to something that varies over time (unstable). It has been determined that there is a relation- ship between this internality or externality component and the action taken in response to the achievement situation. When the actor perceives the attribution to be internal, a success elicits pride and a failure elicits shame (Feather, 1969). "In addition, an observer is more likely to reward or punish an individual who was per- ceived to be responsible for his/her own success or failure" (Krishnan & Valle, 1979, p. 446). Krishnan and Valle (1979) further hypothesized 55 that "the type of attribution made about a product performance will affect the type of response made by the consumer" (p. 446). Applications of attribution theory to the study of consumer complaint behavior on post-purchase dissatisfaction have been con- ducted. In this application, attribution theory is based on the premise that a consumer who feels dissatisfied because he/she was foolish when making a purchase will react differently than one who feels that the manufacturer or retailer was to blame for the problem. Krishnan and Valle (1979) established an empirical typology for consumer complaint behavior. These researchers suggested that various components of post-purchase dissatisfaction can be analyzed using the following typology: 1. The influence of dissatisfaction attributions 2. type of dissatisfaction 3. consumer demographics 0f complaint behavior The researchers found that attributions about cause of dissatisfaction act as important mediators in the explanation of complaint behavior. However, to further the analysis of complaint behavior, researchers should also ascertain: 1. individual consumer attributes importance of purchase characteristics of product/service involved earlier history of bad buying experiences social integration 0301-th awareness of information about consumer protection (Krishnan & Valle, 1979, p. 448) 56 Complaint Behavior of the Elderly In view of the widely accepted notion that the level of dis- satisfaction with consumer products and services is usually greater than the actual number of complaints to retailers, manufacturers, governmental agencies, and such, it is not unreasonable to believe that complaint behavior is "at least modestly representative of the problems consumers actually experience with products and services." Actual voiced complaints can be viewed as potential indicators of market performance, areas needing the attention of producers, and problems requiring governmental intervention (regulation) of markets (Sproles & Geistfeld, 1978, p. 369). Liefeld et a1. (1975) and Warland et a1. (1975) found the characteristics of those individuals who formally complained about a consumer problem to be that they were young, well educated, had a relatively high income, and were above average in social class and group membership. Conversely, those who had experienced dissatis- faction with products and services but didn't complain had a lower income, were less well educated, and were not involved in consumer and political activities (Warland et al., 1975). Bernhardt (1981) stated that although a large number of studies have been conducted on 05/0, problems, and complaining behavior, few have focused specifically on elderly consumers. Wall et a1. (1977) found that the over-65 age cohort had a lower occurrence of communicating complaints than younger consumers. The study also found greater consumer satisfaction among the elderly. Pfaff and Blivice (1977) found that consumers' activism is lowest in the over-65 57 age group. An activist was determined to be a person who complains verbally or takes part in citizen action groups. The proportion of the elderly who were "activists" was one-half the level of other age groups. Koeske and Srivastava (1977) found that older consumers don't know where to obtain reliable consumer information or help with problems, and they are less aware of their rights than younger consumers. Since elderly consumers are heavy users of services, a greater disSatisfaction with services has been reported than for products (Bernhardt, 1981). It has been found that at times the elderly have been reticent regarding consumer problems since they feel that by complaining they may be admitting dependence (Bearden & Mason, 1979). Valle and Lawther (1979) found complaint behavior among the elderly was based largely on psychological variables. Factors like social integration, interpersonal trust, and attribution of respon- sibility may be useful in explaining why elderly consumers complain less (Valle & Lawther, 1979). Elderly consumers who complained after an unsatisfactory buying experience tended to be high in interper- sonal trust and therefore expected truthfulness from others. Government Regulations and Consumer Satisfaction/DTSSatisfaction Several studies have presented findings relative to the rela- tionship of alienation in the marketplace and consumer dissatisfac- tion. Erich Fromm (1962) viewed the tenets of capitalism, mass production, distribution, and consumption as contributing to the consumer's feeling of alienation from the marketplace. 58 . . Man does not experience himself as the active bearer of his own powers and richness, but as an impoverished "thing" dependent on powers outside of himself, unto whom he has pro- jected his living substance. (p. 59) Ireland and Bisner (1968) have determined through measures of meaninglessness, powerlessness, anomic and social isolation that economically disadvantaged consumers have feelings of alienation in the marketplace. Survey findings of Black, Anglo-American, and Mexican-American consumers disclosed that as feelings of alienation from society increased, support increased for intervention of govern- ment in the marketplace (Pruden & Longman, 1972). Hempel and Rosenberg (1975) questioned the role of government in CS/D. They asked whether government should "anticipate or react" to consumer dissatisfaction. A reactive stance would put government in the position of policing the market and not in a role of minimiz- ing potential loss due to inefficient or inferior product selection. An anticipatory mode of operation would permit government to prevent, or at least circumvent, consumer problems. Current directions in regulations associated with CS/D have surpassed the general characteristics of pricing and quality factors. In a 1978 study, consumer concerns were focused on issues such as product labeling, safety, and warranties (Sproles & Geistfeld). Consumer information legislation has manifested in laws regarding product labeling, Truth in Lending, and unit pricing, among others. In a 1976 study, 34 federal agencies were involved in the dissemina- tion of consumer information and education (Wilkie, 1976, p. 334). Some legislation aimed at reducing consumer dissatisfaction may be rather effective while some may, in fact, be increasing the 59 level of dissatisfaction among consumers (Sproles & Geistfeld, 1978). Increased consumer awareness resulting from the mandate for the safety of consumer products has manifested in a variety of responses: "fear of the product, fear of specific brands, boycotting of products and even a general suspicion of . . . products" (Sproles & Geistfeld, 1978). It can be argued that consumer legislation prompts dissatis- faction on the part of the consumer. On the other hand, product labeling, including information and care labeling, has increased the confidence consumers have with specific product industries, therefore, indicating a reduction in consumer dissatisfaction (Sproles & Geist- feld, 1978). Product/service warranties have faced some rather unexpected reactions from consumers also. Sproles and Geistfeld (1978) raised the question of whether or not the addition of a warranty to a product or service will actually decrease consumer dissatisfaction or may even increase dissatisfaction. The researchers cited the Sentry Insurance Study (1977), which revealed that over 50 percent of the consumers surveyed indicated that warranties were "inadequate," almost one-half "felt warranties were written mainly for the protection of manufac- turers, and more than one-half felt warranties were not understand- able" (Sproles & Geistfeld, 1978, p. 390). These authors noted the Gerner and Bryant (1977) study, which found that for those manufac- turers offering a "below average warranty," consumer dissatisfaction may have actually increased. 60 Social Programs and the Elderly The extent of social programs developed for the elderly is broader than that of any other age segment of the population. Gold, Kutza, and Marmor (1976) suggested that the reasons for the variety of programs center on political, economic, psycho-sociological, and ethnical-philosophical motivations. The political rationale is based on the notion of representa- tion in a democratic government. Policy development is motivated through power derived from a voting constituency. The elderly are a significant force in this constituency due to their numbers and excel- lent voting record. It then follows that the "growth of aging programs can be understood as the response to the visibility of older people among political constituencies" (Gold, Kutza, & Marmor, 1976, p. 19). The logic underlying the economic considerations is the potential for imbalance in retirement income and expenses. A strong case is made for assisting the needy aged in procuring basic goods and services (Gold, Kutza, & Marmor, 1976). The psycho-sociological explanation is more esoteric in nature. TWo viewpoints surface in regard to the issue. The first is of a com- pensatory focus; submitting that society in general restructures the work and family environments of the elderly through mandatory retire- ment and mobility of their children and grandchildren. Hence, the social programs established by the government are "aimed at assuag- ing societal guilt" (Gold, Kutza, & Marmor, 1976, p. 19). The second viewpoint envisions social programming in a supportive role. This notion suggests that social programs form part of an ongoing 61 intergenerational support network that is passed on from one genera- tion to the next (Gold, Kutza, 8 Marmor, 1976). The fourth interpretation is of an ethical-philosophical origin and based in principles of reward systems. In this scenario the elderly are viewed as having contributed to society in a variety of ways throughout their lifetime. In their aged years it is now their time to receive back from the propitious society they helped to create (Gold, Kutza, 8 Marmor, 1976). Gold, Kutza, and Marmor (l976) questioned the effectiveness of the general political practice of developing programs based on a constituency group rather than on a problem area. While a problem may reside within a constituency, the efficiency of constituency programming is questionable since benefits are diluted by being spread across the whole class of people, some of whom need the benefits but others of whom do not. (p. 20) These authors suggested that "the more general the government program, the more difficult it is to identify who benefits and who does not." A needs assessment is essential to determine the interests and con- cerns of specific groups to which social programing is directed. In the future, as populations grow the need to know will become of even greater import. Consumer Education Burton and Hennon (1981) conducted face-to-face interviews with 322 members of senior citizen centers in Salt Lake City, Utah; Denver, Colorado; Cedar City, Iowa; and Missoula, Montana, to identify the consumer areas that the elderly would like to know more about 62 through consumer education programs. The rank ordering from highest to lowest in these areas is as follows: 1. d d —l C 12. 13. OkomVOlU‘l-wa Medical services Utilities Taxes--local, federal Food cost, nutrition Repairs--home, general Auto insurance, repairs Insurance, health, hospital Public transportation Social security and Medicare Clothing--cost, selection Government services Eye and dental care Prescription drugs Burton and Hennon (1981) pointed out that a review of current audio- visual consumer education programs indicates that they do not meet the needs of the elderly as expressed in the previous list. Burton and Hennon (1981) indicated that before determining the format of any consumer education program aimed at the elderly, consideration must be given to their ability to process information. Based on the previously discussed information processing concerns, the authors suggested some guidelines to be followed in establishing a consumer education program. 1. A slow pace with planned repetition established when intro- ducing unfamiliar concepts and material (Botwinick, 1973; Canestrari, 1963). 63 2. The amount of new information should be limited in any one session (Canestrari, 1963). 3. Sessions should be short in length (Gounard 8 Hulicha, 1977). 4. A variety of presentation techniques should be used to provide a multisensory approach (Gounard 8 Hulicha, 1975). 5. Subject matter and examples should be relevant to the experi- ences of the elderly (Woodruff 8 Walsh, 1975). Implications and Summary The 1980 census figures indicated that the over-65 group is increasing twice as fast as the under-65. In fact, projections show that: In 1975, approximately five people age 20 to 64 were alive for every individual over age 65. This ratio should remain rela- tively constant to the end of the century. If fertility rates remain at their current low levels, however, the ratio could fall below three to one by the first third of the next century. This is possible even without major increases in longevity. (Burkhauser 8 Tolley, 1978, p. 452) Concomitantly, individuals are retiring earlier, having smaller families, and living longer. These factors in themselves present new economic and social strains on society that must not be merely glossed over. Combining the above demographic changes with a dynamic social, political, and economic environment, there is reason for con- cern on the part of the other segments of society (Business Week, 1971, p. 52). Plonk and Pulley (l977) raised some salient research questions regarding the expenditure patterns of elderly consumers. As people age do they need less income because they have fewer expenses, or do they have less income and must cut expenses? Do wants and necessities decrease with age? Maybe retirement should be divided into periods of early, middle and later years and studied. Possibly each period would have its unique prob- lems with changing income and expenditure patterns. (p. 260) 64 Although the past few years have been plagued by high infla- tion, Robert Binstock (1975) indicated that we can view our future elderly more optimistically. Our future elderly generations "will be healthier, have had more formal schooling, and will have more ade— quate income to live on in retirement (p. 42). In light of the inherent changes in the future elderly population, it becomes increas- ingly apparent that more research is needed to determine the types of programs, facilities, and services that will be effective in meeting the future needs of retirees. In order to place retirees' needs in perspective, it is necessary, however, to analyze ecological changes in our entire societal environment so that there can be a dovetail- ing of social programming. CHAPTER III RESEARCH DESIGN AND METHODOLOGY This chapter describes the sample selection and criteria, procedures for development of the survey instrument, pretest method- ology, data collection procedure, and data analysis for hypothesis testing. Research Design This cross-sectional study used survey research methodology to collect data on the consumer attitudes of retired males and females in six Michigan communities. A structured self-report survey instru- ment was developed specifically for this study. The survey instrument contained several types of questions including five-point Likert-type questions to measure consumer attitudes and complaint behavior, and multiple choice and free-response questions to assess situational and demographic characteristics. The survey instrument was completed by qualified respondents in their place of residence. The questionnaire was sent and returned by mail. Attribution theory methodology was used to analyze the consumer complaint behavior of the respondents. Data were analyzed using descriptive and inferential statistical measures. Descriptive statistics were used to identify demographic characteristics and selected situational characteristics. Inferential statistical analysis was used for hypothesis testing. 65 66 Hypotheses The hypotheses were developed following a review of the lit- erature, interviews with agency and advocacy personnel working with elderly adults, and the researcher's experience in the field of gerontology and consumer affairs. Hypothesis testing was conducted to study the relationship of selected consumer attitudes by stage of retirement. Group 1 There is no significant difference between a retiree's stage of retirement and attitudes toward the marketplace. 1.1 There are no significant differences between retirees' stages of retirement and attitudes toward sales/price perceptions. There are no significant differences between retirees' stages of retirement and attitudes toward product information sources. There are no significant differences between retirees' stages of retirement and attitudes toward price/quality relationship. There are no significant differences between retirees' stages of retirement and attitudes regarding shopping difficulties. There are no significant differences between retirees' stages of retirement and attitudes toward shopping assistance. There are no significant differences between retirees' stages of retirement and attitudes toward sales personnel assistance. There are no significant differences between retirees' stages of retirement and cash/credit payment. There are no significant differences between retirees' stages of retirement and attitudes regarding shopping habits. There are no significant differences between retirees' stages of retirement and attitudes toward brand names. 67 Group 2 There is no significant relationship between selected demographic characteristics and the type of consumer problems experienced by retirees. 2.1 There is no significant relationship between age and the type of consumer problems experienced by retirees. 2.2 There is no significant relationship between the number of years in retirement and the type of consumer problems experi- enced by retirees. Group 3 There are no significant differences between selected demographic and situational characteristics and the method of complaint reso- lut1on ut1l1zed by retirees in different stages of retirement. 3.1 There are no significant differences between age and the method of complaint resolution utilized by retirees. 3.2 There are no significant differences between age and attribu- tion of blame for consumer problems among retirees. 3.3 There are no significant differences between age, attribu- tion of blame, and method of complaint resolution utilized by retirees. Sample Selection The study used a multiphase random sampling technique. The first phase, or primary sampling, involved selecting six cities or towns within Michigan from which the final sample was drawn. A detailed analysis of many communities in the state of Michigan was conducted to select the areas for inclusion in the study. The cri- teria used to make the area selections were racial composition, pre- retirement occupational distribution, income, and urban/rural locations. The six Michigan areas selected for their aggregate representation were Flint, Traverse City, Kalamazoo, Lansing, Trenton, and Jackson. 68 The second phase designated the Specific zip codes to be used in delineating the area within each city or town from which the sample would be drawn. The sampling frame was the mailing list con- taining the names of retirees arranged by zip code. The mailing list was obtained from the occupational mailing lists of the R. L. Polk Company, a national research publishing house. The third phase employed a systematic random sampling tech- nique. An initial sample of 600 was selected in proportion to the number of retirees in each of the six zip code areas. Following the return of the first wave questionnaires, demographic characteristics of the respondents were analyzed. It was determined that the sample was over-representative of white respondents and under-representative of minorities. In an attempt to obtain a racially representative sample, the second wave was drawn to over-represent those areas con- taining larger minority populations. The second wave of respondents contained 266 additional names to whom questionnaires were sent. Sample Criteria Although the questionnaire was addressed and sent by mail to the name selected from the sampling frame, any qualified retiree living in the household to which the survey was sent was permitted to complete the questionnaire. Therefore, the sampling unit was defined as the household, and the sampling element was any retired member in the household to whom the survey was sent. The extent of the sample was the six areas within the state of Michigan from which the sample was selected. 69 Before completing the survey, the retiree was required to have met specific criteria for inclusion in the study. First, a definition of "retired" was provided on the first page of the question- naire. The definition of retired read: "a person who no longer holds a full time job with an employer and receives retirement income of some kind. Widows who never worked outside the home and receive income based upon their husbands' work experiences should also complete the questionnaire." This definition was placed on the questionnaire to indicate clearly the criteria being used in the study for qualifi- cation of an individual as a retiree. As the purpose of this study was to analyze the attitudes of retirees living on retirement income only, additional criteria were necessary to qualify the retirees for participation in the study. The criteria to be met were enumerated in the general directions statement on the questionnaire. The retiree had to: 1. be currently retired 2. have been retired for at least one year be at least 60 years of age 3 4. if married, both husband and wife had to have met these three criteria Any respondent disqualified on the basis of any of the criteria was asked to check the statement, "I do not meet the requirements to complete this survey," and return the unanswered questionnaire to the researcher. The delimiting nature of the criteria provided research con- trols. The first requirement regarding current retirement guarded against the inclusion in the study of any person who had formerly 70 retired and returned to the workforce on a full-time basis. It was believed that a retiree having returned to the workforce on a full- time basis would not exemplify the spending and consumer habits of a person who relied solely on retirement income. The requirement of retirement for at least one year was stipu- lated to standardize the psychological implications of living on retirement income. Recent retirees, for example, who have been living on a pre-retirement income for less than one year may not yet have adjusted their cognitive perception of living on retirement income. Their attitudes regarding consumption decisions may be different from those who have psychologically adjusted spending and consumption behavior to retirement income, which is usually less than pre-retirement income. It was recognized that retirement can occur at an earlier age than 60 years; however, by lowering the age criterion of the respondent below age 60, the chances were increased of including a household where one spouse was retired and the other remained in the workforce. Consequently, the lower limit for the age criterion was set at 60 years. The fourth criterion established that both spouses in married couples must have met all three requirements. Again, this stipulation controlled for the inclusion of households where one person was retired and one remained in the workforce. This standardizing proce- dure assured that only households living on retirement income were included in the study. 71 Sample Characteristics The demographic characteristics of the respondents were com- pared with the Michigan and United States population characteristics for persons aged 60 and older, unless otherwise noted. When compared to population data, analysis of tabulated sample data indicated some similarities and dissimilarities across demographic categories. Over-all, the age distribution of the sample was fairly similar to the population in Michigan that is over age 65 (Table 3.1). The sample under-represented those aged 60 to 64 (12.8 percent in the sample and 30.1 percent in the Michigan population) and those between age 76 and 85 (25.5 percent in the sample and 30 percent in the Michi- gan population). The sample was over-representative of residents age 65 to 75 (50.5 percent in the sample and 43.4 percent for the Michigan population). While the Michigan and U.S. population parame- ters are not available for those classified as retirees, the implica- tion of the age profile of the respondents indicates that the attitudes of younger retirees (age 60 to 64) and older retirees (age 76 to 85) may be under-represented. The sample was fairly representative of the population on the gender variable. In the Michigan population in 1980 females over age 60 exceeded males by nearly 14 percent (Table 3.2). In the sample only 5 percent more females (51 percent) than males (46 percent) completed the interview schedule; therefore, female views as a per- centage of the over-6O population may be under-represented. As with the previous discussion regarding age distribution, Michigan and U.S. population figures are reported for all persons age 60 and above, not 72 .ooua mmwgmm uzm m new .mm .mm .moz .ommp .mugonmg gmwpgmm new amp .oz .F-_wuom on mmwgmm .mpgoqmm xgmpcmsmpaazm .cowpmps on mo msmcmu ommp .mzmcwu mgu mo ammgzm .m.= mg» sage ompsaeooea .owm_ .muwumwgmuumcmcplmcwmao: vcm ukEocoum .meoomntummpmswamm chowmw>oga .pgoqwm zgmpcmempaaam .mcwmso: ucm cowumpzaoa mo msmcmo .mzmcmu mg» we :mmgam .m.= seem umpzaao04 "mmugzom o.oo_ o.oo_ o.oop o.oo_ _apop .. .. o.¢ o.¢ mmcoamm oz <2 <2 .. o._ 28>0 8 _m <2 A+mmv o.m A+mwv 0.0 o.m om-om . . ~.m mm-Pm <2 Rem-mfiv o om Amw-m~v m mm m.o_ om mu Am>oam new mov e._e A¢N-mov a.ms Ame-mcv m.om 8.0m mu ON a.mp mo-mo m.mN F.om w.~_ a.mp ao-om Aowmfiv N Aowmpv & 8 N «acowpmpaaoa .m.: «cowpmpzaom cmmw;Uwz mwwmmwwm< mFQEmm mm< .m>on< use om mm< :owpmpaaoa .m.: use :mmwcu?z o» mpasmm wmcwpmm mo wm< to comwgmasouun.p.m mpamh 73 only retirees age 60 and above. Again, this fact may lessen the effect of the under-representation in female views. Table 3.2.--Comparison of Retiree Sample Age 60 and Above to Michigan and U.S. Population Age 60 and Above by Sex. Sex Sample Michigan (l980)* U.S. (l980)** fl_ % % % Male 90 45.9 43.1 42.0 Female 100 51.0 56.9 58.0 No Response 6 3.0 .. .. Total 196 100.0a 100.0 100.0 Sources: *Computed from U.S. Bureau of the Census, Census of Popula- tion and Housing, Supplementary Report, Provisional Esti- mates of Social, Economic and Housing Characteristics, 1980. **U.S. Bureau of the Census, 1980 Census of Population; Sup- plementary Reports, series PC 80-51-1, 1980. aDoes not equal 100 percent due to rounding. In 1980, 10 percent of the minority population was over 65, while in the sample less than 1 percent was minority (Table 3.3). In spite of efforts to over-sample in the areas identified as having larger minority populations, the high nonresponse rate from minorities resulted in their being under-represented in the sample. This low response rate among minorities may be accounted for by several factors. First, the over-all educational level of minorities is less, and the questionnaire involved the need to read and respond to a long series of questions. Second, there is a general skepticism among minorities about answering questions from a government agency or questions that 74 may be used by a government agency. The researcher feels the only way more minorities could have been included would have been by following up with personal contacts via phone calls or personal interviews in the home or at a senior citizen center where there would be help in com- pleting the interview schedule (U.S. Bureau of Census, 1978). Table 3.3.—-Comparison of Retiree Sample Age 60 and Above to Michigan and U.S. Population Age 65 and Above by Ethnic Background. Ethnic Background .éflmfllg. Michigan* U°S°** % % (Age 65+) % (Age 65+) White 99.0 90.1 89.2 Black .5 9.1 8.0 Hispanic .5 .8 2.8 Tota1 100.0 100.0 100.0 Sources: *U.S. Bureau of the Census, unpublished data, Tables 32 and 33, 1980. **U.S. Bureau of the Census, Current Population Reports, series P—20, No. 365, 1980. The pre-retirement occupation distribution of the sample dif- fered from the population in several categories. The greatest varia- tion between the sample and the population occurred with the craft/ operatives group (16.8 percent for the sample and 32 percent for the Michigan population) and the sales/clerical group (15.3 percent of the sample and 23.3 percent of the Michigan population) (Table 3.4). Accounting for the differences between the occupational characteris- tics of the sample and population, the attitudes of retirees in 75 sales/clerical and craft/operatives pre-retirement occupations may be under-represented. The sample closely paralleled the U.S. Depart- ment of Labor occupational groupings in all other categories. Table 3.4.--Comparison of Pre-retirement Occupation of Retiree Sample Age 60 and Above to Michigan and U.S. Population. . . * Occupational Category §EEEIE- EDEIEEEEL- -lifi§i::— % % (1979) % (1979) Professional, Technical/ Manager, Administrator 26.5 24.8 26.3 Sales Worker/Clerical Worker 15.3 23.3 24.6 Craft and Kindred Worker/ Operatives 16.8 32.0 28.3 Non-farm Laborer 4.1 4.2 4.8 Service Worker (including homemaker, if specified) 15°8 13'9 13’2 Farm Worker 1.0 1.8 2.8 No Response (including 2] 4 homemaker, if unspecified) ' Total 99.9 100.0 100.0 Sources: *U.S. Bureau of Census, Current Pppulation Reports, series P-20, No. 334, Tables 33 and 34, 1979. **Bureau of Labor Statistics, Employment and Earnian, Vol . 27, no. 1 (Washington, D.C.: January 1980), Tables 11 and 22, and similar documents for previous years. aDoes not equal 100 percent due to rounding. Analysis of income distribution within the sample indicates a general under-representation of the lower retirement income house- holds and over-representation of the higher retirement income households 76 (Table 3.5). Those retirees with income less than $10,000 accounted for approximately 45 percent of the sample, while those elderly house- holds having the same income represented about 76 percent of the aged U.S. population in 1978. Conversely, those retirees with incomes greater than $10,000 were represented by about 55 percent of the sample and only approximately 24 percent of the aged U.S. population in 1978. This anomaly may be explained by the fact that the U.S. income statis- tics were reported for those age 65 and above, not age 60 and above as represented in the sample. It is reasonable to expect that those age 60 to 64 may have a higher income than those older persons receiving less generous pensions throughout the years. Also, the U.S. figures are four years old and do not reflect increases in social security benefits in 1979, 1980, or 1981. The educational levels of the retiree sample and the aged population deviated in two groups, the lower (17.9 percent for the sample and 45.0 percent for the population) and upper (19.9 percent in the sample and 8.8 percent in the population) (Table 3.6). Again, this may be explained by the fact that the population educational statistics were available only for those age 65 and above, not retirees age 60 and above as in the sample. Additionally, the under- representation of minorities may contribute to the under-representation of retirees having a lower educational background. Marital status among Michigan retirees in the study differed slightly from the general population in the "married" and "divorced" categories (Table 3.7). The sample had a lower number reporting 77 Table 3.5.--Comparison of Retirement Income of Retiree Sample and U.S. Population Age 60 and Above. Retirement Income Sample -Jifiil % % Under $3,000 6.6 15.9 $ 3,000- 4,999 12.2 24.2 $ 5,000- 6,999 13.3 18.3 $ 7,000- 9,999 12.8 17.3 $10,000—14,999 17.9 12.2 $15,000-19,999 9.2 4.4 $20,000-24,999 6.6 2.4 Over $25,000 9.2 4.7 No Response 12.2 .. a Total 100.07 100.0u Source: U.S. Bureau of Census, Current Popplation Reports, Series P—60, No. 123, 1978. Note: U.S. figures would not reflect income changes from the 1979, 1980, and 1981 social security increases; thus U.S. figures may be lower than the sample data which were collected in 1982. aDoes not equal 100 percent due to rounding. Table 3.6.--Comparison of Educational Levels of Retiree Sample Age 60 and Above to U.S. Population Age 65 and Above. Ed . 1 L 1 Sample U.S. ucat1ona eve % % (Age 65+) Elementary 17.9 45.0 High School 40.3 39.7 (1-3 yr.) Some College 18.4 6.6 (3 yr. or less) Completed College 19 9 8 8 or Graduate School ' ' No Response 3.6 .. Total 100.0a 100.0a Source: Adapted from U.S. Bureau of the Census, Current ngulation Re orts, series P-20, No. 365; series P-23, Nos. 57, 59, and 85; series P-25, No. 271; and series P-60, No. 129, and earlier reports, 1980. aDoes not equal 100 percent due to rounding. 78 married status (49.5 percent in the sample and 58.7 percent in the over-65 age population) and a higher number reporting "widowed" status (36.7 percent in the sample and 32.3 percent in the aged population). The comparison to U.S. population data analyzing the marital status of those 65 and above, not 60 and above as in the sample, may account for the difference. Table 3.7.--Comparison of Marital Status of Retiree Sample Age 60 and Above to U.S. Population Age 65 and Above. Marital Status Sample U.S. Population % % (Age 65+) Married 49.5 58.7 Widowed 36.7 32.3 Divorced, Separated 5.1 3.6 Single, Never Married 6.1 5.5 No Response 2.5 .. Total 100.0a 100.0a Source: U.S. Bureau of the Census, Current Population Reports, series P—20, No. 365; series P-23, Nos. 57, 59, and’85; series P-25, No. 721; and series P-60, No. 129, and earlier reports, 1980. 6Does not equal 100 percent due to rounding. Comparing geographical representation of the sample to the distribution of elderly within Michigan indicates some variation. The relative frequencies of the communities represented in the sample indicate that the views of the elderly in urban areas may be under- represented (Table 3.8). 79 .cowumpaaog :mmwcumz quou mo ucmosmaa .m>onm can om ommm :owuwpzaoa :mmwsuwz pope“ mo pcmugmam .Ommp acowumpsfloa $0 mzmcwo .m.D .mzmcmu 0:». $0 :meDm .m.: "wuszom N.mm oom._w¢ o.oo_ map Payee .. .. .. P.m op mmcoammm oz .. 0. com.“ mmcm>mth agate ¢.om ow «meme save mmtm>mth o.¢_ ©.F Nwm.P~ comxuac m.NF mm Fommq zomxuma o.m_ o.~ ome.o~ OONmEm_ax _.mN Am _oome OONaEapax o.__ o.m www.mm EmcacH\=opmu m.¢_ mm ~_mm¢ mcwmcms 0.2, m.m NmN.Pm mammcmw P.o NF momma “cape o.m. N.N~ ¢-.mmm mesa: ~._P mm mm_me cepcmzp as we :mmflgowz xucaou & mpasmm z muoo awN\mmL< .xpcaou >5 m>on< vcm om mm< :owpm_:aoa :mmeUTE ow m>on< vcm om mm< wFQEmm mmgwpmx %o comwgmasounn.m.m wreck 80 Sample Characteristics by Stage 0? Retirement Location of residence.--To define the sample further, it is interesting to identify from which area retirees in each stage of retirement came. Crosstabulation of responses to Question 3 and Question 63 indicated that of the retirees in the early stage of retirement (ages 60 to 64), the largest percentage came from both Jackson and Traverse City (3.4 percent from each area), and the smallest percentage came from Flint (0.6 percent) (Table 3.9). Retirees aged 65 to 69, in the middle stage of retirement, were most frequently from Lansing (5.6 percent) and least frequently from Flint (1.1 per- cent). The last stage of retirement, containing retirees age 70 and above, most frequently represented retirees from Kalamazoo (23.5 per- cent) and Traverse City (14.6 percent) and least frequently from Flint (3.9 percent). Overall, as the age group 70 and above accounted for a majority of the sample, each geographical area had its greatest concentration of retirees in the 70 and over age group. Retirement income.--Whi1e the over-all assessment of the economic well being of these retirees is fairly high, a closer look at the distribution within groups indicates some dissimilar patterns. It is not surprising to note that the oldest age group is skewed more toward the lower (under $6,999) and middle ($7,000-l4,999) income levels than the higher levels (Table 3.10). Contrasting this situa- tion with the middle stage of retirement, the most frequently reported income level is the $7,000-14,999 (7.2 percent) and the $15,000-24,999 (6.6 percent). The youngest retirees, those falling into the early 81 .mmmmu mcwmmps compcm>mmo .xuwu m to cowuomm Lo .czop Ppwsm .ngznam mmmfionmcwxgozn .wwgm auwuucwccmm Aso.oopv Axw.FNv Ago.epv Axm.omv A&©.mpv Aflo.mv Axm.m_v um“? mm mm 4m mm OF mm Peach . . . . . . . m>o a Axe may AR, mpv Age my Rem va Axm NV Asa my Asa my ace om mmwv up, AN N_ Na m_ A o_ tapas ASN._NV Axm.mv 58m.mv Afim.¢v Axm.mv ASP._V Axm.~v Amo-mo ammv mm o A m o_ N m a_uuPz AN¢.m_v A&¢.mv Aae.mv ARN.NV Afim.~v Axo.ov ASF._V Aeo-oo mmmv «N m o a m P N »_me 58V pmv amv Amy nmv Amv nmv z z z z z z z .|>.1..~|.Fll| HCQEGL mama pagoh mmgm»wch ocomxomn noonEmme nmcwmcm; «paw—u acopcmgh mo macaw .ucmsmgwumm mo mmmum an compmoog mucmuwmmm .mmmgmuwm mo comwgwasoouu.m.m m_aoh 82 .mmmmo mcwmmwe mcouzagwspm Afio.oopv ARN.OPV Asp.mpv A&_.mmv A&¢.mmv 58m.mv amop up on mm ma mp m_mpop . m>o m Aeo.mov Axm.mv Afim.uv Afim.emv Ago.e_v A8m.ev new OM mmmv mop m m, me ow NP tapas Axo.mmv Axo.mv Afio.ov ASN.NV A&¢.mv .. Ame-mo «may mm m __ N_ a .. m_uuwz Aam.m_v Aso.mv Axe.mv Axe. V Axe.mv Axe. V Aeo-oo ammv mm m m P m _ spam“ ARV Amy amv Amy nmv hmv z .luummll. .111Hm11. .llbwll. .11_m11. .llmml. Les ace mom.e~ mmm.ep mam.o ooo.mm pewewmwpmm m_m80F ooo.mmm -ooo.m_w -ooo.ua -coo.ma gave: to a mum msoucm .ucmEmgwumm mo macaw xn msoocH acmemgmwmm to comwgmaeouuu.op.m mpnmp 83 stage of retirement, also had a normally distributed income pattern. This group had about the same number falling above the middle income ($7,000-14,999) as falling below. Gender comparisons.--When comparing gender by stage of retire- ment, it is interesting to note a fairly even split at all stages (Table 3.11). The only slight deviation from this general observa- tion is with those in the later stage of retirement; the distribution indicates that 53 percent are females and 47 percent are males. Table 3.11.--Comparison of Retirees' Gender by Stage of Retirement. Gender T t 1 Stage of Retirement Male Female 0 a H N 6 Early (age 60-64) 13 12 25 Middle (age 65-69) 20 19 39 Later (age 70 and above) 57 54 121 Total 90 95 1853 a . . Eleven cases m1ss1ng. Educational level.--As previously observed from Table 3.6 on page 77, this sample of retirees would be considered well educated. There are more retirees in the later stage of retirement with college or graduate school educations than the other two stages of retirement (Table 3.12). The middle stage of retirement comprised the greatest 84 .mmmmu acrmmwe campgsoum Axo.oc_v ”Rm.omv ASN.6PV Ago.mmv Agw.m_v Ago.kpv mum, Km mm me mm mm _6pop Axm.eov Ago.m_v ANP.NPV ARN.m_V ham.¢PV Axe.mpv 6:6 emmwmmw mp, mm mm 6N mm mm 26564 Aao.NNV Axm.mv Aflm.mv Axk.hv A8m.mv Axm.mv Amo-mo many ca A A a_ o o 6_uuwz ANN.m_V Axm.mv Axm.mv Axm.mv Aflo._v ARQ.PQ Aeo-oosommv am A o m m m F26“ Amy Amy Amy Amy nmv ARV z z z z z z Foogam mamaomgw mmwppou Possum saw: Foogum pawEmgwpmm Pouch Lo mmmppoo meow nwawFQEou sow: meow xgmucwsmpm mo mmmum nwumpaeou cowpmuanu mo Pw>m4 .ucmsmcwpmm mo mmwum xn cowumosum mo comwgmasousu.mp.m mpam» 85 number of retirees having completed high school (7.7 percent) and the fewest having either some high school (3.3 percent) or an elementary school education (3.3 percent). A review of the educa- tional background of those in the early stage of retirement also indicates a high level of educational attainment; about 7 percent had at least some college education. Development of the Survey Instrument The self-report questionnaire was composed of struc- tured questions to be completed by retired males and females in their place of residence. The focus of the questions was on the post- evaluative attitudes of retirees regarding consumption experiences and marketing practices (Appendix A). Relevant survey questions were developed to allow hypothesis testing and descriptive analysis. Some of the attitude items regard- ing marketplace practices were used either verbatim or modified slightly from research previously conducted by Barnes (1978), Day and Ash (1979), Krishnan and Valle (l979), Towle and Martin (1976), and Sproles and Geistfeld (1978). These questions employed a five-point Likert-type statement using the following scale: 1 = Never, 2 = Seldom, 3 = Sometimes, 4 = Usually, and 5 = Always. The attitude assessment questions were clustered into several areas. 1. The marketplace (store reputation, advertising, and sales personnel) 2. Shopping behavior (comparative shopping, shopping assist- ance, credit/cash payment, brand/price/quality perceptions, information processing, enjoyment of shopping, shopping lists, and sales/coupon usage) 86 3. Factors affecting shopping behavior (transportation and health) A second section of the survey instrument focused on the post- evaluative attitudes of retirees concerning satisfaction/dissatisfaction with products and services and complaint behavior. Both closed and open-ended questions were developed to measure the nature and extent of consumer dissatisfaction. Likert-type questions were used to assess the attribution of blame for the product/service problem and the complaint action taken to resolve the problem. The consumer satisfaction/dissatisfaction (CS/D) questions were patterned after those developed by Krishnan and Valle (1979) and Zaltman et a1. (1978). Itemized rating scales were designed to determine the level of ’satisfaction/dissatisfaction with the outcome resulting from the com- plaint action taken to try and resolve the consumer problems. The structured responses to these questions were formed from a study con- ducted by Grainer et al. (1979, p. 498). Finally, a study by Day and Ash (1979) provided background for the questions explaining why no complaint action was taken. The last section concerned questions regarding situational and demographic variables. Several types of items were used in these sections, including itemized rating scales, fill-in, and multiple choice questions. The areas of interest regarding the situational variables included items concerning: 1. social integration of retiree with family, friends, and neighbors 2. living arrangements 3. household shopping responsibilities (fl (DVDS 87 transportation to the marketplace health conditions life satisfaction reasons for retirement pre-retirement financial planning regarding consumption decisions The demographic questions requested information on: 1. GUT-DOOM 9. 10. present age (of both husband and wife) age at retirement (of both husband and wife) education (of both husband and wife) household income (both pre-retirement and retirement) pre-retirement occupation sources of retirement income sex marital status ethnic background zip code (area of residence) References used in the formation of these questions were from Sundling (l979), Kushler (l979), Farace (1978), Shadko (1978), and the Retire- ment History Study from the Social Security Administration. Pretest of the Survey Instrument Before conducting the research study, the explanatory letter and questionnaire along with a discussion of the research design, consent procedure, and risk were sent to the University Committee on Research Involving Human Subjects at Michigan State University. 88 Approval was given for the researcher to initiate the research project. Following approval by the MSU committee, a pretest of the survey instrument was conducted by using a group of elderly citizens from the Older Wiser Learning Students (OWLS) Club at Lansing Commu- nity College in Lansing, Michigan. This group consisted mainly of retired persons within the community who met on a regular basis for various educational and social activities. The researcher attended the December 1981 meeting of this group to discuss the nature and purpose of the study and to solicit qualified volunteers to participate in the pretest. Club members interested in participating in the pretest were given a cover letter describing the purpose and objectives of the study, a copy of the questionnaire, and a self-addressed postage paid envelope in which to return the completed survey instrument. Participants were instructed to complete the questionnaire at home and return it by mail within one week. Comments on specific questions were solicited from respond- ents with regard to question clarity and ambiguity and ease of com- prehension. Few comments were received on individual questionnaire items. Consequently, only minor revisions of the survey instrument were warranted. Although minor in nature, the change impacting most on the results of the actual study was a modification in the wording of the initial qualifying question in the consumer satisfaction/ dissatisfaction (CS/D) section of the questionnaire (Question 14a). The question was changed from: "Based upon your experience, have you 89 (or your spouse) had a consumer problem with a product or service in the last year?" to "Based upon your experience, have you (or your spouse) had a consumer problem with a product or service since you retired?" TWo questions were added, based on the pretest--one to indicate when the consumer problem occurred and the second to separate the service problems from the product problems. These changes were necessitated as few of the pretest respond- ents reported that they had had a consumer problem within the last year. It was determined, however, that the short time interval of one year precluded retirees from responding "yes" to this question(4la). This assumption was borne out in the actual study as the number of complaints reported by retirees increased significantly. The rationale for extending the time period to include any time since retirement was based on the less frequent purchase behavior of retirees. The addi- tion of the question regarding when the problem occurred was to pro- vide the researcher with a time frame within which the problems took place. The separation of the service problems from the product prob- lems was completed so that a more detailed analysis regarding the nature of the problems could be conducted. Another change was made relative to measuring attribution of blame and frequency with which specific complaint actions were taken to correct the problem. A five-point Likert-type scale was added to these questions to indicate the level with which Specific "actors" were perceived to be to blame for the problem and the frequency of actions taken to resolve the problem (Questions 41e and 42a). The scale ranged from "Never" to "Always." Other minor changes were 90 made in the final survey instrument to eliminate ambiguous wording in three questions. Of the 30 pretest surveys given to Club members, 19 usable questionnaires were returned, representing a 63 percent return rate. Several were disqualified on the basis of incomplete or inconsistent responses. Pretest Data Analysis Data from two sections of the pretest survey instrument were statistically analyzed for reliability. The data analyzed were from items measuring retirees' attitudes toward marketplace practices and sh0ppingbehavior,and the life satisfaction measures. Factor Analysis Before conducting the Cronbach's alpha test for reliability, factor analysis was executed to construct attitude scales by deter- mining the underlying patterns of relationships between the variables. As the analysis was conducted on the pretest data with the limitation of a small sample size (N = 19) and a relatively large number of vari- ables, two-factor solutions were forced for each predetermined group of subject related questions. Nie et a1. (1975) outlined three steps in factor analysis: 1. Preparation of the correlation matrix 2. The extraction of the initial factors resulting in pos- sible data reduction 3. The rotation to a terminal solution to search for simple and interpretable factors (p. 469) 91 As factor analysis is a form of multivariate analysis, it permits the intercorrelation within a group of interval level vari- ables. The major purpose is to summarize information from a number of variables into a smaller set of new latent variables by removing redundant information in the original variables (Frank 8 Green, 1967, p. 75). The new factors are best represented by the linear equation where (Xj) = the original manifest variables and F = new factor (Tull 8 Albaum, 1973, p. 216. F = alx1 + aZX2 + ... + aan The factor loadings represent a correlation between the observed manifest variable scores of the individuals in a sample and the factor scores. Factor loadings indicate what percentage of the total variance is accounted for by each factor (Tull 8 Albaum, 1973, p. 216). Finally, loadings are used to determine the number of resulting factors, the grouping of variables within each factor, and the variables to be discarded. For the purposes of the study, R-factor analysis was conducted to determine the correlations between variables. Rotation techniques were employed to simplify the factor structure. Simplification occurs since loadings on rotated factors are not as sensitive to the relative number of variables as factor loadings on unrotated factors (Nie et a1. 1975). Specifically, two types of rotation were used for comparison. Oblique rotation was selected based on the flexibility and realism of the method. Oblique rotation does not require orthogonality or 92 uncorrelation of the factor axes. It is more realistic than unrotated analysis as the underlying dimensions are not assumed to be unrelated to each other. Varimax rotation, the most popular rotation technique, was also used as it simplifies the factor structure by maximizing the variance of the squared loadings in each column, therefore,further simplifying factor loading analysis (Nie et al., 1975, pp. 483-85). Varimax is an orthogonal rotation technique that assumes the resulting factors are uncorrelated to each other (Nie et al., 1975). Loadings obtained from analysis of the pretest data and the resulting scale groupings are displayed in Appendix 8. Thus, each scale grouping represents a set of items or variables that are clustered around the empirical factor underlying the data. Once defined, Cronbach's alpha tests were used to determine the internal consistency of these groupings. Reliability Testing Reliability measures indicate the accuracy, on the average, of the estimate of a true score in a population of objects to be measured (Hull 8 Nie, 1979, p. 74). This concept can be represented by the following equation: X = t + e where: X = single observation t = true score e = error of measurement Further explanation illustrates the relationship of the observed score and errors of measurement (Hull 8 Nie, 1979, pp. 74-75). 93 rho (t) sigma (t)/sigma (o) [sigma (o) — sigma (e)]/sigma (o) l - sigma (e)/sigma (o) where: sigma (e) variance of the errors of measurement sigma (t) - variance of the true scores sigma (o) variance of the observed scores In calculation, when all the variation in observed scores represents errors of measurement, the reliability coefficient will equal zero. If no error of measurement exists, the reliability coefficient will equal one (Hull 8 Nie, 1979). Guttman (1945) outlined the basic assumptions underlying reliability estimation. 1. Reliability is defined as the variation over an indefinitely large number of independent repeated trials of errors of measurement over 2. an infinite population of objects for each item being measured. 3. The observed values of an individual on an item are experi- mentally independent of the observed values of any other individual on that or any item. 4. The observed values of an individual on an item are experi- mentally independent of the observed value for that individual on any other item. 5. The variances of the observed scores on each item and the covariances of the observed scores between items exist in the population. Additionally, a sixth assumption is frequently cited in reliability theory. 6. The items are measuring the same thing. (Hull 8 Nie, 1979, p. 75 94 Cronbach's Alpha Test of Internal Consistency_ To test the consistency, dependability, and stability of the scale groupings derived from the factor analysis, Cronbach's alpha, a widely used test of reliability, was selected (Horowitz 8 Golob, 1979, p. 532). Some limitations should be noted regarding the interpreta- tion of the alpha coefficient.t/Cronbach (1951) indicated that alpha, being sensitive to the number of variables used, increases as the number of test items increases. Additionally, Novick (1966, pp. 1-18) identified a second limitation in that alpha is affected by changes in the scale of the components but not by changes in location. Specifically, coefficient alpha is derived according to the following equation (SPSS-6000 Supplement, 1981, pp. 12-23): 2 a=_k_]_zisi/k=k]-§2 k-l $21 k-l :2? 2 . . . . where: Si = variance of measur1ng instrument (item, subtest, variables, column) 1 and szl = variance of the sum over k items -2 s = average item variance As with other reliability measures, Cronbach's alpha ranges from .00 to 1.00. The alpha coefficient is based on the interitem correlation and the number of test items in the scale, as previously stated. \jTherefore, the alpha coefficient will increase with increases in the number of items in the scale and the average interitem correlation (Carmines 8 Zeller, 1979). 95 The results from the Cronbach's alpha tests of internal con- sistency on the pretest data are given in Appendix C. The reliability coefficients of the scales derived from the factor analysis ranged from .91 to .35. Two questions were reworded due to the low alpha coefficients. It was determined that all questions were essential to the study; therefore, none was deleted on the basis of the low alpha coefficients. Recognizing that reliability is situation specific, Cronbach's alpha was run a second time on the actual test data. It was not until then that a decision was made regarding which groupings would be deleted from final multivariate analysis of variance on the basis of low reliability. Methodology_ When designing a research project, the researcher must con- tinually weigh the advantages and disadvantages of various methodo- logical approaches to the research problem. Following consideration of several techniques of data collection, the mail survey was selected. The advantages of this method include: lower cost and greater efficiency than personal interviews, the possibility of reach- ing persons who might not otherwise be accessible through other research approaches, and increased anonymity as compared to personal interviews or telephone surveys (Emory, 1980; Miller 8 Turner, 1979). However, perhaps the greatest advantage of the mail survey for col- lecting data from the elderly is that they are able to read each question at their own pace, which increases the effectiveness of their information processing (Botwinick, 1973). Additionally, they have 96 more time to think about each question which increases the accuracy of the response (Miller 8 Turner, 1979). Recognition must also be given to the limitations of mail surveys. The single greatest inadequacy results in bias due to nonresponse (Emory, 1980). Several studies have indicated that respondents to mail surveys are typically better educated and have a greater interest in the topic. Further analysis implied that indi- viduals responding to a given survey were those who had responded to previous surveys. Likewise, persons not responding to a given survey were generally "habitual nonrespondents” (Wallace, 1954). Response rates and response bias in mail surveys are variable depending on the subject of the survey, the interest of respondents in the subject, and the sponsorship of the survey (Emory, 1980; Miller 8 Turner, 1979). Brunner and Carroll (1969) found that mail surveys sponsored by a university received a higher response rate than those with a commercial sponsor. However, Miller and Turner (1979) found dissimilar results in the three sponsorship groups in their study; the bank or commercial sponsor elicited the highest, the university the second highest, and the research agency the lowest response rate. These researchers also found no significant effects on response bias due to the sponsorship of the study. A second disadvantage in mail surveys concerns the limitation on type and amount of information to be collected. Mail surveys restrict the researcher from extensive investigation of a research problem as complex questioning results in a lengthy questionnaire. A general rule is that a respondent should be able to complete the survey 97 within 10 minutes (Emory, 1980). Recognizing the open-ended nature of some of the questions in this study, it might be difficult for some elderly persons to complete this questionnaire within 10 minutes. Data Collection Data were collected in two waves during March and April of 1982. The survey instruments were sent by mail in two waves. The first wave of 600 questionnaires was sent out the first week of March 1982, and the second wave of 266 surveys was sent out the last week of April 1982. A second wave was used in an attempt to obtain a sample that was representative of the retired population in the state of Michigan. The second wave was randomly selected from groups who were under-represented based on initial returns. Two weeks after the first wave of questionnaires was received by the respondents, a reminder card was sent. As the second wave was sent shortly after the reminder card, it was difficult to determine the increase in the first wave response rate due to the reminder card. Likewise, two weeks after the second wave was sent, a reminder card was sent. The response rates of the communities represented in the study varied. The unadjusted (for nondelivery and disqualification) return rates by area indicated the following pattern, in descending order: Traverse City (27.6 percent), Kalamazoo (26.8 percent), Lansing (23 percent), Trenton (18.3 percent), Jackson (15.2 percent), and Flint (13.2 percent). This distribution indicates a slight under- representation from the urban and minority areas. The over-all return 98 rate, adjusted for nondelivery and return of questionnaires by unqualified respondents, was 24 percent. As a 30 percent rate is generally considered to be satisfactory, the return rate achieved in the study could be considered to be slightly low (Dillman, 1972). As the currentness of the mailing lists for each geographic area varied, it was felt that this factor may have contributed to the low response rate from certain communities. Data Preparation Editing was executed to detect data recording errors, omissions, consistency, and accuracy. While editing, lists of responses were com- piled for the open-ended questions. From these lists, subsequent codes were developed. The closed questions and attitude scales were pre- coded on the survey form itself; consequently, further coding of these questions was not necessary. Coding of the raw data was recorded on scan sheets for reading by optical scanning equipment at the MSU computer laboratory. Follow- ing the reading of the scan sheets, the data were put on a magnetic tape. Cards were punched at data processing in the Computer Center. Tabulation of the raw data provided frequency counts and contingency tables. Following tabulation and data reduction, statis- tical analysis was conducted on the Control Data Corporation 6500 model computer using Version 8 of the Statistical Package for the Social Sciences (SPSS) at the computer laboratory at Michigan State University. 99 Data Analysis Both descriptive and inferential statistics were used in the study. Descriptive statistics were used to describe the demographic and situational characteristics of the sample. Inferential statis- tics were used to test the stated hypotheses. Crosstabulation and Chi-Square Analysis Crosstabulation and chi-square analysis were conducted to analyze the association of selected demographic and situational vari— ables with the type and extent of consumer problems. Crosstabulation is a useful initial approach to analyzing the association of variables. This technique displays the data associated with each variable and provides direction for further analysis to detect the degree of asso- ciation present (Tull 8 Hawkins, 1980). The matrix derived from the classification of data by two or more characteristics is a contingency table or crosstabulation. Using this method of data analysis, an r x c contingency table would contain a matrix with r rows and c columns (Emory, 1980). A two-way tabulation allows the testing of a relationship between two variables. A three—way tabulation permits the analysis of relationship between three variables. It allows a more indepth investigation of the causal relationship between two variables while holding a third variable constant (Emory, 1980). Chi-square is a popular test of significance for nonpara— metric data. While higher levels of data can be analyzed in chi-square, 100 it is especially useful for nominal or categorical data (Emory, 1980). Chi-square is an analytic technique that determines the proba- bility of association among variables. The computational formula for chi-square is as follows (Tull 8 Hawkins, 1980, p. 481): X2= ; (01-51-12 i=1 E1 where: i = ith cell c = total number of cells 0i = observed value for cell i E = expected value for cell i 1 Following from this formula, chi-square then measures the probability of the differences between the observed values and the expected values occurring due to chance (Tull 8 Hawkins, 1980; Emory, 1980). Statistics related to chi-square are Cramer's V and phi. These provide a measure of the degree of association between vari- ables. Phi (0) is computed for a 2 x 2 table, and Cramer's V is calculated for tables larger than 2 x 2. Phi adjusts chi-square by dividing it by the number of cases. Cramer's V modifies phi based on the number of rows or columns, whichever is smaller, in the contin- gency table. The following represents the formula for phi, 2% ¢=(2‘N—) where: X2 = chi-square, 101 and Cramer's V, 2 _ 9 t V ' (min (r-l,c-l)) where: r - rows 0 ll columns Both phi and Cramer's V range from 0, meaning no association, to +1, meaning perfect association (Nie et al., 1975, pp. 224-25). Analysis of Variance and Multi- variateAnélysis of Variance Multivariate analysis of variance (MANOVA) was employed to detect differences in attitudes toward marketplace practices among the three retiree groups. Analysis of variance is a test of significance that allows the comparison of several groups to determine whether or not the group means differ. The between group variance measures the varia- tion in the means between the groups, while the within group variance determines the mean of the variance within a particular group (Amos, 1965). In practice, then, the total variance in a group of data is broken down into its component sources derived from the factors under study (Emory, 1980). Multivariate analysis of variance (MANOVA) is similar to ANOVA with the exception that MANOVA allows a vector of dependent variables to be used instead of a single dependent variable as in ANOVA. MANOVA is illustrated by the following formula (MSU Computer Laboratory, 1981, p. 38): 102 Y.. = p + ai + Bj + (a8)ij + g. 1jk 1jk where: Yijk = kth observption on the q-component response vector in the ijt cell u = q-component grand mean ai’ Bi’ and (08)1' = q components that represent two treat- ment effecis and an interaction effect, respectively Eijk = q-component vector representing discrepancies or errors between the response vector and the predicted vector The multivariate analogue of F is similar to the following equation (MSU Computer Laboratory, 1981, p. 38): _ MST “116': where: MST = a q x q matrix H of Treatment Sum-of—Squares and cross- products (hypothesis sum of products) MSE = q x q matrix E of Error Sum-of—Squares and cross-products The multivariate analogue of F represents a ratio of H and E and is 1 which are expressed as characteristic roots or eigenvalues of HE- generalized variances. Wilks Lambda statistic was used to test for significance. Wilks Lambda computes the equivalent of an F-ratio upon which the null hypothesis is either accepted or rejected for a given significance level. A nonorthogonal design was specified in the computation of MANOVA as unequal cell sizes were present. Special contrasts were performed on the attitude measures regarding the marketplace to define specific comparisons among the variables. 103 MANOVA requires at least one dependent variable that is measured on interval scale data and one factor (categorical) or one covariate. When only one dependent variable is being analyzed, uni- variate analysis of variance is run, and when more than one dependent variable is present, multivariate analysis is run. Factors are defined as integer valued variables that are categorically scaled. Factors specify the manner in which variables will be grouped in the analysis (MSU Computer Laboratory, 1981). Hypothesis Testing Hypothesis testing was used to (1) detect differences between the population values and the critical values, (2) determine differences between specific populations or sections of populations and certain variables, and (3) make inferences about the relationships among attributes or variables (Tull 8 Albaum, 1973, p. 175). Hypotheses were accepted or rejected on the basis of critical values. CHAPTER IV DATA ANALYSIS AND FINDINGS The objectives of the study were to determine the attitudes of Michigan retirees at different stages of retirement (early, age 60-64; middle, age 65-69; and later, age 70 and above) regarding selected market practices; determine the nature and extent of con- sumer problems most frequently identified by Michigan retirees at different stages of retirement; and determine the relationship of demographic and situational characteristics to methods of consumer complaint resolution. To operationalize the measurement of consumer satisfaction/dissatisfaction among retirees, several measurement techniques were used: Likert-type scales, itemized rating scales, multiple choice questions, and open-ended questions. Both descriptive and inferential statistics were used to analyze the data. Hypothesis testing was used to study the relationship of selected variables. Chi-square analysis, Cronbach's alpha test of reliability, analysis of variance, and multivariate analysis of variance were the statisti- cal tests run on the data. Descriptive analysis was used to report measures of central tendency, frequency counts, and crosstabulation. This chapter presents a discussion of the statistical analysis, hypothesis testing, and the findings of the study. Tables are pre- sented that display the results of both the descriptive and inferen- tial analysis. 104 105 Attitudes of Retirees Toward the Marketplace at Different Stages of Retirement This section of the data analysis is presented in two parts. The first part describes the hypothesis testing of group 1 hypotheses. The second part presents findings from the statistical analysis of three attitude items analyzed as singlets. Respondents' attitudes toward selected marketplace practices were operationally defined by scale groups containing a series of five-point Likert-type statements ranging from 1 = Never to 5 = Always. Individual questions from the survey instrument were clustered to form each of the nine scale groups (Table 4.1). Items 22, 25, 29, and 32 were combined to create a new variable identified as sales/price shopping behavior, scale 1. Items 9, 14, 30, 34, and 35 were clustered to form scale 2, price/quality consciousness. Scale 3, product information sources, was created by combining items 13, 20, 38, and 39. Shopping difficulties, scale 4, was formed by items 6, 24, and 26. Scale 5, salesperson assistance, was developed by clustering items 4 and 5. Items 16, 17, 23, and 33 formed the basis for scale 6, shop- ping assistance. Items 7, 8, 36, and 37 were combined to create scale 7, cash/credit payment. Scale 8, shopping habits, was developed by clustering items 10, 11, and 12. Scale 9, brand names, was formed by items 19 and 28. Factor analysis conducted on pretest data was used to identify the variables to be aggregated for each attitude scale grouping. Cronbach's alpha test for reliability was again computed on the test data to check the over-all reliability and the reliability of each attitude group. 106 Table 4.l.--Attitude Scale Groupings of Survey Instrument Items. figem Items Used for Attitude Measures Scale 1: Sales/Price Shopping Behavior 22. I use discount coupons frequently when making food purchases. 25. I usually watch ads for announcements of sales. 29. I read newspaper advertisements closely looking for bargains. 32. High-priced products are always a better quality. Scale 2: Price/Quality Consciousness 9. I enjoy shopping very much. 14. I shop around a lot to take advantage of specials or bargains. 30. I like shopping at sales. 34. I usually compare prices on grocery products at several different stores before I decide on which to buy. 35. I usually compare prices on non-grocery products at several dif- ferent stores before I decide on which to buy. Scale 3: Product Information Sources 13. I find newspaper advertising helpful in making purchase decisions. 20. I read the newspaper every day. 38. When I read a newspaper I read it thoroughly. 39. I get more consumer information from reading a newspaper than from watching television. 'Scale 4: Shopping Difficulties 6. I usually shop by myself. 24. The condition of my health makes shopping a very difficult process. 26. Transportation problems have restricted my shopping opportuni- ties. 107 Table 4.1.--Continued. figem Items Used for Attitude Measures Scale 5: Salesperson Assistance 4. When shopping I generally seek out a sales clerk for assistance. 5. I find a sales clerk's opinion and evaluation of products help- ful in purchase decisions. Scale 6: Shopping Assistance 16. I prefer to buy products of which my family and friends would approve. 17. I generally rely on my family for new product information. 23. Before making major purchases I frequently consult my family and friends for advice. 33. When shopping I make my purchase decisions based upon the repu- tation of the store. Scale 7: Cash/Credit Payment 7. I pay cash for most of my consumer purchases. 8 I generally maintain a charge account in the stores in which I shop. 36. I buy many things with a credit card. 37. To buy anything other than a house or car on credit is unwise. Scale 8: Shopping Habits 10. Before grocery shopping I make a detailed shopping list and buy only those products on the list. 11. Before shopping for products other thanpgroceries I make a . detailed shopping list and buy only those products on the 11st. 12. Before shopping I set an upper price limit and stay under that limit in actual purchases. Scale 9: Brand Names 19. I feel new products are inferior to older established products. 28. Older established brand names are superior in quality to new, recently introduced products. 108 The results of the Cronbach's alpha tests were used to make a determination regarding the internal consistency of each attitude group. Intuitive judgment was necessary to identify items that might be considered to be unreliable. In a general sense, however, an r equal to i .70 to 1.00 was considered to be highly reliable and t 0.20 to 0.40 was identified as low or unreliable (Amos et al., 1965, p. 64). The alpha coefficient for the entire group of attitude measures was .73 which was considered to be moderately high, using the scale above. A few scale groupings were determined to have a low alpha coefficient; therefore, they were deleted from the multivariate analysis of variance statistical test. Individual alpha coefficients for the subsets of attitude measures ranged from .55 to .72 (Table 4.2). Table 4.2.--Reliability Coefficients for Attitude Scale Groupings: Actual Test Data. Scale Grouping Coezfgfigent 1. Sales/Price Shopping Behavior .69 2. Price/Quality Consciousness .72 3. Product Information Sources .67 4. Shopping Difficulties .59 5. Salesperson Assistance .69 6. Shopping Assistance .55 7. Cash/Credit Payment .70 8. Shopping Habits .63 9. Brand Names .64 109 Multivariate analysis of variance was used to test group 1 hypotheses. The dependent variables were the attitude subsets. As several of the attitude questions were negatively stated, it was necessary to reverse the coding of the responses for purposes of statistical analysis. Responses were reversed to question numbers 8, 15, 24, 26, 28, 31, and 36. The dependent variable, retiree's age, had three groups identified: those retirees age 60 to 64, 65 to 69, and 70 and over. Special contrasts were designated to make comparisons among the groups of retirees. Group 1 Hypptheses Testing There are no significant differences between retirees' stages of retirement and attitudes toward the marketplace. One statistical test was run which combined the following hypotheses, as well as testing them individually. 1.1 There are no significant differences between retirees' stages of retirement and attitudes toward sales/price perceptions. 1.2 There are no significant differences between retirees' stages of retirement and attitudes toward product information sources. 1.3 There are no significant differences between retirees' stages of retirement and attitudes toward price/quality relationship. 1.4 There are no significant differences between retirees' stages of retirement and attitudes regarding shopping difficulties. 1.5 There are no significant differences between retirees' stages of retirement and attitudes toward shopping assistance. 1.6 There are no significant differences between retirees' stages of retirement and attitudes toward sales personnel assistance. 1.7 There are no significant differences between retirees' stages of retirement and cash/credit payment. 1.8 There are no significant differences between retirees' stages of retirement and attitudes regarding shopping habits. 110 1.9 There are no significant differences between retirees' stages of retirement and attitudes toward brand names. Marketplace practices were measured on nine scales including sales/ price perceptions, information sources, price/quality relationship, shopping difficulties, shopping assistance, sales personnel assist- ance, cash/credit payment, sh0pping habits, and brand name perceptions. Multivariate analysis of variance statistical tests indicated that no significant differences existed between the three groups of retirees regarding their attitudes toward the marketplace at a sig- nificance level of .05 (Table 4.3). The null hypotheses were there- fore accepted--that there is no significant difference in attitudes about marketplace practices by stage of retirement. While the hypotheses were not statistically significant, there were some differences in attitudes among the groups. Review of Table 4.3 indicates a profile of attitudes for retirees in each of the three stages of retirement. The younger retirees, age 60 to 64, reported a greater ten- dency than the other two groups toward price consciousness through the use of discount coupons, seeking out bargains and sales (mean score - 3.55), and comparison shopping (mean score = 3.35). This group also indicated the greatest use of advertising, in general, and newspaper advertising, in particular, for product information (mean score = 3.99). Of the three groups, more retirees in the middle stage of retirement (age 65 to 69) expressed the attitude that new products and brands were inferior to older established products and brands 111 .00. u a pm acmowmwcmwm mew: mmcwazoem oz0 0NN00. 00000. 00.N 00.0 00.N 0620: 00000 .0 00N0_. 0_000._ o_.0 00.N 00.N 00000: 00000600 .0 00N0N. 00P0N. 00.0 0N.m 0N.0 pcmesaa 000000\0000 .N 00000. Nmeoo. 00.N 00.N 00.N 660000000< 00060600 .0 NNFNO. 00NN0. _0.N N4.N 00.N 660000000< eomtmammpmm .0 NmepN. F_000.F NN.0 00.0 00.0 060003600000 00000600 .0 . . . . . mogzom 00000 00N0~ P 0N 0 0m 0 00 0 06000526000 063060; .0 . . . . . mmmcmaowumcou NFNOP 00N_0 N 00 N 0_ 0 0m 0 s00.0so\660za .N . . . . . Lo0>mzmm 0N00N 000NN N0 0 00 m 00 m 0:006600 «6026\06P0m .0 :00: :00: com: o . 6>00<-0N 60< 00-00 am« 00-00 60< ‘ mucmwwwwcmwm ANNP NV 0 team; mpuuwz xpcmu mmcwazogw m_mum pavemewpmm mo mmmpm .pcmsmcwpmm to mmmpm >0 wum_aumxemz as» ugmzoh mmczuwpu< to wucmvem> do mwmxpmc< mu0000>0upsz-u.m.¢ mpamh 112 (mean score = 3.0). These retirees reported the greatest incidence of shopping alone and the fewest health and transportation problems restricting shopping activities (mean score = 4.03). Retirees in the later stage of retirement, age 70 and above, indicated the strongest attitudes of the three groups toward paying cash for consumer purchases (mean score = 3.36) and preparing shopping lists and purchasing only those items on the list (mean score = 3.10). This group most frequently reported health or transportation prob- lems hampering shopping activities (mean score = 3.72) and indicated the greatest need for shopping assistance by sales personnel (mean score = 2.61). The frequencies and mean scores of items contained within each of the nine scale groups are presented in Table 4.4. Analysis of the mean scores revealed that, on the average, all groups of retirees had a positive attitude toward price consciousness, newspaper advertising as a source of consumer information, newspapers as an important general information source, paying cash for consumer pur- chases, and transportation problems having restricted shopping oppor- tunities. In general, the three groups of retirees indicated that they did not usually: comparison shop for grocery or nongrocery products, find sales personnel helpful when shopping assistance was needed, consult with family and friends before making major purchases, make a shopping list for nongrocery items and restrict purchases to only those items on the list, and find new products and brands to be inferior to older established products or brands. 1'13 606.66 600.6N6 606.6N6 606.066 606.6N6 .666 66 66606100 6 .666 6 666666 066660 666666666 m.~ mp 06 mm mm 06 6666>60 06 06666666 666666 1:6: co 066066 6666566 >_F6606 H .00 606.66 600.6N0 606.0NV 606.6N6 600.0N6 .666 66 66663.mm.mm6m66 _ 666666 066660 666666666 6.N up mm me me 66 p666>m0 06 00066666 66666 no 066666 6666566 6pp660: ~ .60 606.666 606.0N6 606.0N6 606.666 606.066 6.0 Nm me me mm mm .06660 06 66666660 660. H .60 606.066 600.6N6 666.6N6 606.0N6 666.666 .06666666 66 6.0 0N 60 66 06 6N 0.666660 66 666666666 6666 66 666 6 666666 6660 6 .66 606.N66 606.606 600.606 606.066 606.66 0.0 MN 00 N0 6N 06 .6666 6666 66666660 66666 6 .6 0066066060606 00066:O\6ow66 "N upoum .. 606.6N6 606.606 606.066 666.N6 6.0 .. 60 app 60 6 .xu_pmaa 660066 6 066366 666 00666666 666066-660: .Nm 606.606 600.6N6 606.6N6 600.066 606.06 .06666666 0.m mm me we mN F— 666 666666. 6p60o—u 00662606066>66 66666036: c666 H .mm 606.606 666.006 606.666 666.666 606.60 6.0 60 60 mm mp mp .06—60 66 0066266666666 666 066 6606: 066660: 6 .mm 606.6N6 606.N06 666.6N6 660.N.6 66N.66 .060666666 6.0 N0 .0 66 ea 0 6666 m:_xos 66:: x_u:6:c666 0666666 06666006 60: H .mm 6op>6cmm 66066660 66666\0m—6m up 6,660 066 606 666 666 666 :66: z z z z z 0Emu~ 6:666 upmum 606 666 606 6N6 666 0xmz—< appma0= 065666560 6m>6z 666660 1...... 11' '1 1.. 666666066662 6;» 666306 06660006< .06660661 66 :o0w6oaeou-u.¢.e 6666» 1'14 H00.HH Hue.0HV Hum.mmv H06.~mv Hmo.6HH .06660Hu6u 60666666 :6 H666H66 0.~ m mN H0 00 H0 00666666 66 666066H6>6 666 6606666 0.666H6 06H60 6 6666 H .0 606.66 600.06 606.066 666.006 606.06 .6666606006 0.N N 0H mm mm 0H 666 x66Hu 06H60 6 666 6660 >HH666=66 H 66H66660 :66: .6 666660600< 66066606H60 “0 6H660 600.06 600.06 606.NHH 606.666 606.006 .0666666666666 H.e NH NH em 00 mop 66H66660 6: 6666660066 6>66 0E6H6666 66666066606666 .mm H06.66 666.66 H60.0Hv 606.0N6 H06.N66 .0066666 6.6666666 6.0 6H 0H mm we 66 666> 6 66666660 06666 60H666 65 66 666066666 666 .6N 600.066 606.006 606.666 606.0N6 600.66 N.0 6N 60 N0 H6 66 .666066 06 6660 06.6606 H .0 0666H6606666 66066660 "6 6H660 606.6N6 600.066 606.6N6 666.66 H00.66 .6660666666 6666666: 6666 6666 6.0 cc Hm 06 0H m 66666036: 6 66H6666 5666 :oHuue66666 66560666 6605 666 H .60 666.606 H0N.60H 600.060 666.066 600.06 6.0 60 00 N0 0N 6H .6H66666660 60 6666.H 66666036: 6 6666 H 6662 .mm 600.N66 660.066 HaN.06 666.06 606.06 0.6 60H 0N 6H HH 0 .666 x66>6 666660366 66» 6666 H .om 606.666 606.600 666.006 666.N66 606.66 6.0 HN 66 60 6N 6 .066606666 60666666 666666 66 6666—66 66606666666 6666 H .0. 0666660 6606666666H 0666666 "0 6H660 emv 066 666 006 666 6662 2 z z z z 6 066 666 6 66 H06 H66 H06 6N0 6.6 6H 6 H 0 0603H< xHH660= 066606560 socH6m 66>6z P LIT .666666666--.6.6 6.666 1'15 606.606 606.006 606.6_. 606.66 606.66 .606666 00 6.6 we on 60 6H 0H 666666 66 666 66 60666 6 6666 66666 66666666 666 66 .60 660.006 606.606 606.66 666.66 .. 6.6 66 mm 0H 6H .. .6666 666666 6 666: 066666 6666 666 H .60 606.66 600.066 606.0_6 606.666 600.606 7 .6660 6 6666: 66 6.6 m 6N mm mm 66 066660 666 66 6666666 666666 6 66666666 >HH666666 H .6 606.606 606.606 600.66 606.66 606.66 6.6 66H 66 6H m 6 .060666666 66660666 66 66 6066 666 6066 666 H .6 6666666 666666\606u ”6 6H6um 666.66 Aum.mmv An~.va 666.6Hv Aao.NHV .66660 666 66 6666666666 666 6.6 6 mm 66 06 mm 6666 66066 066606666 60666666 a: 6666 H 66666660 6663 .60 606.666 606.666 606.606 606.666 606.666 .660666 666 0666666 666 6.6666 m.~ Nu Hm mm on 60 66 6H60666 6H6666666m H 060666666 66666 666666 666666 .mm 600.06 606.666 606.606 606.606 600.066 6.~ m HN 66 66 N6 .66666666666 6666666 366 666 6H666w 66 66 6H66 AHH666666 H .6— Aum.6v ANN.mMV 66H.6Hv “66.H~v 566.6Hv .6666666 6H663 0666666 6.0 66 66 00 60 60 666 6H6666 66 6606: 66 06666666 666 66 666666 H .66 666660600< 66666666 ”6 6H6um 666 666 666 666 666 6662 2 z z z z 0666H 6666c 6H66m 60v 666 606 666 666 0>6IH< 6HH660= 06666666m 666F6m 66662 '1' J N -.l 1.' 0| ..N"ll. 1"!" .II‘1|I‘1I.'!‘ Il'sII‘QIIII i 1‘1: .666666666--.6.6 66666 1'16 .c.e mpamh cw ummgm>wg no: use mcopumwaa umumum apm>wuummc o» mmmcoammm "muoz Auv.mv Amm.NNV Anm.mmv Amm.~pv Anm.~v .muuauoga uuuavoguc_ xpucwuwg 3m: o» p.m m ow sop mm c xuwpasa cw govgonam ago muse: vcogn cmgmwpnmumo gmcpo .wm 5am.¢v Axe.mpv Ame.oev Aup.o~v Anm.av .muuacogg m.~ m mm mm on up umzmppnmumm Lmupo o» Lowgmmcw «so muuauoga 3m: pom; u .mp moEmz tango "m mpoum Awo.mv Aao.pmv Aa~.m.v Rum.o~v Aum.o~v .mmmozugaa .oauuu cw »_s,_ ”as“ m.~ or mm mm mm mm Lung: xmum can uwspp muwgg song: so umm a mcvggozm wgoumm .m— .amwp Ago.~v Am¢.omv Au~.omv ARN._~V flap.omv mg“ no muupmogg «mega apco xaa can um._ mcvgaogm umpvmumu m.~ er mm mm mm um a mxae u mm_gmuogm saga gmcuo muuauoga Lou mcwqaogm unamwm .FF Aum.m—V “no.pev Amp.m_v Amp.mpv Auo.wv .umpp us» no muuauogn wmo a x no »:a van m.m om sh mm on mp umwp a=*QQOgm um_,~umc a mxas H m=_nao;m gmuog ugo$mm .op maven: unpaaozm "w upmum Amy ”my amv flmv hmv com: 2 z z z z mewun nzogw mpuum Amv Aev Amv ANV A_v mxm3p< »——m=m: mmevuwsom socpom Lo>oz .umacpucou--.e.e opaah ll7 Other Findings--Attitudes of Retirees Toward theTMarketpTace Three attitude items were not included in the MANOVA test due to low reliability with other variables in each respective scale grouping. As these three items were considered to be important atti- tude assessments, they were analyzed as singlets. Chi-square analysis was used to test the significance of the relationship between stage of retirement, question number 3, and each separate attitude measure, question 27, 3l, or 40. As previously discussed, each attitude ques- tion was measured on a five-point Likert-type scale ranging from l = Never to 5 = Always. The stages of retirement were operational- ized as early, ages 60 to 64; middle, ages 65 to 69; and later, ages 70 and above. Attitudes toward advertising.--The three stages of retirement were crosstabulated with question number 27. Question 27 reads, "In general, advertising presents a true picture of products.“ Chi-square was not significant at the .05 level; therefore, no statis- tically significant association exists between stage of retirement and attitudes toward advertising as a true picture of a product (Table 4.5). While this relationship is not statistically significant, it is interesting to note the apparent attitude differences between individuals in the three stages of retirement. Those age 70 and above stated a more positive attitude toward advertising as a true picture of a product (22.6 percent reported "usually" to "always"). Retirees in the middle stage of retirement, age 65 to 69, indicated a somewhat 118 more negative view of advertising (5.8 percent indicated "never" or "seldom"). Retirees in the early stage of retirement were slightly more negative (3.4 percent indicated "never" or "seldom") in attitude toward advertising. Table 4.5.--Comparison of Attitudes Toward Advertising as a True Picture of Products and Stage of Retirement. Never Seldom Sometimes Usually Always Stage of Retirement (1) (2) (3) (4) (5) Early o o 0 (Age 60-64) l-7% 1.74 8.14 1.74 0 Middle 0 O . a 4 0 (Age 65_69) .64 5.24 9.94 5.44 .54 Later 5.8% 12.2% 23.3% 20.9% 1.7% (Age 70 and above) Totala 8.1% 19.2% 41.3% 29.1% 2.3% Chi-square = 9.05368 df = 8 Cramer's V = .l6223 a = .3378 a . . . Twenty-four m1SS1ng observations. Generally, a slightly more positive attitude toward advertis- ing existed for all stages of retirement. Approximately 3l.4 percent of the sample indicated a positive attitude, 27.3 percent indicated a negative attitude, and 4l.3 percent indicated a neutral attitude toward advertising as a true picture of a product. Attitudes toward television advertising as a new product information source.--The relationship between stage of retirement and attitudes toward television advertising as a source of new product information was measured by crosstabulating question number 3 and ll9 question number 40. Question number 3 identified the stage of retire- ment, and question number 40 read, "I like television advertising because it keeps me informed on new products" (Table 4.6). Table 4.6.--Comparison of Attitudes Toward Television as a Source of New Product Information and Stage of Retirement. Never Seldom Sometimes Usually Always Stage Of Retirement (1) (2) (3) (4) (5) Ear‘y 1 1% 3 3% 5 5% 2 8% 1 1% (Age 60-64) - ° - - - Middle o (Age 55-59) 2.24 5.1% 8.3% 4.4% 1.1% Later 5 1% 18 8% 22 1% 13 8% 3 3% (Age 70 and above) ' ° ' ' ' Totala 9.4% 28.2% 35.9% 21.0% 5.5% Chi-square = .86395 df = 8 Cramer's V = .04885 a = .9990 a . . . . Fifteen m1551ng observations. A chi-square test of significance was conducted on the cross— tabulated data. Chi-square was not significant at a = .05; therefore, it was determined that no statistically significant relationship existed between stage of retirement and attitudes toward television advertising as a source of new product information. Review of Table 4.6 indicated that overall, a slightly more negative attitude existed toward television advertising as a source for new product information (37.6 percent responded "never" or "seldom" and 26.5 percent responded "usually" or "always"). Independently. 120 individuals in each of the three stages of retirement also indicated a slightly more negative attitude toward advertising related to new product information. Attitudes toward shopping only when a need arises.--The syste- matic relationship between attitudes toward shopping only when a need arose and stage of retirement was analyzed by crosstabulating question number 3, stage of retirement, and question number 31, "I only go shopping when I really need something." Chi-square analysis indicated that no statistically significant relationship existed between the two variables at a significance level of .05 (Table 4.7). Table 4.7.--Comparison of Attitudes Toward Shopping Only for Need and Stage of Retirement. Never Seldom Sometimes Usually Always Stage of Retirement (1) (2) (3) (4) (5) Early 4 0 (Age 60-64) 0 .55 5.35 4.8% 2.7% Middle 0 c O O 0 (Age 65-69) .54 .54 4.34 10.74 5.34 Later a 0 (Age 70 and above) 1.6% 3.24 7.5% 33.74 19.3% Total 2.1% 4.3% 17.1% 49.2% 27.3% Chi-square = 12.97007 df = 8 Cramer's V = .18622 a = .1129 From Table 4.7 it can be observed that retirees in the later stage of retirement indicated the strongest attitude toward shopping only when a need arose (53 percent stated "usually" or "always"). 121 Overall, all three groups stated a definitive attitude toward shop- ping when they really needed something (76.5 percent checked "usually" or "always" to question number 31). Consumer Problems of Retirees at Different Stages of Retirement Analysis of group 2 hypotheses was accomplished by independ- ently testing Hypotheses 2.1 and 2.2. The analysis and findings of Hypothesis 2.1 are presented first, followed by the analysis and findings of Hypothesis 2.2. Discussion of the reasons underlying the consumer problems of retirees is presented last. Group 2 Hypotheses Testing There is no significant relationship between selected demographic characteristics and the type of consumer problems experienced by retirees. 2.1 There is no significant relationship between age and the type of consumer problems experienced by retirees. Hypothesis 2.1 was statistically tested to determine whether or not a systematic relationship existed between the retirees' stage of retirement and the types of consumer problems encountered. Ques- tionnaire items assessing product problems (#41b) and service problems (#41c) were independently crosstabulated with question number 3, retiree's age, to operationalize the analysis of this hypothesis. To facilitate chi-square analysis, data were collapsed from 18 to 12 categories for services and from 15 to 10 categories for products. To effect a more detailed analysis of Hypothesis 2.1, the consumer problems reported by retirees were separated into two areas, specifically, product and service problems. The computer program used 122 (SPSS) does not compute chi-square on aggregated variables. There- fore, it was necessary to statistically analyze each of the two product and service problems separately. Chi-square analysis was conducted on the crosstabulated table results. The chi-square statistic indicated that age was not significant at the .05 level; therefore, the null hypothesis was accepted. Hhile chi-square was not significant for either product or service problems by stage of retirement, some interesting trends were identified. For both products and services, the largest number of problems were reported by retirees age 70 and above (55 percent and 52.4 per- cent, respectively) (Tables 4.8 and 4.9). The fewest number of problems for both products and services were reported by the 60-64 age group (16.7 percent and 14.7 percent, respectively). Analysis of consumer product problems by stage of retirement indicated that retirees in the early stage (age 60 to 64) had the most frequent problems with cars (36.4 percent) and food (27.3 percent) (Table 4.8). Individuals in the middle stage of retirement (age 65 to 69) reported the greatest dissatisfaction with food (37.5 percent) and products in the clothing/shoes category (18.8 percent). Those in the later stage of retirement (age 70 and above) were most dissatis— fied with clothing and shoe products (18.2 percent) and major appli- ances (18.2 percent). For the three groups of retirees collectively, the most frequently identified product problem was first related to food (21.7 percent) and second related to clothing/shoes (16.7 percent). The third most frequently stated product problems were with cars and 1223 .mmm2_um2 we gnoLm uwmwumnm game so» umu2oao2 msmpao2g uu=u02g mo Logan: Faye» any we xucmacmgm m>_uapmm n H .mmmgwgog mo ma=o2m mm2gu ppm 2o» umusoqmg mempao2a “usuo2a mo 2885:: Page» mgu $6 aucmauoLm m>muopmm u A V "muoz 252642, 53.2022 520683 2822 NNNN. u a Newmm. u > 4.222828 A28 8.2 ooooo.a_ u 228354-228 282342, 8658522 “4222 2522 mNNN. » 5 «NNNN. u > 4.252428 A25 8.2 Nm_NN.N u 22a=am-_2u Aao.oo_2 2N8._v 2N8.22 2N8.Fv Ano.o22 Aao.m_v ANN.F_V ANN._NV ANN.8_2 Auo.mv Auo.m_v om _ 2 p o m N NF o_ N a 283°» "No.oo_u HNO.NH ”NC.NH .. HN_.NPH ”NP.N_H HNN.mFH HN_.NPH ”Nfl.m_u Hu_.oH HNN.N_H A o . . . .. . . . . . . . a, 22 ANo mmv AN8 22 Ana _2 Age 82 AN8 82 ANN N2 Rao 82 2N8 822 ANN N2 Ago o_v ecu oN am2mm mo Logan: pope» on» we aucmaam2w m>wum_mm u m .mmm2vuo2 mo maao2a wmggu p—m 2o» umu2oam2 mem~a02q oup>gmm mo 2mnszc pave» any we aucmaaosu m>wuo—mm u A "muoz Avmumwp ou_>2mm ucoumm 2omv woom. u a NONoo. u > m.2wsm29 Ayn opv ooooN.m u m2uscm1_;u Auwumpp mup>2mm umgwm 2on2 FeNN. u a MNmoc. u > m.2msm2u Amt oNV N—muw.mp u o2oaamuw2u Aao.oopv Au—.NV Au¢.o—v “um.mv Aum.mv Au—.Nv An_.NV Anm.ov Aum.mmv Aum.va Aus.opv me p m n v — p m op c m Pouch mao.oo_H muo.eH Huo.N~H muo.mH Huo.NFH .. .. muo.wH Huo.NmH muo.N~H muo.NFH :3.va 32.3 38.8 38.: 38.3 .. .. 38.: 32.22: 38.3 :3: Amsmmaamuw mN P m N m .. .. N w m m Loam; $8.8: .. EN: .. .. :83 38.3 32.3 3:: GEN: :88: :3 N2 :Ni .. . 32.3 32.3 32.3 3N: 3N4; 38.3 38.8 2.5 o_ .. N .. .. — p p o N m upcvwz mao.oopg .. .. Hum.epu mum.e~H .. .. .. Hum.mNH mum.¢FH muo.mNH E .2: :2 NV :2 N2 .. 3N4; $th 33.: 38.8 as N o. o. —. _. o. o. .o N —. N NFLGU Amy Amy amv Amy AMV ANN Amv ANN Amy fimv amv z z z z z z z z z z z a: 2m moan :o acm5w2_umm m a o ”was; 12:m:~ 1wmmu xmwwMW2 z2mpwpmo «UTWme -LoQMWMWP Lwoamm 2_maoa mu_>2om mo macaw _ u p -u2oa< supmm: ._.u= u a z p? u m o_pn:a FNLocmu 2mg capao: Em_no2a muw>2wm “AIIPlsuIIIUAsW1I‘111‘l1 ll .u:m&m2wumm No mmmum .mmm2_umm xn mempao2m mow>28m mo xucmzcm2m 2o :oNPLmasouuu.a.c mpaap 126 2.2 There is no significant relationship between the number of years in retirement and the type of consumer problems experi- enced by retirees. Hypothesis 2.2 was analyzed for significant relationships between product and service consumer problems and the number of years a retiree has been in retirement. The testing of this hypothesis was conducted by first independently crosstabulating survey question numbers 41b and 41c (product and service problems) with question num- ber 2 (years in retirement). As with Hypothesis 2.1, categories of product and service problems were collapsed to facilitate statistical testing. The number of years in retirement were grouped into five categories: group 1, retirees of l to 2 years in retirement; group 2, retirees of 3 to 5 years in retirement; group 3, retirees of 6 to 10 years in retirement; group 4, retirees of 11 to 20 years in retirement; and group 5, retirees of 21 or more years in retirement. Chi-square analysis was calculated on the crosstabulated data for products as one group and services as a second group. The chi- square statistics for both products and services were not statistically significant at the .05 level (Tables 4.10 and 4.11). The null hypothe- sis of no significant relationship was consequently accepted. The absolute and relative frequencies indicating product and service problems are presented in Tables 4.10 and 4.11. It is inter- esting to note the difference in frequency of total consumer problems between those who have been retired for a longer period of time (Tables 4.10 and 4.11) and those who are older (Tables 4.8 and 4.9). Retirees who are 70 years and above (later stage) and those who have been retired from 11 to 20 years (group 4) reported the greatest 127 number of consumer problems (58 and 41 problems, respectively). The fewest total problems were identified by retirees in the youngest age group (N = 18) and by those who were either retired for 1-2 years, group 1 (N = 3), or 21 years and over, group 5 (N = 2). Analysis of product problems by number of years in retirement indicated a profile similar to the findings from Hypothesis 2.1 (Table 4.10). No product problems were reported for retirees having been retired l to 2 years (group 1). Retirees having been retired 3 to 5 years (group 2) stated that car problems (23.1 percent) gave them the greatest difficulty. Problems related to food (31.6 percent) and major appliances (21.1 percent) were most often indicated by per- sons in group 3 (retired for 6 to 10 years). Retirees in group 4 (retired for 11 to 20 years) designated the clothing and shoes category (29.4 percent) as the most problematic product area. As there were only two product problems identified by persons having been retired for 21 years or more (group 5), there was no one particular product area that could be cited as a significant problem. Retirees having been retired for 6 to 10 years (group 3) reported the greatest frequency of product problems (37.3 percent), and those having been retired for 21 years or more (group 5) reported the fewest product problems (3.9 percent). This situation may be accounted for by the fact that persons who have been retired for a long period probably buy fewer products overall and therefore have fewer product problems. Comparison of service problems by the number of years in retirement presented findings somewhat similar to those in Hypothesis 2.1 12E) .555..ao. .o 5=o25 u...5555 sumo .o. cougoao. 455.5o.5 uuovo25 .o 2555:: .muou 5;. mo 5555555.. o>.uo.o¢ u n .455..552 .o ma:o.m 552:. ..5 5o. 55.25552 455.5o25 «oooo25 mo .5555: .muo» 5;» .o hocooomgw 5>.55.o¢ u A "ouoz .55.4.. 4555525 555554 255. .5.N. 5 5 .5..5. n > 4.255525 .55 5N. 554...N u 525554-.;5 .5544.. 5555525 442.. 255. .55.. u 5 54N55. n . 4.255525 .55 .N. 5.N.... u 525554-.55 .4.55.. .45.N. .45.N. .45.N. .45.4. .45.... .45.5. .45..N. .45.5.. .45.5. .4..5.. .4 . . . 5 5 4 .. 5. 5 . .545. .45.55.. .. .. .. .. .45.54. .45.54. .. .. .. .. . .. .. .. .. . . .. .. .. .. 525: 25 .45 5. .45 N. .45 N. .42. .N N .5 so 5. 5. —. F .o .o o. oo WQDOLQ .45.55.. .45.4. .. .45.4. .. .45.... .45.... .45.... .45.5N. .45.4. .45.4. .45.55. .45.N. .. .45.N. .. .45.4. .45.5. .45.4. .45.5.. .45.N. .45.N. .42. 5N-.. .. . .. . .. 5 N 5 4 . . 5 55525 .45.55.. .. .45.4. .. .44.5.. .44.5.. .. .45..5. .45.4.. .45.4. .4...N. .45..5. .. .45.N. .. .45.5. .45.5. .. .45.... .45.4. .45.N. .45... .42. 5.-5 5. .. . .. N N .. 4 5 . 5 5 55525 .45.55.. .. .. .. .4.... .4..5N. .45.4.. .45.4.. .45.4.. .4.... .45.4.. .44.4N. .. .. .. .45.N. .45.4. .45.5. .45.5. .45.5. .45.N. .45.5. .42. 4-5 5. .. .. .. . 5 N N N . N N52:5 .4. 55. .4. .4. 55. 55. .5. .5. .5. 55. 5.. z z z z z z z z z z z muoaoo25 455:5 muoooog5 moo mo... moocm ucosogpuoz 455.54.: 5.525.. ...5355 -..55< 5259 5o.uoo..5 woo. gm ou=< -..55< a. 5255. uoaoo25 mo 55.. .ucoso..uo¢ c. 4.55» .5 555.5o.5 uoouoL5 mo .u:o=oo.. mo :om.2ooeou-u.o..e 5.55. '129 .4552..5. .o nooLm 5.5.5555 2555 2o. 55.2ooo2 455.5o25 ou.>2om .o 2552:: .muo. 5:. .o .u:5=oo.. 5>.uo.5¢ u H . .4552..o. .o mgao.m ..5 2o. 55..oao2 425.5o2o mu.>.5m .o 2555:: .muou 5:. .o aucoaoo.. o>.uo.om u . . "ouoz .55.4.. 55.>254 555554 .55. 5.55. n 5 55555. n > 4.255525 ..5 N.. ..55... u 525554-.55 .5554.. ou.>.om .5... Lou. m..m. n 5 m55.5. u > 5.25552u ..o .N. mmm.m... u 5.55:5-.zu 45.55.. .4N.N. .4..... .45.5. .45.5. .4N.N. .4N.N. .45.5. .45.55. .45.5.. .44.4.. 45 . 5 N 5 . . 5 4. 5 . .555. .45.55.. .4.... .4.... .45.5.. .45.5.. .. .. .45.5.. .45..N. .45.5.. .4.... .45.55. .4N.N. .4N.N. .45.5. .45.5. .. .. .45.5. .45.5. .45.5. .4N.N. .42. 5N-.. 5. . . N N : .. N 5 N . 552:5 .45.55.. .. .44.5.. .. .45.4. .45.4. .45.4. .45.4. .4..N5. .45.4.. .44.5.. .45..5. .. .45.5. .. .4N.N. .4N.N. .4N.N. .4N.N. .4..... .45.5. .45.5. .42. 5.-5 5. .. N .. . . . . 5 5 N 5 555.5 .45.55.. .. .4N.NN. .. .4..... .. .. .. .4N.NN. .4..... .45.55. .45.5.. .. .45.5. .. .4N.N. .. .. .. .45.5. .4N.N. .45.5. .42. 4-5 m .5 N .5 P o. 5. o. N _. m N Q=°Lw HNO.OO—H_ .. .. .. .. .. .. .. H&N.00H .. HNm5mmH afiooov .5 o. o. o. a. o. o. ARQ5¢V o. AflNoNv 5WL> NIP m o. oo o. oo o. o. no N o. P PQ=°LQ .4. .5. .5. .5. .5. .5. .5. .5. .5. .5. .5. z z z z z z z z z z 2 55.5.5 1wwmm. 45.. 45x5. .:5>..oo 5o.>.5m -.MMWWMW. 2.555: 2.555. ou.>.om u:mfiomwwwm 4.5.o. moongmo £5.55: 1...»: .ugmao25 4:52 ..5uom o..o:5 .5.o:oe .mu :u.5o= mo.2onsaz ou.>.5m .o 55.. .111 “1-11-1-1- 111,1...”111'1‘ 11-11, 1.1-41-41! ~1“lrllll ‘1 ..:5eo2..oa :. 4255. an 425.5o25 mo.>2om .o 55:55:52. mo :o5.25aeoo-....e 5.55. 130 (Table 4.ll). Retirees in group l (retired from l to 2 years) reported only three service problems; two problems related to general repair and one problem with health service. Group 2 retirees (retired for 3 to 5 years) indicated that health service (33.3 percent) was their most frequent service problem. Individuals having been retired for 6 to 10 years (group 3) cited general repair (42.1 percent) as the most frequent service problem. Respondents in group 4 (retired for ll to 20 years) stated that general repair gave them the most frequent dissatisfaction (2l.4 percent). The greatest frequency of service problems was reported by the group of retirees having been retired for 6 to 10 years, group 3 (41.9 percent). Those retirees who have been retired for the shortest time, group l, reported the fewest service problems (6.6 percent). Other Findings--Consumer Problems Ef'Retirees Another dimension in the analysis of consumer problems is the reason underlying the specific problem. Absolute and relative fre- quencies were tabulated for the nature of the problem (Table 4.l2). Response categories were collapsed from 14 to ll reasons. The most frequently identified reasons for the reported consumer problems were the collapsed category of poor quality and design of the product (35 percent) and the poor quality of the service (27 percent). It is interesting to note that while mentioned as reasons for dissatisfac- tion, the price of the product or service did not appear to be as significant a problem as one might expect for persons living on rela- tively fixed income. 131 Table 4.lZ.--Frequency of Type of Product or Service Problem. Frequency of Problem Type of Problem fl_ % Price of Product Too High 8 8 Poor Quality Product 34 35 Poor Quality of Sales Service 2 2 Poor Quality Service 26 27 Didn't Meet Agreed Service Terms l0 l0 Rent Raised Too Frequently l l Took Longer Than Agreed 4 4 Rebate Not Honored l l Different Product Price 1 1 Price of Service Too High 9 9 Comfort/Fit of Product l l Total 97 iooa aDoes not equal lOO percent due to rounding. Complaint Behavior Among Retirees at Different Stages of Retirement The findings of the group 3 hypotheses testing are presented first in this section, followed by descriptive analyses of the attri- butions of blame and complaint actions taken regarding consumer problems. Discussion of additional findings relative to the level of satisfaction with the corrective action taken and the reasons for retirees taking no corrective action on consumer problems is presented at the end of the section. 132 Group 3 Hypotheses Testing There are no significant differences between selected demographic and situational characteristics and the method of complaint resolution utilized by retirees in different stages of retirement. 3.1 There are no significant differences between age and the method of complaint resolution utilized by retirees. 3.2 There are no significant differences between age and attri- bution (yfblame for consumer problems among retirees. 3.3 There are no significant differences between age, attribu- tion. of blame, and method of complaint resolution utilized by retirees. Collectively, the group 3 hypotheses measured significant mean differences in the complaint behavior of retirees at different ages or stages of retirement based on perceived attributions of blame. To operationalize the testing of the group 3 hypotheses, four types of complaint behavior were identified. The types of complaint behavior (dependent variable) and the associated actions from questionnaire item number 42a are shown in Table 4.13. The attribution of blame factors (independent variable) are presented in Table 4.14. Both complaint actions and attribution of blame were measured on a five- point Likert-type scale ranging from 1 = Never to 5 = Always. Stage of retirement (item number 3 on the questionnaire) was also treated as an independent variable with three levels. To decrease the number of cells for analysis of variance (ANOVA) testing, the four types of attributions of blame were col- lapsed into three. Factor 2, the retailer or salesperson, and factor 3, the designer or manufacturer, were combined into one (Table 4.14). Also, the five-point scale for measuring the attribu- tions of blame was collapsed into a three-point scale. 133 Table 4.13.--Complaint Actions Resulting From Consumer Problems. Item 42a: Please describe the action(s) taken by you or your spouse to correct those consumer problems listed in the previous questions. Circle the number next to each statement which best describes how often you have used the action to cor- rect consumer problems. Factor 1: Noncomplaining Type Didn't do anything at all Factor 2: Private Action Complained to my family Complained to my friends and urged them to avoid the product/service Decided never to purchase/use the product again Factor 3: Remedial Action Seeking Type Complained in a letter to a newspaper or magazine Returned the merchandise for a replacement/refund Contacted the store/dealer to complain Contacted the manufacturer to complain Contacted the BBB Contacted a consumer advocate to suggest an investigation Factor 4: Legal Action Type Contacted my lawyer about possible legal action Complained to my congressman Stopped payment or refused to pay Complained to a local consumer agency Contacted the Office of Consumer Affairs in Washington, D.C. Contacted a federal regulatory agency Filed complaint in Small Claims Court 134 Table 4.l4.--Attribution of Blame in Consumer Problems. Item 4le: In your opinion, how often were each of the following people or companies to blame for the consumer problems you listed in the previous questions? Circle the number next to each statement which best describes how often each was to blame. Factor 1: Myself My spouse, friend, or other person involved in the purchase Factor 2: The retailer or company who distributed or sold the product or service The salesperson or serviceperson involved The designer or manufacturer of the product Factor 3: The advertisers of the product or service Analysis of variance (ANOVA) using a fixed-effects, four-factor model was run on the four independent variables and one dependent vari- able. ANOVA statistical analysis indicated only one significant inter- action effect at the .006 level of significance (Table 4.15). The interaction of stage of retirement with the retailer, manufacturer, or advertiser to blame for the problem has a significant effect on an individual using a form of legal action to resolve the consumer problem. While other main and interaction effects did not prove to be significant, there is an indication that the interaction of attri- bution of blame and stage of retirement have an effect on the complaint action taken. 0n the basis of the findings not being significant across all complaint levels, group 3 null hypotheses were accepted-- that no significant differences exist at the .05 significance level. 135 Table 4.15.--Summary of Analysis of Variance in Complaint Action by Stage of Retirement and Attribution of Blame. . . Sum of Mean F Source of Variation Squares df Square Value Noncomplaining Type Main Effects 2.4339 7 .348 2.119 Interaction Effects 2.331 14 .166 1.012 (Z-way) Interaction Effects .003 l .003 .020 (3-way) Private Action Type Main Effects 1.597 7 .228 .873 Interaction Effects 3.393 14 .242 .928 (2-way) Interaction Effects .395 l .395 1.511 (3-way) Remedial Action Type Main Effects 1.014 7 .145 .691 Interaction Effects 1.818 14 .130 .691 (Z-way) Interaction Effects .001 1 .001 .007 (3-way) Legal Action Type Main Effects .494 7 .071 1.642 Interaction Effects .635 14 .045 1.054 (Z-way) Interaction Effects .371 1 .371 8.626* (Age x Manufacturer/ Retailer x Advertiser) *p < .05. Outside the realm of hypothesis testing, further descriptive analysis of data can prove to be instructive. From Table 4.16 it can be detected that, on the average, the two "actors" in the marketplace blamed most frequently for consumer problems are the retailer 136 (mean = 3.05) or the designer/manufacturer of the product (mean = 3.02). Employing attribution theory, both of these attributions would be psychologically distant from the retiree. It is also interesting to note that as a total group, those attributions considered to be psychologically close to the retiree, specifically self and spouse or friend, were considered to be, on the average, infrequently to blame for the product or service problem (means = 1.73 and 1.23, respec- tively). Table 4.l6.--Summary of Absolute and Relative Frequencies and Means for Attribution of Blame in Product or Service Problem. Frequency of Response Some- Never Seldom times Usually Always Attribution of Blame (1) (2) (3) (4) (5) Mean T T T T— N (%) (7) (%) (3%) (3%) Designer/Manufacturer 20 3 5 12 16 3.02 of Product (36%) (5%) (9%) (21%) (29%) Advertiser 35 3 6 9 3 1.96 (63%) (5%) (11%) (16%) (5%) Retailer or Company 16 4 1] 1] 14 3 05 Selling Product/ 0 o o o o ' Service (29%) (7%) (20%) (20%) (25%) Sales or 22 3 14 8 9 2.63 Service Personnel (39%) (5%) (25%) (14%) (16%) Spouse, Friend 47 5 4 1.23 or Other (84%) (9%) (7%) Self 37 6 7 3 3 1.73 (66%) (11%) (13%) (5%) (5%) Total 177 24 47 43 45 137 On the average, the most frequently reported complaint action for the entire group of retirees was "to contact the store or dealer to complain" (mean = 2.82), "complained to family" (mean = 2.4), and "decided to never purchase or use the product again" (mean = 2.35). Again, these actions profile a psychologically closer action, "Private Action" or "Remedial" rather than "Legal Action Types" (Table 4.17). Other Findings--Complaint Behavior Among Retirees For those retirees indicating that an action was taken to resolve the problem, more than one-half (58 percent) stated that they were "not at all satisfied" with the results of the action. Those retired consumers who "were not completely satisfied but did get something" were represented by 6 percent, those who were "not com- pletely satisfied but the solution was acceptable" accounted for 22 percent, and the group who "received more than they asked for and were completely satisfied" totaled 14 percent (Table 4.18). When asked why no action was taken to resolve a particular consumer problem, the most frequent response was, "I didn't think I could get anyone to do anything about the problem" (36 percent) (Table 4.19). The response, "I didn't think it was worth the time or effort," was given 33 percent of the time, and "I didn't know what to do about it" was stated 21 percent of the time. 1138 Table 4.17.--Actions Taken to Resolve Consumer Problem. Frequency of Response . Never Seldom Sometimes Usually Always Action (1) (2) (3) (4) (5) Mean N N N N r_1 (%1 (31 (i) (7) (%) Complained to Family 34 2 10 12 9 2.40 (51%) (3%) (15%) (18%) (13%) Complained to Friends and Urged Them to (32w) (2“) (13w) (3“) (1gw) 2.29 Avoid Product/Service % b ” " ” Complained to Local 49 4 7 3 3 1.59 Consumer Agency (74%) (6%) (11%) (5%) (5%) Complained to 60 4 2 1.03 Congressman (91%) (6%) (3%) Complained in a Letter 58 5 3 1.17 to Newspaper or Magazine (88%) (8%) (5%) Stopped Payment or 62 3 . 1 1.09 Refused to Pay (94%) (5%) . (2%) Decided to Never 29 2 9 11 15 2.35 “mesa/”5‘? the (44%) (3%) (14% (17%) (23%) Product Again Returned the Merchandise 37 5 7 7 10 2.21 for a Replacement/Refund (56%) ( %) (11%) (15%) Contacted the Store/ 26 4 9 10 17 2.82 Dealer to Complain (39%) (6%) (14%) (16%) (27%) Contacted the Manufac- 51 5 1 3 6 1.61 turer to Complain (77%) (8%) (2%) (5%) (9%) Contacted the Better 58 3 3 2 1.26 Business Bureau (88%) (5%) (5%) (3%) Contacted Lawyer About 63 1 1 1 1.11 Legal Action (95%) (2%) (2%) (2%) Contacted the Office of Consumer Affairs, (33%) ( 5%) ' 1'00 Washington, D.C. ' ' Contacted a Consumer 62 1 1 2 1.17 Advocate for Investigation (94%) (2%) (2%) (3%) Contacted a Federal 65 1 1.03 Regulatory Agency (99%) (2%) Filed Complaint in 66 . 1-00 Small Claims Court (100%) ' Didn't Do Anything 36 3 9 10 8 2.26 (55%) (5%) (14%) (15%) (12%) 139 Table 4.18.--Leve1 of Satisfaction With Action Taken to Correct Problem. Frequency of Response Level of Satisfaction N_ % Received More Than I Asked for: Completely Satisfied 7 14 Not Completely Satisfied But Solution Was Acceptable 11 22 Not Completely Satisfied But Did Get Something 3 6 Not at A11 Satisfied 29 58 Total 50 100 Table 4.l9.--Exp1anation of Why No Action Was Taken to Resolve Consumer Problem. Frequency of Response Level of Satisfaction N_ % It Wasn't Worth the Time/Effort 11 33 Didn't Think Could Get Anyone to Do Anything About Problem 12 36 Didn't Know What to Do About It 7 21 Other Reasons 3 9 Total 33 99a aDoesn't equal 100 percent due to rounding. CHAPTER V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS This chapter is organized in four sections. The first section is a summary of the research methodology, data analysis, and findings of the study. The second section is a presentation of the limita- tions of the research study. The third section is a discussion of the conclusions that can be drawn from the research. The fourth section contains the recommendations for marketers, educators, and public policy makers and the recommendations for future research. Summary of Research Methodology and Data Analysis The purpose of the study was to investigate and analyze the attitudes of retirees toward the marketplace, consumer problems, and consumer complaint behavior. The objectives of the research were to determine the attitudes of Michigan retirees at three stages of retirement toward marketplace practices and the nature and extent of consumer problems and complaint resolution. The research was cross-sectional in nature and used survey research methodology. The self-report survey instrument was sent to, and returned by, respondents through the mail. Any qualified retired member of the household could complete the questionnaire. Four spe- cific requirements were established to qualify a respondent for 140 141 inclusion in the study. The criteria specified that, upon receiving the survey, the respondent must have been at least 60 years of age, currently retired, and retired for at least one year. If married, both the respondent and spouse must have met the criteria in order to participate in the study. The research instrument was developed using various types of response formats. Likert-type scales were used to measure atti- tudes toward the marketplace and complaint behavior. Closed and free-response questions were developed to assess the nature and extent of consumer dissatisfaction, demographic and situational variables. I The sample consisted of randomly selected retirees from six different areas within the state of Michigan. The areas from which the sample was drawn included Lansing, Jackson, Kalamazoo, Flint, Traverse City, and Trenton. Respondents were randomly selected from the occupational mailing lists of the R. L. Polk Company, a national publishing house. . During the months of March and April 1982, the survey instru- ment was sent in two separate mailings to 866 randomly selected retirees. Following the deletion of questionnaires for incomplete responses and unqualified respondents, 196 questionnaires were deter- mined to be usable in the study. The overall return rate, adjusted for nondelivery, unquali- fied respondents, and incomplete questionnaires, was 24 percent. The relatively low response rate was due to variability in the cur- rentness of the mailing lists and the fact that some retirees were 142 residing at a winter residence different from the address to which the survey instrument was sent. In some instances, these retirees did not receive the questionnaire until after the data collection period ended as they had not been forwarded by the postal service. While census data reporting the demographic characteristics of retirees as a separate group were not available, the character- istics of the sample were compared to Michigan and U.S. census data for persons age 60 and above or 65 and above, depending upon availa- bility. The sample was determined to be representative of the retired male/female population in Michigan who are age 65 and above, middle income, and high school educated. Both descriptive and inferential analyses were employed to examine the data. Descriptive statistics were used to characterize the sample according to the demographic and situational data. Hypothe- sis testing and inferential statistical analysis were used to test for significance among variables. The specific statistical tests con- ducted were analysis of variance (ANOVA), multivariate analysis of variance (MANOVA), and chi-square analysis. Attitude items were formed into scales through factor analysis and tested for reliability using Cronbach's alpha. The overall coefficient alpha for the atti- tude scales was .73. Summary of Findings While statistically significant differences among retirees were not generally supported through hypothesis testing, descriptive analysis indicated that some differences existed between retirees in 143 the three stages of retirement. One reason that the findings may not have been found to be statistically significant was due to the small sample size. Another reason may have been related to the large number of products and services being analyzed in the consumer dissatisfaction section of the study. Attitudes of Retirees Toward the Marketplace 0f the three groups of retirees, the younger retirees (age 60 to 64) had a greater tendency toward price consciousness and com- parison shopping. They also found advertising, in general, and newspaper advertising, in particular, to be helpful in providing product information. The attitudes of those retirees classified into the middle stage of retirement (age 65 to 69) indicated that they frequently shopped alone and had the fewest health and transportation problems restricting shopping activities. This group also viewed new products and brands as being inferior to older established products. Retirees in the later stage of retirement (age 70 and above) indicated the strongest attitudes of the three groups toward paying cash for consumer purchases, preparing shopping lists, and purchasing only those items on the list. Additionally, older retirees tended to be more reliant on sales personnel for assistance, shopped only when they really needed something, and indicated that health and transportation problems made shopping difficult. Retirees in this group stated a more positive attitude toward advertising as a true picture of a product than retirees in the other two stages. 144 On the average, all groups of retirees stated a positive attitude toward: price consciousness, advertising in general, news- paper advertising as a source of consumer information, newspapers as an important general information source, paying cash for consumer purchases, transportation problems having restricted shopping oppor— tunities, and shopping only when they needed something. In general, the three groups of retirees indicated that they did not usually: comparison sh0p for grocery or nongrocery products, find sales per- sonnel helpful when shopping assistance was needed, consult with family and friends before making major purchases, make a shopping list for nongrocery items and restrict purchases to only those items on the list, find television advertising to be a helpful information source for new products, and find new products and brands to be inferior to older established products or brands. Consumer Problems of Retirees To operationalize the comparison of consumer problems among retirees, two types of measurement were used. The first measurement used the retiree's age for categorization into one of three stages of retirement: early, middle, and later. The second used the number of years in retirement to classify a retiree into one of five groups: retirees of l to 2 years, 3 to 5 years, 6 to 10 years, 11 to 20 years, and 21 years or more. Analysis of consumer problems indicated general agreement between findings generated through measurements using retirees' age and years in retirement. The largest number of product and service 145 problems were reported by retirees who had been retired 11 to 20 years and were 70 and above. The fewest number of consumer problems were reported by younger retirees, age 60 to 64, and those who had been retired for either 1 to 2 years or 21 years and above. Specifically, retirees age 60 to 64 identified the product categories of cars and food and the service categories of general repair and health service as the consumer problem areas of greatest concern. Those in the middle stage of retirement (age 65 to 69) stated that food and clothing/shoe products and general repair and health services presented the greatest dissatisfaction. Respondents age 70 and above (later stage) reported that problems related to clothing, shoes, and major appliance products and general repair services were the most frequently identified areas of concern. For all groups of retirees, product problems were more fre- quently reported than service problems. Product problems related to the food and clothing/shoes areas were the most frequently mentioned by all retirees. Repair services were by far the greatest service problem for retirees. Health service problems were ranked second in frequency by all retirees. The most frequently cited reasons for the reported consumer problems were the poor quality or design of the product and the poor quality of the service. While high prices of the products and ser- vices were mentioned as reasons for consumer dissatisfaction, the findings indicated that high prices were not a significant problem for retirees. 146 Complaint Behavior Among Retirees The findings indicated consistency between the attribution of blame for the consumer problem and the resulting complaint action for the entire group of retirees. The most frequently reported attributions of blame were identified as the retailer or manufacturer. The complaint action most frequently taken was to complain to the store or dealer. Of the retirees who took some type of action to resolve the reported consumer problem, more than one-half reported that they were not satisfied with the results of the action. For those respondents who chose not to pursue resolution of their product or service prob- lem, two reasons were most frequently mentioned. The first was that the retiree didn't think that anyone would do anything about it, and the second was that it was not worth the time nor the effort. Limitations of the Study The nonexperimental nature of survey research introduces some potential limitations that must be carefully addressed when designing a study. Henretta et a1. (1977) submitted that when review- ing these limitations, the adequacy of the research methodology is a compromise between accuracy and cost. The first limitation to be reviewed was that of sample size. Bernhardt and Kinnear (1976) raised some concerns regarding the representativeness of small sample sizes, those ranging from 100 to 200 respondents. Henretta et a1. (1977) discussed a related question, that of whether or not the data gathered are representative of the 147 socially isolated elderly. The researchers suggested that concerns regarding the data quality of survey samples may be assessed through comparison with census or Current Population Survey (CPS) population distributions (p. 161). Comparison of sample characteristics with census data for persons aged 60 and over indicated that the sample was over-representative of retirees who are white, widowed, more highly educated, and who have higher annual incomes. Additionally, the pre-retirement occu- pation classifications that were under-represented included the sales/clerical, craft and kindred, and operative workers. As income is generally linked to education and occupation, it was not surpris- ing that the sample was also under-representative of lower income households. Therefore, the findings of the study cannot be accu- rately generalized to retirees who are age 60 to 64, minority, married, and have less education. Some variation between sample and population statistics can be accounted for by the higher nonresponse rates in certain commu- nities due to variability in the currentness of the mailing lists. Additionally, the sample characteristics were compared with census parameters for all persons age 60 and above or 65 and above, not merely retirees. The second limitation of survey methodology was related to the time frame in which the data were collected. As the study was cross-sectional in design, it can be questioned whether or not cross- sectional data should be used to estimate longitudinal relationships between variables. A longitudinal study of specific individuals 148 before and at several points after retirement would strengthen the understanding of the relationship between stage in retirement and consumer attitudes, problems, and complaint resolution. The third limitation was related to the measurement of selected consumer attitudes among retirees. The study was limited to certain consumer attitude measures that were assumed to be repre- sentative of consumption behavior among the retired population. The fourth limitation was the survey instrument itself. Sev- eral researchers have questioned the reliability of scale research instruments used in the sociological, psychological, and social- psychological fields for measuring the construct of consumer dissatis- faction (Lundstrom & Lamont, 1976). Concomitantly, Bell and Emory (1971) raised concerns about the application of the conceptual frame- work of "consumer satisfaction" due to the difficulties of definition and measurement. To broach this concern, Cronbach's alpha tests were conducted to measure the reliability of the scaled attitude questions. An overall alpha of .73 was determined for the attitude questions. In the fifth limitation, the researcher recognized the poten- tial for physical impairment among retirees which might have affected the accuracy of their responses. Health problems, poor eyesight, and restricted arm or hand movement might have increased inaccuracy or nonresponse among the elderly. The sixth limitation relates to error in measurement. Regard- less of the attention to detail and control on the part of the researcher, no study using survey research methodology is without 149 error. Potential for error emanated from the instrument itself, the respondent, and the researcher. The last limitation identified the potential effects of economic conditions upon marketplace perceptions. The economic cli- mate in the state of Michigan may possibly have influenced the percep- tions of retirees in the study. It could be argued that the poor state of the economy might have resulted in respondents having more negative views of the marketplace than they might otherwise have had. This concern was recognized by the researcher; however, literature suggested that retirees are not as greatly impacted by the conse- quences of a downturn in economy, specifically, unemployment (Michman, 1979). Conclusions Application of decision-making theory, attribution theory, and the human ecological perspective to the findings of the research provides greater insight into the conclusions to be drawn from the study. The first conclusion is that retirees' attitudes toward the marketplace are not significantly different by age or stage of retirement. The findings indicate that retirees' attitudes toward the marketplace are heterogeneous regardless of stage of retirement. Analyzing the heterogeneity of consumer attitudes through the human ecological perspective supports the notion that consumer atti- tudes and decisions are a function of several interrelated factors: 5°C161’ psychological, physiological, and economic. Age or stage of 150 retirement relates to only one factor in the development of atti- tudes. The heterogeneity of the elderly is supported through research by Allan (1981), Towle and Martin (1975), Reinecki (1964), and Phillips and Sternthal (1977). From the findings of the study, it is possible to draw a second conclusion regarding the attitudes of retirees toward the marketplace. The oldest retirees, age 70 and above, had the greatest reliance on sales personnel and others for shopping assistance due to health and transportation problems which restricted shopping activities. This conclusion is consistent with decision-making theory as applied to decisions throughout the life cycle. Ultimate consumption decisions are affected by an interchange between many behaviorally altering factors, often beyond a consumer's control. This conclusion is generally supported by Sproles et al. (1977), Lambert (1979), and Payne (1960). A third conclusion is that, overall, retirees support price consciousness and paying cash for consumer purchases. Studies con- ducted by Bernhardt and Kinnear (1975) and Waddell (1975) arrived at a similar conclusion. A fourth and related conclusion is that retirees, in general, do not frequently comparison shop nor make shopping lists before entering the marketplace. It appears that while economy or price consciousness is a priority, other factors mediate the retirees' total commitment to value for the dollar. Some intervening factors may be 151 health and transportation difficulties that affect shopping behavior. Bernhardt and Kinnear (1975) concurred with this conclusion. A sixth conclusion is that retirees use newspapers as a primary information source for consumer and general information. The research by Steiner (1963) and Bernhardt and Kinnear (1975) concluded the same findings. While findings from the study indicated that more product problems were reported than service problems, repair services were the most frequently cited category of dissatisfaction. This seventh conclusion is not surprising, given the reliance of the elderly on services, in general, and repair services, in particular. The find- ings of this part of the study are not generally consistent with those of Bernhardt (1981), who found that "all services" accounted for the greatest dissatisfaction among elderly consumers. The interrelatedness of decision-making theory and attribu- tion theory is particularly helpful in drawing conclusions related to consumer dissatisfaction and complaint behavior among retirees. Mizerski and Green (1978) stated that attribution theory is frequently used to analyze "how consumers process market information in order to make consumption decisions" (p. 126). The eighth conclusion is that retirees view the perceived blame for consumer problems to be external to their locus of control. TWo conditions support this conclusion using attribution theory. First, the most frequently reported attributions of blame were identi- fied as the retailer or manufacturer. Both are considered to be external attributions or psychologically distant from the consumer. 152 Second, the most frequently identified reason for the problem was the poor quality of the product or service. An attribution theory approach would categorize these underlying reasons as "unstable" or varying over time. The variable nature of the problem is again perceived to be outside the consumer's perceived control. This conclusion is supported by the findings of Krishnan and Valle (1979) and Valle and Wallendorf (1977). A ninth and related conclusion is that the findings indicate consistency between the perceived externality of the attribution of blame and the resulting complaint action taken by retirees. The most frequently used complaint behavior in this study was to contact the retailer. Applying an attribution theory approach, this complaint action would be considered to be a "remedial action." Using a remedial complaint action in attempting to resolve an externally or psycho- logically distant attribution of blame is appropriate and consistent. Studies by Krishnan and Valle (1979) provide support for this con- clusion. A tenth conclusion is that retirees do not actively pursue resolution of consumer problems. While retirees appear to be taking some types of consumer complaint action, they are not being effec- tive in satisfactorily resolving particular problems. More than one-half of the retirees who complained about a consumer problem were not satisfied with the solution. It appears that decision-making and management skills concern- ing complaint behavior need to be more fully developed in retirees. Retirees need to be able to effect a rational decision regarding the 153 attribution of blame and then apply appropriate management skills to seek a satisfactory solution. Awareness of the private, remedial, and legal avenues for complaint resolution is a requisite to sound complaint management. Recommendations Generally, the research findings regarding retirees' attitudes toward the marketplace, consumer problems, and complaint resolution were not statistically significant by stage of retirement. Detailed review of the descriptive analysis, however, indicated some implica- tions and recommendations for marketers, educators, and public policy makers. Recommendations for Advertisers From the findings of the study, the majority of the retirees indicated that they usually read newspapers for sales and bargain information. However, they do not actively engage in comparison shopping. These data would suggest that while economy is important to retirees, other factors such as health and transportation problems intervene to minimize pre-purchase price comparison at different stores. Consequently, given the importance of newspapers as an information source, advertisers should focus attention on newspapers as a primary advertising medium to disseminate product and pricing information to retirees. Additionally, greater attention should be paid to the information processing abilities and behavior of the elderly when advertisements are developed. 154 Recommendations for Retailers In general, the elderly respondents indicated a somewhat negative attitude toward retail sales personnel. Specifically, opin- ions and product information provided by sales clerks were not thought to be helpful to retirees in making purchase decisions. These find— ings indicated that retailers must make a greater effort to assist elderly shoppers in the retail environment. In view of the current trend toward catalog and mail-order shopping, retailers who do not provide adequate assistance to retirees will lose sales to these growing and dynamic retail forms. Consideration of the need for increased assistance for elderly shoppers is also necessary when retailers are contemplating the adoption of centralized checkout facilities to reduce sales personnel labor costs. Retailers or dealers were most frequently identified as being to blame for consumer problems among retirees. Further analysis indicated that,while the complaint action most frequently taken was to complain to the store or dealer, the majority of retirees were not at all satisfied with the corrective action taken. Perhaps greater care and assistance must be afforded elderly consumers by retailers to aid them in reaching an acceptable complaint solution. Recommendations for Manufacturers As a majority of the retirees' responses ranged from negative to neutral on questions indicating the superiority of new products or brands to older established products, it appears necessary that manufacturers need to more effectively target advertising and new 155 product information at elderly shoppers. Promotional campaigns that assuage the risk factors sometimes associated with new products may move elderly consumers closer to purchasing new brands and products. Findings from the consumer problem/dissatisfaction section of the survey provide additional recommendations for manufacturers. The most frequently reported source of product related problems by retirees came from the food and clothing/shoes categories. Spe- cifically, the causes 3 § 3 E 5 High.prlced products are 3 .i i g g alwaysabetter 8 3 3 quality. [44) 1 2 3 4 5 When shopping i make my purchase decisions based upon the reputation of the store. [45'] 1 2 3 4 5 I usually compare prices on r products at several different stores before i decide on which to buy. [46] 1 2 3 4 5 i usually compare prices on 0.91-9th01 91060018 at several different stores before I deCIde on which to buy. [47] 1 2 3 4 5 i buy many things with a credit card. [48] 1 2 3 4 5 To buy anything other than a house or car on credit is unwise. [49] 1 2 3 4 5 When I read a newspaper l readit thoroughly. [50) 1 2 3 4 5 i get more consumer in- formation from reading a newspaper than from watching television. [51) 1 2 3 4 5 I like television advertising because it keeps me in- formed on new products. [52) 1 2 3 4 5 JNLEss orwenwfse iNbICZTE‘b,iLEASE CHECK THE RESPONSE wmcw BEST answchs EACH or THE REMAINING QUESTIONS IN THE sunvev. 41a. Based upon your own experience. have you (or your spouse) had a cogsumer problem or dissatisfaction with a product or service since. VOLLEIIEQ'? [53) 1: Yes (it Yes. continue with question 41 b) (1) D No (If No. skip to question 43s) (0) D Don’t know (Skip to question 43a) (9) 415. Since retirement. which products have given you 1'10 01041991 summed» or the most gigiggh problems? List 1 or 2 products In the spaces provided. (Some examples might be related to clothing. appliances. food products. etc.). Please also Indicate when this problem occurred. PLEASE BE SPECIFIC. [54-59) Date Problem Occurred Product 1 . 2. 41c. Since retirement. which services have given you the greatest dissatisfaction or the most difficult problems? List 1 or 2 services In the spaces previded. (Some examples might be related to health. transportation. repair services. etc.). Please also indicate when this problem occurred. [”05] PLEASE BE SPECIFIC. service Date Problem Occurred 1 . 2. 41d. For the products or services listed In the previous Questions. write in the spaces provided the two of dissatisfaction or problemts) you (or yOut spouse) had with the product or service. (Some examples might be related to the quality. price. or Information given with the product or service. etc.) [do-73] 1._ 2. 3. a 164 41a. in your opinion. how often were each of the following people or companies to blame for the consumer problems you listed in the previous questions? Circle the number ”my statement which best describes how ofteibsach was to blame. a > . g E s g If ( a .. g g S a 3 e -The designer or manu- facturer of the product 1 2 3 4 5 -‘l'he advertisers of the product or service 1 2 3 4 5 -The retailer or company who distributed or sold the the product or service 1 2 3 4 5 -The salesperson or service- persOn involved 1 4 5 -My spouse. friend or other person involved In the purchase 1 2 3 4 5 -Myself 1 2 3 4 5 -Other (please list)__ 1 2 3 4 5 . Please describe the action(s) taken by you or your spouse to correct those consumer prob lems listed in the previous questions. Circle the number gestuto‘gchgstatemeunt which best de- cribes how often vOu have used the action to correct consumer problems [BS-102] fl 2 > .. I g a- : = tar ( > ... g I s s 3 a: Complained to my family 1 2 3 4 5 -Complained to my friends and urged them to avoid the product/service 1 2 3 4 5 Complained to a local consumer agency 1 2 3 4 5 ~Cornplained to my con- gressman 1 2 3 4 7 5 Complained In a letter to a newspaper or magazine 1 2 3 4 5 -Stopped payment or refused to pay 1 2 3 4 5 -Decided to never purchaser use the product again 1 2 3 4 5 ofieturned the merchandise for a replacement/refund 1 2 3 4 5 -Contacted the store to complain 1 2 3 4 5 Contacted the manufacturer to complain 1 2 3 4 5 -Contacted the Better Business Bureau 1 2 3 4 5 -Contscted my lawyer about possible legal action 1 2 3 4 5 -Contacted the Office of Consumer Affairs in Washington. D.C. 1 2 3 4 5 -Contscted a consumer advocate to suggest an investigation 1 2 3 4 5 .Contactsd a federal regulatory agency (ex. Federal Trade Commission. Consumer Product Safety Commission etc.) -Flled complaint in Small Claims Court 1 -Didn‘t do anything at all 1 -Other (please list) NM” N 090 9 .‘fi ‘ 0‘.“ 0 42b. 42c. if you indicated in the previous question that yOu (or your spouse) “didn‘t do anything at all" to correct the product or service problem. check the one reason which best explains why. [10:] D i didn't think it was worth the time and effort (1) D i wanted to do something about it but never got around to it (2) Di didn't think that i could get anyone to do anything ab0ut it (3) DI didn't know what to do about it or who to go to for help (4) C Other (please list) [3 Doesn't apply (9) How aetisified were yOu (or your spouse) with the aCtions taken to correct the product or service problems? [104] D I received more than i asked for (1) D i was completely satisfied (2) D I was not completely satisfied but the solution was acceptable (3) i: l was not completely satisfied but i did get something (4) [I I was not at all satisfied (5) D Other (please list) [3 Don't know (9) 43b. . Please write in the number of children you Mve_____ [105-107) Please write In the number of children you hsvs living at home with yo_u [100) 44b. 45a. . Please write in the space provided the number of times per week that you MB on the LOLODDQQQ with r_s_istives (other than your Children). friends or mighbgrs? times per week [109] number of Please write in the space provided the number of times per week that you v_islt In person with relative; (Other than yOur children). friends or neighbors? [110] times per week number of Please write in the space provided the number of times per week that yOu M on the mphons with y0ur Mm? [111] times per week number of 45b. Please write In the space provided the number of times per week that you visit in person with y0ur children? [112] times per week number of 45. i presently live with (Check one response): (113) C Spouse (1) [3 Alone (2) [3 Children (3) C Relative (other than child) (4) C Friend (5) C Other (please list) 47a. Regarding your housing costs. do you; [114] C. Pay all the rent/mortgage payment yourself (1) [3 Receive more than half of the rentl mortgage payment from a family member (other than spouse) (2) 165 Ci Receive less than half of the rentl mortgage payment from a family member (other than spouse) (3) C3 Become a government subsidy (4i [3 Share the rent/mortgage with another non-relative (5) D Other (please list) Ci Don’t know (9) 47b. Please cheek below whether you £191 or m the place in which you live. [115) C l rent (If rent. answer question 47C) (1) D i own or am buying (if own. answer questions 47d and 47s (2) {2' Don‘t know (9) 47c. if you gn_t. do you live In a(n): (ire-1171 [2 Single family dwelling (1) C] Hall of two family home (duplex) Or flat (2) 0 Apartment (3) C Room in someone else‘s home (4) C Home for the elderly (5) O Mobile home (6) D Townhouse (7) C Other (please list) [I Don't know (9) 47d. If you own: [715] C 00 y0u own your place outright (1) I Are yOu Currently making mortgage payments (2) I: Don't know (9) 47s. It you 9_w_n. do you live in a(n): [119] C Single family dwelling (1) C Hall of two family home (duplex) or flat (2) 2 Apartment (3) , C Mobile home (4) ‘ C Cooperative (5) C Condominium (6) C Other (please list) I 48a. Please indicate the person who does most of I the weekly shopping (food. clothing. house- , wares. etc.) for the household. [120) i C} Husband (1) i C Wife (2) l C Children (3) C Relative (other than children) (4) C Friend (5) ('3 Sell (6) [3 Other (please list) 48b. How does the person who does the weekly household shopping My get to the store? [121-122) C Walk (1) C Drive self (2) C Ride with friend/neighbor (3) 5 Ride with adult child (4) C Bus (5) D Taxi (6) C Subway (7) C Other (please list) [3 Don’t know (9) 48c. 00 you have any health conditions or physical disabilities that limit yOur ability to gel to the store? [12 D Yes (if yes. answer question 45d) (1) D No (If no. skip to question 49) (0) 48d. If so. please describe this health condition or physical disability' [124-125] 49 Taking all things together. would you say y0u‘re. very happy. pretty happy or not too happy these days? [125‘] f: Very happy (1) 2 Pretty happy (2) I Not too happy (3) D Don't know (9) 50 Generally. how satisfied are y0u with the way yOu are living now - that is. as far as money and what yOu are able to have are concerned? Wauld yOu say the way you are living is: [127] C More than satisfactory (1) Z Satisfactory (2) 2 Less than satisfactory (3) [2 Very unsatisfactory (4) [2 Don't know (9) 51. Would ydu say the way you are living Is better than. worse than. or abdut the same as that of yOur friends and acquaintances? [128] C Better (1) C Worse (2) .2 Same (3) 2 Don‘t know [9) 51a WOuld you say the way you are living (during retirement) is better than. worse than. 0r about the same as W? [129] C Better (1) : Worse (2) C Same (3) : DOn't know (9) 52 Please write in the age at which y0u and L0,)! spouse (It alive at the time) _— retired'? — [1”135] a. _ b. Sell SpOuse 53. Please indicate why m and your spOuse (ii alive at the time) retired. Place a "1" next to the ma)or reason for retirement and a “2" next to any other reasons that apply. [res-144) [145-153] a. Self b. Spouse ~Health reasons a Cl -Mendatory re- tirement age for employer C] D -Adequate re- tirement in- come 0 C1 -Forced early retirement (:1 D -interest In travel/leisure hobbies D 0 -Good early re- tirement offer from employer C) D -Tired of working [3 D -Spouse retired D D -Other (please list) ‘Does not apply {3 Ci 54a. Before you retired. did you start to "Cut back" on any of yOur living expenses? [154] l: Yes (if Yes. answer question 54b) (1) C‘ No (if No. skip to question 55a) (0) (I: Don't know (Skip to. question 55a) (9) 166 if you did Cut back. in what areas? (Check all that apply) [155-165] [1 Food (1) 8 Clothing (2) C Hausing (3) E: Entertainment (4) C Transportation (5) (3 Household expenditures (6) Z Other (please list) E Dont know (9) 5a. In the two years before you retired. did you have any outstanding bills? [:55] D Yes (if Yes. answer questions 55b and 55c) (1) D No (if No. skip to question 56a) (0) D Don't know (skip to question 56a) (9) 55b. Outstanding film carer; Retirement [167-176] [2 Past due gas bills (1) (home heating/cooling) [3 Past due electric bills (2) 2 Past due telephone bills (3) a Bankcard bills (4) (Visa, Master Charge. etc.) :3 Store credit card bills (5) C Medical bills (5) 8 Mortgage (7) C Property Taxes (3) C Car (10) C Other (please list) 55c. OutstindingBilis Paid Off [glare Retirement (177-186] D Past due gas bills (1) (home heating/cooling) L" Past due electric bills (2) [3 Past due telephone bills (3) C Bankcard bills (4) (Visa. Master Charge etc.) [2‘ Slow Credit card bills (5) [3 Medical bills (6) [2 Mortgage (7) (3 Property taxes (8) C Ger (l0) 0 Other (please list) 56a. In the two years before you retired. did you buy any major items? (see list in quesnon 56:2) [187] C1 Yes (if Yes. answer question 56b) (1) ('3 No (it No. skip to question 57) (0) ('3 Don't know(Skip to questions 57) (9) 56b. Which items did you buy? [103.197] (3 Refrigerator or stove (1) C‘ Washer or dryer (2) [3 Furniture (3) C TOIBVISIOR (4) D CarlTruck/Van (5] U Carpeting (5) D Boat (7) D Camper/recreational vehicle (8) [3 other (please list) [3 Don‘t know (9) 57. Please check the highest level of education that you and ycur spouse (if married) have completed. (res-red) a. Your b. Spouse's Education Education .some elementary school [3 C] (1] -Cornpieted elementary school C] D (2) -$ome high school D D (3) —Completed high school (3 Cl (4) -Some college D D (5) Completed college (4 year degree) D D (6) ~$ome graduate school (Masters or Professional 099'”) Cl C] (7) ~Completed graduate pro- 9'3'“ D Ci (8) Does not apply C] C] (9) Please write in the spaces provided the occup- ations held by yo_u and your spouse (if married) before retirement. [200.203] YOur Occupation 59. Spouse's Occupation Please check the bracket which indicates your total annual family income 1 year before I retirement. 204-205] L: Under 55.000 (1) Z 55.000-57.499 (2) : 5750059999 (3) : 510,000-514,999 (4) T. 515000519999 (5) T. 520000424999 (6) C 525000535999 (7) Z 536000449999 (8) E 550.000 and ever (10) Please check the bracket which indicates yOur total annual family income after W (20st C Under 53.000 (1) C 53000-54999 (2) C 55000-56999 (3) C 57000-59999 (4) C 510000514999 (5) 2 515,000-519,999 (B) I: 520,000-524,999 (7) I Over 525.000 (8) 167 B1 . From the following list of retirement income scurces. please check those sources from which you receive income. Place a “1" next to the two__s_ources VNCN Provide the ma)or part (50 percent or more) of your total Income and a "2" next to those sources which provide mg income. [207-221] .Sociai Secuth D ~Govemmenl employee pension C] oPension (from previous non— D government employer such as private businesses. unions or corporations) -Pension (Keogh. IRA. etc.) -Veterans benefits ~Supplementai Security income -Savlngs or interest from savings -Real estate rentals or sales -lnvestments (stocks. bonds) -Money from children -Money from relatives other than Children or spouse -Trust Fund -Llfe insurance policies converted to annuity -Rallroad Pension(s) -Other (please list) U DO DDDDDDDD Please indicate the ethnic origins of both yo_u and not 5991!! (If married). [222.23] a. Self b. Spouse White 0 Cl (1) Black C) D (2) Asian or Pacific Islander D D (3) American Indian or Alaskan Native D D (4) Hispanic [3 D (5) Other (please list) Does not apply 0 D (9) Please write your zip code on this line 1224'“ Please indicate yOur sex [226] 2 Male (1) F: Female (2) THANK YOU VERY MUCH FOR YOUR COOPERATION IN COMPLETING THIS SURVEY. Please return your completed questionnaire in the addressed postage paid envelope. APPENDIX B VARIMAX ROTATED FACTOR MATRIX: PRETEST DATA 168 169 Table B-l.—-Varimax Rotated Factor Matrix: Pretest Data Question Number Factor Loadings ¥ Factor 1 Factor 2 Scale 1: Sales/Price Perceptions 22 .50554 * .33740 25 .91911 .19917 29 .99440 .05578 18 .05079 .62135 32 -.l4807 .80640 Explained Variance Per Factor 69.4 30.6 Scale 2: Product Information Sources 17 .4677l .21147 20 .65380 .lll4l 38 .79449 .l828l 39 .50085 .34923 27 -.090 .14445 40 .12879 .00165 Explained Variance Per Factor 58.5 41.5 Scale 3: Price/Quality Relationship 12 .53227 * .52346 14 .54562 .54197 34 .90393 .2866l 35 .6335 .28050 30 .29182 .95762 Explained Variance Per Factor 86.7 13.3 Scale 4: Shopping Difficulties 6 -.64756 * .30290 24 .88018 .49869 26 .77018 .19305 9 -.0386 .44349’ 31 .07182 .42384 Explained Variance Per Factor 79.3 20.7 170 Table B-l.--Continued. Question Number Factor Loadings Factor 1 Factor 2 Scale 5: Shopping Assistance 4 .08668 .73195 5 -.07649 .98729 13 -.25252 .02225 16 1.00973 8 .13869 23 .36233 .23833 33 .60842 .05129 Explained Variance Per Factor 55.6 44.4 Scale 6: Cash/Credit Payment 7 -.40509 .74199 37 -.19206 .72502 8 .84068 * -.32519 36 .95289 .31123 Explained Variance Per Factor 83.8 16.2 Scale 7: Shopping Habits 10 -.41842 * .11209 19 .99186 .01416 28 ..86033 .11613 11 .40096 .47697 15 -.11974 .59240 21 .25518 .96797 Explained Variance Per Factor 61.3 38.7 a Indicates the highest loading in each row. Measuring more than one construct. APPENDIX C RELIABILITY COEFFICIENTS--COEFFICIENT ALPHA: PRETEST DATA l7l Table C-1.--Reliabi1ity Coefficients--Coefficient Alpha: Pretest Data 172 Scale Grouping Coefficient Alpha 10. ll. 12. l3. l4. Sales/Bargain Shopping Behavior Price/Quality Consciousness Product Information Sources TV Advertising/Product Information Price Consciousness Newspaper Ads Shopping Problems (Physical) Enjoyment of Shopping Salesperson Assistance External Information Sources Cash Purchases Credit Purchases Shopping Habits Brand Names .91 .4l .65 .35 .80 .72 .69 .59 .77 .73 .78 .78 .44 .50 REFERENCE LIST l73 REFERENCE LIST Abdel-Ghany, Mohamed. “Quality of Life From the Perspective of the Elderly." Home Economics Research Journal 6,1 (1977): 38-47. Allan, Carole 8. "Measuring Mature Markets." American Demographics 3,3 (1981): 13-17. "Americans Change." Business Week 2522 (February 20, 1978): 64-69. "America's War on Poverty--Is It a No-Win Struggle?" U.S. News and World Report 86,3 (1979): 20-23. Amos, Jimmy R.; Brown, Foster Lloyd; and Mink, Oscar R. Statistical Concepts, A Basic Program. New York: Harper & Row, Publishers, 1965. Andrews, F., and Withey, 5. "Developing Measures of Perceived Life Quality." Social Indicators Research 1 (1974). Atchley, Robert C. "Retirement and Work Orientation." The Geron- tologist 11, Part I (1971): 29-32. Barfield, Richard E., and Morgan, James N. “Trends in Planned Early Retirement." The Gerontologist 18,1 (1978): 13-18. . "Trends in Satisfaction With Retirement." The Gerontolo- gist 18,1 (1978): 19-23. Barksdale, Hiram C., and Darden, William R. "Consumer Attitudes Toward Marketing and Consumerism." Journal of Marketing 36 (October 1972): 28-35. Barnes, James G. "A Hierarchical Model of Source Effect in Retail Newspaper Advertising." In Advances in Consumer Research, Vol. 5, pp. 235-42. Edited by H. Hunt. Ann7Arbor: PAssociation for Con- sumer Research, 1978. Barnhill, J. A. "Market Injustice: The Case of the Disadvantaged Consumer." Journal of Consumer Affairs 6,1 (1972): 78-83. Bearden, William 0.; Gustafson, A. William; and Mason, J. Barry. "A Path-Analytic Investigation of Life Satisfaction Among Elderly Consumers." In Advances in Consumer Research, Vol. 6, pp. 386-91. Edited by William L. Wilkie. Ann Arbor: Association for Consumer Research, 1979. 174 175 Bell, Martin L., and Emory, William C. "The Falterin Marketing Concept." Journal of Marketing 35 (October 1971): 37-42. Bernhardt, Kenneth L. "Consumer Problems and Complaint Actions of Older Americans: A National View." Journal of Retailing 57,3 (1981): 107-23. , and Kinnear, Thomas C. "Profiling the Senior Citizen Market." In Advances in Consumer Research, Vol. 3, pp. 449-52. Edited by Beverlee B. Anderson. Cincinnati: Association for Consumer Research, 1976. Best, Roger, and Hawkins, Del I. "The Effect of Varying Response Intervals on the Stability of Factor Solutions of Rating Scale Data." In Advances in Consumer Research, Vol. 6, pp. 539-41. Edited by William L. Wilkie. Ann Arbor: Association for Con- sumer Research, 1979. Binstock, Robert. "Planning for Tomorrow's Urban Aged, An Analyst's Reaction." The Gerontologist 15,1 (1975): 42-43. Birren, J. "Age and Decision Strategies." In Interdisciplinary, Topics in Gerontology, Vol. 4, pp. 23-26. Edited By A. T. Welford. NewTYork: ’S. Korger Publishers, 1969. Blitstein, D. Human Social Development. New Haven, Connecticut: College and University Press Services, Inc., 1971. Block, Joyanne. "The Aged Consumer and the Marketplace." Marguette Business Review (Summer 1974): 73-80. Botwinick, J. Aging and Behavior. New York: Springer Publishing Co., 1973. Brotman, Herman 8. "Income and Poverty in the Older Population in 1975." The Gerontologist 17,1 (1977): 23-26. Brunner, Allen 6., and Carroll, Stephen J. "The Effect of Prior Notification on the Refusal Rate in Fixed Address Surveys." Journal of Advertising Research 9 (March 1969): 42-44. Bureau of Labor Statistics. In Carole Allan. "Maturing Markets." American Demographics 3,3 (1981): 13-17. Burkhauser, Richard V., and Tolley, G. S. "Older Americans and Market Work." The Gerontologist 18,5 (1978): 449-53. Burton, John R., and Hennon, Charles 8. "Consumer Education for the Elderly." Journal of Home Economics 73,2 (1981): 24-28. 176 Butler, Robert N. Why Survive: Being Old in America. New York: Harper & Row, 1975. Campbell, Angus; Converse, P. E.; and Rogers, W. L. The Quality of American Life: Perceptions, Evaluations, and Satisfactions. New York: Russell Sage Foundation, 1976. Canestrari, R. E., Jr. "Paced and Self-Paced Learning in Young and Elderly Adults." Journal of Gerontology 18 (1963): 165-68. Carmines, Edward G., and Zeller, Richard A. Reliability and Validity Assessment. Beverly Hills, California: Sage Publications, 1979. Clark, Robert, and Spengler, Joseph. "The Implications of Future Dependency Ratios and Their Composition." In Aging and Income, pp. 55-89. Edited by Barbara Pieman Herzog. New York: Human Sciences Press, 1978. Clark, Robert; Kreps, Juanita; and Spengler, Joseph. "Economics of Aging: A Survey." Journal of Economic Literature 16 (1978): 919-62. Conference Board Report, 1981. In Rosalie Greenfield. "The Maturing Market." Stores (November 1981): 30-34. Cravens. David W. "An Exploratory Analysis of Individual Information Processing." Management Science 16,10 (1970): 656-70. Cronbach, L. J. "Coefficient Alpha and the Internal Structure of Tests." Esychometrika 16 (1951): 297-334. Cumming, Elaine, and Henry, William E. Growing Old: The Process of Disengagement. New York: Basic Books, 19611 Cumming, Elaine, and McCaffrey, D. "Some Considerations Associated With Morale Among the Aging." In Psychopathology of Aging. Edited by P. Hoch and J. Zubrin. New York: Grune and Stratton, 1961. Davis, Joseph S. "Standards and Content of Living." American Economic Review 35 (1945): 1-15. Day, George 5. "Assessing the Effects of Information Disclosure Requirements." Journal of Marketing 40 (April 1976): 45-52. , and Aaker, David A. "A Guide to Consumerism." Journal of Marketing 34 (July 1970): 12-19. Day, Ralph L., and Ash, Stephen 8. "Consumer Response to Dissatisfac- tion With Durable Products." In Advances in Consumer Research, Vol. 6, pp. 438-44. Edited by William L. Wilkie. Ann—Arbor: Association for Consumer Research, 1979. 177 Day, Ralph L., and Landon, Laird, Jr. "Collecting Comprehensive Consumer Complaint Data by Survey Research." In Advances in Consumer Research, Vol. 3, pp. 263-68. Edited by Beverlee78. Anderson. Cincinnati: Association for Consumer Research, 1976. Diamond, Steven; Ward, Scott; and Faber, Ronald. "Consumerism: Analysis of Calls to a Consumer Hot Line." Journal of Marketing 40 (January 1976): 58-62. Dillman, D. A. "Increasing Mail Questionnaire Response in Large Samples of the General Public." Public Opinion Quarter1y_36 (1972): 254-57. Dodge, Robert E. "Purchasing Habits and Market Potentialities of the Older Consumer." Law and Contemporary Problems 27 (1962): 142-56. "Elliot Calls for Research to Counteract Consumerism." Advertising Agg_44 (October 22, 1973): 10. Emory, C. William. Business Research Methods. Homewood, Illinois: Richard D. Irwin, Inc., 1980. Enis, Ben M. Marketing Principles: The Management Process. 2d ed. Santa Monica,7California: Goodyear, 1977. Farace, Bettie F. "Factors Predicting Retirement Decision-Making for Michigan Farm Families." Ph.D. dissertation, Michigan State University, 1978. Feather, N. T. "Attribution of Responsibility and Valence of Success and Failure in Relation to Initial Confidence and Task Perform- ance.“ Journal of Personality and Social Psychology_13 (1969): 129-44. Flint, Jerry. "The Old Folks." Forbes 125,4 (1980): 51-56. Frank, Ronald E., and Green, Paul E. ggantitative Methods in Marketing. Englewood Cliffs: Prentice-Hall, 1967. Fromm, Erich. "Alienation Under Capitalism." In Man Alone: Alienation in Modern Society, pp. 56-73. Edited by Erich Josephson and’Mary Josephson. New York: Dell Publishing Company, Inc., 1962. Gelb, Betsy. "Exploring the Gray Market Segment." MSU Business Topics (Spring 1978): 41-46. Gerner, Jennifer L., and Bryant, Keith W. "Consumer Economics of Appli- ance Warranties." Working Paper, Department of Consumer Economics and Public Policy, Cornell University, 1977. 178 Gitter, A. George, and Knoche, E. "Importance Ratings of Sixteen Aspects of Life." Communication Research Center Report No. 59. Boston: Boston University, 1971. Gold, Byron; Kutza, Elizabeth; and Marmor, Theodore. "United States Social Policy on Old Age: Present Patterns and Predictions." In Social Policy, Social Ethics, and the AginggSociety. Edited by B. Neugarten and R. Havighurst. National Science Foundation Research Association. Washington, D.C.: Superintendent of Docu- ments, 1976. Goodman, John. In Ann Harvey. "Consumerism." Public Relations 35,2 (1979): 47. Gounard, B. R., and Hulicha, C. M. "Maximizing Learning Efficiency in Later Adulthood: A Co nitive Problem Solutions Approach." Educa- tional Gerontology 2 %1977): 424. Grainer, Marc A.; McEvoy, Kathleen; and King, Donald. "Consumer Problems and Complaints: A National View." In Advances in Con- sumer Research, Vol. 6, pp. 494-500. Edited by William L. Wilkie. Ann Arbor: Association for Consumer Research, 1979. Graney, M., and Graney E. "Communications Activity Substitutions in Old Age." Journal of Communications 24,4 (1974): 88-96. Greenfield, Rosalie. "The Maturing Market." Stores (November 1981): 30-34. Greyser, Stephen A., and Diamond, Steven L. "Business Is Adapting to Consumerism." Harvard Business Review 51 (September-October 1974): 38-40+. Gross, Bertram M. Social Intelligence for America's Future: Explora- tions in Societal Problems. Boston: Allyn and Bacon, 1969. Guttman, Louis. "A Basis for Analyzing Test-Retest Reliability." Psychometrika 10,4 (1945): 255-82. Haines, George H., Jr. "Three Papers on Consumer Satisfaction/ Dissatisfaction: A Comment." In Advances in Consumer Research, Vol. 6, pp. 450-52. Edited by William L. Wilkie. Ann Arbor: Association for Consumer Research, 1979. Harvey, Ann. "Consumerism." Public Relations 35,12 (1979): 47. Hempel, Donald J., and Rosenberg, Larry J. "Consumer Satisfaction: A Neglected Link?" In Advances in Consumer Research, Vol. 3, pp. 261-62. Edited by Beverlee B. Anderson. PCincinnati: Association for Consumer Research, 1976. 179 Henretta, John C.; Campbell, Richard T.; and Gardocki, Gloria. "Survey Research in Aging, An Evaluation of the Harris Survey.“ The Gerontologist 17,2 (1977): 160-67. Holton, Richard H. "Government-Consumer Interests: Conflict and Prospects." Changing Marketing Systems--Consumer, Corporate and Government Interfaces. Edited by Reed Moyer. Proceedings of the American Marketing Association (Winter 1967): 15. Horn, Bob. In Rosalie Greenfield. "The Maturing Market." Stores (November 1981): 30-34. Horowitz, Abraham 0., and Golob, Thomas F. "Survey Data Reliability Effects on Results of Consumer Preference Analysis." In Advances in Consumer Research, Vol. 6, pp. 532-38. Edited by William E. Wilkie. Ann Arbor: Association for Consumer Research, 1979. Hudson, Robert. "The 'Graying' of the Federal Budget and Its Conse- quences for Old-Age Policy." The Gerontologist 18,5 (1978): 428-40. Hull, C. Hadlai, and Nie, Norman H. SPSS Update. New York: McGraw- Hill Book Company, 1979. Hustad, Thomas P., and Pessemier, Edgar A. "Will the Real Consumer Activist Please Stand Up: An Examination of Consumers' Opinions About Marketing Practices." Journal of Marketing Research 10 (August 1973): 319-24. Hutton, R. Bruce; McNeill, Dennis L.; and Wilkie, William L. "Some Issues in Designing Consumer Information Studies in Public Policy." In Advances in Consumer Research, Vol. 5, pp. 131-37. Edited by H. Hunt. Ann Arbor: Association for Consumer Research, 1978. "The Inflationary Squeeze on the Over-65$." Business Week 2535 (May 22, 1978): 148-50. Irelan, Lola M., and Bisner, Arthur. "Low Income Outlook of Life." In Low Income Life Styles, pp. 1-9. Edited by Lola M. Irelan. Washington, D.C.: Government Printing Office, 1968. Jacoby, Jacob; Speller, Donald E.; and Kohn, Carol A. "Brand Choice Behavior as a Function of Information Load." Journal of Market- ing Research 11 (February 1974): 63-69. James, Richard 0. "Measuring the Quality of Life." The Wall Street Journal, May 1972. Juster, F. Thomas. "The Psychology of Inflation." Economic Outlook U§A_6,l (1978-1979): 16-18. 180 Kelley, H. H. "Causal Schemata and Attribution Process." In Attri- bution: Perceiving the Causes of Behavior, pp. 151-74. Edited by El E. Jones et a1. (Morristown, New Jersey: General Learning Press, 1972. Klippel, Eugene R., and Sweeney, Timothy W. "The Use of Information Sources by the Aged Consumer." The Gerontologist 14 (April 1974): 163-66. Koeske, Randi D., and Srivastava, Rajindra. "The Sources and Handling of Consumer Complaints Among the Elderly." In Consumer Satisfac- tion, Dissatisfaction, and Complaining Behavior, pp. 139-43. Edited"by Ralph L. Day. ‘Bloomington: Indiana University School of Business, 1977. Krishnan, S., and Valle, Valerie A. "Dissatisfaction Attributions and Consumer Complaint Behavior." In Advances in Consumer Behavior, Vol. 6, pp. 445-49. Edited by WilliamAL. Wilkie. Ann Arbor: Association for Consumer Research, 1979. Kushler, Martin Gregory. "Alternative Modes of Conducting Outreach to Low Income Elderly: An Experimental Examination." M.A. thesis, Michigan State University, 1977. Lambert, Zarrel V. "An Investigation of Older Consumers' Unmet Needs and Wants at the Retail Level.“ Journal of Retailing 55,4 (1979): 35-57. Lawther, Karen. "Social Integration and the Elderly Consumer: Unfairness Awareness, Complaint Actions and Information Usage." In Research Frontiers in Marketing: Dialogues and Directions, pp. 341-45. Edited by Subhash C. Jain. 1978 Educators, 1978, Series 43. Leavitt, Clark. "Consumer Satisfaction as the Ultimate Life Force." In Advances in Consumer Research, Vol. 3, pp. 252-58. Edited by BeverleelB. Anderson. Cincinnati: Association for Consumer Research, 1976. Liefeld, J. P.; Edgecombe, F. H. C.; and White, Linda. "Demographic Characteristics of Canadian Consumer Complainers." Journal of Consumer Affairs 9 (Summer 1975): 73-80. Linden, Fabian. "The Consumer's View of Value Received--l974." The_ Conference Board 11 (1974): 48-53. . "Consumer Markets: Midlife and Beyond." Across the Board 13,12 (1976): 50-52. 181 Lundstrom, William J., and Lamont, Lawrence M. "The Development of a Scale to Measure Consumer Discontent." Journal of Marketing Research 13,4 (1976): 373-81. Main, Jeremy. "Retire in Haste, Repent at Leisure." Money 3,11 (1974): 49-52. Margolius, Sidney. "The Consumers' Real Needs." Journal of Consumer Affairs 9 (Winter 1975): 129-38. Mason, J. 8., and Smith, 8. "An Exploratory Note on the Shopping Behavior of the Low Income Senior Citizen." Journal of Consumer Affairs (Summer 1973): 204-10. McGuire, J. "Some Internal Psychological Factors Influencing Consumer Choice." Journal of Consumer Research 2 (1976): 302-19. Michigan State University Computer Laboratory. SPSS-6000 Supplement. East Lansing: Michigan State University, 1981. Michigan Statistical Abstract. 16th ed. Edited by D. I. Verway. Detroit: Bureau oleusiness Research, School of Business Admin- istration, Wayne State University, 1981. Michman, Ronald 0. "Changing Patterns in Retailing." Business Horizons 22,5 (1979): 33-38. Miller, John A. "Who Is the Discontented Consumer." In Combined Proceedings, pp. 486-90. Edited by Thomas V. Greer. Chicago: AmericanAMarketing Association, 1974. Miller, Kenneth E., and Turner, Marilynn. "The Effects of Sponsorship on Mail Survey Response and Evaluation Bias." In Advances in Consumer Research, Vol. 6, pp. 542-44. Edited by William L. Wilkie. Ann Arbor: Association for Consumer Research, 1979. Mizerski, Richard, and Green, Stephen. "An Investigation Into the Causal Links Between Attribution Schema and Decision-Making." In Advances in Consumer Research, Vol. 5, pp. 126-30. Edited by H. KéitthUnt. Ann Arbor: Association for Consumer Research, 1978. Nagy, Marilyn P. "Consumer Attitudes Toward the Prevention and Reso- lution of Consumer Complaints." Ph.D. dissertation, Michigan State University, 1976. Nahemow, L. "Persuasibility, Social Isolation and Conformity Among Residents of a Home for Aged." Ph.D. dissertation, Columbia University, 1963. 182 National Clearing House on Aging. Facts About Older Americans--1977. Washington, D.C.: U.S. Department of Health, Education, and Welfare, 1977. Neugarten, Bernice L., and Havighurst, Robert J., eds. Social Policy, Social Ethics and the Aging Society, Washington, D.C.: Superin- tendent of Documents, National Science Foundation--Research Association, 1976. Nichols, Barbara 0., and Dardis, Rachel. "Consumer Satisfaction With Home Furnishings." Textile Chemist and Colorist 5 (September 1973): 23-27. Nie, Norman et a1. Statistical Package for the Social Sciences. 2d ed. New York: McGraw-Hill Book Company, 1975. Novick, Melvin R., and Lewis, Charles. "Coefficient Alpha and the Reliability of Composite Measurements." Psychometrika 32,1 (1967): 1-13. Office of Services to the Aging, State of Michigan. The Michi an Comprehensive Plan on Aging, East Lansing, Michigan: 1973 State Plan on Aging--1981-1983. East Lansing, Michigan, ———1980. Oyer, Herbert J., and Dyer, E. Jane. A in and Communication. Baltimore: University Park Press, 6. Paolucci, Beatrice; Hall, Olive A.; and Axinn, Nancy. Family Decision Making: An Ecosystem Approach. New York: John Wiley and Sons, 1977. Payne, R. "Some Theoretical Approaches to the Sociology of Aging." Social Forces 38 (May 1960): 359-62. Peterson, Vicki. "Information Processing of the Elderly." Unpub- lished paper, Michigan State University, 1981. Pfaff, Martin, and Blivice, B. "Socio-Economic Correlates of Consumer and Citizen Dissatisfaction and Activism." In Consumer Satisfgg: tion, Dissatisfaction, and Complaining Behavior, pp. 115-23. Edited by Ralph Day. Bloomington: Indiana University, School of Business, 1977. Phillips, Lynn W., and Sternthal, Brian. "Age Differences in Infor- mation Processing: A Perspective on the Aged Consumer." Journal of Marketing Research 14,4 (1977): 444-57. Plonk, Martha, and Pulley, Mary Ann. "Financial Management Practices of Retired Couples.“ The Gerontologist 17,3 (1977): 250-61. 183 Plummer, Joseph T. "Life Style, Social and Economic Trends Influenc- ing Consumer Satisfaction." In Conceptualization and Measurement of Consumer Satisfaction and Dissatisfaction. Edited by H. Keith Hggt. Cambridge, Massachusetts: Marketing Science Institute, Pruden, Henry 0., and Longman, Douglas S. "Race, Alienation and Consumerism." Journal of Marketing 36,3 (1972): 58-62. Quinn, Joseph F. The Early Retirement Decision: Evidence From the Retirement History Study. Social Security Administration, 978. . The Microeconomics of Early Retirement: A Cross-Sectional View. Washington, D.C.: U.S. Department of Health, Education and Welfare, 1975. Rabbitt, P. "An Age Decrement in the Ability to Ignore Irrelevant Information." Journal of Gerontology 20 (April 1965): 233-38. Rao, C. P., and Kiser, G. E. An Evaluation of Consumer Inflation Psychology and an Assessment of Income Effects. Proceedings of the Southern Marketing Association, 1977. Regional White House Conference on Aging. Proceedings. Lansing, Michigan: March 14, 1981. Reinecke, John A. "The 'Older' Market--Fact or Fiction?“ Journal of Marketing 28,1 (1964): 60-64. Rones, Philip L. "Older Men--The Choice Between Work and Retirement." Monthly Labor Review (November 1978): 3-10. Rosen, Jacqueline L., and Neugarten, Bernice L. "Ego Functions in the Middle and Later Years: A Thematic Apperception Study of Normal Adults." Journal of Gerontology_15,B (January 1960): 62-69. Rosow, I. "Old People: Their Friends and Neighbors." American Behavioral Scientist (September/October 1970): 59-6 . Russo, J. Edward. "Consumer Satisfaction/Dissatisfaction: An Outsider's View." In Advances in Consumer Research, Vol. 6, pp. 453-55. Edited by William L. Wilkie. *Ann Arbor: Associa- tion for Consumer Research, 1979. Sargent, Tony. "Social Security: The Program That Delivered Too Much." Enter rise Publication of National Association of Manu- facturers) 6,§ (1982): 2-3. 184 Schlater, Jean Davis, Director. National Goals and Guidelines for Research in Home Economics. Washington, D.C.: iAmeriEan Home Economics Association, 1970. Schramm, W. "Aging and Mass Communication." In Aging and Society, pp. 352-75. Edited by M. Riley and J. Riley. New York: Russell Sage Foundation, 1969. Schulz, James H. The Economic Status of the Retired Aged in 1980: Simulation Proiections. Washington, D.C.: U.S. Department of Health, Education, and Welfare, Social Security Administration, #24, 1968. . The Economics of Aging. 2d ed. Belmont, California: Wadsworthinblishi 9 Co., Inc., 1980. Sentry Insurance. Consumerism at the Crossroads. Stevens Point, Wisconsin: Sentryiinsurance Company,’1977. Shadko, Jeannette Bond. "Changes in Activity of Older People: Benzie County, Michigan." Ph.D. dissertation, Michigan State Univer- sity, 1967. Shanas, E., and Hauser, P. "Zero Population Growth and the Family Life of Old People." Journal of Social Issues 30,2 (1974): 79-92. Sheppard, Harold L. "The Economics of Population, Mortality, and Retirement." In The Economics of Aging, pp. 1880-83. A National Journal Issues Book. ’WaShington, D.C.: The Government Research Corporation, 1978. , and Rix, Sara E. The Graying of Working America--The Comin Crisis of Retirement-Age Policy. New York: ThelFree Press, 197 . Sheth, Jagdish N., ed. Multivariate Methods for Market and Survey, Research. Chicago: AmeriCan Marketing Association, 1977. Siegel, Jacob S. "Demographic Aspects of Aging and the Older Popula- tion in the United States." Current Population Reports. Special Studies Series P-23, No. 59. Washington, D.C.: Govern- ment Printing Office, 1976. Sproles, George 8., and Geistfeld, Loren V. "Issues in Analyzing Consumer Satisfaction/Dissatisfaction With Clothing and Textiles. In Advances in Consumer Research, Vol. 5, pp. 383-91. Edited by H. Hunt. Ann Arbor: PASSOCiation for Consumer Research, 1978. ; and Badenhop, Suzanne. "Efficiency of Consumer Performance." In The Proceedin s from the American Council on Consumer Interest, p. 61. Denver, Olorado: 1977. 185 Steiner, G. The People Look at Television. New York: Alfred Knopf, Inc., 1963. Steiniger, Lynn 8., and Dardis, Rachel. "Consumers' Textile Com- plaints." Textile Chemist and Colorist 3 (1971): 33-37. Stokes, Raymond C. "Consumerism and the Measurement of Consumer Dissatisfaction." In Attitude Research Bridges the Atlantic. Edited by Philip Levine. Chicago: American MarketingiAssocia- tion, 1975. Sturdivant, Frederick 0. "Better Deal for Ghetto Shoppers." Harvard Business Review (March-April 1968): 132. Sundling, Betty Jeanne. "An Exploratory Study of the Financial Adjustment of a Selected Group of Single Female Retirees to Their Retirement Income." Ph.D. dissertation, Michigan State University, 1979. Szafran, J. "Psychophysiological Studies of Aging in Pilots." In Human Aging and Behavior. Edited by G. Talland. New York: Academic Press, 1968. Terleckyj, Nestor E. National Goals Accounting_(Draft #1). Washing- ton, D.C.: National Planning Association, 1972. Torda, Theodore. "The Impact of Inflation on the Elderly." Economic Review (October-November 1972): 3-17. Towle, Jeffrey, and Martin, Claude R., Jr. "The Elderly Consumer: One Segment or Many?" In Advances in Consumer Research, Vol. 3, pp. 463-68. Edited by Bevehlee B. Anderson. Cincinatti: Association for Consumer Research, 1976. "Translations in Gerontology--From Lab to Life: Psychophysiology and Speed of Response." American Psychologist 29 (November 1974): 808-15. Tull, Donald S., and Albaum, Gerald S. Survey Research: A Decisional Approach. New York: Intext Educational Pbblishers, 1973. Tu11, Donald S., and Hawkins, Del 1. Marketing Research. 2d ed. New York: Macmillan Publishing Co., Inc., 1980. Union Mutual Life Insurance Company. In Jeremy Main. "Retire in Haste, Repent at Leisure." Money 3,11 (1974): 49-52. U.S. Bureau of the Census. U.S. Census of Population: 1980. Washington, D.C.: Government Printing Office, 1980. 186 . 1980 Census of Population. Supplementary Reports, Series PC 80-51-1. WaShington, D.C.: Government Printing Office, 1980. Census of Population and Housing, Supplementary Report, Provisional Estimates of Social, Economic, and Housing Char- acteristics. Washington, D.C.: Government Printing Office, 1980. . "Demographic Aspects of Aging and the Older Population in the United States." Current Population Reports. Special Studies Series, P-23, No. 59. Washington, D.C.: Government Printing Office, 1976. . "The Social and Economic Status of the Black Population in the United States: An Historical View, 1790-1978." Current Pppulation Reports. Special Studies Series P-23, No. 80. Washington, D.C.: Government Printing Office, 1980. . Statistical Abstract of the United States: 1981. 102 ed. Washington, D.C.: Government Printing Office, 1981. , Bureau of Labor Statistics. Employment and Earnings, Vol. 27, No. 1. Washington, D.C.: Government Printing Office, January 1980. U.S. Department of Health and Human Services, Social Security Adminis- tration. Retirement History Study. SSA Bulletin No. 13-11700 (November 1980). U.S. Department of Health, Education, and Welfare. Annual Report of the Board of Trustees of the OASDI Trust Funds. In Richard S. Foster. “Demographic Trends and Social Security." Kansas City, Missouri: Society of Actuaries Regional Meeting on Pensions, May 24-25, 1979. Valle, V., and Lawther, Karen. "Interpersonal Trust and Attribution of Responsibility Determinants of Elderly Consumer Complaining." In Proceedings of the Division 23 Program of the 1978 American PeychologiEal Association Meeting,71978, pp. 64-67. Valle, V., and Wallendorf, M. "Consumer Attributions of the Cause of Their Product Satisfaction and Dissatisfaction." In Consumer Satisfaction, Dissatisfaction and Complaining Behavior. Edited by R. L. Day. Bloomington: SChool of Business,ilndiana Univer- sity, 1977. Waddell, Fred E. "Consumer Research and Programs for the Elderly-- The Forgotten Dimension." Journal of Consumer Affairs 9,2 (1975): 164-75. 187 Wall, Marjorie; Dickey, Lois E.; and Talarzyk, W. Wayne. "Predicting and Profiling Consumer Satisfaction and Propensity to Complain." In Consumer Satisfaction/Dissatisfaction and Complaining Behavior, pp. 91-101. Edited by Ralph L. Day. Bloomington: Indiana Uni- versity, School of Business, 1977. Wallace, D. "A Case for and Against Mail Questionnaires." Public Opinion Quarterly 18 (1954): 40-52. Warland, Rex H.; Herrmann, Robert 0.; and Willits, Jane. "Dissatisfied Consumers: Who Gets Upset and Who Takes Action." Journal of Con- sumer Affairs 9 (Winter 1975): 148-63. Weiner, B. et a1. "Perceiving the Causes of Success and Failure." In Attribution: Perceiving the Causes of Behavior, pp. 95-120. Edited by E) E. Jones et a1. Morristown, Neleersey: General Learning Press, 1972. Wilde, Louis L. "The Economics of Consumer Information Acquisition." Journal of Business 53,3, pt. 2 (1980): 5143-58. Wilkie, William L. "Analysis of Effects of Information Load." Journal of MarketingyResearch 11.4 (1974): 462-66. . "Consumer Information Acquisition." In Advances in Con- sumer Research, Vol. 3, p. 334. Edited by Beverlee B. Anderson. Cincinnati? Association for Consumer Research, 1976. "Consumer Information Processing and Public Policy Research." In Marketing and the Public Interest, Proceedings of Marketing Science Institute. Cambridge, Massachusetts: Marketing Science Institute, 1978. . "Consumer Information Processing Research: Product Labeling.I Working Paper. Report No. 75-104. Cambridge, Massachusetts: Marketing Science Institute., March 1975. Webster's Seventh New Colle iate Dictionary. Springfield, Massachu- setts: G E’C Merriam ompany,'Pub1ishers, 1967. Woodruff, D. S., and Walsh, D. A. "Research in Adult Learning: The Individual." The Gerontologist (October 1975): 424-30. Zaltman, Gerald; Srivastava, Rajendra K.; and Deshpande, Rohit. "Perceptions of Unfair Marketing Practices: Consumerism Impli- cations." In Advances in Consumer Research, Vol. 5, pp. 247-83. Edited by H. Hunt. Ann Arbor: Association for Consumer Research, 1978. "I11111111111411.11111“