Customers Complain-Businesses Make Excuses: the Effects of Linkage and Valence

Donna J. Hill, Bradley University
Robert Baer, Bradley University
ABSTRACT - Company reactions to customer complaints were examined. A framework, based upon a psychological theory of excuse making (Higgins and Snyder 1990) was advanced. Excuses were classified as either linkage (diminishing the individual's perceived causal role in producing the event) or valence (diminishing the perceived negative consequence of the act). Subjects sent complaint letters to companies who had caused them dissatisfaction. Company excuse responses were categorized according to the excuse framework. The effects of different excuse types were measured. The results showed that valence excuses lead to less perceived harm and less perceived blame than linkage excuses. Valence excuses were also rated more acceptable than linkage excuses. The effects of different kinds of linkage and valence excuses were also examined. The implications for complaint management were discussed.
[ to cite ]:
Donna J. Hill and Robert Baer (1994) ,"Customers Complain-Businesses Make Excuses: the Effects of Linkage and Valence", in NA - Advances in Consumer Research Volume 21, eds. Chris T. Allen and Deborah Roedder John, Provo, UT : Association for Consumer Research, Pages: 399-405.

Advances in Consumer Research Volume 21, 1994      Pages 399-405

CUSTOMERS COMPLAIN-BUSINESSES MAKE EXCUSES: THE EFFECTS OF LINKAGE AND VALENCE

Donna J. Hill, Bradley University

Robert Baer, Bradley University

ABSTRACT -

Company reactions to customer complaints were examined. A framework, based upon a psychological theory of excuse making (Higgins and Snyder 1990) was advanced. Excuses were classified as either linkage (diminishing the individual's perceived causal role in producing the event) or valence (diminishing the perceived negative consequence of the act). Subjects sent complaint letters to companies who had caused them dissatisfaction. Company excuse responses were categorized according to the excuse framework. The effects of different excuse types were measured. The results showed that valence excuses lead to less perceived harm and less perceived blame than linkage excuses. Valence excuses were also rated more acceptable than linkage excuses. The effects of different kinds of linkage and valence excuses were also examined. The implications for complaint management were discussed.

The decade of the 1980's may be regarded by historians as the decade of customer rebellion (Desatnick 1987). Customers everywhere are demanding that they get what they pay for, whether it is on-time delivery, courteous treatment at the point-of-purchase, or a clean hotel room. Concerned with service and product quality, companies by the scores are adopting customer orientations that consider customer satisfaction the key to long-term profitability. Customer oriented firms are contributing to the customer rebellion by energetically encouraging their customers to complain. As consumers become more vocal about their dissatisfaction, the management of their complaints will play an increasingly important role. Every marketing manager will have to develop approaches for effectively responding to disgruntled customers. Although there is now a large body of literature on consumer complaining behavior, relatively little is known about complaint management. What is known is that a rapid company response, apologies, financial compensation (e.g., coupons, refunds and rebates), demonstrating a genuine interest in relieving the problem, and directly addressing the problem can all help recapture a dissatisfied customer (Folkes 1984; Gilly and Gelb 1977; Clark & Kaminski and Rink 1992; Goodwin and Ross 1992; Krentler and Cosenza 1987).

Garrett, Meyers and Camey (1991) claim that despite the significant advancements that have been made in our understanding of consumer complaints, very little attention has been devoted toward understanding the role of communication (i.e., what is said) in complaint interactions. This is unfortunate since the complaint process is, after all, a communication process. Disappointed consumers communicate their dissatisfaction to company representatives who, in turn, communicate their response. One aspect of the communication process that is worthy of more attention, they maintain, is account analysis. Accounts are excuses or explanations that accused actors give to minimize the severity of their predicament. When things go wrong in an organization, employees may try to soothe disgruntled customers by excusing either their own or their firm's behavior (Bitner 1990). Hill, Baer and Kosenko (1992) contend that managers should be cognizant of the excuse making activities of their employees. Rather than handling complaints in their own haphazard way, front line personnel should be trained in how to verbally respond to complaints.

The study of excuses may help enlighten the complaint management process. The marketing literature has been almost silent about excuse making. It is as if marketers believed that firms do not make excuses for their mistakes. As consumers, however, we know that firms do make excuses. This should come as no surprise. People have a natural tendency to present themselves in a positive light (Goffman 1963) and excuses can help serve that purpose. The desire to put oneself in a good light should be heightened whenever aspersions are cast about one's character, conduct, motives or skill. Not only individuals, but organizations too are vitally concerned with nurturing positive images (Higgins and Snyder 1990); negative perceptions of a firm are hazardous to its long-term viability.

The purpose of the present paper was to explore company excuse making in the face of a customer complaint. Specifically, we advanced a framework for understanding excuses based upon a psychological theory of excuse making (Higgins and Snyder 1990, 1991; Snyder 1985; Snyder and Higgins 1988; Snyder, Higgins and Stucky 1983). A structure for categorizing excuse types is presented and consumer reactions to the different excuse types are measured.

Previous definitions of excuses (Scott and Lyman 1968; Schonbach 1980) have made a distinction between excuses and justifications. Excuses have been considered as explanations as to why one should not be held so completely responsible for unfortunate acts or consequences; justifications have been regarded as attempts to minimize the perceived undesirability of those events. In their psychological theory of excuse making, Higgins and Snyder (1990) have proposed a broad definition of excuses that subsumes under one over arching process the traditional excuse and justification distinction. Higgins and Snyder define an excuse as the "motivated process of (a) diminishing the perceived negativity of esteem-threatening outcomes and (b) shifting causal attributions from sources that are relatively more central to the person's sense of self to sources that are relatively less central" (p.74). Within the traditional framework, excuses shift causal attributions, whereas justifications diminish the perceived negativity of the event.

According to excuse theory (Higgins and Snyder 1990, 1991; Snyder 1985; Snyder and Higgins 1988; Snyder, Higgins and Stucky 1983) excuse making is triggered whenever individuals or organizations perceive themselves to be linked to acts or outcomes that are perceived as undesirable. The tendency to offer an excuse is based upon two independent assessments (linkage-to-act and valence-of-act) that are made whenever individuals or organizations are confronted with information that is threatening to a positive view of oneself. The linkage-to-act assessment represents the degree to which the individual perceives him or herself to be linked (from no linkage to total linkage) to a particular act or outcome. The valence-of-act assessment represents the individual's qualitative assessment of the positiveness of the act or outcome (from positive to negative). Factors that contribute to either the perceived negativity of the event or to a sense of responsibility for the event will enhance the tendency to engage in excuse making. If excuse making is caused by the perception that one is linked to a negative event, then for an excuse to be effective it must temper the negative repercussions of the act by either distancing the excuse maker from the undesirable act (linkage excuse) or by altering the perception that the event was negative (valence excuse). If the individual successfully lessens the perceived negativity of the event, the motivation to cut one's link to the act may decrease. On the other hand, if the individual effectively reduces their perceived connection to the negative event, the need to alter the perceived negativeness of the act will correspondingly diminish.

We propose that there are three types of linkage excuses (Denial, Deflection and Explanation) and two types of valence excuses (Minimization and Justification).

Linkage Excuses. (1) Denial excuses attempt to eliminate the perceived linkage to the act by verbally denying any implication of guilt (e.g., "I didn't do it"). Denials do not repudiate the fact that something wrong occurred. Rather, they deny that they had anything to do with it. (2) Deflection excuses go beyond denial by pointing out who or what was to blame (e.g., "He did it"). By pinning the blame on someone or something else, the excuse-maker can further disconnect him or herself from having any apparent responsibility. (3) Explanations seek to lower the sense of condemnation for the act by identifying extenuating circumstances that contributed to the negative event. With explanation, the accused does not completely deny responsibility for the event but implies that the event is, for the reasons they provide, not a permanent reflection on their motives, skill, character or conduct (e.g, "I had a bad day"). Explanations are successful to the extent that they encourage external attributions for the dissatisfying event.

Valence Excuses. (1) Minimization excuses aim to underestimate the perceived unpleasantness of the event by suggesting that the event was not quite as bad as the complainer describes it (e.g., "its not that bad"). Minimization excuses do not deny that an error has occurred or that the organization is responsible. Instead, minimization excuses assert that the event was not so bad after all. (2) Justification excuses also assume responsibility for the event but go beyond minimization by boldly asserting that not only was the event not so bad but it was, in fact, quite good and that the customer is better off as a result (e.g., "Its better for everyone if you wait your turn"). Justification is the claim that, contrary to the perceiver's opinion, the action taken was a positive one.

In an earlier study (forthcoming), we found considerable support for this excuse classification scheme. In that study, students were asked to write complaint letters to companies who had recently caused them genuine dissatisfaction. Company replies were analyzed in the context of the excuse framework discussed above. Two judges coded excuses in terms of the five different excuse types. We found that 51.9% of the firms who responded to the written complaint letter did so by offering at least one excuse. In fact, companies who gave an excuse averaged 1.45 excuses per letter. The most common excuse type given was deflection excuses (47.4% of all excuses), followed by explanations (30.9%), denials (12.4%), justifications (5.2%) and minimizations (4.1%).

HYPOTHESES

In the present study we examined the impact of the different excuse types that we had collected earlier. Subjects rated each excuse in terms of a) their acceptability, b) the extent to which the excuse maker was blameworthy, and c) the amount of harm experienced by the complaining party. Based upon excuse theory, hypotheses were formed. We propose that linkage and valence excuses should effect consumers differently. Because they challenge the perceived negativity of the event, valence excuses should primarily influence the perceived positiveness of the event.

H1: Valence excuses would cause the event to be perceived as less harmful than linkage excuses.

Linkage excuses do not attack the valence of the act. Instead, they seek to diminish the extent to which one is perceived as connected to that negative event.

H2: The excuse maker should be perceived as less blameworthy after a linkage excuse than a valence excuse.

Specific hypotheses can also be made regarding comparisons between the different valence excuses and different linkage excuses. Justification excuses go beyond minimization excuses by asserting that not only was the event not too bad, but that the customer is actually better off as a result of the company's actions.

H3: Between the two types of valence excuses, justifications should result in less perceived harm than minimizations.

Explanations admit one contributing role in the undesirable event, but deny that this has any bearing on one's ability to do well in the future.

H4: Among the three types of linkage excuses, explanations should result in the highest ratings of blame.

Deflections are more effective than denials because rather than merely disavowing one's association with the negative event, deflections name the culprit who was responsible.

H5: Among the linkage excuses, deflections should result in the lowest ratings of blame.

METHOD

Collecting Organizational Excuses

To test our hypotheses, we used genuine excuses actually offered by businesses to complaining customers. These excuses were collected in a study previously described. In that study, we asked students to write letters of complaint to companies who had recently caused them genuine dissatisfaction. Students complained about a variety of different products and services representing an extensive range of prices ($5.00 - $2000.00). A content analysis of the company responses was conducted to determine the types of excuses provided. The coding scheme consisted of identifying excuse statements and categorizing them by excuse type (denial, deflections, explanations, minimizations, and justification). Two business-school graduate students were trained as coders. After a brief practice session, all company response letters were coded by both judges. The inter-judge reliability for the five excuse types ranged from 71% (deflections) to 98% (denials). When the coders disagreed, a panel of two faculty members jointly categorized the excuse type.

Questionnaire Design and Procedure

Since the excuses we received covered a broad range of products, services and price ranges, we modified the excuses to fit one complaint situation. For this purpose, we created an automobile repair scenario. The scenario described a customer, whose warranty had just expired, seeking service from an automobile dealership. The customer was charged for labor and replacement parts. However, the customer claimed that the repairs failed to fix the car. The customer complained because additional charges will now be incurred to repair what was wrong in the first place. Each excuse that we collected was adapted to this scenario. However, excuses whose meaning was unclear or implausible were eliminated. In cases in which the base message was reiterated by a previous excuse, the most typical excuse of that type was used. A total of 37 statements (types of excuses) was developed in this manner. Table 1 shows each excuse statement used in the present study. Of these 37 excuses, 2 were justifications, 4 were minimizations, 5 were denials, 13 were deflections and 13 were explanations.

A questionnaire was then constructed and administered to a total of 51 junior and senior level business students enrolled in a marketing course. Subjects were first instructed to read the description of the car repair dissatisfaction incident. Following this reading, each of the 37 excuse statements was presented. Subjects were asked to rate the 37 statements on a scale of 1 (not at all) to 7 (completely) for the following three items: 1) "To what extent was the customer harmed by this experience?" 2) "To what extent was the dealership deserving of blame?" and 3) "To what extent should the customer accept this answer?" Item 1 was chosen to measure the extent that the company response impacted the perceived valence of the dissatisfaction. Item 2, on the other hand, was a measure of the impact on the perceived linkage between the dissatisfying act and the car dealership. Additionally, item 3 assessed how acceptable these excuses were to the rater.

TABLE 1

LIST OF MODIFIED EXCUSES FROM COMPANY REPLIES TO CUSTOMER COMPLAINTS

TABLE 2

MEAN RATINGS FOR ALL EXCUSES

RESULTS

Table 2 shows the average harm, blame and acceptance ratings for each of the 37 different excuses used in the present study. The table shows considerable variation between the different excuses on all three dependent variables (harm, blame and acceptance). Harm ratings ranged from 3.37 ("We shared your concerns with our staff") to 6.08 ("The mechanics were inexperience but are now more seasoned"). Blame ratings ranged from 3.78 ("Replacing worn parts before they fail helps keep our customers safe") to 6.40 ("The mechanics who worked on your car were inexperienced but are now more seasoned"). Acceptance means ranged from 1.60 ("Our mechanic was sick that day and did not perform at his best") to 4.60 ("We shared your concerns with our entire staff").

To test the hypotheses, a series of a priori contrasts were conducted between the different excuse types for each dependent variable. The first hypothesis states that valence excuses would lead to less perceived harm than linkage excuses. Table 3 shows the mean harm, blame and acceptance ratings for the combined valence (minimization and justification) and combined linkage excuses (denial, deflection, and explanation). The table indicates that the customer was perceived as harmed less after a valence (M = 4.11) than a linkage excuse (M = 4.95, t(43) = 6.13, p < .001). Hypothesis 1 was thus supported. According to hypothesis 2, linkage excuses should have resulted in less attributed blame than valence excuses. Table 2 indicates that not only was this hypothesis not supported, but the means were in the opposite direction: Linkage excuses resulted in more blame than valence excuses: M = 5.33 and M = 4.57, respectively, t(49) = 6.27, p < .001.

TABLE 3

MEAN RATINGS FOR LINKAGE AND VALENCE EXCUSES

TABLE 4

MEAN RATINGS FOR FIVE EXCUSE TYPES

Inspection of the top portion of Table 4 indicates support for hypothesis 3. Justifications resulted in less perceived harm (M= 3.59) than minimization excuses (M = 4.38, t (49) = 4.84, p < .001).

Between the linkage excuses, it was predicted that explanations would result in the most blame (hypothesis 4) and that deflection would result in the least blame (hypothesis 5). As illustrated in the bottom portion of Table IV, both of these final hypotheses were supported. Explanations resulted in more blame (M = 5.59) than denials (M = 5.38, t (49) = 2.05, P < .05). Denials resulted in more blame than deflection: M = 5.38 and M = 5.05, respectively (t (49) = 3.18, P < .01).

Although no hypothesis concerning the acceptability measure were made, means tests were also conducted to determine differences in acceptability. Table 2 indicates that valence excuses were perceived as more acceptable (M = 3.32) than linkage excuses (M = 2.87, t (49) = 4.18, p < .001). Justifications were the most acceptable excuse of all and differed significantly from minimizations (M = 3.14, t (49) = 3.44, p < .001). Denials (M = 2.57) and explanations (M = 2.63, ns) were the least acceptable excuses. Both were accepted to a significantly less extent than deflections (M = 3.22), t (49) = 5.45, p < .001 and t (49) = 3.18, p < .01, respectively (see Table 4).

DISCUSSION AND IMPLICATIONS

This paper presented a framework for studying organizational excuse making based upon a psychological theory of excuses. We have argued that excuses work by "shifting causal attributions" (Higgins and Snyder , 1991, p. 74). Excuses are communication strategies that are aimed at soothing or pacifying dissatisfied customers by providing plausible causal attributions designed to alleviate their displeasure. These attributions, in turn, may moderate the customer's perception of the actual harm done or of who or what was responsible for the bad outcome. The basic premise of this framework is that people will react differently to a transaction outcome depending upon the type of attribution they make. An examination of the effects of the thirty-seven different excuses provides evidence that excuses can have a striking impact on consumers' perceptions. All subjects in our study ware exposed to an identical scenario describing a customer dissatisfied with their automobile service. Yet, widely differing ratings of harm, blame and acceptance were obtained depending on the particular excuse. Thus, different excuses can cast the same situation in vastly different lights.

Using this framework, excuses can be classified as either linkage or valence excuses. Linkage excuses (denial, deflection and explanation) were purported to diminish the extent to which the accused actor was connected to the negative event. Valence excuses (minimization and justification) reduce the perceived negativity of the event.

Results provide support for four of the five stated hypotheses. As we had anticipated, linkage and valence excuses provoked different reactions among our research participants. Valence excuses led to perceptions of less harm than linkage excuses (H1). The least amount of harm was associated with justification excuses (H3). Finally, among linkage excuses, explanations resulted in the highest ratings of blame and deflection excuses resulted in the lowest ratings of blame (H4 & H5).

The combination of reducing the perceived harmfulness and blameworthiness of the event was also associated with higher ratings of acceptability. Even so, the highest mean acceptability rating that we obtained (justification excuses) were still below the mid-point of the rating scale (M = 3.68 on a 7-point scale), indicating that subjects were reluctant to accept any excuse to a great degree. Although it seems likely that lower blame and lower harm ratings caused higher levels of acceptability, it may also be the case that higher excuse acceptability ratings influenced perceptions of harm or blame. An experimental design is needed to examine the causal relationship between these variables.

When faced with an angry customer there is a natural tendency to give an excuse. Our findings indicate that the various types of excuses offered may have different effects. Managers need to be cognizant of this and train employees to respond with excuses in an appropriate manner. Secondly, we suggest that excuses could be combined with speedy response, apologies, compensation and or other forms of financial restitution to improve chances of reinstating a dissatisfied customer. For example, if a customer is complaining about a partially filled bag of chips, the company could respond with an apology for the dissatisfaction, coupons to recapture the customer, and a justification (partially filled bags are better because the chips are more prone to breakage if placed in a smaller bag) to reduce the perceived harm. This type of three pronged response would have a better chance of succeeding in the long run than merely offering coupons. Excuses can serve an additional function of reducing the perceived harm as well as educating the consumer and possibly even modifying future expectations to be more realistic.

The results also suggest that explanations and deflections were not very effective in terms of alleviating perceptions of harm and blame and were not very acceptable. Yet, in our preliminary study, these two types of excuses were found to be, by far, the most common types of excuse responses (deflection represented 47.4% and explanations 30.9% of all excuses). In light of our findings, managers would be wise to re-evaluate their use of these two types of excuses.

Hypothesis 2 was not confirmed by our study. According to hypothesis 2, linkage excuses should have resulted in less attributed blame than valence excuses. Contrary to our expectation, valence excuses resulted in more blame than linkage excuses. Thus valence excuses were more effective than linkage excuses in alleviating both harm and blame. According to excuse theory, valence and linkage should be independent judgements that have a differential impact on perceptions of harm and linkage. However, the present study found that harm and blame ratings for four of the five excuse types were highly correlated (correlations ranged from .64 to .81). Explanations were the one type of excuses where blame and harm were not highly correlated (.26). In retrospect, the correlation between harm and blame should not have been surprising. As harm decreases, it seems natural that there should be a corresponding decrease in blame. Without harm, there is little to find blameworthy. Despite these results, we would still conjecture that linkage and valence are independent. Perhaps this was not the case in the present study because, regrettably, we measured blame instead of linkage. Blame and linkage are two different constructs. Blame is an attribution made after the perceiver assesses and does not accept the validity of the excuse (Shaver 1985; Shaver and Drown 1986; McGraw 1987). Blame is a judgment of disapproval and fault. It involves accusations and criticism. By contrast, linkage reflects the extent to which an internal rather than an external attribution can be made for the cause of an action (Higgins and Snyder 1991). It is independent of the severity of any consequences that action produces and makes no recriminations. From these perspectives, linkage and blame are quite different. Although harm and blame are related, harm and linkage should be independent. For example, consider justifications (valence excuse). Justification excuses represent a high degree of linkage and a low level of blame. The accused admits that the act was internally caused (high linkage) but denies that the action was wrong or the outcome was harmful. By contrast, explanations involve a high degree of harm because there is no denial of the harmful consequences, but seek to provide external attributions (low linkage) for those harmful consequences.

This paper provides evidence that excuses serve a worthwhile function after a dissatisfying experience. Clearly business organizations in our study responded to complaining customers with excuses. They showed an obvious preference for deflections followed by explanations. The results showed that to mitigate the negative effects of the customer's displeasure, events should be managed to make the act appear less disagreeable. These results support findings by Giacalone and Pollard (1987) who found that personnel managers in their study who were accused of breaching the confidentially of an employee could temper the negative repercussions of the breach by down playing the actual harm suffered. Smith and Whitehead (1988) point out, however, that this strategy will be effective only to the extent that the audience for the excuse does not have knowledge regarding the actual negative consequences.

Business should be strategic in their use of excuses. If honored, they can successfully present the organization in a favorable light. Although it has not been fashionable in marketing to talk about organizational excuse making, excuses are not necessarily lies. As Mehlman and Snyder (1985) point out, excuses tend to operate in gray areas where there are few objective yardsticks for measuring the "truth" about varying explanations of events. Garrett, Bradford, Meyers and Becker (1989) argue that organizations should give excuses. There is a popular misconception, they maintain, that the company is always wrong and the consumer is always right. Customers have the right to protest or complain, businesses have the right to respond. Organizational excuses fulfill a vital function. They allow the customer to weigh all information before passing judgment. By providing information concerning what went wrong, excuses serve to educate consumers about the complexities of the marketplace. In turn, consumers can use this information to make informed decisions about repeat purchases and how forgiving they should be.

REFERENCES

Clark, Gary L. and Peter Kaminski and David R. Rink (1992), "Consumer Complaints: Advice on How Companies Should Respond Based on an Empirical Study," Journal of Services Marketing, 6, (Winter), 41-50.

Desatnick, Robert L. (1987), Managing to Keep the Customer, San Francisco CA: Jossey-Bass Publishers.

Folkes, Valerie (1984), "Consumer Reactions to Product Failure: An Attributional Approach," Journal of Consumer Psychology, 13, (March), 534-539.

Garrett, Dennis E., Jeffrey L. Bradford, Renee A. Meyers and Joy Becker (1989), "Issues Management and Organizational Accounts: An Analysis of Corporate Responses to Accusations of Unethical Business Practices, Journal of Business Ethics, 8, 507-520.

Garrett, Dennis E., Renee A. Meyers, and John Camey (1991), "Interactive Complaint Communication: A Theoretical Framework and Research Agenda, Journal of Consumer Satisfaction, Dissatisfaction, and Complaining Behavior, 4, 62-79.

Giacalone, Robert A. and Hinda G. Pollard (1987), "The Efficacy of Accounts for a Breach of Confidentiality by Management," Journal of Business Ethics, 6, 393-397.

Gilly, Mary and Betsy D. Gelb (1982), "Post-Purchase Consumer Processes and the Complaining Consumer," Journal of Consumer Research, 9, (December), 323-328.

Goffman, E. Behavior in Public Places, (1963), New York: Free Press.

Goodwin, Cathy and Ivan Ross (1990), "Consumer Evaluations of Responses to Complaints: What's Fair and Why," Journal of Services Marketing, 4, (Summer), 53-61.

Higgins, Raymond L. and C. R. Snyder (1990), "The Business of Excuses," in R. A. Giacalone and Paul Rosenfield (Eds.), Impression Management in the Organization (pp 73-85).

Higgins, Raymond L. and C. R. Snyder (1991), "Reality Negotiation and Excuse-Making," in Handbook of Social Psychology: The Health Perspective, eds. C. R. Snyder and R. Donelson, New York: Pergamon Press, 79-95.

Hill, Donna, Robert Baer and Rustan Kosenko, "Organizational Characteristics and Employee Excuse Making: Passing the Buck for Failed Service Encounters," in Advances in Consumer Research, Vol. 19, eds. John F. Sherry and Brain Sternthal, Provo, UT: Association for Consumer Research, 673-677.

Krentler, Kathleen and Robert Cosenza (1987), "Redress Response and its Effects on Company Impression," Public Relations Review, 13, (Fall), 33-45.

McGraw, Kathleen (1987), "Guilt Following Transgression: An Attribution of responsibility Approach," Journal of Personality and Social Psychology, 53, 247-256.

Mehlman, Rick C., and C. R. Snyder (1985), "Excuse Theory: A Test of the Self-Protective Role of Attributions," Journal of Personality and Social Psychology, 49, 994-1001.

Schonbach, Peter (1980), "A Category System for Account Phases," European Journal of Social Psychology, 10, 195-200.

Scott, Marvin B. and Stanford M. Lyman (1968), "Accounts," American Sociological Review, 23, 46-62.

Shaver, Kelley (1985), The Attribution of Blame: Causality, Responsibility and Blameworthiness, New York: Springer-Verlag.

Shaver, Kelley and Debra Drown (1986), "On Causality, Responsibility, and Self-Blame: A Theoretical Note," Journal of Personality and Social Psychology, 50, 697-702.

Smith, Stephanie H. and George I. Whitehead III (1988), "The Public and Private Use of Consensus-Raising Excuses," Journal of Personality and Social Psychology, 56, 355-371.

Snyder, C. R. (1985), "The Excuse: An Amazing Grace?" In ed. B.R. Schlenker, The Self and Social Life (pp 235-260), New York: McGraw- Hill.

Snyder, C. R. and Raymond L. Higgins (1988), "Excuses: Their Effective Role in the Negotiation of Reality," Psychological Bulletin, 104, 23-35.

Snyder, C. R., Raymond L. Higgins, and Rita J. Stucky (1983), Excuses: Masquerades in Search of Grace, New York: John Wiley and Sons.

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