An Examination of the Structure of Buying-Selling Interactions Among Insurance Agents and Prospective Customers

James L. Taylor, University of Alabama
Arch G. Woodside, University of South Carolina
ABSTRACT - Wilson's model of a multi-stage sequence of buyer-seller interactions is supported in a field observation study. In-depth interviews with life insurance salesmen, sales managers and customers, followed by content analysis of 40 recorded sales encounters, revealed the existence of six distinct stages of the insurance buying-selling process.
[ to cite ]:
James L. Taylor and Arch G. Woodside (1980) ,"An Examination of the Structure of Buying-Selling Interactions Among Insurance Agents and Prospective Customers", in NA - Advances in Consumer Research Volume 07, eds. Jerry C. Olson, Ann Abor, MI : Association for Consumer Research, Pages: 387-392.

Advances in Consumer Research Volume 7, 1980     Pages 387-392

AN EXAMINATION OF THE STRUCTURE OF BUYING-SELLING INTERACTIONS AMONG INSURANCE AGENTS AND PROSPECTIVE CUSTOMERS

James L. Taylor, University of Alabama

Arch G. Woodside, University of South Carolina

ABSTRACT -

Wilson's model of a multi-stage sequence of buyer-seller interactions is supported in a field observation study. In-depth interviews with life insurance salesmen, sales managers and customers, followed by content analysis of 40 recorded sales encounters, revealed the existence of six distinct stages of the insurance buying-selling process.

INTRODUCTION

Recent marketing literature reveals several attempts by marketing scholars to formulate a comprehensive buyer-seller interaction model. Most of these undertakings have relied on the use of a deductive methodology. A deductive methodology requires the development of theory prior to data collection. Conceptualizations of the buying-selling process by Sheth (1976), Wilson (1976), O'Shaughnessy (1971-72), and Levy and Zaltman (1975) have resulted from this approach.

Sheth's model (1976) is based on a communication perspective. The basic postulate of this model is that the quality of the interaction between buyer and seller is a function of the compatibility between the two parties with respect to both the style and the content of communication. A satisfactory interactive transaction between the buyer and seller presumably occurs if and only if they are compatible in regard to both the content and the style of communication. If the buyer and seller are incompatible with respect to communication style and content, no interaction usually takes place. An inefficient transaction takes place when the participants are compatible with respect to communication style (content) but are incompatible with respect to content (style).

O'Shaughnessy founded a model on the notion that selling is an interpersonal influence process. This model is derived from a fusion of Kelman's (1961) model on the process of opinion change with Etzioni's (1961) model of power and its correlative compliance. O'Shaughnessy's conceptualization regards retail salesmanship as the use of symbolic means to influence the customer, in which case the salesclerk's source of influence is his or her attractiveness (i.e., similarity or likeability) and credibility. Several marketing studies have supported this model in that salesman credibility and similarity to customer consistently have emerged as important ingredients to selling success (Brock, 1965; Busch and Wilson, 1976; Woodside and Davenport, 1974).

The models advanced by Wilson (1976) and Levy and Zaltman (1975) view dyadic buyer-seller interaction as an exchange process. The basic premise underlying the formulation of these models derives from Homan's (1958) contention that all elementary social behavior is an exchange of material and non-material goods.

Wilson postulates that the dyadic relationship goes through a number of stages during which a bundle of attributes valuable to the buyer is developed by the seller. The proposed stages are (1) source legitimization, (2) information exchange-problem identification, (3) attribute delineation, (4) attribute value negotiation, and (5) relationship maintenance. Two basic assumptions of Wilson's model are (1) that the buyer attempts to secure a bundle of attributes, both tangible and psychological, from the seller, and (2) that the buyer-seller relationship develops over a period of time.

Levy and Zaltman's conceptualization is founded on several basic propositions of Homan's social exchange theory. Homan's (1961) propositions include the concepts of (1) frequency of reward (punishment), (2) likelihood of action, (3) similarity of stimuli, (4) value of a reward, (5) recency of last critical mass (satiation) of the rewards, (6) discrepancy of expected reward (punishment) with realized reward (punishment) and (7) degree of approving (disapproving) behavior. Levy and Zaltman have recast these propositions in a marketing context and have presented the results as a set of selling propositions particularly applicable to the area of buyer-seller interaction.

A common element in these modeling efforts has been the deductive approach used in the formulation. Yet Olshavsky (1976) suggests that a second approach should be utilized in the development of a comprehensive model. The method advocated is an inductive approach in which data is collected via a two-step procedure prior to the formulation of a theory of buyer-seller interaction. Olshavsky's approach is founded on the notion that descriptive studies of individual decision-making in buying-selling situations may be an important prelude to selling theory development.

Several descriptive studies have examined the buying-selling processes for different products and different buying situations; nevertheless, the results of these studies have not yet become input into theory formation. Most of these descriptive studies followed the example of Cyert, Simon and Trow's (1956) description of the anatomy of a major business decision of an industrial firm. Robinson, Faris and Wind (1967) studied the new-task buying situations of three industrial firms and found that eight distinct stages of problem solving characterized the purchasing process for these companies. Similarly, Wilson and Bambic (1977) partially tested and found support for the adequacy of Wilson's five stage decision process model of buyer-seller interactions in industrial marketing situations. Woodside and Taylor (1978) and Woodside, et al (1977) demonstrated the feasibility of examining and describing the individual decision processes of buyers and sellers of a consumer good--life insurance.

The purpose of the present article is to report the results of an empirical study to describe the buying-selling behavior of actual life and health insurance salesmen and their prospective customers. The study reported was a descriptive research effort in that no attempt was made to manipulate specific predetermined independent variables. The testing of specific hypotheses was not intended in the study hut the findings discussed from prior research were used to develop several questions to direct the investigation. Specifically, the questions that formed the focus of this research were:

1.  Is the buying and selling process for life and health insurance characterized by a series of stages with each stage having different content?

2.  If so, what are the distinguishing features describing the content of each identifiable stage?

3.  What can be generalized to other sales encounters regarding the buying-selling process for life and health insurance?

METHODOLOGY

A variety of data sources were used to examine the structure of insurance buying-selling interactions. Among these were:

1.  Discussions with the agency manager and associate manager of a local headquarters of an insurance company in Columbia, South Carolina.

2.  Individual in-depth interviews with five health and life insurance salesmen.

3.  Individual in-depth interviews with five prospective customers for health and life insurance.

4.  Content analysis of forty recorded sales encounters involving actual salesmen (n=3) meeting with prospective buyers (n=57).

Three types of data were collected. First, in-depth personal interviews were conducted with the agency manager, the associate manager and five salesmen to determine insight into the structure of the typical sales encounter. Second, forty encounters between salesmen and prospects were tape recorded; and finally, postpurchase in-depth interviews were conducted with members of three randomly selected selling and buying centers.

The discussion with the agency manager, associate manager and salesmen focused on discerning the existence of various distinct stages in the life and health insurance buying-selling process. These sales people were asked to describe what takes place in the typical sales encounter and what percentage of time is spent on each activity. While acknowledging a sequence of objectives in the sales encounter, these persons noted that each salesman had his own selling "style." The differences in selling styles were attributed to specific personality traits, the type customers generally sought by each salesman and the diversity in previous training and sales experience associated with each representative. Yet, despite alleged individual differences, the following sequence of events and relative times were common to all discussions regarding the typical sales encounter:

Prospecting                                                      3   -  9 %

Establishing a common ground with prospect   4  -  7 %

Determining prospect's needs                         35   -  50 %

Presenting alternatives                                    20   -  45 %

Closing                                                          10   -  20 %

Follow-up                                                        4   -  5 %

These events formed the basis of subsequent analysis of recorded transactions of actual sales encounters.

Forty prospective life insurance buying centers agreed to cooperate and permit tape recordings of the meetings with sales representatives (only two buying centers contacted refused to have the meetings tape recorded). In 17 (43 percent) buying centers, two people met with the salesman during the sales encounter; while in 23 (57 percent) of the encounters only one prospect met with the salesman. Thus, prospects actually meeting with salesmen totaled 57.

Three salesmen were utilized in the data collection in order to ensure diversity among target clients; this diversity among prospects should permit greater generalization of study results. The salesmen's experience with the cooperating company ranged from 1-5 years and their ages from 28-34.

The data collection method involved a variation of the Willett and Pennington (1966) field study in that audio recordings were made of actual transactions between salesman and prospect as they took place in field settings. The conversations between salesmen and potential buyers were recorded by the researchers while accompanying the salesman when calls were made on potential clients. In 32 (80 percent) of the initial meetings the buyer(s) did not know that a researcher would be accompanying the salesman before arrival. This was a result of the difficulty in synchronizing work schedules. Additionally, the salesmen believed that permission to record the transaction was better obtained when the buyer(s) could meet and evaluate the researcher in person.

Once permission was granted, the salesman immediately engaged the prospect in conversation and started the sales encounter. The researcher started the recorder and positioned himself and recorder out of the prospect's and salesman's view. This latter action was taken in order to minimize the impact of any artificialities introduced into the exchange setting by the researcher's and recorder's presence.

Prospects were questioned as soon after the sales meeting as possible to determine (1) the extent of their awareness of the researcher's or the recorder's presence and (2) to what degree they felt this presence affected their communication with the salesman. Only those prospects who stated that they quickly forgot about the researcher and recorder once they became involved in conversation with the salesman are included for analysis in this study. This represented 97% of the prospects whose conversations were recorded.

Content analysis (Kassarjian, 1977) was utilized to analyze the recorded data. The content analysis was followed by "retrospective" analyses with five randomly selected insurance customers and three salesmen. In the retrospective analyses, the tape recordings were played back to the salesmen and prospects separately and the participants were requested to comment on the meanings and strategies behind their own statements in the interactions. The retrospective analyses provided an evaluation of the consistency between participant's reported strategies throughout the sales encounter and inferences drawn by the researchers from the content analysis regarding such meanings and strategies.

The content analysis focused on discerning the existence of the various stages in the life and health insurance buying-selling process as suggested in prior discussions with the selling personnel. Consequently, this analysis included an examination of the bargaining behavior of both prospects and salespersons and the use of certain selling strategies by the salespersons throughout the sales encounter. If the typical insurance buying-selling encounter consists of six distinct stages as suggested, the content of each stage should exhibit some distinguishable differences. Thus, the total time of each sales encounter was segmented into six stages and the content of each segment examined. The percentage of the total time allocated to each stage was as follows: stage 1 (5 percent), stage 2 (5 percent), stage 3 (40 percent), stage 4 (30 percent), stage 5 (15 percent), stage 6 (5 percent).

The examination of bargaining behavior among the participating salesmen and prospects required that specific measurable components of the bargaining process first be identified. Pennington (1968) has formulated four such components of bargaining behavior from theoretical works of Kuehn (1963) and Schelling (1956). This conceptualization of bargaining behavior was adopted for this study (the bargaining components used are defined in Pennington 1968, p. 256). The specific components are (1) the making of direct offers, (2) the presentation of concession limits, (3) the determination of concession limits, and (4) attempts to change concession limits.

A review of personal selling trade literature revealed several selling activities that consistently are held to be important to selling success. Sales representatives are advised to (1) make effective use of questions; (2) provide the prospects an abundance of information; (3) make frequent references to persons, places or objects familiar to both salesman and prospect, that is, make references to "familiar others"; (4) attempt to establish similarity with the prospect; (5) attempt to establish expertise; (6) make references to potential catastrophes; and (7) refer frequently to the prospect's loved ones. For example, Frank Bettger, one of the most successful and highest paid salesmen in America, ascribes much of his success to just one of these selling activities--the effective use of questions (Bettger, 1949).

Most of these activities are applicable to the selling of many, if not all, products. Two, however, are primarily germane to the selling of life and health insurance--references to potential catastrophes and to the prospect's loved ones. References to the possibility of facing personal catastrophes are held to heighten the felt need of the product while references to loved ones evoke the prospect's concern for these persons.

The relative frequencies in which each salesman or prospect relied on use of a bargaining variable were determined in the content analysis. The recorded sales encounters were played back by two judges working independently and the number of times each salesman or prospect used each bargaining variable counted. Similarly, the use of each selling activity by salesmen was counted. Bargaining acts were counted for both prospects and salesmen. The frequencies with which prospects asked questions and provided information also were counted. A selling or bargaining act (i.e., use of a selling or bargaining variable) was measured in terms of a single declarative or interrogative sentence. That is, the unit of analysis for the study was the sentence.

The count data for each bargaining and selling variable obtained by the two judges working separately were correlated to determine the extent of agreement between judgments. Product moment correlations between the two sets of count data were equal to or greater than .95 for all of the variables; thus, considerable agreement between judges was obtained concerning the extent to which bargaining and selling acts occurred. Minor differences between judges were resolved by using an average of the two counts for analysis. No major discrepancies were obtained.

FINDINGS

The content and retrospective analyses confirmed several focal points in the interactions studied with each focal point characterized by different exchange behavior on the part of participants. The buying-selling processes examined included some idiosyncratic dialogue and exchange behavior; however, six focal points emerged that were common to all 40 interactions. Figure 1 reflects the six stages of the life and health insurance buying and selling process observed among the participants of the study. The stages have been renamed by the authors to more accurately reflect their contents.

The existence of the six stages does not imply the use of structured sales presentations on the part of participating salesmen. In fact, the opposite was the case; the sales representatives relied on completely unstructured or interactive sales presentations. The six stages simply capture the changing goals, strategies and activities of participants as the interaction progressed.

FIGURE 1

STAGES IN THE LIFE AND HEALTH INSURANCE SELLING PROCESS OBSERVED AMONG FORTY BUYING AND SELLING CENTERS

Stage One: Content Initiation

Typically, the exchange was initiated by the salesman calling on the prospective buyer in person or by telephone. Specifically, twenty-two (55 percent) of the buying centers were contacted initially by the salesman via telephone, six (15 percent) by direct mail, and twelve (30 percent) by "cold call." A "cold call" refers to a salesman's unannounced visit to a prospect's place of business or home.

The purpose of this first step was for the salesman to make initial contact with the prospective customer and to arrange an appointment to discuss the prospect's insurance needs; thus, no product selling took place in this initial stage. Generally, the salesman attempted to "sell" himself to the customer and to persuade the customer of the need to discuss his or her insurance needs. Some attempts by the salesman to establish his expertise and similarity with the prospect were made. References to familiar others were common since many initial contacts were prompted by "leads" supplied to the salesman by individuals known to the prospect. This stage ended when the appointment was granted or denied or when the prospect permitted the salesman to initiate the interaction on the spot. A total of 87 prospects were initially contacted in order to acquire the 40 interactions used for subsequent analysis.

Stage Two: Rapport Building

The opening moments of the exchange were spent with both salesman and prospect attempting to establish rapport with the other party. Conversation was generally social in nature with the bulk of the salesman's references to familiar others and attempts to establish similarity with the customer being made during this period. The salesperson typically made frequent comments designed to establish his expertise. The retrospective analyses indicated that possible evaluations of the salesman by the prospect occur in this stage.

The existence of a rapport building phase of the buying-selling process was supported in both the content and retrospective analyses. In the retrospective analyses, all salespersons described their initial goals and strategies in terms of rapport building. The following is a typical description by a salesman of the first five minutes of his encounter with one prospect (all names are hypothetical):

"I arrive at Kathy's home as scheduled. My first goal is to build rapport with the prospect. My strategy for doing this calls for talking about things of interest to Kathy. We talk about Columbia and the local university. I discover a common ground between us. Kathy knows and likes a professor that I once worked for. I made several references to this fact."

Some typical examples of comments characterizing this stage revealed in the content analysis are:

"Do you know a guy who works there named Cohn (a familiar other) ?"

"Well, I used to be somewhat of a bowler myself (establishing similarity)."

"I am an ACE consultant with the Small Business Administration and see these problems all the time (establishing expertise)."

Although rapport building was essentially a salesman initiated activity, the retrospective analyses indicated that the prospective buyers desired information regarding the salesman. Consequently, prospects undertook evaluative activities in the opening moments of the sales encounter to get to know the salesman. Typical were the comments of one prospect describing her first five minutes of interaction with the salesman:

"The salesman arrives as scheduled. My first goal is to find out more about him. I ask how long he has lived in Columbia. I find that he graduated from the local university where I attend as a part-time student. I ask if he has ever met a particular professor that I know and like. I am surprised to discover that the salesman once worked for the professor."

Stage Three: Exchange of Information

An exchange of information followed attempts to establish rapport. In this stage of the sales encounter bargaining acts by the salesman and prospect were first observed. High frequencies of attempts to determine and to present concession limits were characteristic of this stage for both prospect and salesman. The salesman attempted to take measure of the prospect's needs and expectations while the prospect sought knowledge of the company's offering. Consequently, a high frequency of questions were asked and information provided by both salesman and prospect during this phase. Some references by the salesman to potential catastrophes were observed. Some examples of typical comments recorded during this phase include:

"How much money could you put into a policy (salesman determining concession limits)?"

"I would like to keep my payments under $20 per month (prospect presenting concession limits)."

"Approximately, what are you saving in the credit union (salesman determining concession limits)?"

"How much would the monthly premium be (prospect determining concession limits)?"

"You have the option of converting a group term policy to permanent life insurance within 30 days (salesman providing information)."

Stage Four: Persuasion Attempts

Although the salesman typically discussed several insurance options during the third stage, little or no "selling" was observed; that is, no persuasive efforts were undertaken to convince the prospect to purchase a particular policy until the salesman had made an assessment of the prospect's needs and expectations. This assessment, in which the sales representative typically sought to determine the likelihood of making a particular sale, was the principal activity in the "exchange of information" stage.

After the salesman had taken measure of the prospect's needs and expectations, persuasion attempts were begun. The transition point to this stage was when the salesman ceased to discuss the availability of a number of purchase options and began to focus on a particular purchase option; that is, the sales representative in this stage started to make recommendations to the prospect as to which option should be purchased.

Bargaining by the salesman was intensified in that the sales representative typically increased the frequency in which concession limits were presented. More specific information regarding the insurance option being promoted was given. An increase in the frequency of references to potential catastrophes occurred in this phase. Bargaining by prospects continued to involve attempts to determine concession limits of the salesman. Thus, the specific bargaining activities encountered in this stage were not much different than those observed in the information exchange phase. The essential difference was in the goals of the participants--especially the sales representative. In the third stage, the discussion involved exchanging information regarding customer needs and possible purchase options whereas, in the fourth stage, the conversation shifted to the relative merits of the particular policy advocated by the salesman.

Stage Five: Close Attempts

After persuasion attempts were made, the salesman typically tried to close the sale. If objections were met from the prospect, bargaining by the salesman shifted from presenting and determining concession limits to greater reliance on attempts to change the prospect's expectations or concession limits. The prospect usually responded to the salesman's attempts to close the sale by either (1) offering objections; (2) expressing the desire to postpone the decision, that is, to "think the decision over"; (3) refusing the salesman's offering; or (4) making a purchase. A purchase was made by 15 (38 percent) of the buying centers participating in the study.

The salesmen attempted to close the sale in a variety of ways. The most direct approach observed was when the salesman simply asked the prospect, "Well, what do you think?" or, "What would you like to do?". Yet, the most common close attempt involved the salesman presenting the prospect with a choice between two things--both involving purchase (e.g., "Would you prefer to go with the $20,000 or $30,000 policy?"). The prospect's answer to such questions informed the salesman as to whether the purchase decision had been made. Another common close attempt involved the salesman simply requesting information from the prospect necessary to complete the purchase; that is, the salesman would produce a printed form and request information necessary for its completion.

Close attempts generally ended when the purchase decision was made (i.e., purchase or nonpurchase) or when the salesman was certain that the prospect was not ready to make a commitment. The salesmen made, on the average, three close attempts per interaction. When prospects desired to postpone the purchase decision, the salesmen typically reiterated the major selling points that had been raised and then tried to obtain a future appointment.

Stage Six: Rapprochement

Rapprochement attempts were normally made by both parties as they ended the encounter; that is, the interaction ended with the conversation returning to being social in nature. This transition occurred regardless of the purchase outcome as evidenced by its presence in ali 40 recorded transactions. The participants in the exchange seemed to seek a return to the less business-like atmosphere that characterized the initial stages of the interaction--particularly the rapport building phase. For the salesman, attempts to develop the long term relationship (if made) were undertaken in this last stage.

CONCLUSIONS

The purposes of this article have been limited. The focus has been directed toward identifying possible stages in the buying-selling process for one product--life and health insurance. The analysis of the recorded transactions and the subsequent retrospective analyses suggest the existence of six stages having different content. The rapport building and rapprochement stages observed for these interactions indicate that much of life and health insurance buying and selling is social in nature. Bargaining or product related exchange behavior was limited to only three stages in the process--the exchange of information, persuasive attempts, and close attempts stages.

Salesman bargaining behavior was characterized by a high frequency of attempts to determine the prospect's concession limits in early stages of the exchange; the salesman then shifted to greater use of presenting concession limits; and finally, the salesman increased attempts to change the prospect's concession limits if resistance was met in closing the sale. Bargaining by prospects was almost equally divided between (1) seeking information from the salesman regarding purchase options or terms, and (2) communicating to the salesman expectations regarding price, payment terms or purchase intentions. The prospect sought information with greater frequency in earlier stages of the exchange; he or she then shifted to expressing expectations with more frequency as the exchange progressed.

The stages of the buying-selling process identified for the 50 buying and selling centers in this study can be interpreted as limited support for the dyadic sales process model advanced by Wilson (1976). Support is limited in that only 40 recorded interactions involving a single product were examined; thus, the results may not generalize to a larger number of sales encounters or to other products. Additionally, the study was not intended by the writers to be a test of Wilson's conceptualization of the buying-selling process. Nevertheless, the results support Wilson's model in that two sets of researchers working independently--one using a deductive approach and the other an inductive methodology-- have come to essentially the same conclusion regarding the existence and content of distinct stages in the buying-selling process.

Wilson's model includes source legitimization, information exchange-problem identification, attribute delineation, attribute value negotiation, and relationship maintenance stages. The postulated objective of source legitimization (rapport building) stage is to establish the salesman as a legitimate and credible partner in the dyadic interaction process. The information exchange-problem identification (exchange of information) stage involves uncovering the need to be fulfilled through purchase. In the attribute delineation (persuasion) stage, the buyer and seller develop the bundle of attributes that are to be exchanged. Bargaining takes place in the attribute value negotiation (closing attempts) stage where buyer and seller attempt to reach agreement on the terms of exchange. The buyer-seller dyad seeks to maintain and build upon their relationship in the final relationship maintenance (rapprochement) stage. This close parallel between Wilson's conceptualization and the empirical findings of this study suggests that future research attempts to validate the Wilson model are warranted.

REFERENCES

Bettger, Frank (1949), How I Raised Myself From Failure to Success in Selling, Englewood Cliffs, NJ: Prentice-Hall, Inc.

Brock, Timothy C. (1965), "Communicator-Recipient Similarity and Decision Change," Journal of Personality and Social Psychology, 1, 650-654.

Busch, Paul and Wilson, David T. (1976), "An Experimental Analysis of a Salesman's Expert and Referent Bases of Social Power in the Buyer-Seller Dyad," Journal of Marketing Research, 13, 3-11.

Cyert, R. M., Simon, H. A., and Trow, D. B. (1956), "Observation of a Business Decision," Journal of Business, 29, 237-248.

Etzioni, Emitai (1961), A Comparative Analysis of Organizations, Glencoe, Ill.: Free Press.

Homans, George C. (1958), "Human Behavior as Exchange," American Journal of Sociology, 63, 597-606.

Homans, George C. (1961), Social Behavior: Its Elementary Forms, New York: Harcourt Brace Jovanovich.

Kassarjian, Harold H. (1977), "Content Analysis in Consumer Research," Journal of Consumer Research, 4, 8-18.

Kelman, Herbert C. (1961), "Processes of Opinion Change" Public Opinion Quarterly, 25, 57-78.

Kuehn, Alfred (1963), The Study of Society: A Unified Approach, Homewood, Ill.: Richard D. Irwin and the Dorsey Press.

Levy, Sidney J. and Zaltman, Gerald (1975), Marketing, Society and Conflict, Englewood Cliffs, NJ: Prentice-Hall, Inc.

Olshavsky, R. W. (1976), "Consumer Decision Making in Naturalistic Settings: Salesman-Prospect Interaction," in Beverlee B. Anderson, ed., Advances in Consumer Research, Cincinnati: Association for Consumer Research, 364-369.

O'Shaughnessy, John (1971-72), "Selling as an Interpersonal Influence Process," Journal of Retailing, 47, 32-46.

Pennington, Allen L. (1968), "Customer-Salesman Bargaining Behavior in Retail Transactions," Journal of Marketing Research, 5, 255-262.

Robinson, Patrick, Faris, Charles, and Wind, Yoram (1967), Industrial Buying and Creative Marketing, Boston: Allyn and Bacon.

Schelling, Thomas C. (1956), "An Essay on Bargaining," The American Economic Review, 46, 281-306.

Sheth, Jagdish N. (1976), "Buyer-Seller Interaction: A Conceptual Framework," in Beverlee B. Anderson, ed., Advances in Consumer Research, Cincinnati: Association for Consumer Research, 382-386.

Willett, Ronald P. and Pennington, Allen (1966), "Customer and Salesman: The Anatomy of Choice in a Retail Setting," Proceedings, Fall Conference, American Marketing Association, 598-616.

Wilson, David T. (1976), "Dyadic Interaction: An Exchange Process," in Beverlee B. Anderson, ed., Advances in Consumer Research, Cincinnati: Association for Consumer Research, 394-397.

Wilson, David T. and Bambic, Peter J. (1977), "Conceptual Models of Buyer/Seller Interaction," Paper presented at the 85th Annual Convention, American Psychological Association, San Francisco.

Woodside, Arch G. and Davenport, J. William (1974), "The Effect of Salesman Similarity and Expertise on Consumer Purchasing Behavior," Journal of Marketing Research, 11, 198-202.

Woodside, Arch G., and Taylor, James L. (1978), "Observations of Buyer and Seller Transactions," in H. Keith Hunt, ed., Advances in Consumer Research, Ann Arbor: Association for Consumer Research, 643-652.

Woodside, Arch G., Taylor, James L., Pritchett, S. Travis, and Morgenroth, William M. (1977), "Transactions Among Buying and Selling Centers," in Proceedings of the Fall Educators' Conference, Chicago: American Marketing Association, 1977, 59-64.

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