The Effect of Information Presentation Format and Decision Frame on Choice in an Organizational Buying Context

ABSTRACT - We conducted a series of experiments in order to explain anomalous findings from a previous set of experiments. In our previous work, we expected individual organizational buying decisions that were framed as gains to lead to cautious supplier choices and individual organizational buying decisions that were framed as losses to lead to risky supplier choices. We found the opposite. Namely, gain frames led to risky choices. Subsequent experiments suggested that the information presentation format had an effect on subjects' choices. We found that subjects in our experiments used the information as presented and did little information transformation. It seems that adjusting the information presentation format led to changes in the choices that subjects made when decisions were framed as gains or losses. These results support the "concreteness" principle (Slovic 1972) which suggests that individuals use information as presented and do little, if any, information transformation before making decisions.

Citation:

James E. Stoddard, University of New Hampshire Edward F. Fern (1996) ,"The Effect of Information Presentation Format and Decision Frame on Choice in an Organizational Buying Context", in NA - Advances in Consumer Research Volume 23, eds. Kim P. Corfman and John G. Lynch Jr., Provo, UT : Association for Consumer Research, Pages: 211-217.

Advances in Consumer Research Volume 23, 1996      Pages 211-217

THE EFFECT OF INFORMATION PRESENTATION FORMAT AND DECISION FRAME ON CHOICE IN AN ORGANIZATIONAL BUYING CONTEXT

James E. Stoddard, University of New Hampshire

Edward F. Fern, Virginia Polytechnic Institute and

State University

[This research was partially funded by a grant from The R.B. Pamplin College of Business.]

ABSTRACT -

We conducted a series of experiments in order to explain anomalous findings from a previous set of experiments. In our previous work, we expected individual organizational buying decisions that were framed as gains to lead to cautious supplier choices and individual organizational buying decisions that were framed as losses to lead to risky supplier choices. We found the opposite. Namely, gain frames led to risky choices. Subsequent experiments suggested that the information presentation format had an effect on subjects' choices. We found that subjects in our experiments used the information as presented and did little information transformation. It seems that adjusting the information presentation format led to changes in the choices that subjects made when decisions were framed as gains or losses. These results support the "concreteness" principle (Slovic 1972) which suggests that individuals use information as presented and do little, if any, information transformation before making decisions.

Prospect theory has been used by researchers to explain the effect of decision frames on organizational buying behavior. Findings from much of this research support the theory's predictions. At the individual buyer level, organizational buying decisions framed as gains tend to lead to cautious supplier choices and those framed as losses tend to lead to risky supplier choices (e.g., Puto 1985, 1987; Qualls and Puto 1989).

In our own work we have attempted to replicate and extend these findings to buying groups. First, we expected that individual buying decisions that were framed as gains would lead to cautious choices and those framed as losses would lead to risky choices. Second, we wished to extend the gain/loss frame effect to the group buying context, hypothesizing that buying group interaction would increase the magnitude of these decision frame effects (McGuire, Kiesler, and Siegel 1987; Neale et al. 1986).

Initially, we conducted two replication experiments to test prospect theory predictions at the individual buyer level. However, contrary to previous research in the area we found that individual organizational buying decisions framed as gains led subjects to choose the risky supplier and those decisions framed as losses led subjects to choose the cautious supplier.

Prospect theory suggests that the individual choice process consists of two phases, editing and evaluation (Kahneman and Tversky 1979). During the initial editing phase the choice options are hypothetically reformated in order to simplify subsequent evaluation and choice. During the editing phase actual choice outcomes are coded into gains and losses relative to some psychologically determined reference point. The determination of the reference point and subsequent coding of choice outcomes into either gains or losses is proposed to be affected by the formulation of the offered prospects (Kahneman and Tversky 1979, p. 274).

The present paper describes and demonstrates the effect of several information presentation format factors. Our work suggests that subjects do little editing to simplify choice alternatives. Rather, subjects used only explicitly displayed information in the form in which it was displayed, supporting the concreteness principle advanced by Slovic (1972).

BACKGROUND

Research in organizational buying demonstrates that framing affects the individual buyer's choices as prescribed by prospect theory (Kahneman and Tversky 1979). Accordingly, a buyer's preference for a supplier is partially determined by the way the purchasing decision is framed. Should a buyer be faced with a decision between two suppliers, one whose outcome is known with certainty and another whose outcome is probabilistic, the buyer will tend to choose the cautious supplier when the decision is framed as a gain and the risky supplier when the decision is framed as a loss.

This framing effect is explained by the value function in prospect theory. The value function represents the relation between objectively determined gains and losses and the subjective value a person places on these gains and losses (Whyte 1989). When faced with a choice between a cautious supplier and a risky supplier, the buyer's reference point shifts to that of the cautious supplier. The buyer compares the subjective value associated with the risky supplier's possible outcomes to the subjective value associated with the certain outcome of the cautious supplier. Since the value function is concave for gains, the difference between the subjective value associated with the risky supplier's more positive objective outcome and the subjective value of the cautious supplier's objective outcome is less than the difference between the subjective value of the risky supplier's less positive objective outcome and the subjective value of the cautious supplier's objective outcome (assuming that the expected value of the certain and the risky suppliers' offers are the same). From this new reference point (i.e., the subjective value of the cautious supplier), it appears that the buyer will gain little and faces the possibility of losing a lot by choosing the risky supplier. Therefore, the buyer will choose the cautious supplier.

When decisions are framed as losses, the buyer should choose the risky supplier. When faced with this type of decision the buyer's reference point once again shifts to the subjective value associated with the cautious supplier. However, since the value function is convex for losses, the difference between the subjective value associated with the risky supplier's less negative outcome and the subjective value of the cautious supplier's outcome is greater than the difference between the risky supplier's more negative outcome and the cautious supplier's outcome. From this new perspective it appears that the buyer's gain may be great with little to lose by choosing the risky supplier over the cautious supplier.

Empirical research has found support for these framing effects in the organizational buying context. Puto (1985, 1987) found that an easy-to-attain reference point (i.e., high budget, gain frame condition) tended to cause buyers to choose a supplier which offered a guaranteed outcome while a difficult-to-attain reference point (i.e., low budget, loss frame condition) tended to cause buyers to choose a supplier which offered two possible outcomes (i.e., the risky supplier). Schurr (1987) has demonstrated similar effects in a negotiation and bargaining setting. He found that bargaining teams thinking "a gain is at stake" made less risky bargaining agreements than teams thinking "a loss reduction is at stake."

TABLE 1

FRAMING RESULTS FOR GOODS AND SERVICES

Qualls and Puto (1989) examined hypotheses relating certain factors to the initial reference point, decision frame, and subsequent choice of buyers. They found that the procurement budget was related to the initial reference point, the initial reference point was related to the decision frame, and the decision frame was related to subsequent choice. Buyers who framed alternatives as gains tended to choose the certain (a sure thing) alternative, and buyers who framed the alternatives as losses tended to choose the risky (probabilistic) alternative.

EXPERIMENTAL RESULTS

We attempted to replicate the results of this previous research before extending the theoretical notions to buying groups. We constructed four organizational buying procurement scenarios within which were embedded our experimental manipulations (see a sample in Appendix 1). Of the four scenarios, two were goods-based and two were service-based. Since the subjects were students, we designed the purchasing scenarios to look like typical case studies that are frequently used in business courses. The subjects were told that the experimenters were testing the cases for possible inclusion in an industrial marketing course.

We conducted two separate experiments to test the effect of decision frame on supplier choice, one goods-based experiment and one service-based. Each experiment was a within-subjects design where subjects were either exposed to a gain (loss) frame scenario first and then a loss (gain) frame scenario. In all, two-hundred and fifty-six undergraduate business student subjects from a major southeastern university responded to the experimental stimuli.

As can be seen in Table 1, subjects made choices opposite to those predicted by prospect theory a majority of the time. In these experiments subjects tended to choose the risky supplier when the decision was framed as a gain and the cautious supplier when the decision was framed as a loss.

DISCUSSION

In this section we will discuss plausible theoretical explanations for the results of experiment one. We believed that the primary explanation for our inconsistent results was that the operationalization of gain and loss frames was somehow different from that of previous work in the area. We addressed this issue by examining break-even effects, the effect of wording, and the effect of probabilities associated with the risky supplier.

Recent research by Thaler and Johnson (1990) and Ross (1991) suggests that decisions framed as losses do not always cause individuals to choose the risky alternative. Specifically, they propose that the way in which a gain or loss frame is operationalized can affect the choices people make. There are at least two different ways in which a gain or loss frame can be operationalized while maintaining equal expected value between a cautious and risky prospect. In one method, one of the possible risky outcomes equals the original reference point. In the other method neither risky outcome equals the original reference point.

Thaler and Johnson (1990) and Ross (1991) provide evidence to suggest that when decisions are framed as losses and neither risky outcome allows the individual to reach the original reference point, the individual may choose the cautious alternative. In addition, they propose the opposite result in the gain frame condition. That is, if both risky outcomes are above the initial reference point the individual may choose the risky as opposed to the cautious alternative. Thaler and Johnson (1990) call this the "break-even effect."

One theoretical notion that could explain this "break-even effect" has do with perceived risk. Perceived risk has been conceptualized as a multiplicative function of decision-related uncertainty and the aversive consequences associated with the decision (Webster and Wind 1972, Sheth 1973). This formulation of perceived risk implies that if either: (1) decision-related uncertainty regarding a possible supplier is low, or (2) the aversive consequence associated with selecting a supplier is low, then the perceived risk associated with selecting that supplier would be low. Presumably, in choosing between suppliers, buyers would select the supplier with the lowest level of associated perceived risk.

TABLE 2

RESULTS FROM THE BREAK-EVEN EFFECT

In the situation where a buyer has identical information about each of two suppliers in a choice set, the buyer is likely to evaluate the perceived risk of choosing a supplier solely on the two suppliers' outcomes. This could be the case since an equal amount of information about each supplier implies an equal level of uncertainty regarding each supplier's offer. Therefore, the only pieces of information left for the buyer to use in differentiating between the two suppliers' offers are their respective outcomes.

When a buying decision is framed as a gain and the uncertainty between two suppliers is the same, it is unlikely that the buyer would feel decision making risk because the aversive consequences associated with a choice between gains is likely to be low. The buyer is in a no-lose situation. This could cause a buyer to select a risky supplier over a cautious supplier because there is little to lose and there is an even greater possible gain associated with choosing the risky supplier.

Conversely, when a buying decision is framed as a loss and the uncertainty between the two suppliers is the same, it is likely that the buyer will feel a high level of decision making risk because the aversive consequence associated with a choice between losses is likely to be high. Here, the buyer would lose no matter which choice the buyer makes. Rather than risk even greater losses, the buyer might choose the cautious supplier to minimize down side risk.

We decided to explore whether the lack of an opportunity for the subjects in our first two studies to "break-even" was an explanation for our inconsistent results. Also, we thought that the use of a within-subjects design may have exacerbated the opposite choice effect by forcing subjects to make opposite choices when exposed to a different frame after they were first exposed to an initial decision frame and made an initial choice. To test for the break-even effect and for the effect of our experimental design, we conducted a third experiment. In this experiment we changed the outcomes associated with the risky supplier's offer so that one of the outcomes met the procurement budget (a reference point) in both the gain and loss conditions. In addition, we utilized a between-subjects design rather than a within-subjects design. Subjects were 56 undergraduate business majors at a major northeastern university (see Table 2).

As can be seen in Table 2, subjects exposed to the gain frame tended to choose the cautious supplier when one of the risky supplier's possible outcomes equaled the reference point (i.e., budget). However, those in the loss frame also tended to choose the cautious supplier. Our results were consistent with predictions of prospect theory for the gain frame condition when there existed a possibility to break-even, and opposite to the predictions of prospect theory in the loss frame condition. Tests for differences in proportions for cells A through D for experiments one versus three and two versus three were conducted at the .10 level of significance (i.e., z_crit=1.28). No differences were found.

We thought that the nature of the sample could be influencing our results. Perhaps the organizational buying scenarios which we had constructed were too complex for subjects that had no experience in making such decisions. Therefore, we decided to readminister the experimental stimuli developed for the third experiment using a sample of MBA students, assuming that MBA students would have more experience with making these type of complex organizational buying decisions. The forth experiment was also a between-subjects design with a sample of twenty-four MBA students from a major northeastern university (see Table 2). Tests for differences in proportions were not significant.

The results of experiment four were consistent with those of the third experiment. Namely, decisions that were framed as gains tended to elicit cautious supplier choices when one of the risky supplier's possible outcomes equaled the reference point (i.e., budget). However, decisions framed as losses also led to cautious supplier choices. Therefore, differences in the sample did not seem to account for the results that we obtained..

We conducted post experimental interviews with the sample of MBA's so that we could determine how their decisions were being made. During these interviews we asked the subjects to make a choice between the following two prospects:

A: 50% chance to win $1000

50% chance to win $0

B: Win $500 for sure

and also the following two prospects:

A: 50% chance to lose $1000

50% chance to lose $0

B: Lose $500 for sure

Consistent with findings from previous research (e.g., Kahneman and Tversky 1979), an overwhelming majority of subjects chose the cautious prospect in the first instance and the risky prospect in the second.

We reworded our decision frame manipulations to be more consistent with the manipulations found in Kahneman and Tversky (1979). Specifically, we included a table within each scenario that summarized the decision to be made, in terms of the gains and losses that would accrue from each choice. In addition, the wording of the scenarios was simplified and they were administered in a between-subjects design with a sample of forty undergraduate business students.

TABLE 3

SCENARIOS WITH SUMMARY TABLES

TABLE 4

RESULTS OF SIGNIFICANCE TESTS FOR DIFFERENCES IN PROPORTIONS BETWEEN EXPERIMENTS FIVE AND SIX

The addition of the summary table delineating the actual gains and losses associated with the two supplier choice alternatives enabled us to approximate the results achieved in earlier research in the area (see experiment five results in Table 3). Specifically, subjects in the gain frame condition chose the cautious supplier significantly more than those in the loss frame condition (t=-1.32, df=38, p<.10). However, a majority of subjects in the loss frame condition still chose the cautious supplier. In addition, tests for differences in proportions for cells A through D between experiments four and five were not significant. Finally, we altered the probabilities associated with the outcomes of the risky supplier while maintaining the condition of equal expected value between the cautious and risky suppliers' offers. The revised stimuli were administered in a sixth between-subjects experiment to a sample of thirty-seven undergraduate business students (see Appendix 2 and Table 3). The results in reported in Table 3, experiment six, are consistent with previous research. A majority of subjects in the gain frame condition chose the cautious supplier and a majority of subjects in the loss frame condition chose the risky supplier (t=-5.03, df=35, p<.01). Furthermore, tests for differences in proportions in cells A through D between experiments five and six were significant (see Table 4). We readministered the same stimuli to another sample of thirty-seven undergraduate business students in a seventh experiment to insure that our results from experiment six did not capitalize on chance (see Table 5). These results replicate those from the sixth experiment. A majority of subjects in the gain frame condition selected the cautious supplier and a majority of subjects in the loss frame condition selected the risky supplier (t=-1.99, df=35, p<.05).

SUMMARY

In our experiments we found several effects of information presentation format on choice. Initially, we will provide an overview of the results obtained in the experiments. Subsequently we will discuss their theoretical relevance.

First, individuals tended to choose the risky supplier under the gain frame and the cautious supplier under the loss frame when there existed no possibility for one of the risky supplier's potential outcomes to equal the initial reference point.

Second, including the possibility for one of the risky supplier's outcomes to break-even resulted in a shift in choice for decisions framed as gains. When the risky supplier was offered the opportunity to break-even, choices framed as gains caused subjects to choose the cautious supplier.

Third, the inclusion of the break-even possibility under the loss frame condition did not appreciably change subjects' choices. They continued to choose the cautious supplier. This effect did not seem to be caused by the sample that was used in the experiment.

Fourth, inclusion of a summary table in the scenarios that specified the actual gains and losses that would accrue as a result of each decision tended to make choices under the gain frame condition more extreme. In experiment three under the gain frame condition, 57% of subjects chose the cautious supplier and in experiment four 67% of subjects chose the cautious supplier. The addition of the summary table in experiment five resulted in 75% of subjects choosing the cautious supplier.

Fifth, the inclusion of the summary table specifying the gains and losses that would accrue as a result of each decision tended to cause a larger proportion of subjects to choose the risky supplier although this effect was not large. In experiment three 38% of subjects chose the risky supplier and in experiment four 42% of subjects chose the risky supplier under the loss frame condition. The addition of the summary table in experiment five resulted in 45% of subjects choosing the risky supplier.

TABLE 5

REPLICATION OF EXPERIMENT 6 RESULTS

Sixth, altering the probabilities associated with the risky supplier's outcomes while maintaining equal expected value between the risky and cautious suppliers' offers appeared to have had the greatest effect on choices made under the loss decision frame condition. In experiment five (equal probabilities for each possible risky supplier outcome), 45% of subjects chose the risky supplier under the loss frame. In experiment six, 68% chose the risky supplier and in experiment seven 55% chose the risky supplier. It appears that individuals prefer to be risk seeking under the loss frame condition when there is a small probability of a large loss and a high probability of no loss, even though the expected value of the risky supplier's offer equals that of the cautious supplier. Only when we altered the probabilities of the risky supplier's outcomes did we notice a majority of our subjects making the risky supplier choice under the loss frame. Now we will discuss the relevance of our research to prospect theory.

THEORETICAL RELEVANCE

Prospect theory proposes that an individual's choice process consists of two-stages, an initial editing stage and a subsequent evaluation stage (Kahneman and Tversky 1979). In the first stage, individuals engage in a preliminary analysis of the choice alternatives. In the second stage, the edited choice alternatives are evaluated and the alternative with the highest value is chosen. We would like to present our research results within the context of this two-stage choice process.

During the editing phase of a decision, individuals are posited to organize and reformulate the choice alternatives in order to simplify the choice. One of the major editing operations that individuals engage in during this phase is hypothesized to be the coding of objective outcomes as gains or losses relative to some subjectively determined reference point (Kahneman and Tversky 1979).

Thaler and Johnson (1990) have presented several editing rules that individuals might use to simplify and encode choice alternatives. One of the proposed editing rules is based on the concreteness principle, originally suggested by Slovic (1972). According to Slovic (1972), concreteness is the notion that individuals tend "to use only the information that is explicitly displayed in the stimulus object and will use it only in the form in which it is displayed (p. 14)." This suggests that subjects do no active editing, but simply make decisions based on the information as it is actually presented to them (Thaler and Johnson 1990).

Although Thaler and Johnson (1990) ruled out concreteness as a possible explanation for their results, our studies lend support for concreteness. Specifically, given no break-even possibility, subjects in the gain frame chose the risky supplier and those in the loss frame chose the cautious supplier. Including the possibility to break-even led subjects in the gain frame to choose the cautious supplier but had little effect on choices in the loss frame condition. The addition of a summary table delineating the amount saved or lost and the probabilities of occurrence associated with each choice alternative resulted in a larger proportion of subjects making a cautious supplier choice under the gain frame and a larger proportion choosing the risky supplier under the loss frame. Finally, altering the probabilities associated with the risky supplier yet maintaining the equal expected value condition between the cautious and risky supplier's offers resulted in a larger proportion of subjects making a cautious supplier choice under the gain frame and a larger proportion of subjects choosing the risky supplier under the loss frame.

In summary, we have found that the effect of a gain or loss decision frame on the choice between a cautious and risky supplier is sensitive to how the information is presented to subjects. In each scenario, the expected value of the cautious supplier whose outcome was known with certainty was equal to the expected value of the risky supplier with probabilistic outcomes.

In the gain frame condition, three factors caused the buyers' choices to vary (1) the "break-even effect," (2) inclusion of a summary table, and (3) altering the probabilities associated with the risky supplier's offer. It appears that the greatest effect on choices under the gain frame was the ability of subjects to break-even. If neither of the risky supplier's outcomes equaled the initial reference point, the overall value of its offer appeared to be greater than that of the cautious supplier. However, if one of the risky supplier's outcomes afforded the possibility of reaching the initial reference point, its overall value was less attractive than that of the cautious supplier.

Under the loss frame condition, choices varied as a result of summarizing information (in terms of gains and losses with associated probabilities) and by altering the probabilities associated with the outcomes of the risky supplier. The greatest effect on choices under the loss frame seemed to be the altering of the probabilities associated with the outcomes of the risky supplier. Shifting from equal to unequal probabilities for each outcome tended to make the risky supplier's offer more attractive to subjects. A small possibility of a large loss was outweighed by a large possibility of a small loss (or no loss) and resulted in a majority of subjects choosing the risky supplier over the cautious supplier.

More generally, under a gain frame, choices may vary from cautious to risky and under a loss frame choices may vary from cautious to risky. It seems that the way information is presented to subjects affects their choices. These results lend support to the notion that decision makers tend to use information that is explicitly displayed, and only in the form in which it is displayed. Information that must be inferred or transformed seems to be discounted or ignored (Slovic 1972).

APPENDIX 1

SAMPLE ORGANIZATIONAL BUYING SCENARIO

CONCLUSION

The purpose of this paper was to demonstrate that the way in which information is presented to subjects affects their choices given a gain or loss decision frame. The empirical results that we obtained in seven experiments demonstrate that choices within a decision frame vary based on break-even effects, information presentation effects, and probability effects. The magnitude of choice variation under a gain frame was greatest for inclusion or exclusion of break-even opportunities. The magnitude of choice variation under a loss frame was greatest for changes in the probabilities associated with the risky alternative's possible outcomes. These results support the notion that individuals tend to use information in the format in which it is presented, and do little information transformation.

APPENDIX 2

REVISED ORGANIZATIONAL BUYIN SCENARIO-UNEQUAL PROBABILITIES

REFERENCES

Kahneman, Daniel and Amos Tversky (1979), "Prospect Theory: An Analysis of Decision Under Risk," Econometrica, 47 (March), 263-90.

McGuire, Timothy W., Sara Keisler, and Jane Seigel (1987), "Group and Computer-Mediated Discussion Effects in Risk Decision Making," Journal of Personality and Social Psychology, 52 (5), 917-30.

Neale, Margaret A., Max H. Bazerman, Gregory B. Northcraft, and Carol Alperson (1986), "Choice Shift Effects in Group Decisions: A Decision Bias Perspective," International Journal of Small Group Research, March, 33-42.

Puto, Christopher P. (1985), "The Framing of Industrial Buying Decisions," unpublished doctoral dissertation, Duke University.

Puto, Christopher P. (1987), "The Framing of Buying Decisions," Journal of Consumer Research, 14 (December), 301-15.

Qualls, William J. and Christopher P. Puto (1989), "Organizational Climate and Decision framing: An Integrated Approach to Analyzing Industrial Buying Decisions," Journal of Marketing Research, 26 (May), 179-92.

Ross (1991), "Performance Against Quota and the Call Selection Decision," Journal of Marketing Research, 28 (August), 296-306.

Schurr, Paul H. (1987), "Effects of Gain and Loss Decision Frames on Risky Purchase Negotiations," Journal of Applied Psychology, 72 (3), 351-358.

Sheth, Jagdish N. (1973), "A Model of Industrial Buyer Behavior," Journal of Marketing, 37 (October), 50-56.

Slovic, Paul, (1972), "From Shakespeare to Simon: Speculations-And Some Evidence-About Man's Ability to Process Information," Oregon Research Institute Bulletin, Vol. 12, No. 2 (April), 1-28.

Thaler, Richard H. and Eric J. Johnson (1990), "Gambling With the House Money and Trying to Break Even: The Effects of Prior Outcomes on Risky Choice," Management Science, 36 (6) (June), 643-60.

Webster, Fredrick E. Jr. and Yoram Wind (1972), Organizational Buying Behavior, Foundations of Marketing Series, Eugene J. Kelley, ed., Englewood Cliffs, NJ: Prentice-Hall.

Whyte, Glen (1989), "Groupthink Reconsidered," Academy of Management Review, 14 (January), 40-56.

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