On Integrating Consumer Needs For Variety With Retailer Assortment Decisions

ABSTRACT - Interest in consumer variety seeking behavior has recently gained momentum. A consumer's variety drive must be satisfied by the marketer's assortments of products and services. This paper calls attention to an important problem; the necessity for retailer strategies which satisfy consumer needs for variety. More specifically, it investigates four issues: (1) factors influencing consumer's need for variety and their impact on the retailer's assortment strategies; (2) the utility consumers derive from the assortment; (3) the dimensions and factors which define the framework within which retailers must develop their assortment strategies; (4) the likely existence of an "assortment range" which can respectively satisfy the needs and constraints of both consumers and retailers.

Citation:

Moshe Handelsman and J. Michael Munson (1985) ,"On Integrating Consumer Needs For Variety With Retailer Assortment Decisions", in NA - Advances in Consumer Research Volume 12, eds. Elizabeth C. Hirschman and Moris B. Holbrook, Provo, UT : Association for Consumer Research, Pages: 108-112.

Advances in Consumer Research Volume 12, 1985      Pages 108-112

ON INTEGRATING CONSUMER NEEDS FOR VARIETY WITH RETAILER ASSORTMENT DECISIONS

Moshe Handelsman, The University of Santa Clara

J. Michael Munson, The University of Santa Clara

ABSTRACT -

Interest in consumer variety seeking behavior has recently gained momentum. A consumer's variety drive must be satisfied by the marketer's assortments of products and services. This paper calls attention to an important problem; the necessity for retailer strategies which satisfy consumer needs for variety. More specifically, it investigates four issues: (1) factors influencing consumer's need for variety and their impact on the retailer's assortment strategies; (2) the utility consumers derive from the assortment; (3) the dimensions and factors which define the framework within which retailers must develop their assortment strategies; (4) the likely existence of an "assortment range" which can respectively satisfy the needs and constraints of both consumers and retailers.

INTRODUCTION

Research into consumers' need for variety has gained considerable momentum in the last few years,(e.g., McAlister and Pessemier 1982; Hoyer and Ridgway 1983). Consumers' need for variety is satisfied by the assortment" of products and services that are provided by manufacturers, wholesalers, and retailers. However, the "assortment provider's" perspective has been ignored by researchers; moreover, no attempt has been made to combine the consumers' and providers' assortment perspectives. Hence, major thrusts of this paper are twofold: (1) to increase the awareness among marketers of the necessity to match consumer variety needs with retailer assortment strategies, and (2) to investigate how each of these two groups perceive the assortment problem and its implications for retail assortment strategy.

A recent survey of retail executives (Achabal, McIntyre, and Kriewall (1983)reported that the current economy was forcing management and employees to be more critical of their organizational productivity. Executives felt that the potential for future "front-end" labor (sales staff) reductions is extremely limited. Future productivity gains must come from "behind-the-scenes improvement," especially in the area of improved assortment. However, this practical use of assortment to achieve "managerial goals" (ex., productivity, profitability) has not been complemented by the development of an adequate theoretical framework to explain either the consumers' and/or retailers' conception of assortment size. This is one reason why retail management textbooks tend to treat merchandising as an "art" as opposed to a science (Mason and Mayer (1981). There is a great need for research that will explain and quantify consumers' and retailers' behavior concerning assortment size.

This paper will define "assortment size" as having two major dimensions: (1) Breadth: the number of different product categories offered by the retailerS (2) Depth: the array of product choices within the individual categories. The paper attempts to investigate four aspects relevant to assortment strategy: (1) the consumers' variety seeking behavior, its causes and their implications for retail assortment strategies; (2) the possible existence and shape of the consumer's assortment utility function; (3) the dimensions and factors which influence and might serve to define the framework within which retailer's must make assortment decisions; (4) the possible existence of an "optimal assortment, range" as seen from dual perspectives - the consumers and the retailers' and its implications for retail strategy.

The following section discusses aspects of "varied behavior" among consumers. Consumers are divided into two purchase behavior segments: one whose purchase behavior can be anticipated by retailers, and a second whose purchase behavior is less predictable the variety seeking segment.

CONSUMERS' PERSPECTIVE OF VARIED BEHAVIOR

Segments Whose Purchase Behavior Can Be Anticipated

Consumer populations can be viewed from the retailers' perspective as being composed of several different purchase behavior segments. For example, the brand loyal, bargain hunter, and the variety-seeker segments differ from each other with respect to their purchase motivation and therefore with regard to the applicable assortment strategy. Retailers may establish a different assortment strategy for each segment to which they cater. The amount of difficulty involved in setting a strategy and the probability of success in implementing it depend on both the ability to predict each segment's behavior and to understand its motives.

Consumer segments whose purchase behavior is relatively easy to anticipate include the bargain hunters, the innovators, and brand loyal segments. "Bargain hunters" purchase a brand from a product line assortment based on which brand is on sale. These consumers are not primarily motivated by the need for variety. Retailers' assortment strategies will accommodate this segment whenever at least one brand in the product class is on sale.

Another segment includes the novelty (innovation) seekers. Some consumers seek new brands which they have never tried before. To satisfy their need a retailer has to include the latest brand in his assortment (e.g., high fashion retailing).

Many consumers are brand loyal to one brand over long periods of time. This segment's size amounts to 40%-70% for appliances (Newman 1977), and 30%-39% for product classes with shorter repurchase periods (Handelsman 1982). These consumers-have no need for the total assortment offered by a retailer. They patronize a retail outlet only if their favorite brand is there. How to satisfy the brand-loyal segment is obvious; however, the selection of which specific brands to include in the assortment is not.

The Variety Seeking Segment

Generating a retail marketing strategy to satisfy the variety seeking segment is more complex than for the aforementioned segments. This is due to the complexity of variety seeking behavior and the difficulty of anticipating it. Studies which have been done suggest that variety seeking behavior is inconsistent across product classes; i.e., it is a product specific consumer characteristic. Attempts to explain and predict variety seeking are exacerbated by the fact that it is difficult to assign an index of "variety seeking" to consumers.

The reasons for varied behavior are numerous. However, most variety seeking behavior can be categorized into two types: (1) behavior "derived" as an outcome of external or internal motives, independent of pursuing variety for its own sake; or (2) "direct" behavior as an end in itself, where consumers pursue variety because it is inherently satisfying in and of itself. Each of these major categories are discussed below.

Derived Causes of Varied-Behavior

Satiation, multiple needs, and multiple conditions are included here. One may categorize these causes of varied behavior into a finer classification, where satiation's elements are defined as endogeneous causes, while multiple needs and conditions are defined as exogeneous forces. Multiple needs include such factors as multiple users, contexts, and uses, while multiple conditions include changes in the feasible set of alternatives, changes in tastes, and changes in constraints (McAlister and Pessemier 1982. Laurent 1978).

It may be helpful to divide satiation into two categories: First, satiation at a given point in time in conjunction with attribute balancing (Farquhar and Rao 1976, McAlister 1979). Second, as a result of continuous consumption of objects which leads to boredom (Howard and Sheth 1969, Brickman and D'Amato 1975).

Direct Causes of Varied Behavior

Intrapersonal Causes. A lumber of direct causes of varied behavior can be named: curiosity, novelty, information seeking, balancing, etc. These have all been viewed as desirable goals of the individual because they are "inherently satisfying" (Maddi 1968). Psychologists have been investigating these phenomena since the beginning of this century (Dashiell 1925, McDougall 1908).

More recently the concept of drive reduction (people being motivated toward minimizing stimulation) has begun to be replaced by a newer concept--the Optimal Level of Stimulation (OLS) as an explanation of variety seeking behavior. Noteworthy OLS researchers include several (e.g., Berlyne 1960, Driver and Streufert 1964). Venkatesan (1973), McAlister and Pessemier (1982), and Raju (1981) have summarized and compared the OLS approaches. Basically these approaches suggest that as stimulation (novelty, complexity, etc.) increases past the individual's subjective "ideal level," he will try to simplify the input. When the stimulation falls below the subjective ideal level, the individual experiences boredom and tries to seek greater stimulation.

The heterogeneity of consumers with respect to OLS has direct bearing on the assortment provider's decisions regarding assortment size. Researchers have suggested that maintaining a certain level of stimulus change, novelty and complexity is essential for effective human functioning. They also assumed that individuals differ in their optimal stimulation level and that OLS is related to personality characteristics. However, the specific retail strategy which might be deduced from knowledge about any consumer's OLS is much less clear than are strategies for addressing the purchase behavior of segments such as the bargain hunters, brand loyal, or novelty seekers. One might argue for example, that the same assortment of X items will be perceived differently by consumers whose OLS's differ. A consumer with a relatively low OLS will perceive that assortment as overwhelming; while another with a higher OLS will evaluate it as just right, and a third with a relatively high OLS will perceive it as unsatisfactory.

Another factor that complicates the situation is that each consumer's optimal stimulation Level may change with time and conditioning level (i.e., exposures to the stimulation). At a specific point in time, several people with the same OLS will experience different levels of stimulation from the assortment depending on their prior exposure to stimuli. Thus, even if researchers may be able to develop an accurate index of OLS for consumers by employing some measure of OLS (Raju 1980), the problem remains of determining "how much" stimulation has already occurred prior to the consumer's entering a store.

Novelty Seeking as a Cause for Variety Seeking Behavior

Variety-seeking is an innate drive that may be satisfied by an idiosyncratic level of stimulation (Handelsman 1982). To reach this optimal stimulation level, a consumer has to explore (i.e., gain continuous contract with) novel items. A particular item loses its novelty as the consumer's experience with it increases, so that the consumer has to switch from one item to another, and by this he expresses "varied buying behavior." To maximize variety, a consumer has to simultaneously switch among brands perceived to be maximally dissimilar, and in each switch to choose that brand with which he has minimal experience.

Information Seeking as a Cause of Variety-seeking Behavior

Howard and Sheth (1969) suggest that active seeking of information occurs when ambiguity of brands is felt by a consumer, i.e., the consumer is not sure of the benefits offered by each brand and is not capable of discriminating among them. This seems to be consistent with Hirschman's (1980) hypothesis that motives for varying choice among "known" stimuli do not concern information needs. One way to collect information is through trial of different brands, which results in varied buying behavior.

Interpersonal Causes of Varied Behavior

Croup affiliation and personal identification are two interpersonal motives of varied purchase behavior. Regarding group affiliation, some of an individual's variety seeking behavior may arise from desires to conform to salient group norms. Affiliation needs can often lead to imitative purchase behavior. Much of research on social class and fashion behavior supports this observation (McAlister and Pessemier 1982). Regarding the need for personal identification, much variety seeking behavior may derive from the individual's attempts to communicate "unique" aspects of self through using a specific product (e.g., Landon, 1974; Munson & Spivey 1981).

In sum, retail marketing strategies for satisfying variety-seekers' assortment needs are much more difficult to formulate, and are much less clear-cue than are the strategies for dealing with the bargain hunters, brand loyal, and innovator segments.

CONSUMER'S UTILITY FUNCTION FOR ASSORTMENT LEVELS

One can implicitly assume that a consumer's "need for variety" is often satisfied by the various assortments of products and services offered in the marketplace. However, this assumption muse be validated by answering a key question: Do consumers consider "assortment size" as a salient store attribute in patronage decisions? If not, there is no point investigating the shape of the consumer's utility function derived from various assortment sizes.

This section addresses the following three questions: Q1. How important is assortment size as a determinant of store patronage behavior? Q2. Is there a relation between assortment size and the consumer's perceived utility derived from the assortment? Q3. Is there a "range" of assortment sizes which will optimize a consumer's utility?

Regarding Q1, several studies indicate that assortment size does influence retail patronage behavior (ex., Cardozo 1975; Arnold, Ma and Tigert 1978). Assortment size (together with quality) was found by Claxton and Ritchie (1979) to be by far the single most important factor influencing choice of an outlet for clothing and footwear retailers. They note that providing a wide choice of easily accessible items appeared to be the "fundamental approach to satisfying the needs of (these) consumers." In a review of 26 studies of retail store patronage, Lindquist (1975) found that merchandise selection or assortment were the most frequently mentioned store attributes (mentioned by 42: of the studies). For comparison, merchandise quality and pricing were mentioned in 38%.

Regarding Q2 above, at first glance one might expect that the utility of assortment to the consumer should be a monotonic increasing function; i.e., the more the better. Baumol and Ide (1956) formalized this point of view through their "probability of success" notion: shoppers to not know in advance whether they will find what they want when entering a store; however, they will assume that the greater the number of items carried by the store, the greater their expectation that the shopping trip will be successful. However, recent research has revealed that shoppers often are confused when confronted with too many brands from which to choose (Wall Street Journal. 19825.

These two contrasting effects, the need for wide assortment and the confusion aroused by too much variety, were combined in an effort to model consumer utility from store assortment (Baumol and Ide 1956). In their model the consumer decides which store to patronize by weighing the probability of a successful shopping trip to a store--which is a function of: distance to the store, time, effort, and opportunity costs--against the "difficulty" of shopping in the store itself. "Difficulty is a function of the assortment size and is expressed as the square root of the assortment size (..."the more items stocked the further we must walk to get to the spot where some items are kept"). Results from the model show that increasing a given store's assortment will increase consumer utility only up to a point, and then utility will decline; the store becomes difficult to shop mainly because of confusion and fatigue. Thus, the consumer's utility function for assortment has a bell shape.

This notion of a bell-shaped utility curve is also supported by Brown (1978) who found that "The pattern of patronage is approximately bell-shaped over different store size intervals...(it) is certainly not monotonically increasing as expected... 'Average' sized stores received more patronage than either larger or smaller stores.

Given that the consumer's utility function for assortment size is apparently adequately modeled by a bell-shaped curve, then the answer to Q3 above regarding the existence of an "optimal range of assortment size" is apparent. That is, consumers should favor assortment sizes which fall within close proximity to the peak of their "individual" assortment utility curves (a situation similar to consumer's OLS, as discussed earlier).

RETAILER'S PERSPECTIVE OF ASSORTMENT

The retailer's assortment strategy is driven by the consumers' "need for variety." In prior sections we have suggested that much of variety seeking is an innate drive and that consumers strive to satisfy it through varied purchase behavior. In this section we investigate the issue of assortment size from the other side of the coin--the retailer's.

Who provides consumers with assortments--manufacturers, wholesalers, and retailers of course. However, each of these "assortment providers" is likely to experience differing degrees of difficulty regarding their assortment decisions. Figure 1 points out the differences between these three groups in the degree to which total annual sales tend to be concentrated across different product lines. It is clear that sales concentration is higher among manufacturers (both technological and industrial) than among wholesalers and retailers (where sales concentration is the lowest).

FIGURE 1

SALES CONCENTRATION ACROSS ASSORTMENT PROVIDERS

Figure 1 implies that retailers must provide a wider assortment than either manufacturers or wholesalers. Thus, the assortment issue is likely to be more important and more problematic to retailers. It is primarily for this reason that we have assumed a retailers perspective when investigating assortment issues from the provider' 5 standpoint. Although our discussion below generalizes the assortment issue across retailer types, the reader is reminded that various store types can differ in their sales concentration level depending upon the products carried (e.g., hardware vs. clothing).

Assortment decisions, as other managerial decisions, are part of the retailer's strategy which is usually aimed at increasing sales to current and/or new customers. To achieve this objective a retailer must have a comparative advantage over competition. This advantage should be important and visible to shoppers, as well as make the greatest contribution to the retailer's return on investment. The major dimensions on which a "retailer's assortment" might be analyzed appear in disparate form in the marketing literature. Figure 2 represents an attempt to integrate these dimensions into a more cohesive framework together with the factors influencing a retailer's assortment decision making.

The Figure suggests four major dimensions:

1. Completeness of assortment--the number of items and the consumer's perceived diversity and appropriateness of the items carried.

2. Price/quality/value--the perceived level of prices, quality, and the overall value offered in each product class.

3. Fashion level--the perceived level and the appropriateness of the fashion level.

4. Brand emphasis--consumer satisfaction with the extent of private-brand and national-brand representation. (See Pessemier, 1980 for elaboration).

FIGURE 2

DIMENSIONS OF AND FACTORS INFLUENCING RETAILER ASSORTMENT DECISIONS

The retailer's assortment marketing strategies may emphasize one or more of the aforementioned dimensions. It should be noted that these dimensions may be "substitutable". For example, one approach to retail strategy may be to concentrate on customers who display high store loyalty but little brand loyalty.

Factors which can influence the retailer's assortment decisions regarding each dimension are shown outside the circle in Figure 2. A key notion suggested by the framework is that there exists a multiplicity of factors which individually or jointly can impact each dimension. Some highly complex interaction patterns of factors x dimensions can serve to increase the difficulty of assortment decisions for the retailer. Although discussion of the individual factors is beyond the scope of this paper, the interested reader can refer to Kotler (1980), Mason and Mayer (1981), Bishop (1978), and Pessemier (1980) for more detail.

MATCHING CONSUMER AND RETAIL ASSORTMENT NEEDS

Thus far, we have described the retailer's and consumer's perspectives of the major factors which influence assortment size. Each arty is motivated by a different se: of goals and constraints. Additionally, prior research supports the notion that both the consumer's and retailer's utility functions for assortment size approximates a bell-shaped curve. The key question to now consider is whether there is an "assortment size" which will simultaneously satisfy both the consumer's and retailer's needs?

Unfortunately, several factors may preclude an easy matching of these two groups' respective assortment needs; major ones include the following:

1. The optimal assortment size may vary across consumers due to heterogeneity in their needs for variety. Building an assortment to optimize some "average" assortment utility function will mean the retailer satisfies only one segment. All other segments whose optimal range lies below or above the "average" will perceive the offered assortment as inadequate.

2. The possible dynamic nature of each consumer's utility function, both within across product categories. Little research exists regarding how fast or how frequently a person's range or optimal assortment sizes may change. That the range might shift over time suggests a "conditioning hypothesis": consumers can be conditioned by the assortment offered in their environment, and they will adjust their desired level of assortment accordingly. As a case in point, Arnold, Oum and Tigert (1983) found that in the Tampa, Florida area where "supercombo" stores dominate the market on both food and non-food assortments, "...consumers place greater salience on the 'assortment' attribute and less on price in store patronage decisions comPared to other areas where super-combos do not exist.

3. The retailer's cost structure (i.e., overhead, required profit objectives, etc.) may necessitate that he carry an assortment which is either larger or smaller than that desired by his primary target market.

An optimal assortment size that will satisfy both consumers T needs and retailers' constraints is not easily attainable for the aforementioned reasons. However, in view of the large potential payoffs to both retailers and consumers from a more optimal matching of their respective needs, we encourage future research on this aspect.

CONCLUSION

This paper has reviewed the major forms of "varied" consumer behavior (i.e., direct, derived) and indicates that each can require a different type of retail assortment strategy. It also presents a framework delineating the major factors and dimensions which define the arena in which retailers must develop their assortment strategies. Additionally, evidence is presented which indicates that the shape of the utility function for assortment size is bell-shaped for both the consumer and the retailer.

Hopefully this paper will increase our understanding of the dynamic interplay between the numerous forms of a consumer's "varied behavior," (especially variety seeking) and their subsequent impLications for the retailer's assortment strategies. Rather than treat each of these activities in isolation as has been done in past research, this paper suggests the need for their integration. Viewing the consumer's need for varied behavior and assortment size decisions as "opposite sides of the same coin" should lead to more effective marketing strategies by increasing our understanding of each area.

Regarding future research, the notion that the consumer's utility function for assortment sizes may be dynamic both within and across product classes, suggests the necessity to investigate the extent to which assortment-related problems and strategies differ across retail store types. The consumer's perspective should be considered by investigating the need for variety in various product classes.

REFERENCES

Achabal, D., S.H.McIntyre and M. Kriewall (1983), "Issues and Perspectives on Retail Productivity," Working Paper, Retail Management Institute, The University of Santa Clara, Santa Clara, CA.

Arnold, S.J., S. Ma, and D.J. Tigert (1978), "A Comparative Analysis of Determinant Attributes in Retail Store Selection," Advances in Consumer Research, 5, 663-667.

Arnold, S.J., T.H. Oum, and D.J. Tigert, (1983), "Determinant Attributes in Retail Patronage: Seasonal, Temporal, Regional, and International Comparisons," Journal of Marketing Research, 20, 149-157.

Baumol, W.L., and E.A. Ide (1956), "Variety in Retailing," Management Science 3, 93-101.

Berlyne, D.E. (1960), Conflict, Arousal, and Curiosity, New York: McGraw-Hill.

Bishop W. (1978), "Limited-Assortment Stores Multiply Sell to Cost-Aware Market Segment," Marketing News, 29 December 1978, (p. 13).

Brickman, P., and B. D'Amato (1975), "Exposure Effects in a Free Choice Situation," Journal of Personality and Social Psychology, 32, 3, (415-420).

Brown, D.J. (1978), "An Examination of Consumer Grocery Store Choice: Considering the Attraction of Size and the Friction of Travel Time," Advances in Consumer Re- search, 5, 243-246.

Cardozo, R.N. (1975), "How Images Vary by Product Class," Journal of Retailing, 50, 85-98.

Claxton, J.D. and J.R.B. Ritchie (1979), "Consumer Prepurchase Shopping Problems: A Focus on the Retailing Component," Journal of Retailing, 55, 24-44.

Dashiell, J.F. (1925), "A Quantitative Demonstration of Animal Drive," Journal of Comparative Psychology, 5, 205-208.

Driver, M.J. and S.S. Streufert (1964), "The 'General Incongruity Adaptation Level' (GIAL) Hypothesis: An Analysis and Integration of Cognitive Approaches to Motivation," Paper #114 Institute for Research in the Behavioral Economic and Management Sciences, Krannert Graduate School of Management, Purdue University.

Farquhar, P.H. and V.R. Rao (1976), "A Balance Model for Evaluating Subsets of Multiattributed Items," Management Science, 22, #5, 523-539.

Handelsman, M. (1982), "Measurement of Varied-Buying Behavior Employing Perceptual and Purchase Diary Data," unpublished Ph.D. dissertation, Purdue University, W. Lafayette, IN.

Hirschman, E.C. (1980), "Innovativeness, Novelty Seeking and Consumer Creativity," Journal of Consumer Research, 7, 283-295.

Howard, J.A. and J.N. Sheth (1969), The Theory of Buyer Behavior. New York: John Wiley & Sons.

Hoyer, W.D. and N. Ridgway (1983), "Variety Seeking as an Explanation for Exploratory Purchase Behavior: A Theoretical Model," Advances in Consumer Research, 11, 114-119.

Kotler, P. (1980), Marketing Management: Analysis, Planning, and Control, Englewood Cliffs, NJ: Prentice-Hall, Inc.

Landon, E.L., Jr. (1974), "Self Concept, Ideal Self Concept, and Consumer Purchase Intentions," Journal of Consumer Research, 1, 44-51.

Laurent, G. (1978), "A Study of Multiple Variant Consumption for Frequently Purchased Consumer Products," unpublished Ph.D. dissertation, Sloan School of Management, M.I.T.

Lindquist, J.D. (1975), "Meaning of Image: A Survey of Empirical Hypothetical Evidence," Journal of Retailing, 50, 29-38.

Maddi, S.R. (1968), "The Pursuit of Consistency and Variety," in Theories of Cognitive Consistency: A Sourcebook," eds. Robert P. Abelson et al., Chicago: Rand McNally.

Mason, J.B. and M.L. Mayer (1981), Modern Retailing, Theory and Practice, Business Publications, Inc. Plano, TX.

McAlister, L. (1979), "Choosing Multiple Items from a Product Class," Journal of Consumer Research, 6, 213-224.

McAlister, L. and E.A. Pessemier (1982), "Variety Seeking Behavior: An Interdisciplinary Review," The Journal of Consumer Research, 9, 311-322.

McDougall, W. (1908), An Introduction to Social Psychology, London: Methuen.

Munson, J.M. and W.A. Spivey (1981), "Product and Brand-User Stereotypes Among Social Classes," Journal of Advertising Research, August, 37-46.

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

Pessemier, E.A. (1980), "Retail Assortments - Some Theoretical and Applied Problems," Report #80-111, Marketing Science Institute, Cambridge, Mass.

Raju, P.S. (1980), "Optimum Stimulation Level: Its Relationship to Personality, Demographics, and Exploratory Behavior," Journal of Consumer Research, 7, 272-282.

Raju, P.S. (1981), "Theories of Exploratory Behavior: Review and Consumer Research Implications," Research in Marketing, 4, JAI Press Inc., 223-249.

Venkatesan, M. (1973), "Cognitive Consistency and Novelty Seeking," in Consumer Behavior: Theoretical Sources, S. Ward and T.S. Robertson (eds.), Englewood Cliffs, NJ: Prentice-Hall.

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