Information Search For Services: the Maturity Segment

ABSTRACT - This study investigates information search for services by individuals 55 years of age and older. Services investigated were financial institutions, barber/beauty shops and doctors. Results indicate that search behavior of older adults for the services investigated is strongly influenced by the number of alternatives considered (size of evoked set) and by individual characteristics such as sex and education. In addition, personal sources were identified as very important information sources for this group.

Citation:

Teresa A. Swartz and Nancy Stephens (1984) ,"Information Search For Services: the Maturity Segment", in NA - Advances in Consumer Research Volume 11, eds. Thomas C. Kinnear, Provo, UT : Association for Consumer Research, Pages: 244-249.

Advances in Consumer Research Volume 11, 1984      Pages 244-249

INFORMATION SEARCH FOR SERVICES: THE MATURITY SEGMENT

Teresa A. Swartz, Arizona State University

Nancy Stephens, Arizona State University

ABSTRACT -

This study investigates information search for services by individuals 55 years of age and older. Services investigated were financial institutions, barber/beauty shops and doctors. Results indicate that search behavior of older adults for the services investigated is strongly influenced by the number of alternatives considered (size of evoked set) and by individual characteristics such as sex and education. In addition, personal sources were identified as very important information sources for this group.

Adults over the age of 55 comprise an important market segment in terms of size, growth rates and economic power (Facts and Figures on Older Americans, Gelb 1978; and The Economist, March 5, 1977). Furthermore, some researchers have speculated that people tent to become heavier consumers of services as they grow older (Gelb 1978). A number of studies have examined the maturity segment to gain an understanding of older consumers' behavior (Bartos 1980; Bernhardt and Kinnear 1976, Martin Jr. 1976; Stephens 1980; and Towle and Martin Jr. 1976). However, very few studies have been reported on the information search activities of older adults (Klippel and Sweeney 1974; Schiffman 1971).

Information search activity (of no particular age group) has been the focus of numerous research studies in the past. A review of the literature has failed to reveal an investigation of the search process as it relates to services. However, there is a growing body of literature which supports the contention that there exist differences between the characteristics of services and tangible products. (See, for example, Donnelly 1976; Rathmell 1974; and Shostack 1977, 1978.)

In light of these differences it seems reasonable to question whether the amount and type of information search undertaken by a consumer, particularly an older consumer, will be the same for services as for tangible products. With a product an individual can go to a retail outlet and examine it prior to purchase. Such behavior is usually not possible when purchasing a service. Furthermore, due to the nature of services and their lack of specifications, one might propose that there exists more hidden variance with services than with tangible items. The purpose of the present research is to address this issue by determining the types of sources employed by older individuals in the selection of providers of services. In addition, interest centers around identifying the determinants of how many different source types will be consulted. Only by investigating information search for a variety of "products," tangible and intangible, can a complete understanding of information search and its role in the consumer decision-making process be achieved. In contributing toward this end, the present study focuses on an important growth segment, individuals 55 years of age and older. This sample population was selected due to its heavy consumption of services (Gelb 1978), and to provide a basis of comparison with previously reported information search for products by older adults (Klippel and Sweeney 1974). The goal was to gain an initial view of information search for services. If search for services is identical to that of products then further exploration may be unwarranted.

HYPOTHESES

Most research concerned with external search has identified types of sources used by individuals, regardless of the main thrust of the study. (See, for example, Claxton, Fry and Portis 1974; Hirschman and Mills 1980; Katona and Mueller 1955; Klippel and Sweeney 1974; Newman and Staelin 1973; Schiffman 1971; and Westbrook and Fornell 1979.) In addition to determining the type and number of sources used, some research has sought to determine the importance of various information sources (Hempel 1969; Hirschman and Mills 1980; Katona and Mueller 1955; and Klippel and Sweeney 1974). Past research seems to indicate that many people tend to use few information source types, with personal sources being perceived as the most important. If information search is the same for products and services then support for the following hypotheses would be expected:

H1: More individuals will use one or two source types of information than use three or more source types of information.

H2: Personal sources of information will be perceived as more important than other sources of information.

A number of variables have been identified in past research as having an impact on external search (i.e., Bettman 1979; Bucklin 1966; Bennett and Mandell 1969; Cunningham 1966; Katona and Mueller 1955; Lanazetta and Kanareff 1962; Moore and Lehmann 1980; Newman 1977; and Newman and Staelin 1972). Given the nature of services ar.d the present study, four categories of determinants of search identified in previous studies were investigated. They included: market environment (number of alternatives considered); situational variables including time pressure and special considerations; potential payoff/service importance as determined by perceived risk and number of attributes considered; and individual differences such as age, education, confidence in ability to select service provider and enjoyment of shopping. The following hypotheses were formulated and tested to identify the major determinants of information search for services:

H3: Level of external search is positively related to size of evoked set.

H4a: Time pressure is negatively related to amount of external search.

H4b: Financial pressure is positively related to amount of external search

H4c: Specific service requirements are positively related to amount of external search.

H5: Perceived risk is positively related to amount of external search.

H6: Education level is positively related to amount of external search

H7: Enjoyment of shopping is positively related to amount of external search.

H8: Confidence in ability to select service provider is negatively related to amount of external search.

METHOD

Subjects

Subjects were drawn from a convenience sample of 120 newcomers, 55 years of age and older, who lived in three different neighborhoods of a major southwestern metropolitan area. Ages of respondents ranged from 55 years to 81 years, with a median age of 62 years. The major requirement for the sample was that subjects had to have lived in the area for less than two years. The residency requirement was employed because it was believed that such a sample would be most likely to engage in external information search. Since the subjects were newcomers to the area, they lacked past experience with local service providers which could have served as an information base for internal information search. Lacking past information on specific local service providers, the subjects would have to rely on external sources of information.

Services

Four main criteria were employed in the selection of services to be investigated. First, the services studied were ones that most individuals in the sample population would utilize. Second, the services considered were ones that most subjects would require shortly after moving to the area. These first two requirements were used to insure the compatibility of services tested with the sample as users. Third, the services selected were diverse and different from one another. Factors considered included the cost of the service, frequency of use, and the ease of switching among service providers. Finally, the services were ones that did not engage in extensive franchising or distribution. The final criterion was adopted to further insure that internal search would not be undertaken by the subjects. Three services meeting each of the above criteria were selected: financial institutions (banks or saving and loan associations), barber/beauty shops and physicians.

Procedure

The measuring instrument was administered through the use of personal interviews. Specifically, four interviewers (two female, two male) conducted in-home interviews which lasted approximately 20 minutes. This procedure resulted in 120 completed questionnaires.

Measuring Instrument

A number of different measures have been used to determine external search. Most have involved survey-based recall measures or some type of search index. While realizing weaknesses in self-report recall data, it was the researchers' belief that self-report measures were most appropriate for the investigation of information search for service providers. Specific considerations centered around the lack of observable "store shopping" by the consumer for the purposes of gathering information prior to the selection of a service provider. Therefore, both aided and unaided recall measures of information sources used were included. Measures on the perceived importance of each source used were collected utilizing a seven point Likert-type scale with the end points labeled "very unimportant" and "very important." Perceived risk of each service was measured utilizing Brody and Cunningham's measures of social and performance risk (Brody and Cunningham 1968). A seven point Likert-type scale with end points of "not-at-all able' and "very able" was used to measure the subject's perceived ability to select the particular service provider. In addition, measures on the number of alternatives considered, shopping behavior and various demographic factors were collected.

RESULTS

Hypothesis Testing

Of interest was the amount of external information search for each service and its relationship to a number of predictor variables. Amount or level of information search was determined by summing the number of different information sources a subject used. To determine whether or not an information source was used by a subject, unaided and aided recall data were collected and combined. Table 1 presents a summary of the type and number of information sources used for each service. While the original sample size was 120, not all subjects had been involved in the selection of all three services. Responses from individuals not participating in the selection of the particular service provider were not included in that analysis. As a result, sample size varied for the three services (financial institution," = 91; barber/beauty shop," = 84; and doctors," = 71).

It should be noted that a number of individuals identified a source of information other than those listed in Table 1. A review of these other sources indicated that the majority of those cited centered around the subject seeing the service firm while out walking or driving. The next most frequently mentioned other source was a guide or pamphlet provided by a third party other than the service provider (i.e., Welcome Wagon, Chamber of Commerce).

TABLE 1

NUMBER OF RESPONDENTS USING EACH TYPE OF INFORMATION SOURCE BY NUMBER OF SOURCES USED (BASED ON COMBINED UNAIDED AND AIDED RECALL)

Hypothesis 1: Number of Sources Used. More subjects used only one source type than any other (see Table 1). The maximum number of source types used in the selection of a barber/beauty shop or a doctor was three, while up to five different source types were employed for a financial institution. Due to the nature of the data, chi-square analyses were conducted to compare the level of search for each service. Support for H1 was exhibited by all three services at the .001 significance level (financial institution, x2 = 43.6; barber/beauty shop, x2 = 65.2; doctor, x2 = 49.6).

Hypothesis 2: Importance of Personal Sources. Table 2 presents the mean importance ratings for each source type used. Mean difference tests (one-tail) were conducted to determine which ratings were significantly different from personal sources. For financial institution personal sources were significantly more important than print (t = 3.138, df = 67, p = .0025) and broadcast advertisements (t = 2.056, df = 78, p = .025). However, contacts to the financial institution were significantly more important than personal sources (t = -2.192, df = 75, p = .025). There were no significant differences between the importance rating of Yellow Pages and other sources and personal sources. For barber/beauty shop, a significant difference was observed only between personal sources and print advertisements (t = 2.932, df = 48, p = .005). The mean difference between the importance of Yellow Pages and personal sources approaches significance (t = 1.633, df = 48, p = .10). Personal sources were perceived as significantly more important than the Yellow Pages or contacts by doctors to prospective patients. There was no significant difference between the importance ratings for personal sources and the other source types used in the search for a doctor. However, it should be noted that for the selection of a doctor, print advertisements and contacts to doctors were rated higher on importance than personal sources, although these differences were not significant. Only partial support was exhibited for R2, with more support being observed for financial institution than for barber/ beauty shop or doctor.

TABLE 2

COMPARISON OF MEAN IMPORTANCE RATING OF SOURCES

Hypothesis 3: Size of Evoked Set. Pearson's Product Moment Correlation was used to examine the relationship between information search and size of evoked set, perceived risk, education and ability to select service provider. A positive relationship between size of evoked set (number of service providers considered) and the amount of information search was observed for financial institutions (r = .469, p = .000) and barber/beauty shops (r = .520, p = .000). However, only a weak relationship between size of evoked set and amount of search was observed for doctors (- = .191, p = .056). Therefore, only partial support for H3 was observed.

Hypotheses 4a, 4b, 4c: Situational Variables. Subjects were asked to identify any special considerations for either financial institution or barber/beauty shop. Because of the low response rate, 8 percent for financial institution and only 6 percent for barber/beauty shop, no conclusions could be drawn concerning the effects of special considerations such as time or financial pressure, or specific requirements, on the amount of information search for these services.

In contrast with financial institutions and barber/beauty shops, 32 percent of the subjects involved in search for a doctor indicated that there were special considerations involved. The types of factors/considerations mentioned fit into one of four categories: doctor's ability; doctor's personality; doctor's specialty (able to treat a specific problem); and doctor's location (close/convenient). The specific identification of such factors by subjects in response to an open-ended question seems to indicate that situational variables have a greater impact on the search/ selection process for a doctor than for the other two services investigated.

Hypothesis 5: Perceived Risk. No relationship between perceived social risk and amount of information search was observed (financial institution r = .022, p = .844; barber/ beauty shop r = .032, p = .777; doctors r = .014, p = .457). A very weak relationship between perceived performance risk and amount of search was exhibited for barber/beauty shop (r = .210, p = .09), while no relationship existed for financial institution (r = .137, p = .303), and doctor (r = .099, p = .270). In view of these results little support exists for H5.

Hypothesis 6: Education. A positive relationship was observed between education level and search for financial institutions (r = .316, p = .002). However, no significant relationship was observed for barber/beauty shops (r = .047, p = .658) or doctors (r = .067, p = .291). Therefore, only partial support was present for H6.

Hypothesis 7: Shopping Habits. The extent to which an individual usually shops around before making a purchase was found to be independent of the amount of information search engaged in prior to the selection of a bank or barber/beauty shop (financial institution, x2 = 1.15, df = 2, p = .562; barber/beauty shop, x2 = .543, df = 3, p = .909). However, the relationship approached significance for the selection of a physician (x2 = .0781, df = 1, p = .078). In general, support for H7 was not present.

Hypothesis 8: Confidence in Own Ability. No relationship between confidence in one's ability to select the service provider and amount of external search was observed (financial institution, r = .095, p = .384; barber/beauty shop, r = .003, p = .978; doctor, r = .030, p = .407).

Additional Analyses

Of special interest was identifying the factors which determine the amount of information search in which an individual engages prior to selecting a financial institution, barber/beauty shop, or doctor. To gain further insight, multiple stepwise regression analysis was performed for each service using amount of information search as the criterion variable. A summary of the standardized beta weights and associated F values for each of the predictor variables for the final regression models is presented in Table 3.

TABLE 3

PREDIOTOR VARIABLES FOR AMOUNT OF SEARCH

Dummy variables were used for the dichotomous variables, sex and marital status. For financial institutions all predictor variables were significant at p < .05, with the model possessing an adjusted R2 = .597. For barber/beauty shops the predictor variables were all significant at p < .10, and an adjusted R = .348. The regression model for doctors had an adjusted R2 = .364, with the predictor variables significant at the p < .07 level.

Other independent variables were examined through regression analyses but were found to be insignificant as predictor variables. The additional variables examined included: perceived social and performance risk, ability to judge service provider, confidence in various information sources, and income.

DISCUSSION

Hypotheses

Significant support was found for the first hypothesis, indicating that older people do tend to use only one or two information sources in their selection of services. Results also showed that subjects engaged in more search for a financial institution than for a barber/beauty shop or doctor. This search behavior could be a function of type of commitment involved in purchasing the service. For example, a financial institution probably involves more of a long-term commitment on the part of the consumer than a barber/beauty shop, or, in some cases, a doctor. In addition, it might be said that the number and types of services and their associated costs tend to vary more widely among the different types of financial institutions than might be the case for barber/beauty shops or doctors. In any case, consumers' knowledge of the wide variance might be greater in the case of financial services than in the case of barber/beauty shops and doctors, due to the heavy advertising such services have received in recent years. The evidence seems to indicate that level of search varies from service to service, as is true for different products.

Further insight into the importance of information sources can be gained from the results if one dichotomizes the sources used into marketer-dominated and nonmarketer-dominated sources. For financial institutions, personal sources were significantly more important than the marketer-dominated sources of print and broadcast advertising and contacts made by the financial institution to the potential customer. The only marketer-dominated source for financial institutions which approached personal sources in importance was the Yellow Pages. Some importance may be placed on the Yellow Pages by the user because she/he has had to actively seek out the information. Support for such a view may be found in the fact that contacts made to the financial institution by the consumer were more important than all other sources, personal sources and marketer-dominated. This last result suggests that if one takes the time and trouble to seek out information through direct contact with a company, it will be considered a more important source than any other.

Examining the marketer-dominated sources used by barber/ beauty shop consumers provides further indication of the importance of personal sources. Personal sources were significantly more important than print advertising, while the difference between the importance of personal sources and the Yellow Pages approached significance (with personal sources being more important).

Personal sources were significantly more important than Yellow Pages or contacts by doctors. Personal sources may have been significantly more important than the Yellow Pages for doctors because the types of information they provide is so different. The Yellow Pages gives only impersonal, non-evaluative data, while personal sources might be expected to give more evaluative data. There was no significant difference between importance ratings for personal sources and print advertisements, the other marketer-dominated source.

In general, it appears that for the selection of a service provider, personal sources are more important than marketer-dominated sources. Furthermore, contacts to providers were perceived as most important for each of the services investigated.

These results were consistent with those reported by Klippel and Sweeney (1974), who investigated the information sources used by older consumers (55 years and older) in learning about their present headache remedy and television set. For both products, friends, neighbors and members of the immediate family were the information sources which received the highest mean importance rating. Both of these source groups would be classified as personal sources of information.

As was shown to be true in past research with durable and nondurable goods, the market environment as represented by evoked set does have an impact on search. The more alternatives considered, the greater the level of information search.

No relationship was observed between perceived risk and search, with the exception of a weak, positive relationship with barber/beauty shop's performance risk. The lack of relationship is probably more a function of the instrument used to measure risk than the fact that no relationship exists. Many of the subjects expressed confusion over, or inability to respond to, the risk questions. Therefore, the lack of support for Hypothesis 5 should be viewed as an artifact of measurement as opposed to a lack of concern over potential payoff or the cost-benefit tradeoff associated with acquiring information.

Education was positively related to search for financial institutions but not for barber/beauty shops or doctors. The relationship between education and search for financial institutions could be reflective of more complex financial service needs. Since education and income are often positively related, subjects with higher education may also have higher income. Higher income individuals often require more varied services than the basic checking and saving accounts usually associated with financial institutions. As a result of more complex needs, they may devote more time and attention to the search for a financial institution.

Services cannot be shopped for as one might shop for a car or appliance. As a result, there may be no real reason to believe that this factor will impact search for services. Just as services and products differ, so may their predictors of search behavior.

Only for doctors was total independence between information search and enjoyment of shopping not observed. In other words, subjects who enjoyed shopping also tended to use more sources of information prior to selecting their doctor. This result couLd be a function of the special factors/considerations identified in connection with the search for a doctor.

The final individual characteristic tested was confidence in one's ability to select service provider. Subjects rated themselves high on ability to select the desired service provider (financial institution X = 5.73; barber/beauty shop X = 5.96; doctor X = 5.69).

Additional Analyses

Further insight into the impact of individual characteristics on information search for services can be gained by examining the regression analyses. All independent variables in the models, except one, were individual characteristics. Sex and education level were significant factors in the models for two services. This evidence provides strong support for the fact that individual characteristics do impact search. Even more evidence can be found in the fact that size of household and age were significant for financial institutions. Since this study examined a homogeneous age group and age still proved to be a significant factor, further research on the search patterns for services of younger Persons would seem to be justified.

The final independent variable for financial institution was confidence in Yellow Pages. This was an overall confidence measure in the source and it is unclear why it was significant. One reason may be that it is perceived as nonmarketer dominated and such sources were shown to be important in the analyses of earlier results.

One interesting note is the fact that shopping habits was the final variable in the model for barber/beauty shop. The information search for barber/beauty shop may be more closely related to shopping than other services, especially in light of the high proportion of subjects who identified barber/beauty shops as a result of driving or walking around. Such shopping for a barber/beauty shop could be a function of location, since these shops are often located in shopping malls or Plazas.

The factors which impact information search for doctors were more unique than those variables identified for financial institutions and barber/beauty shops. Size of evoked set was the only variable common to all three models. In each case it explained the most variance. Except for shopping habits, the other predictor variables were unique to the doctor model. Special considerations (such as doctor's ability, personality and location), marital status and the number of special considerations which were mentioned, constituted the remaining predictor variables. Because of the apparent relationship between two of the variables, special considerations and the number of special considerations, the model was checked to insure that multi-collinearity did not occur. There was no significant correlation between these two variables, or any of the other variables in the model. (Each of the other models was also checked to insure that multi-collinearity was no; present.)

CONCLUSIONS

Past information search studies have concentrated on examining external search for tangible goods, with no consideration of the differences between goods and services. However, these differences lead one to question whether information search for services is the same as for tangible products. Past studies have also examined consumers of all ages. Since older consumers are growing rapidly as a group and since they are heavy services consumers, it is useful to look at their particular search habits. In an effort to address these issues, the present study examined external search for three services, financial institutions, barber/beauty shops, and physicians. Given the constraints of the services investigated, it appears that older consumers' search behavior for services is strongly influenced by evoked set and individual characteristics such as sex and education. Furthermore, older individuals utilize fewer sources with greater importance being placed on nonmarketer-dominated sources.

It should be stressed that the present study provides only a starting point in the investigation of information search for services. Additional research needs to be conducted to determine if the patterns observed in the present study are consistent with those for other services and sample populations.

REFERENCES

Barak, Benny and Schiffman, Leon G. (1980), "Cognitive Age: A Nonchronological Age Variable," Advances in Consumer Research, Vol. 8, ed. Kent B. Monroe, Ann Arbor, MI: Association for Consumer Research. 602-606.

Bartos, Rena (1980), "Over 49: The Invisible Consumer Market," Harvard Business Review, 58, 140-148.

Bennett, Peter D. and Mandell, Robert M. (1969), "Prepurchase Information Seeking Behavior of New Car Purchases The Learning Hypothesis," Journal of Marketing Research, 6, 430-433.

Bernhardt, Kenneth L. and Kinnear, Thomas C. (1976), "Profiling the Senior Citizen Market," Advances in Consumer Research, Vol. 3, ed. William D. Perrault, Atlanta: Association for Consumer Research, 449-452.

Bettman, James R. (1974), An Information Processing Theory of Consumer Choice, Reading, MA: Addison-Wesley Publishing Company.

Brody, R. and Cunningham, Scott M. (1968), "Personality Variables and the Consumer Decision Process," Journal of Marketing Research, 5, 50-57.

Bucklin, Louis P. (1966), "Testing Propensities to Shop," Journal of Marketing, 30, 22-27.

Claxton, John D., Fry, Joseph N., and Portis, Bernard (1974 "A Taxonomy of Pre-purchase Information Gathering Patterns, Journal of Consumer Research, 1, 35-42.

Cunningham, Scott M. (1966), "Perceived Risk as a Factor in the Diffusion of New Product Information," Science Technology and Marketing, ed. Raymond M. Haas, Chicago: American Marketing Association, 698-721.

Dommermuth, William P. (1965), "The Shopping Matrix and Marketing Strategy," Journal of Marketing Research, 2, 128-132.

Donnelly, James R. (1976), "Marketing Intermediaries in Channels of Distribution for Services," Journal of Marketing, 40, 55-57.

Facts & Figures on Older Americans (n.d.), "State Trends, 1950-1970," Washington, D.C.: Government Printing Office, 1.

Gelb, Betsy D. (1978), "Exploring the Gray Market Segment," MSU Business Topics, (Spring), 41-46.

Hempel, Donald J. (1969), "Search Behavior and Information Utilization in the Home Buying Process," Marketing Involvement in Society and the Economy, ed. Philip R. McDonald, Chicago: American Marketing Association, 241-249.

Hirschman, Elizabeth C. and Mills, Michael R. (1980), "Sources Shoppers Use to Pick Stores," Journal of Advertising Research, 20, 47-51.

Holbrook, Morris B. and Maier, Karl A. (1978), "A Study of the Interface Between Attitude Structure and Information Acquisition Using a Questionnaire Based Information-Display," Advances in Consumer Research, Vol. 5, ed. H. Keith Hunt, Chicago: Association for Consumer Research, 93-99.

Jacoby, Jacob, Chestnut, Robert W., and Fisher, William A. (1978), "A Behavioral Process Approach to Information Acquisition in Nondurable Purchasing," Journal of Marketing Research, 15, 532-544.

Katona, George and Mueller, Eve (1955), "A Study of Purchase Decisions," Consumer Behavior: The Dynamics of Consumer Research, Vol. 1, ed. Lincoln Clark, New York: New York University Press, 30-87.

Klippel, R. Eugene and Sweeney, Timothy W. (1974), "The Use of Information Sources by the Aged Consumer," The Gerontologist, (April), 163-166.

Lanzetta, John T. and Lanareff, Vera T. (1962), "Information Cost, Amount of Payoff, and Level of Aspiration as Determinants of Information Seeking in Decision Making," Behavioral Science, 7, 459-473.

Martin, Claude R., Jr. (1976), "A Transgenerational Comparison--The Elderly Fashion Consumer," Advances in Consumer Research, Vol. 3, ed. William D. Perrault, Atlanta: Association for Consumer Research, 453-456.

Moore, William L. and Lehmann, Donald R. (1980), "Individual Differences in Search Behavior for a Nondurable," Journal of Consumer Research, 7, 296-307.

Newman, Joseph W. (1977), "Consumer External Search: Amount and Determinants," Consumer and Industrial Buyer Behavior, eds. Arch G. Woodside, Jagdish N. Sheth, and Peter D. Bennett New York: North Holland Publishing Company, 79-94.

Newman, Joseph W. and Staelin, Richard (1973), "Information Sources of Durable Goods," Journal of Advertising Research, 13, 19-29.

Newman, Joseph W and Staelin, Richard (1972), "Prepurchase Information Seeking for New Cars and Major Household Appliances," Journal of Marketing Research, 9, 249-257.

Rathmell, John M. (1974), Marketing in the Service Sector, Cambridge, MA: Winthrop Publishers.

Schiffman, Leon G. (1971), "Sources of Information for the Elderly," Journal of Advertising Research, (October), 33-37.

Shostack, G. Lynn (1977), "Breaking Free from Product Marketing," Journal of Marketing, 40, 73-80.

Shostack, G. Lynn (1978), "The Service Marketing Frontier," Review of Marketing, eds. Gerald Zaltman and Thomas Bonoma, Chicago: American Marketing Association, 373-388.

Stephens, Nancy (1980), "Media Use and Media Attitudes Changes with Time and with Age," Journal of Advertising, 10, 38-47.

The Economist (1977), "Silver Threads Among the Gold," 262, 57.

Towle, Jeffery C. and Martin, Claude R., Jr. (1976), "The Elderly Consumer: One Segment or Many," Advances in Consumer Research, Vol. 3, ed. William D. Perrault, Atlanta: Association for Consumer Research, 463-468.

Westbrook, Robert A. and Fornell, Claes (1979), "Patterns of Information Source Usage Among Durable Goods Buyers," Journal of Marketing Research, 16, 303-312.

----------------------------------------