Decision Rule Uncertainty, Evoked Set Size, and Task Difficulty As a Function of Number of Choice Criteria and Information Variability

ABSTRACT - If consumer brand choices are implemented in terms of decision rules, applying more choice criteria to information which contain more variability and a lower mean level should, (1) increase the uncertainty of constructing an effective decision rule, (2) increase the likelihood of achieving a smaller evoked set, and (3) decrease the difficulty in evaluating brands within a product class. Three 2 x 3 factorial experiments to test these hypotheses are reported. Results support the hypotheses.

Citation:

Joseph J. Belonax, Jr. (1979) ,"Decision Rule Uncertainty, Evoked Set Size, and Task Difficulty As a Function of Number of Choice Criteria and Information Variability", in NA - Advances in Consumer Research Volume 06, eds. William L. Wilkie, Ann Abor, MI : Association for Consumer Research, Pages: 232-235.

Advances in Consumer Research Volume 6, 1979      Pages 232-235

DECISION RULE UNCERTAINTY, EVOKED SET SIZE, AND TASK DIFFICULTY AS A FUNCTION OF NUMBER OF CHOICE CRITERIA AND INFORMATION VARIABILITY

Joseph J. Belonax, Jr., Western Michigan University

ABSTRACT -

If consumer brand choices are implemented in terms of decision rules, applying more choice criteria to information which contain more variability and a lower mean level should, (1) increase the uncertainty of constructing an effective decision rule, (2) increase the likelihood of achieving a smaller evoked set, and (3) decrease the difficulty in evaluating brands within a product class. Three 2 x 3 factorial experiments to test these hypotheses are reported. Results support the hypotheses.

PROBLEM

Chaffee and McLeod (1973) present a generalized model that conceptualizes the consumer as an information processor and decision maker. The consumer is seen as making a buying decision according to some decision rule which combines information from an attribute-by-brand matrix. Within the context of this broad model, research on the decision rule aspect of information processing has taken many forms (summarized in Bettman, 1977).

Most often the studies involve fitting one or more decision rule models to brand evaluations made by persons who have been exposed to an attribute-by-brand Matrix of information. The somewhat mixed results of studies attempting to fit various models may reflect individual and situational differences in those studies.

The consideration of which internal factors affect the choice of decision rule is well researched in the marketing literature. Park (1976) has considered prior familiarity. Wright (1975) considered the effect of such factors as the desire to simplify, and the desire to optimize choice strategies. Popielarz (1976) suggests that breadth of categorization as one dimension of cognitive style may also determine the relevant processing rule.

Previous investigations on environmental factors have focused on what effects the actual or perceived situation have on the choice of decision rule (Hansen, 1976). Studies on the actual situation have focused on such factors as, information load (Jacoby, Speller, and Kohn, 1974), task complexity (Reilly, and Holman, 1978), time pressure and distractions (Wright, 1974), structure and format of product information (Van Raaij, 1977). Effects of the perceived situation have centered on such variables as, perceived risk (Bettman, 1973), and perceived alternatives (Howard and Sheth, 1969).

Although individual differences may in part, account for mixed results of studies attempting to fit various models to brand evaluations, the distinction between the actual and perceived situation may be the more important factor to consider.

First, the distinction suggests that attempts to fit a single model to a total decision process, especially when the buying situation has one or more relatively unfamiliar aspects, may be overlooking an important intermediate stage. The concept of evoked set suggests that choices are made only after the consumer has constructed a set of "acceptable" brands from his awareness set (Narayana, Markin, 1975). The implied two process model, with information first used to form a set of perceived alternatives (evoked set) is consistent with the results reported by Pras and Summers (1975). They found the conjunctive model to be the best predictor of preference rank order when all alternatives were considered, but the linear compensatory model to predict best when only the acceptable alternatives were considered.

Second, the distinction suggests that any attempt on the part of the consumer to construct a decision rule to evaluate brands may depend more on how the situation is perceived rather than on the actual stimuli present in the environment. As Bettman (1973) has pointed out in his theoretical model of inherent risk, the likelihood of constructing an effective decision rule depends upon the perceived distribution of quality over the brands in a product class.

Third, the distinction between perceived and actual situations is related to the distinction between anticipated and actual situations (Hansen, 1976). Since we believe that peoples' perceptions govern their behavior it is important to have measures for the before (perceived uncertainty) and the after (difficulty of the evaluation task).

In summary, individual factors are likely to influence consumer perceptions of the environment. In turn, these perceptions may influence the construction of a decision rule to evaluate brands, determine which of the brands will be perceived as acceptable alternatives, and ultimately affect the difficulty encounter in the evaluation task. This study will examine what effect the distribution of quality over brands in a product class and the number of choice criteria used to evaluate the brands have on the consumers' perceived uncertainty of rule construction, size of the evoked set, and degree of task difficulty.

HYPOTHESES

Although all consumers may be exposed to the same set of alternative brands in an experimental setting or in the market place, the distribution of quality over brands in a product class is likely to influence the consumers' perceived distribution of quality. According to the decision rule component in Bettman's (1973) inherent risk model, the greater the perceived variation in quality and the lower the mean level of quality distribution the less the likelihood of constructing an effective decision rule. Initially, consumers should feel less certain they can develop a rule to evaluate brands when the brands differ markedly in quality than when they differ very little. Further, Wright (1975) suggests that an anticipated increase in information load should increase the consumer's need to simplify his choice strategy. Initially, consumers applying more choice criteria to their awareness set should feel a greater need to engage in a simplified choice strategy. In turn, they should feel more certain that such a decision rule can be constructed.

Since the purpose of evoked set construction is to simplify the ultimate choice process, it seems likely that a person would eliminate brands which failed to meet some minimum level on one or more evaluative criteria. It would follow that, as the variability in quality increased and the general level of their desirability decreased, the likelihood of finding a reason for eliminating brands would increase. Further, as the consumer applied a larger number of evaluative criteria to the awareness set, the likelihood of finding a reason for eliminating brands should be greater.

A greater variability of attribute values facilitates the actual differentiation of the choice alternative on that attribute. We may expect an increase in quality variability and decrease in the general level of desirability, to increase the likelihood of differentiating the brands on their attributes. This being the case, the difficulty in evaluating the brands should decrease. Further, consumers applying more evaluative criteria are likely to be engaged in a greater number of comparisons. In turn, their evaluation procedure should be more difficult. Therefore, specifically hypothesized in this study are the following:

Hypothesis 1: The uncertainty of constructing a decision rule is inversely related to the number of choice criteria used in the evaluation task.

Hypothesis 2: The uncertainty of constructing a decision rule is directly related to the variability and inversely related to the mean level of the attribute ratings presented in the evaluation task.

Hypothesis 3: The size of the evoked set is inversely related to the number of choice criteria used in the evaluation task.

Hypothesis 4: The size of the evoked set is inversely related to the variability and directly related to the mean level of the attribute ratings presented in the evaluation task.

Hypothesis 5: The difficulty in evaluating the brands is directly related to the number of choice criteria used in the evaluation task.

Hypothesis 6: The difficulty in evaluating the brands is inversely related to the variability and directly related to the mean level of the attribute ratings presented in the evaluation task.

METHOD

To test these hypotheses, three 2 x 3 factorial experiments were designed with "number of evaluative criteria" and "variability and mean level of presented attribute information" as independent variables and the reported uncertainty of constructing a decision rule, size of the evoked set, and task difficulty as the dependent variables. The preliminary design considerations are discussed in detail elsewhere (Belonax and Mittelstaedt, 1977). However, they will be presented briefly before the operationalization of the variables are discussed.

First, micro-wave ovens were selected because the product is sufficiently complex to allow the use of several evaluative criteria, and one in which participants would be somewhat interested.

Second, to insure that attributes of importance were selected the "direct dual questioning technique" described by Alpert (1971) was used. The six attributes selected included browning uniformity, cooking precision, cooking uniformity, cooking versatility, oven capacity and warranty protection.

Third, to manipulate the variable of the attribute ratings "Standard squares" were used. Since there were six attributes, information was presented on six "brands" to each participant.

Fourth, control for product familiarity was achieved by selecting 300 students in an introductory nutrition course which had not discussed micro-wave ovens in class. Participants were randomly assigned to treatment levels. The "brands" were presented as "real", but disguised and designated by the letters "A through F".

One independent variable, the variability and mean level of attribute ratings, was held constant within treatment levels, and systematically manipulated between treatment levels. The mean level and variability are from the "low variability" condition (X= 8.0; s² = 5.9), through the "intermediate variability" condition (X = 7.0; s² = 8.4), to the "high variability" condition (X = 6.0; s² = 14.0).

The other independent variable was a stratification according to the number of choice criteria used. After examining the brand ratings the participants were asked to indicate: (a) how certain they were of constructing a decision rule on a twelve-point scale, from "extremely certain" 1 to "extremely uncertain" 12. (The answer to this question is the first dependent variable). After making an overall evaluation for each brand, participants were asked to indicate: (b) the importance of each attribute to the evaluation decision, (c) the minimum rating on each attribute which an oven would have to receive to be evaluated as "acceptable", (d) the number and designation of brands considered to be acceptable for purchase, (The answer to this question is the second dependent variable), (e) how difficult it was to evaluate the brands on a twelve-point scale, from "extremely easy" 1 to "extremely difficult" 12. (The answer to this question is the third dependent variable). Answers from questions (b) and (c) were combined and a mean calculated for each attribute. Any participant's combined answer on each attribute was designated "employed choice criteria" and the number of such was determined for each participant. Previously it had been determined to consider those using 1, 2, or 3 criteria as the "few choice criteria" group and those using 4, 5, or 6 as the "many choice criteria" group.

Since the other independent variable was randomized across the entire group of participants, the two variables are "crossed" and the design factorial. Data were analyzed using the ANOVA subroutine of MULTIVARIATE ANALYSIS (Finn, 1972). Although there are some differences in cell size, reordering the variables for unequal cells did not affect significance.

RESULTS

Table 1 shows the mean level of uncertainty for the various conditions.

An examination of the means shows that, as predicted by Hypothesis 1, those employing "more" choice criteria were more certain of constructing a decision rule. Hypothesis 2 predicted that uncertainty would increase as the variability and mean level of attribute increased and decreased, respectively. The means are in the predicted direction for both the "many" and "few" choice criteria groups and, therefore, in total. Table 2 presents the results of the two-way univariate analysis of variance. Both main effects are significant while the interaction is not. The results confirm Hypotheses 1 and 2.

TABLE 1

MEAN LEVEL OF UNCERTAINTY FOR ALL CONDITIONS

TABLE 2

UNIVARIATE ANALYSIS OF VARIANCE

Table 3 shows the mean evoked set size for the various of the study. [Results of this analysis are presented elsewhere (Belonax and Mittelstaedt, 1977).]

TABLE 3

MEAN EVOKED SET SIZE FOR ALL CONDITIONS

An examination of the means shows that as predicted by Hypothesis 3, those employing more choice criteria constructed smaller evoked sets. Hypothesis 4 predicted that the evoked set would decrease between treatment levels. The means are in the predicted direction. Table 4 presents the results of the two-way univariate analysis of variance. Both main effects are significant while the interaction is not. The results confirm Hypothesis 3 and 4.

TABLE 4

UNIVARIATE ANALYSIS OF VARIANCE

Table 5 shows the mean task difficulty for the various conditions.

The means show that as predicted by Hypothesis 5, those employing more choice criteria reported more task difficulty. Hypothesis 6 predicted that task difficulty would decrease between treatment levels. The means are in the

TABLE 5

MEAN LEVEL OF TASK DIFFICULTY

predicted direction. Table 6 presents the results of the two-way univariate analysis of variance. Both main effects are significant while the interaction is not. Hypotheses 5 and 6 are confirmed.

TABLE 6

UNIVARIATE ANALYSIS OF VARIANCE

DISCUSSION

Although no attempt was made to determine the separate effects of attribute variability and mean level of attribute quality, the results confirm the view that taken together they engender greater uncertainty on the part of the participants. The uncertainty is less when it is anticipated that many criteria will be applied to information. This suggests at least two observations.

First, that people appear to exhibit greater uncertainty dependent on mean level and variability of attribute quality implies that greater attention should be paid this phenomenon in information processing studies.

Second, although no model was fitted to preference orderings, the anticipated increase in information load brought about by the greater number of attributes to be considered per brand appears to increase the relative urgency of simplifying their choice strategy. Thus, it is likely that consumers anticipate using simplification heuristics in situations of perceived information overload and felt more certain of executing the strategy.

That the size of the evoked set seems to depend on the number of choice criteria used and the nature of available information is consistent with the expectation that participants would engage in a simplification strategy to construct an evoked set. It seems reasonable to assume that "overall ratings" (or rankings) of brands which are not included in the evoked set (i.e., are in the "inept" or "inert" sets) would be less valid and/or less reliable than those of brands included in the evoked set. Studies which attempt to fit models to "overall evaluation" ratings (or rankings) for all brands in the awareness set (or about which information is provided in an experimental setting) should probably closely consider the salience of the attributes presented and the variability and mean levels of the information to which participants are exposed.

The results confirm the view that, task difficulty appears to depend on the conditions of this study. This is consistent with the expectation that participants find it easier to differentiate brands that differ markedly in quality than brands that differ little. Further, it seems that the use of more criteria increases the actual difficulty of the evaluation task.

REFERENCES

Mark I. Alpert, "Identification of Determinant Attributes: A Comparison of Methods," Journal of Marketing Research, 8 (May 1971), 184-91.

Joseph J. Belonax, Jr. and Robert A. Mittelstaedt, "Evoked Set Size as a Function of Number of Choice Criteria and Information Variability," in H. Keith Hunt, ed., Advances in Consumer Research, Vol. 5, Proceedings of the 8th Annual Conference of the Association for Consumer Research, 1978, 48-51.

James Bettman, "Perceived Risk and Its Components: A Model and Empirical Test," Journal of Marketing Research, 10 (May 1973), 184-90.

James R. Bettman, "Data Collection and Analysis Approaches for Studying Consumer Information Processing," in W. D. Perreault, Jr., ed., Advances in Consumer Research, Vol. 4, Proceedings of the 7th Annual Conference of the Association for Consumer Research, 1977, 342-48.

Steven H. Chaffee and Jack M. McLeod, "Consumer Decisions and Information Use," in S. Ward and T. S. Robertson, eds., Consumer Behavior: Theoretical Sources, Englewood Cliffs, N. J. Prentice-Hall, 1973, 385-415.

Jeremy D. Finn, Multivariance: Univariate and Multivariate Analysis of Variance, Covariance and Regression. Chicago, National Educational Service, 1972.

Fleming Hansen, "Psychological Theories of Consumer Choice," Journal of Consumer Research, 3 (December 1976), 117-142.

John A. Howard and N. Sheth Jagdish, The Theory of Buyer Behavior. New York: Wiley, 1969.

Jacob Jacoby, Donald E. Speller, and Carol A. Kohn, "Brand Choice Behavior as a Function of Information Load: Replication and Extension," Journal of Consumer Research, 1 (June 1974), 33-42.

Clem L. Narayana and Rom J. Markin, "Consumer Behavior and Product Performance: An Alternative Conceptualization,'' Journal of Marketing, 39 (October 1975), 1-6.

Whan C. Park, "The Effect of Individual and Situation-Related Factors on Consumer Selection of Judgmental Models," Journal of Marketing Research, 13 (May 1976), 144-51.

Donald Popielarz, "An Exploration of Perceived Risk and Willingness to Try New Products," Journal of Marketing Research, 4 (November 1967), 349-54.

Bernard Pras and John Summers, "A Comparison of Linear and Nonlinear Evaluation Process Models," Journal of Marketing Research, 12 (August 1975), 276-81.

Michael Reily and Rebecca H. Holman, "Does Task Complexity of Cue Interrelation Affect Choice of an Information Processing Strategy? An Empirical Investigation," in W. D. Perreault, Jr., ed., Advances in Consumer Research, Vol. 4, Proceedings of the 7th Annual Conference of the Association for Consumer Research, 1977, 185-90.

Peter Wright, "The Harassed Decision Maker: Time Pressure, Distractions, and the Use of Evidence," Journal of Applied Psychology, 59 (October 1974), 555-61.

Pete R. Wright, "Consumer Choice Strategies: Simplifying Vs. Optimizing," Journal of Marketing Research, 12 (February 1975), 60-67.

Fred Van Raaij, "Consumer Information Processing for Different Information Structures and Formats," in W. D. Perreault, Jr., ed., Advances in Consumer Research, Vol. 4, Proceedings of the 7th Annual Conference of the Association for Consumer Research, 1977, 176-184.

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