Proceedings of the Third Annual Conference of the Association for Consumer Research, 1972 Pages 417-427
CONSUMER REACTION TO RESTRICTION OF CHOICE ALTERNATIVES
Michael B. Mazis, University of Florida
Robert B. Settle, University of Florida
[The research reported in this paper was supported by the College of Business Administration, University of Florida. The authors are indebted to Dennis C. Leslie for assistance in questionnaire design and in data processing.]
[Assistant Professor of Marketing, University of Florida and Assistant Professor of Marketing, San Diego State University, respectively.]
On January 1, 1972, Dade County Florida began prohibiting the sale, possession, or use of any detergent or cleaning agent containing phosphates. For the next several weeks, Miami housewives who visited their local supermarket to purchase laundry detergent found that their choice alternatives were diminished from a few dozen brands to two or three, and in some cases there was no choice whatsoever. There appeared to be three basic responses which the consumer could make to this diminished choice situation: (1) she might react negatively toward the new, no-phosphate detergent, (2) she might prefer the new detergent to the phosphate detergent, or (3) she might be indifferent to the old and the new, regardless of the diminished choice.
The dramatic changes in the consumer's choice alternatives created by the legal restraints against phosphates appeared to provide an opportunity to test the theory of psychological reactance in the field situation, and in relation to a significant marketing problem. The formulation and investigation of psychological reactance has been associated principally with experimental research and in the laboratory, and this association has necessarily created some questions concerning the external validity and generalizability of the concept. A study of consumer reaction to real choice limitations in the market appeared appropriate.
The theory of psychological reactance, as formulated by Brehm (1966) states that it is very important to the individual to maintain his choice alternatives. If he is able to do so, he will be better able to maximize the rewards he obtains from his behavior. As a consequence of this association between the number of alternatives available and the reward value of action, the individual tends to prefer a wide variety of alternatives, and learns to protect these options even under circumstances where he may not wish to exercise them. Thus, the diminishment of a person's choice alternatives will lead to a motivational state such that the person will try to reinstate the lost alternatives; he will respond in a reactive manner.
Several examples of reactive behavior are easily available to the casual observer, and they make the concept intuitively clear. The small child, told to avoid the oven door because it is hot, will often proceed immediately to put his hand on the stove. The woman who enters an ice cream shop and is told that guava and rhubarb sherbet is not available today might be heard to mutter about today being the day she was going to try it. In the extreme situation, a person who is forced to accept the alternative he would otherwise have selected may try to refuse it and attempt to obtain an alternative he would never have chosen under a free choice situation. All are examples of reactive behavior.
Antecedent and Consequent Conditions of Reactivity
The degree of reactance experienced by an individual is affected by a variety of factors. Principal among them are: (1) the strength of the rewards, mediated by the behavior alternatives, (2) the number of alternatives eliminated and the number remaining, (3) the perceived power and intent of the force which limits the choice alternatives, and (4) the perceived threat to the remaining alternatives and to those in other choice situations. In addition to these factors, the propensity to be reactive might prove to be an individual difference variable operative over a wide variety of situations.
While the factors listed constitute the antecedent conditions of a motivational state termed reactance, there are several behavioral possibilities open to the individual which can be recognized as the consequent conditions. The most common modes of response to reactance are: (1) to attempt to reestablish the alternatives which were removed, (2) to re-evaluated upward the value of foregone alternatives and re-evaluate downward the value of the remaining alternatives, (3) to exercise alternatives in other choice situations parallel to those which were removed, (4) to downgrade the "necessity" or reason for the choice limitation, (5) to manifest increased negativity and opposition to the entity which removed the choice alternatives, and (6) to encourage other people to exercise the denied choice alternatives.
To the extent that the antecedent conditions of reactance are present in the Miami area after the ban on phosphate detergents, the shopper can be expected to react along the lines described as consequent conditions, if reactance theory correctly describes the psychological process associated with a diminishment of choice. Other psychological theory, however, suggests rather different modes of response. The theory of cognitive consistency predicts that the shopper in this situation would tend to accept the new no-phosphate detergent and perhaps prefer it because she has learned of the contribution of phosphates to pollution. The law might be seen as an input of new information, and the negative pronouncements concerning phosphates would lead the consumer to down-grade the old and up-grade the new detergent. Similarly, she might exaggerate the threat created by phosphates and to enhance her positive opinion of the authority which has removed the phosphate detergent from the market,
For a variety of reasons, the consumer may manifest behavior which supports neither psychological theorY. For example, she may perceive little difference between phosphate and no-phosphate detergent because she perceives few differences among brands of detergent, because the item is not particularly important or expensive, or because she is not ego-involved in a socially invisible good. Any such reason for detachment from this situation may lead to a failure to discriminate.
THE FIELD SURVEY
A five part questionnaire was constructed, with the first section designed to obtain the necessary information on the brand and type of detergent being used by the respondent presently and in the past. If the respondent indicated use of a no-phosphate detergent, she was also asked why she had switched, to measure whether or not she perceived herself as having a choice. This section also obtained the respondent's opinion concerning the amount of no-phosphate detergent and amount of extra ingredients used, compared to that used with the old deterRent.
The second section of the questionnaire obtained the respondent's ratings of the phosphate and no-phosphate detergent on seven evaluative criteria, together with an overall rating of each. Ratings were made separately, and no comparison was solicited or suggested. Obviously, only those respondents using no-phosphate detergent were asked to rate this product. Respondents were also asked in this section what products were banned in Dade County, to measure their awareness of the law. They were also asked their feelings concerning the law banning the sale or use of detergents containing phosphates. Lastly, this section of the questionnaire obtained information about stocking up (hording) and about the purchases outside of Dade County (smuggling).
Section three of the questionnaire dealt with the respondent's attitudes toward: (1) water pollution, (2) the contribution of phosphates to pollution, (3) the role of the government in protection from pollution, and (4) legal restrictions against the sale of phosphate detergent. Respondents were given a card containing a series of five statements on these topics and asked to indicate if they strongly agreed, agreed, were neutral, disagreed, or strongly disagreed with each. A second card contained four statements concerning attitudes toward washing, and the respondent answered in a similar manner.
The fourth section of the questionnaire solicited information about the respondent's values and personality. First she was given a card containing five social goals or objectives: equality, freedom, family security, a clean environment, and social recognition, adopted from Rokeach's Value Survey (1968). Each was followed by a short explanatory phrase, and the respondent was asked to rank them in their order of importance to her. Second, she was given a card containing five pairs of statements taken from Rotter Internal-External Locus of Control Scale (1966). Each pair contained one statement identifying an internal attribution and one statement of an external attribution. Respondents were asked to indicate which of each pair they believed most strongly.
The last section of the questionnaire was concerned with the respondents judgement of two pieces of terrycloth material. The two were identical, however, one was clearly labeled "phosphate" and the other, "no-phosphate." The respondent was told that both had been washed several times, each in a different kind of detergent. She was asked to judge which was cleaner, which was softer, and which was fresher. Finally, the necessary demographic data was obtained.
The data were obtained in personal interviews in the subjects' homes. The four female interviewers employed in Miami and the two in Tampa completed 84 and 45 interviews, respectively. The interviewers were instructed in the use of the questionnaire and the nature of the study, but remained ignorant of the specific hypotheses. They were permitted to select a judgement sample within constraints on age, sex, family, geographic area, dwelling area status, and house type. The data were collected within eight weeks after the law became effective.
RESULTS AND CONCLUSIONS
This report focuses on the patterns of interaction between three main elements: area of residence (Miami versus Tampa), respondent attitudes, and perceptual distortion. In addition to these, differential patterns of response are examined between those in the Miami group who were required to switch brands because their favorite brand was no longer available and those who were able to maintain brand continuity. The relationship between the factors cited above and the two personality variables, value orientation and locus of control, were also investigated; however, these did not prove to be systematically related. Several elements did prove sufficiently sensitive, however, to test the basic propositions of the theory of psychological reactance.
The theory of psychological reactance suggests that limitations in choice alternatives might lead to a motivational state to re-establish the alternatives. If this is not possible, one would expect the frustration to result in attitude change. Table 1 depicts an index of attitudes toward washing clothes resulting from answers to four questions, for respondents in Miami and those in Tampa. Only a third of Miami residents maintained positive attitudes toward washing, while nearly twice that proportion in Tampa were favorable. The difference in the distribution proved to be significant
INDEX OF ATTITUDES TOWARD WASHING, BY AREA GROUP
The attitude of respondents toward government protection of clean water is displayed in Table 2. While only five percent of Tampa residents were hostile toward the government's role, a quarter of Miami residents so indicated. Reactance theory suggests that negativity toward the source of choice deprivation is a consequent condition of the motivational state, thus these results tend to support a reactance interpretation.
ATTITUDE TOWARD GOVERNMENT ROLE, BY AREA GROUP
The attitude of respondents toward legal restrictions against phosphate detergents are depicted in Table 3. Again, those in the choice deprivation condition manifested significantly more negative attitudes, supporting the hypothesis that they have experienced reactance.
ATTITUDE TOWARD LEGAL RESTRICTIONS, BY AREA GROUP
Tables 4, 5, and 6 contain the results of the questions concerning which piece of material was cleaner, softer, and fresher. One was labeled as having been laundered in phosphate detergent and the other in non-phosphate detergent. Table 4 indicates the distributions of response are significantly different, and the Miami group shows a higher proportion selecting the no-phosphate material as cleaner. Since this result, if different from chance, would be antithetical to a reactance explanation, it was examined further. The chi square value for the distribution of response without the "same" category did not prove to be significant, and consequently it can be concluded that the differences in the distributions result from the fact that more than twice as many of the Tampa respondents felt both pieces of material were equallY clean.
MATERIAL JUDGED CLEANER, BY AREA GROUP
Table 5 also indicates significantly different distributions of response between Miami and Tampa, however, in this case, where respondents judged the softness of the two pieces of identical material, Miami residents chose the phosphate detergent product in a markedly higher proportion. When the "same" category was eliminated, and a two-by-two chi square was obtained, it proved to be significantly different beyond the .01 level. Thus, a significantly larger proportion-of-Miami residents selected the phosphate detergent product as superior, and this constitutes support for the reactance theory explanation.
MATERIAL JUDGED SOFTER, BY AREA GROUP
The freshness of the material was judged in the third question of this section, and the results are shown in Table 6. The distributions are significantly different with only marginal certainty, however, the table indicates that again a larger proportion of Miami respondents perceived a difference in the identical pieces of material. It appears that the Miami group, having been exposed to information concerning the relative effectiveness of the two types of detergent, were more likely to discriminate between detergents. In one case, for the judgement about softness, respondents were significantly more favorable toward the phosphate detergent, as reactance theory would predict.
MATERIAL JUDGED FRESHER, BY AREA GROUP
Tables 7 and 8 contain the data on perceptual distortion cross-tabulated with the respondents' attitudes on the danger of phosphate pollution. Table 7 contains the distributions of response on which material was judged to be cleaner. A higher percentage of those who felt that phosphates did not constitute an important threat also perceived the phosphate detergent-laundered material as being cleaner. Table 8 depicts the distributions on the material judged softer, according to the same attitude. Nearly all those who agreed phosphates were not an important threat also perceived the phosphate-laundered material as softer. Neither result constitutes a test of reactance, however, both demonstrate the consistency of attitudes and percePtual distortion.
MATERIAL JUDGED CLEANER, BY ATTITUDE TOWARD PHOSPHATE
MATERIAL JUDGED SOFTER, BY ATTITUDE TOWARD PHOSPHATE
Within the Miami experimental group, nine of the 86 respondents continued to use phosphate detergents when the survey was taken. These individuals either hoarded the product or smuggled it in from a nearby county. Thus, over ten percent of those interviewed can be regarded as highly reactive, since they risked possible prosecution to maintain their choice alternatives.
Of those respondents from Miami using no-phosphate detergent, some found it necessary to switch brands because their favorite brand was no longer available. The vast majority of these switchers were customers of either Colgate or Proctor and Gamble, both of whom failed to provide a no-phosphate substitute carrying the same brand names as the popular phosphate products. Lever Brothers did provide such a substitute, and their customers constitute a major part of those who maintained brand continuity while switching to a no-phosphate product.
Respondents' estimates of the amount of detergent used per washload, the amount of extra ingredients used, and the cost of no-phosphate detergent were divided according to these "switcher" and "non-switcher" categories. The data are presented in Tables 9, 10, and 11. Table 9 indicates that nearly a third of those who had to_s itch brands felt the no-phosphate detergent required more per washload, while only four percent of the non-switchers thought so. Table 10 indicates a similar disparity regarding the amount of extra ingredients used. Table 11 indicates that over two thirds of the switchers believed no-phosphate detergent cost more to use while less than a third of the non-switchers felt this was the case.
It should be noted that in the case of both the switcher and the non-switcher, the consumer is using a totally new product, however, in one case she is also using a new brand, while in other she maintains brand continuity. The differentiaL response according to brand continuity indicates the importance of the brand to the consumer, and in this case the loyalty appears to override the radical changes in the physical and chemical contents of the product, even when this change is obvious to the consumer.
The results of this analysis tend to support the reactance theory because non-switchers perceived little limitation in their choice, while switchers were obviously limited. These differences in choice alternatives were directly manifested in evaluations of the amount of detergent used, the extra ingredients used, and the cost of no-phosphate detergent. In all three cases the differences are as predicted by reactance theory.
AMOUNT OF DETERGENT USED BY BRAND CONTINUITY
AMOUNT OF EXTRA INGREDIENTS USED BY BRAND CONTINUITY
COST OF NON-PHOSPHATE DETERGENT BY BRAND CONTINUITY
If switchers are more reactive than the non-switchers, one would expect them to be more negative toward the law which limited their choices. Attitudes toward the law, by brand category, are presented in Table 12. While the majority of switchers favored the law, the proportion was significantly lower than the 91% of non-switchers who also indicated a favorable attitude. In this case, even though the data-provide only marginal certainty of a significant difference, the results appear to be supportive of the theory of psychological reactance explanation of behavior by Miami respondents.
ATTITUDE TOWARD THE LAW BY BRAND CONTINUITY
This report focused on differential patterns of response to questions concerning perception, attitudes, and evaluations, according to the respondent's area of residence and brand continuity. It appears that the theory of psychological reactance received general support from the data, and none of the results of the study directly contradicts the predictions of the theory. In view of the relative unimportance of laundry detergent, compared to other consumer products, the existence of signs of psychological reactance is indicative of the consumer's need to maintain her choice alternatives in the market. It seems likely that consumer reactance would be magnified directly with the increase in the number and kind of limitations imposed on consumer choice, and with the importance of the products subjected to restriction.
Jack W. Brehm. A Theory of Psychological Reactance New York: Academic Press, 1966.
Milton Rokeach, Beliefs, Attitudes and Values. San Francisco: Jossey Bass, Inc., 1968, 156-178.
Julian B. Rotter. Generalized Expectancies for Internal Versus External Control of Reinforcement. Psychological MonoGraphs, 1966, 80 (whole no. 609), 1-28.