Correlations between regret and decision avoidance
...; however, two new variables were added to the situation. In the first experiment, participants read a description of an executive who invested in new stock or opted to keep their old stock that caused them to lose twelve hundred dollars. The participants were also told whether the executive was a risk seeker or risk avoider. Participants were then asked to indicate the level of regret they felt was associated with the decisions the executives had made, and how wise of a decision they felt the executive had made using a 101-point scale where 0 represented not at all and 100 represented very much. Regret was the dependent variable, and wisdom and desirability were the independent variables in the study. The second experiment used the same vignette from experiment one; however, the participants were not given information about the consequences of the executives decision until after rating the consistency and desirability of the event in which a decision was being made. The rating was on a 101-point scale where 0 represented very inconsistent or very undesirable, and 100 represented very consistent or very desirable. After rating the consistency and desirability, participants were given information about the results of the decision and then asked to rate the level of regret on a 101-point scale. After calculating the data, Seta et al (2001) found that more regret was associated with inaction in the risk seeker condition and more regret was associated with action in the risk avoider condition. Because the decision not to act was assumed undesirable from the perspective of the risk seeker, high levels of regret were expected to be associated with the risk seekers inaction. Wisdom does not have an effect like that of desirability, however, and can be explained by the notion that when one focuses on the merits of the decision itself; the focus is no longer on the decision makers risk seeking or risk adverse orientation, rather on whether the decision was good or not. Data from the second experiment showed that there was no longer a relationship between inaction, action, and regret, rather a relationship between desirability and regret. Action and inaction no longer predicted judgments of regret because there was no known outcome. Because of this, as desirability of ones decision increased, regret decreased. The more a person believes in his actions or inactions, the less the actual outcome will affect the amount of regret one will feel. In conclusion, desirability of ones decisions is a greater cause of regret than whether a person acts or fails to act when making a decision, and desirability has a negative correlation with regret whereas action has a positive correlation with regret. Because of the contradicting studies on omission bias, psychologists decided to focus their attention on some alternate causes of regret in decision-making processes, and began research on responsibility and desirability. Zeelenberg (1998) focused on Seta et al’s (2001) finding on responsibility and hypothesized that responsibility would have a small effect on disappointment and regret when making decision. A person would experience greater disappointment or regret when an active decision made by a computer has a negative outcome than when an active decision is made by ones self has a negative outcome. When a person is more responsible for their actions, in most cases they will have a higher desirability for the decisions that person makes, decreasing feelings of regret. Zeelenberg presented sixty Dutch undergraduate students with a scenario involving three students who were placed randomly into three different sections of a required class for their major with no prior knowledge of how easy or difficult the sections would be, but with knowledge that they would all be different. Section A ended up being the easy section, Section B was the mediocre section, and Section C was the difficult section. Due to unforeseen circumstances all three students are reassigned to Section B. The computer reassigned Alan to section B, Bob was not reassigned and stayed in section B, and Chuck reassigned himself to section B. After gaining knowledge of how the different section turned out, the participants were asked to rate the happiness on an 11 point scale, (-5) representing very unhappy, and (5) representing very happy; amount of regret and disappointment on a 9 point scale, 1 representing no regret or no disappointment, and 9 representing much regret or much disappointment. The participants were also asked to rate on a 9-point scale how responsible each student was for being in section B on a separate sheet of paper, 1 representing not responsible, and 9 representing very responsible. After calculating the data, Zeelenberg found that the participants perceived the students to be more responsible when the student made an active or inactive effort to be in section B, and they were less responsible when the computer assigned the students to section B. Happiness had no significant explanation because the ratings were based on personal opinion. The results also show that when the students obtained outcomes that are worse than those forgone, disappointment is not affected by regret, but by an alternative emotion. The student who was switched by the computer into Section B from Section A felt more disappointed than the student who switched by choice. Regret on the other hand was more intense for the student who chose to switch into the class. Zeelenberg’s study shows the difference between disappointment and regret, and how a person without knowing the outcome of an event would be more susceptible to let the computer make a decision for them rather then being held responsible for their own actions. Ordonez and Connolly (2000) wrote a response confirming Zeelenberg’s findings, but addressing regret in more detail. Ordonez and Connolly were primarily checking for between study differences in how regret and other decision related emotions might be assessed. They also hypothesize that choice is not a necessary condition for the experience of regret. Ordonez and Connolly’s experiment was the same as Zeelenberg’s, but the study was done with eighty-one Undergraduate business students at the University of Arizona who all spoke English. The students were asked to rate Alan, Chuck, and Bob’s level of happiness, regret, disappointment, and responsibility on the same scales as in Zeelenberg’s study; the only difference were some minor changes in the wording of the questions. Ordonez and Connolly’s study produced many interesting and conflicting results. The first finding was that regret is affected by decision responsibility. The effect may not be large, but it exists and supports the emotional amplification role of decision responsibility on regret. They also found that the computer assigned students whose fortunes declined experienced regret just as the students whose fortunes declined after making the decision on their own. Responsibility does appear to increase the expectation of regret; however it is not a precondition for it. Lastly data suggest that the student-choice scenario was interpreted as a decision on whether to take action or to maintain the status quo and the previous notion that all three students would experience equal regret was false; regret or rejoicing in turn is based on the comparison of the outcome with the initial assignment. It is evident that active decision-making, lessened desirability, and responsibility are all causes of regret, and in turn they increase the likely hood of the occurrence of decision avoidance. All of these causes do in fact aid in the onset of regret however, the factor that influences the onset of regret the greatest is outcome knowledge. Ritov and Baron (1995) conducted a study to show the connection between omission bias, outcome knowledge, and regret. They hypothesized that because acts tend to be seen as more causal than omissions and because regret depends on the perception of causality, regret can effect decisions, by causing people to avoid them. The feeling of regret is mostly effected by outcome knowledge, which means that regret is anticipated when a decision maker expects after making a choice that the outcome of the choice not taken will be more advantageous than the choice taken, causing the decision maker to again avoid making a decision. Ritov and Baron (1995) conducted a study to support their hypothesis. In the study, subjects were presented with a hypothetical managerial decision expected to increase efficiency. Each subject was randomly assigned to an action or omission version where a company was considering implementing a certain change in the production system, designed to increase efficiency. Each of the two managers had to decide whether to implement or not implement the proposed new procedure. The subjects in the act version were told that the manager decided to implement the change and those in the omission version did not implement the change. Subjects were then asked to rate the manager’s satisfaction with his decision on a 9 point scale ranging from ”little satisfaction” to “very high satisfaction.” Ratings were done in order under three knowledge conditions: the outcome is unknown, the outcome is known but the alternative option outcome is not, and both outcomes are known. The outcome of the manager’s decision is positive however, the alternative option turns out to be better. The results showed that anticipated regret is greatest under full knowledge, which is the knowledge of what would have happened if the foregone option had been chosen. Under partial knowledge, regret effects might still be obtained, but they are significantly less than under full knowledge. They also found that before any outcome was known; the managers felt less regret with acts than with omissions. This is possible because the subjects expected positive outcomes. With more knowledge, the results reversed, and actions were associated with higher regret than with omissions when it was revealed that the outcome of the forgone option was better. Ritov (1996) narrowed her research on outcome knowledge and did a study on the anticipation of uncertainty resolution in choice and the regret that is associated. Ritov (1996) hypothesizes that anticipation of uncertainty resolution or the probability of a positive or negative outcome for all alternatives in the choice set, affects preferences among some options, due to the probability of regret. Ritov’s (1996) study consisted of two conditions where three pairs of gambles, which were presented by boxes containing marked and unmarked balls. The gambles were used with expected values ranging between 8 and 12.5 shekels. In the single lottery, each choice concerned one box. One box had a certain value of grey and white balls, the second had white and X-marked balls, and the third had all three types of balls. The participants where given ticket options, and asked which two tickets they would prefer to buy out of a possible three tickets. The second version or the single lottery version was set up the same as the first; however, the games were played simultaneously. The results showed that subjects were generally risk averse and chose the bet with less risk, despite a lower return on the bet. Subjects did prefer, however, the higher risk bet in the single-lottery version because of the simultaneous resolution of both gambles. This study also supports the idea that preferences are affected by the potential for regret if the mechanism of resolution entails simultaneous resolution in both gambles. If subjects can obtain knowledge about both bets simultaneously, feelings of regret will decrease because they do not have to wonder how they would have done if they had chosen the other ticket. This finding contradicts that in Ritov and Baron’s (1994) study where regret increases with more knowledge of a decision outcome. Larrick and Boles (1995) is similar to Ritov and Baron’s (1996) study, but they add negotiation as a new independent variable. They hypothesized that the key factor underlying the anticipation of regret is whether or not people expect feedback on forgone alternatives, and that if they do expect feedback the anticipation of regret will be at its greatest hindering decision makers abilities to make negotiations. To support their hypothesis, Larrick and Boles (1995) conducted a study where subjects were randomly assigned to play the role of a recruit from Northwestern University, or the vice president of personnel for a consulting company called ALPHA. Both were asked to do as well as possible in the negotiation they would soon face. The recruits were told they had debt to pay off, and vice presidents were told that they had shrinking budgets. The recruits had to choose between two companies, ALPHA and BETA, and that if they chose ALPHA they were in a binding agreement, but if they did not choose ALPHA they were offered a one time non-negotiable offer from BETA. Two conditions were given, the “feedback only after impasse” condition where information about BETA’s offer was given to the recruit if he or she made a deal with BETA after passing up ALPHA’s offer, and the “feedback regardless of outcome” condition where information about BETA’s offer was given regardless of which company the recruit chose. Both conditions were the independent variables, and whether they made a negotiation was the dependent variable. All recruits were given information about what offer they could expect from BETA, but were only distribution offers from previous years. The vice presidents were given similar conditions as the recruits, but from the opposite end of the deal. Following the negotiation recruits and vice presidents were asked to complete a questionnaire asking them if they ...