The Effects of Choice Complexity and Decision Freedom on Consumer Choice Behavior

ABSTRACT - The applicability of decision freedom and complexity decision concepts to a consumer soft drink choice problem was assessed. The number of choice alternatives and the quantity of soft drink were manipulated in three treatment conditions. Results indicated the amount consumed and time required to make the decision varied across these treatment conditions.


John R. Walton and Eric N. Berkowitz (1979) ,"The Effects of Choice Complexity and Decision Freedom on Consumer Choice Behavior", in NA - Advances in Consumer Research Volume 06, eds. William L. Wilkie, Ann Abor, MI : Association for Consumer Research, Pages: 206-208.

Advances in Consumer Research Volume 6, 1979      Pages 206-208


John R. Walton, University of Minnesota

Eric N. Berkowitz, University of Minnesota


The applicability of decision freedom and complexity decision concepts to a consumer soft drink choice problem was assessed. The number of choice alternatives and the quantity of soft drink were manipulated in three treatment conditions. Results indicated the amount consumed and time required to make the decision varied across these treatment conditions.


To an increasing degree psychological theories of decision making have been applied to consumer choice problems. Without question, these applications have expanded our knowledge about consumer choice behavior (Hansen, 1976).

At the same time, it should be recognized that consumer decision making issues (e.g., amount of consumption) differ from typical psychological research concerns. Therefore, consumer researchers should be interested in synthesizing and extending these psychological approaches to particular consumption problems. The purpose of this research is to determine the applicability of two psychological decision concepts to a particular consumer choice problem.


As Engel, Kollat, and Blackwell (1978) have pointed out, consumer decision making may be viewed as a process. Process elements include problem recognition, search, alternative evaluation, choice, and post-choice evaluation. This research is concerned with the concepts of choice set complexity and decision freedom as they relate to the alternative evaluation stage.

At this stage it is assumed that the individual consumer has identified a choice set consisting of N objects. Hendrick, Mills, and Kiesler (1968) used the concept of choice set complexity to explain the amount of time the individual will take to choose one of these N alternatives. Specifically, these authors hypothesized that the complexity of the choice set increases as the number of equally valued alternatives increases and/or the number of attributes used to describe those alternatives increases. The authors predicted an inverted U relationship between complexity and time to decision. The results of the study supported the prediction.

Decision freedom refers to the perception of personal volition on the part of the decision maker. Steiner (1970) asserted that decision freedom increases as the number of equally desirable alternatives in the choice set increases.

Neither the complexity nor the decision freedom explanation, however, makes specific predictions concerning the actual consumption of the chosen alternative. Reibstein, Youngblood, and Fromkin (1975) have extended the decision freedom concept to this issue.

These authors presented undergraduate students with either two brands or four brands of soft drinks. In each condition subjects were instructed to choose one of the alternatives and then consume as much of that alternative as they wished. Individual subject preferences toward each alternative were not assessed beforehand. Results showed that perceived decision freedom was higher and consumption of the chosen alternative was greater in the four brand condition.


The objective of this study is to attempt to replicate and extend these earlier findings. Specifically, it is unlikely in all choice situations that more alternatives will result in greater consumption. Furthermore, the presentation of equally valued alternatives (either two or four) and/or the quantity of product available for consumption may have some impact on choice behavior. These issues are addressed in the methodology described below.

Students in an introductory marketing class at a large Midwestern university were pretested to determine their preferences toward eight different flavors of a nationally known soft drink mix. These preferences were measured on a three point scale (dislike, neutral, like).

Approximately one week later subjects reported individually to a laboratory for the second phase of the study. Subjects were randomly assigned to one of three treatment conditions. Condition one consisted of exposure to two flavors of the soft drink with one two-quart pitcher of each flavor. Condition two consisted of two flavors of the soft drink with two two-quart pitchers of each flavor. Finally, condition three consisted of four flavors of the soft drink with one two-quart pitcher of each flavor. Pretest data allowed for the construction of equally preferred alternative sets for each individual.

The experimenter prepared the alternative set before the subject arrived. The pitchers of soft drink were placed on a table directly in front of a chair on which the subject would sit. A partition was placed around the area so that the table could not be seen by the subject until the experimenter was ready.

Upon arrival the subject's current level of thirst was measured. The subject was then informed of the nature of their choice task. They were told that as soon as they chose a flavor they should indicate that choice to the experimenter. Further, they were informed that they could drink as much of that flavor as they liked.

As soon as the subject was seated, the experimenter started a stopwatch in her pocket. When the subject announced his/her choice, the watch was stopped. After the subject had consumed as much as he/she wished, they were asked to complete a questionnaire measuring decision freedom and their perception of the purpose of the experiment. After the subject left, the experimenter measured the amount of liquid that had been poured from the pitcher. Since, in every case, all liquid poured had been drunk, this measure represented the individual's total consumption.

After reviewing the questionnaire item concerning the experimental purpose, it was clear that no subject was aware of the true nature of the experiment. After all experimental trials were concluded subjects were debriefed in class.

To summarize, dependent variables in this study include seconds to decision, perceived decision freedom, and amount of soft drink consumed. In the case of each of these dependent variables, the null hypothesis predicts no differences among the three treatment conditions. One way analysis of variance was used to test these hypotheses. The results of these analyses are described in the following section.


The choice complexity explanation predicts that for this relatively simple choice situation, seconds to decision should be greater in the four alternative treatment condition. Furthermore, no differences between the two alternative, one pitcher and two alternative, two pitcher conditions should be evidenced.

As shown in Figure 1, the results support these expectations [F(2,67) = 6.20, p < .01]. Specifically, the Newman-Keuls test indicated that mean seconds to decision in the four alternative case (X = 10.15 sec.) was significantly higher than the two alternative, one pitcher (X = 6.17 sec.) and two alternative, two pitcher (X = 6.78 sec.) treatments. No difference between the two alternative conditions was observed.



The degree of perceived decision freedom also varied across treatment conditions [F(2,69) = 18.29, p < .0001). The highest level of perceived decision freedom was observed for the four alternative condition (X = 5.17). This level of decision freedom was significantly greater than found for the two alternative, one pitcher (X = 2.78) and two alternative, two pitcher (X = 3.21) treatments. Figure 2 presents a graphical portrayal of these results.



The finding that perceived decision freedom is greater for four alternatives rather than two supports the results of Reibstein et. al. Subsequently, these authors found that a larger amount of soft drink was consumed in the four alternative choice situation. An attempt to replicate this relationship between decision freedom and amount consumed was tested.

Furthermore, a quantity effect was also tested. Treatments 2 and 3 (two alternatives, two pitchers and four alternatives, one pitcher) exposed subjects to twice as much soft drink as the remaining treatment group. Significantly greater consumption in treatments 2 and 3 would support this quantity explanation.

Mean consumption levels for each treatment are depicted in Figure 3. While significant differences in amount consumed were found [F-(2,67) = 4.07, p < .03], neither the decision freedom effect nor the quantity effect was observed. Specifically, the highest consumption was found in the two alternative, two pitcher condition (X = 2.3 cups). The Newman-Keuls test indicated that this level of consumption was higher than was observed in the four alternative, one pitcher treatment (X = 1.39 cups),-but not in the two alternative, one pitcher treatment (X = 1.64 cups).



It was expected that random assignment would preclude thirst difference between groups. However, a significant thirst difference was observed [F(2,69) = 4.68, p < .02]. Therefore, the analysis of variance using amount consumed as the dependent variable was rerun with level of thirst as a covariate. Significant differences among the treatments were still observed [F (2,66) = 3.30, p < .05]. Furthermore, the direction of those differences were as previously described.


The applicability of decision freedom and complexity decision concepts to a particular consumer food choice problem was assessed. The results indicated that direct application of these concepts to consumer decision problems may not be warranted.

Specifically, the findings of Reibstein et .al., that more soft drink would be consumed when perceived decision freedom was greater, was not supported. Indeed, the opposite result was found.

Results were more promising with the complexity construct. Significantly less decision time was required in both of the two alternative conditions. Furthermore, since more soft drink was consumed in these conditions, a relationship between complexity and amount consumed is tenable.

The approach taken in this research may be appropriate for a variety of consumer food decisions. The decision process of the out-of-home food consumer is one example. The consumer makes one decision in selecting a particular restaurant or fast food outlet. A second decision is required in choosing one alternative from the range of alternatives offered on the menu.

This study suggests that the manner in which information about these alternatives is presented may impact on amount ordered (and, therefore, consumed). Presumably appropriate psychological concepts such as conflict and decision freedom will help restaurant management ascertain the most effective manner to present the alternatives to the consumer. Further research on these concepts and others will be necessary to determine the specific nature of this environment.


James F. Engel, Roger D. Blackwell, and David T. Kollat, Consumer Behavior, 3rd ed., (New York: Holt, Rinehart and Winston, 1978).

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

Clyde Hendrick, Judson Mills, and Charles A. Kiesler, "Decision Time as a Function of the Number and Complexity of Equally Attractive Alternatives," Journal of Personality and Social Psychology, 8 (June 1968), 313-318.

David J. Reibstein, Stuart A. Youngblood, and Howard L. Fromkin, "Number of Choices and Perceived Decision Freedom as a Determinant of Satisfaction and Consumer Behavior," Journal of Applied Psychology, 60(August 1975), 434-437.

Ivan D. Steiner, "Decision Freedom," in L. Berkowitz (Ed.), Advances in Experimental Social Psychology, Volume 5, (New York: Academic Press, 1970).



John R. Walton, University of Minnesota
Eric N. Berkowitz, University of Minnesota


NA - Advances in Consumer Research Volume 06 | 1979

Share Proceeding

Featured papers

See More


Ineffective Altruism: Giving Less When Donations Do More

Joshua Lewis, University of Pennsylvania, USA
Deborah Small, University of Pennsylvania, USA

Read More


Give Me Something of Yours: The Downside of Digital (vs. Physical) Exchanges

Anne Wilson, Harvard Business School, USA
Shelle Santana, Harvard Business School, USA
Neeru Paharia, Georgetown University, USA

Read More


The Neutral Face of Blue: How Color Can Make Consumers Stay Sensitive

Sung Hee Wendy Paik, University of Oregon, USA
Jiao Zhang, University of Oregon, USA
Aparna Sundar, University of Oregon, USA

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.