The Influence of Decision Task on the Magnitude of Decoy and Compromise Effects in a Travel Decision

Abstract

This research assesses the effects of choice alternatives on the travel destination decisions of travelers. The decoy effect involves the addition of a new inferior alternative into a choice set, thereby increasing the choice of an existing option. Meanwhile, the compromise effect involves the addition of a new alternative into a choice set that increases selection of an existing option with nonextreme attributes, and decreases selection of options with extreme attributes. In this study, a series of scenario-based experiments is performed to determine if the decoy and compromise effects influence travel destination decisions. Results show that the decoy effect is stronger in a choice (vs. rejection) task, whereas the compromise effect is stronger in a rejection (vs. choice) task when deciding travel destinations.

Keywords

decision choice destination decoy effect compromise effect experiment

Introduction

The tourism literature suggests that travelers initially create a list of destination options under their consideration, then remove some of these options prior to making a final choice (Crompton 1992; Crompton and Ankomah 1993; Kozak, Kim, and Chon 2017). Imagine a traveler wants to make a reservation for a hotel in an unfamiliar foreign country. Because of uncertainty of the specific location and her travel schedule, she may book three hotels from many options. As the travel date approaches, she will cancel two of her reservations by identifying the two least attractive options and eliminating them.

This anecdote clearly illustrates that choice (e.g., choosing one option from many alternatives) and rejection (e.g., cancelation of a booking or rejecting an unattractive option) are key elements of the decision making process for travelers. Even though research on different decision tasks (i.e., choice vs. rejection) has been documented in such fields as psychology (e.g., Chernev 2009; Shafir 1993) and marketing (e.g., Laran and Wilcox 2011; Nagpal, Lei, and Khare 2015; C. W. Park, Jun, and MacInnis 2000), the tourism literature contains few studies that address the issue of the different decision tasks and their influence on travelers’ decision-making processes.

In addition, it should be noted that when the traveler searches destination options, all options are not found at the same time. A new option may often be found at a later time after the initial options are identified. The new option’s characteristics will influence travelers in assessing the attractiveness of their existing options. The contextual effects associated with the change in choice-set by adding new options have been examined in the fields of psychology and marketing in research into decoy effects, whereby people select dominating options (e.g., Huber, Payne, and Puto 1982; J. Park and Kim 2005), and compromise effects, whereby people select intermediate (middle) options given a new option is added to a choice set (e.g., Simonson 1989). Nonetheless, few studies have reported how these contextual effects influence travelers’ decision-making processes in tourism.

To address the aforementioned gaps in the tourism literature, the goal of this research is to investigate the impact of the different decision tasks of choosing versus rejecting new options on travelers’ decision-making processes. In particular, the current study examines the influence of different decision types on two contextual effects: the decoy effect and the compromise effect. Specifically, this study attempts to test whether the decoy effect will be stronger than the compromise effect in the choice task, and vice-versa in the rejection task. Results of the study provide theoretical insights into travel decision making and suggest managerial implications for effective promotion. The following sections articulate the literature of travelers’ decision making processes, two contextual effects (i.e., decoy and compromise effects) and two different decision tasks (i.e., choice vs. rejection) followed by research hypotheses.

Theoretical Framework

Travelers’ Decision Making

The decision-making process of travelers regarding their destination choice has elicited considerable attention in the tourism literature (Crompton 1992; Crompton and Ankomah 1993; Um and Crompton 1990; Woodside and Sherrel 1977). The destination choice model (Crompton 1992; Crompton and Ankomah 1993) elaborates on the destination selection process of travelers as follows. Tourists initially come up with a list of destinations for their consideration (i.e., awareness set or early consideration set). Subsequently, they omit some destinations to narrow down the list into a late consideration set. Thereafter, a destination will finally be selected from the reduced consideration set. This selection procedure is described as a funnel-like process, in which tourists continuously screen and eliminate destinations from an early consideration set until they make a final destination choice (Sirakaya and Woodside 2005).

Throughout this funnel-like process, travelers rely on destination attributes when selecting or rejecting destinations, assuming that the attributes used during selection are the same as those used during rejection (Hu and Ritchie 1993; Meyer and Johnson 1995). However, the competing literature (van Middelkoop, Borgers, and Timmermans 2003) asserts that the attributes used in selection may differ from those used in rejection. Perdue and Meng (2006) established how the reasons (attributes) for selecting a ski resort destination differ from the reasons for rejecting a destination. Their findings implicitly suggest that the attributes for selection will likely assign a destination to an evoked set, whereas the attributes for rejection can be used to finalize a destination choice. On the basis of this logic, we suggest that different decision-making tasks (i.e., selection vs. rejection) may considerably affect decision outcome.

In addition, we further investigate the possibility that the characteristics of the choice task itself could influence the choice outcome above and beyond the preference for the options. This assumption is against the traditional one that travelers have strong preferences for their traveling options and will not change their decisions across varying situations. The view that decision makers can have flexible preferences, based on decision contexts, is assumed in the “preference construction paradigm” (e.g., Bettman, Luce, and Payne 1998; Payne, Bettman, and Johnson 1993). Payne and his colleagues assume two important goals of decision makers: (1) to maximize the accuracy of decision making and (2) to minimize the effort of decision making, and suggest that decision makers show different decision patterns based on the strength of these two motivations. For instance, if the decision is important, decision makers will carefully consider/compare many options and find the best option among many alternatives. By contrast, if the decision is trivial, decision makers will consider few options and find the suboptimal option, while saving on effort. In addition, Payne and his colleagues have also suggested that decision makers try to find decision rules or heuristics, adaptively based on the given situation (e.g., time pressure, number of options available, or the location of options). These rules or heuristics can significantly influence a decision outcome.

A further example regarding preference construction concerns the number of options available for decision makers. Traditionally, researchers assumed the concept of “more is better” in that decision makers preferred a large number of options since they could find their ideal option from large assortments (i.e., a preference matching principle; Loewenstein 1999). However, recently, researchers (e.g., Iyengar and Lepper 2000; B. Schwartz 2000) have suggested the opposite concept of “less is better,” based on the fact that decision makers could feel mental confusion and regret when exposed to a large number of alternatives. Similarly, J. Y. Park and Jang (2013) introduce the concept of choice overload in selecting tourism products, stating that individuals may be less satisfied with products or may be unable to select a product when they are surrounded by too many options (Scheibehenne, Greifender, and Todd 2009). Given that travelers today are exposed to numerous tourism products online, J. Y. Park and Jang (2013) investigate the effect of choice overload using a scenario-based experimental method. They find that under overabundance of tourism products, travelers making a choice feel more regret about their selected tourism product than those making no choice, as travelers’ likelihood of foregoing better options grows.

In sum, following the line of previous research in the preference construction paradigm, travelers may demonstrate flexible and changeable preferences based on the decision context. One important factor influencing preference construction is the characteristics of the option. We will cover this issue in the next section in detail.

Decoy Effect and Compromise Effect

A key feature in economic assumptions lies in the independence of individual decision making, suggesting that the attractiveness of one alternative is independent of the remaining alternatives in the choice set (Keller, Markert, and Bucher 2015; Schoemaker 1982). The decoy and the compromise effects are well-known examples of the context effects that violate the assumption of independence from irrelevant alternatives.

The decoy effect (or the attraction effect) involves the addition of a new alternative into a choice set, thereby increasing the choice of the existing alternative, which dominates the newly added and inferior alternative (Huber, Payne, and Puto 1982; J. Park and Kim 2005). For example, let us assume that there are two equally attractive travel destination options for travelers as illustrated in Figure 1, and two options are different in terms of two attributes such as attributes X and Y. Further assume that the current relative market share of two options, A and B, is 50%:50%. In this situation, when a new option D1 is available for the travelers, this new option significantly influences the relative share of the existing options. Specifically, no one chooses the new option because new option D1 is dominated by option A. However, there is no dominance relationship between options B and D1. This asymmetric dominance relationship could influence the relative share of existing options. Previous literature (Huber, Payne, and Puto 1982) has found that new option D1 helps option A in that the relative share of option A over option B could increase, for example, to 60%:40%. The increase in the relative share of option A due to the new relatively inferior option (i.e., P{A; A, B}_{_}_{ABD1 condition} – P{A; A, B}_{_AB condition} = 60% – 50% = +10%, where P is the probability) is the decoy or attraction effect. In the same logic, when a new option D2 is introduced in the choice set, it will help the relative share of option B and show a decoy effect for option B.

Figure 1.

Examples of the decoy/attraction effect and compromise effect.

A different type of context effect is the compromise effect. The compromise effect suggests that the addition of a new alternative into the choice set could increase the potential of selecting the existing alternative with a nonextreme attribute and could decrease the options with an extreme attribute (J. Kim, Spence, and Marshall 2018; Simonson 1989). For example, going back to the previous situation when options A and B are available for the travelers, a new option C1 could be added to the choice set. Some travelers would move to the new option C1 since the new option is located in the market efficiency line. However, C1 significantly influences the relative market share of options A and B in that the relative share of option A over option B could increase, for example, to 60%:40% by the impact of introducing the new option. The increase in the relative share of option A due to option C1 (i.e., P{A; A, B}_{_ABC1 condition} – P{A; A, B}_{_AB condition} = 60% – 50% = +10%) is the compromise effect. In the same lines, new option C2 will generate the compromise effect for option B over option A.

Since the two context effects were first introduced in the marketing field (i.e., Huber, Payne, and Puto 1982; Simonson 1989), the decoy and compromise effects have been investigated in other fields such as psychology (Wedell and Pettibone 1999), animal studies (e.g., Scarpi 2011), political science (e.g., Herne 1997, 1999), or medical studies (J. A. Schwartz and Chapman 1999). In the travel literature, the decoy effect was introduced by Josiam and Hobson (1995). They showed that the decoy effect was stronger for the high-price and high-quality (vs. low-price and low-quality) option. However, further development of the decoy effect in the travel area is quite limited (with few exceptions; e.g., see Gonzalez-Prieto et al. 2013). The compromise effect has so far not been investigated in the travel and tourism area.

Several literature sources in other fields show that other factors (e.g., level of justification; Simonson 1989) can strengthen both the decoy and compromise effects. By contrast, other factors (e.g., self-regulatory focus [Mourali, Böckenholt, and Laroche 2007] or resource depletion [Pocheptsova et al. 2009]) show opposite patterns of the magnitude of the context effects. For instance, Pocheptsova et al. (2009) showed that under resource depletion, the decoy effect increases, and the compromise effect decreases in strength. Mourali, Böckenholt, and Laroche (2007) also provided empirical evidence that the decoy effect is stronger in a promotion-oriented condition, whereas the compromise effect is more powerful in a prevention-oriented condition. Although existing research has examined how various factors have an influence on (decoy or compromise) context effects, there exists little research relating to how different decision tasks influence context effects.

Additionally, an important issue relating to context effects lies in the typical format adopted for decision tasks. The typical format for decision tasks with respect to the compromise and decoy context effects favors choice rather than rejection. However, rejection is an important decision element for travelers. For example, some decision heuristics are mainly based on rejection (e.g., the elimination-by-aspects rule and the conjunctive model; Bettman, Luce, and Payne 1998). This additional issue provides an opportunity in this study to investigate the influence of different decision tasks (choice vs. rejection) on context effects.

Two Types of Decisions: Choice versus Rejection

When individuals make decisions, they are usually faced with two contrasting decision tasks: choice versus rejection. The literature on choice versus rejection suggests that decision-making processes and decision outcomes are dependent on the decision task in which the individual engages, arising from the task’s characteristics (e.g., Chernev 2009; Huber, Payne, and Puto 1982; Laran and Wilcox 2011; Shafir 1993). In regard to differences in the decision-making process between choice and rejection tasks, research has shown that people tend to consider the choice task more important than the rejection task, given they require stronger justifications for their choice among alternatives (e.g., Ganzach 1995; Wedell 1997). Huber et al. (1982) also suggested that stricter standards are needed for those who choose an option over others as they may regard the choice task more consequential compared to the rejection task.

With respect to differences in the decision outcomes between choosing and rejecting tasks, research has reported that people search for different kinds of information when they choose versus reject options available (e.g., Ganzach 1995; Nagpal, Lei, and Khare 2015; Shafir 1993; Wedell 1997). In particular, Shafir (1993) proposed that in the context of a choice task, individuals place more emphasis on the positive features of the options, since the goal is to choose the option that is most appealing to justify their choice decision. In a rejection task, the goal is to eliminate options that individuals do not want. As such, they place more weight on the negative features of the options to justify their rejection decision. Similarly, Laran and Wilcox (2011) suggested that individuals are likely to focus more on preference-consistent alternatives in the choice task condition, while they tend to focus on preference-inconsistent features in the rejection task. For example, when an individual’s goal is to buy a mobile phone for business, the individual focuses more on business-use mobile plans in the choice task and focuses more on student-use mobile plans in the rejection task (Laran and Wilcox 2011). Recently, Mourali and Nagpal (2013) argued that people with high power are more likely to focus on the positive side of options resulting in option selection, whereas those with low power tend to focus on the negative side of options resulting in option rejection. In summary, individuals primarily weigh the positive attributes in a choice task and the negative attributes in a rejection task.

Sokolova and Krishna (2016) suggested that rejection and choice tasks differ in terms of information processing level. In particular, their study provided empirical evidence that participants engaged in a rejection task tend to use a more deliberative information-processing strategy that requires cognitive resource or effort. By contrast, participants involved in a choice task rely on a less deliberative information processing strategy that is unaffected by limitations in cognitive resource.

Unlike such disciplines as psychology (e.g., Chernev 2009; Shafir 1993) and marketing (e.g., Laran and Wilcox 2011; Nagpal, Lei, and Khare 2015; C. W. Park, Jun, and MacInnis 2000; Sokolova and Krishna 2016), there have been few studies that address the issue of decision tasks (i.e., choice vs. reject) and their impact on travelers’ decision making in the tourism literature. In fact, the terms choice and rejection have, by default, often been treated as homogenous in the tourism literature, with little attention paid to the differences between choice and rejection tasks (e.g., Karl, Reintinger, and Schmude 2015; Woodside and King 2001). For instance, Karl, Reintinger, and Schmude (2015) assumed that the reasons for rejection of destinations in a choice task are same as the reasons for destination selection, in their analysis of tourists’ destination choice processes. Likewise, Woodside and King (2001) also considered choice (buying) decisions not different from rejection decisions in travelers’ decision-making processes. Furthermore, the issue of booking cancelation (cf., a rejection task) has often been examined in the tourism literature from the operator’s perspective (e.g., Murphy and Chen 2016; Smith, Parsa, and Bujisic 2015) rather than the customer’s.

However, there are a few exceptional studies in the tourism literature that treat option choice differently from option rejection (e.g., Oppewal, Huybers, and Crouch 2015; Perdue and Meng 2006). For instance, Perdue and Meng (2006) examined the impact of choice and rejection tasks on the importance of destination attributes in a decision consideration set. They found that when travelers were asked to choose their winter vacation destination, quality attributes (e.g., snow quality, lodging quality) were more important than others, while price attributes (e.g., lodging prices, price of access) were more important when they were asked to reject destination options. Oppewal, Huybers, and Crouch (2015) noted that prescreening information in the initial stage of decision making (cf. a rejection task) is not as important to the traveler as the information considered in the final stage of decision making (cf. a choice task). In short, compared with other disciplines, the topic of different decision tasks has received little attention from tourism researchers.

Main Prediction: The Impact of the Decision Task on the Context Effect

The literature sources above demonstrate how the two different types of decision tasks might influence the decoy effect and the compromise effect. In this paper, it is postulated that the decision type will influence the magnitude of the decoy effect and the compromise effect.

On the one hand, we expect that the decoy effect will be stronger in a choice versus a rejection task. The literature on the decoy effect has shown that the decoy effect is generally stable and positive. The most prevalent method of expressing preference is choice rather than rejection. Based on the results of previous literature showing a strong decoy effect in a choice task, we expect that the decoy effect in the travel situation will be stronger in a choice task as well. However, when travel decision makers attempt to reject alternative(s) in the final decision process, the dominated option (i.e., inferior option) is the first option to be rejected. It then follows that individuals faced with two options comprising equal preference will find it difficult to reject one or the other option (Simonson 1989). In other words, decision makers under a rejection task will face the exact same two options in the last stage of selection whether they were initially exposed to either (1) a two-trade-off-option set (e.g., options A and B in Figure 1) or (2) a three-option-with-a-decoy set (e.g., options A, B, and D1 in Figure 1). In summary, it is proposed that the decoy effect will be strong in the choice (vs. rejection) task.

Hypothesis 1: The decoy effect will be relatively stronger in the choice versus rejection response mode.

On the other hand, the compromise effect will be stronger in the rejection task than in the choice task, based on two theoretical grounds. First, as Shafir (1993) explains, in the rejection task, individuals place more weight on the negative attributes of options. Hence, they will exclude extreme options that contain negative values on one attribute. Put differently, the extreme aversion could be stronger for a rejection (vs. choice) task since the option located on the extreme will be rejected frequently. In addition, Luce, Bettman, and Payne (1997) suggest that a rejection task elicits negative emotions, as individuals have to sacrifice options. Therefore, to avoid unnecessary loss, the information is processed more thoroughly. Similarly, Sokolova and Krishna (2016) argued that choice tasks rely on a less deliberative information processing strategy, whereas rejection tasks require a deliberative information processing strategy. The current literature has suggested that the compromise effect requires relatively extensive processing in information search and option comparison, where the decoy effect requires relatively less extensive processing in information search and judgment (Dhar and Simonson 2003; Khan, Zhu, and Kalra 2011).

On the basis of two theoretical support sources, we propose that the compromise effect will be higher in the rejection response condition rather than in the choice condition. The formal prediction follows.

Hypothesis 2: The compromise effect will be relatively stronger in the rejection versus choice response mode.

In this article, we will report two main studies and one follow-up study showing empirical evidence regarding the two hypotheses. Each study will use different stimuli, decision sets, and sample populations.

Methodology and Results of Main Experiment 1 (Travel Decision Tasks and Context Effects)

The empirical study examines how decision tasks influence context effects in travel decision making. Specifically, this study manipulates different travel decision tasks (i.e., choice vs. rejection) in relation to compromise and decoy effects using various stimuli (e.g., vacation spot choices) from the existing literature (J. Kim 2017; Shafir, Simonson, and Tversky 1993).

Method: Subjects, Design, and Procedure

One hundred twenty-nine undergraduate students (43.4% female) at a large university in the USA participated in the study. Participants were randomly assigned to one of 2 (decision task: choice vs. rejection) × 3 (choice set: AB vs. ABC–with the compromise option vs. ABD–with the decoy option) between-subjects experimental conditions.

First, participants were asked to imagine that they had planned a week-long vacation. Next, they were asked to choose one vacation spot from two or three alternatives based on the random assignment of the experimental conditions. The vacation spots were different in terms of “weather and beaches (0 = worst to 10 = best)” and “quality of hotel (1= worst to 5 = best).” All information in this study was presented in a numerical information mode, as shown in Figure 2. In the choice condition, participants were asked to choose one vacation spot (e.g., “Which vacation spot would you prefer? Please look carefully and choose one”). Conversely, participants in the rejection condition were informed that they could not retain a current reservation for two (in the AB choice set condition) or three (in the ABC or ABC choice set conditions) options and were asked to indicate their rejection option(s) (e.g., “Which reservations would you decide to cancel? Please look carefully and reject two. Please indicate the rejected ones”). The operationalization of the choice set is shown in Figure 2B and C.

Figure 2.

Examples of choice sets in experiment 1. (A) Choice set: AB. (B) Choice set: ABD with the decoy option (option D = Spot C). (C) Choice set: ABC with the compromise option.

Results

No participant in the decoy condition (i.e., ABD) chose the dominated option, option D. Eight participants in the compromise condition (i.e., ABC) chose option C. We excluded these 8 participants from further analysis (new n=121)¹ to focus on the relative share of option B over A in order to calculate the decoy and compromise effects.

A binary logistic regression analysis (i.e., DV = choice of option B or C, IVs = decision task, choice set, and decision task × choice set) found a significant interaction effect between the two factors, decision type and choice set (Wald = 6.29, p<.05). Specifically, as shown in Table 1, the results were the same as the expectation. Specifically, the decoy effect was significant only in the choice response mode (option B’s share under the ABD vs. AB condition = 75.0% vs. 30.0%; decoy effect = +45.0%, χ²[1] = 8.12, p<.01), whereas the decoy effect was not significant in the rejection mode (option B’s share under the ABD vs. AB condition = 50.0% vs. 42.9%; decoy effect = +7.1%, χ²[1] = 8.12, p>.10). In sum, the results support hypothesis 1.

Table 1.

Results of Main Experiment 1: Decision Task and the Context Effect.

Choice SetOptions	Choice Response Mode			Rejection Response Mode
Choice SetOptions	AB	ABD	ABC	AB	ABD	ABC
A	70.0% (14/20)	25.0% (5/20)	44.0% (11/25)	57.1% (12/21)	50.0% (11/22)	19.0% (4/21)
B	30.0% (6/20)	75.5% (15/20)	40.0% (10/25)	42.9% (9/21)	50.0% (11/22)	62.0% (13/21)
C or D	–	0.0% (0/20)	16.0% (4/25)	–	0.0% (0/22)	19.0% (4/21)
Decoy effect (ABD vs. AB)	45.0%, p < .01 (75.0% [15/20] vs. 30.0% [6/20])			7.1%, p = .64 (50.0% [11/22] vs. 42.9% [9/21])
Compromise effect (ABC vs. AB)	17.6%, p = .28 (47.6% [10/21] vs. 30.0% [6/20])			33.6%, p < .05 (76.5% [13/17] vs. 42.9% [9/21])

Note: Numbers in parentheses show the raw choice data and cell sizes.

Decoy effect = P{B; A, B}_{under the ABD condition} – P{B; A, B}_{under the AB condition}.

Compromise effect = P{B; A, B}_{under the ABC condition} – P{B; A, B}_{under the AB condition}.

In contrast, the compromise effect was not significant in the choice response mode (option B’s share under the ABC vs. AB condition = 47.6% vs. 30.0%; compromise effect = +17.6%, χ²[1] = 1.34, p>.10), but significant in the rejection response mode (option B’s share under the ABC vs. AB condition = 76.5% vs. 42.9%; compromise effect = +33.6%, χ²[1] = 4.35, p<.05). In conclusion, the results support hypothesis 2 as well.

Follow-up Test

In order to check the elimination order from the rejection response mode, a follow-up study (n=26, 61.5% female undergraduates) was conducted. Participants in this study were asked to indicate their rejection options sequentially in the ABC or ABD condition of the main study. Specifically, they were first asked to reject the worst option, and then to reject the next worst option. It was predicted that participants would reject the dominated option in the decoy setting (i.e., the ABD condition), since they could easily identify the dominant relationship. In contrast, it was predicted that participants would reject the extreme options in the compromise setting (i.e., the ABC condition), since more weight is placed on the extreme attributes of the options.

The results confirmed expectations that participants in the decoy setting rejected the dominated option in the first rejection. The majority of participants in the decoy setting rejected the dominated option in the first rejection stage (92.3%=12/13 vs. random 33.3%, χ²[1]=9.68, p <.01). However, participants in the compromise setting rejected the extreme options rather than the nonextreme option in both stages (in the first rejection stage: 76.9%=10/13) and in the second rejection stage: 84.6%=11/13).

Methodology and Results of Main Experiment 2 (Testing Hypotheses with the General Public)

In this study, we replicated experiment 2 with a few modifications. First, we used the general public rather than a student sample and increased the sample size to reduce the limitations encountered in study 1. Second, similar to previous literature (e.g., S. A. Kim and Kim 2016), a different method to manipulate the compromise and decoy effects was employed. Specifically, we compared two 3-alternatives choice sets in order to calculate the context effects. Third, one weakness of study 1 was the potential lack of mundane realism primarily because of the direct use of stimuli from the existing literature (J. Kim 2017; Shafir, Simonson, and Tversky 1993). In the current study, we not only increased the number of attributes but also used more realistic ones for the target travel products to extend the external validity of the previous findings and to provide convergent evidence to support our argument. Fourth, the sample size of study 1 was relatively small. We contrived that the empirical setting (i.e., data collection based on a small group setting) exerts considerable influence on the decoy and compromise effects. In the current study, we used a relatively large sample size from the general public. Finally, one theoretical explanation for the different compromise effects between choice and rejection tasks is the varying information-processing levels (Sokolova and Krishna 2016). In particular, a relatively elaborate processing of rejection (vs. choice) task can enhance the compromise effect. We measured a decision time to verify this explanation for the underlying mechanism of context effects.

Subjects, Design, and Procedure

Three hundred eighteen US adults from an online panel (i.e., Amazon MTurk) (50.6% female, average age= 38.35 years) participated in the study. Participants were randomly assigned to one of 4 (choice set²: ABC vs. BCD [for the compromise effect], B^dBC vs. BCC^d [for the decoy effect]) × 2 (decision task: choice vs. rejection) between-subjects experimental conditions.

The main task was similar to that of experiment 1. First, participants were asked to imagine that they had planned a week-long vacation over the spring break³. Next, they were informed that they currently have three options, but they could no longer retain their reservation. To extend external validity, each vacation spot was different across six attributes as shown in Figure 3. In the choice condition, participants were asked to choose one vacation spot (e.g., “Which reservation would you decide to choose? Please look carefully and choose one”). In contrast, participants were asked to indicate their rejection option(s) (e.g., “Which reservations would you decide to cancel? Please look carefully and reject two”). Then, they were asked to indicate their first and second option to cancel. After answering their decision, all participants were asked to rate the importance of the six attributes along a 5-point scale (1 = not at all important to 5 = very important). Then, they were asked to respond their previous experience of taking a vacation trip in the last 24 months. The majority of participants (i.e., 75.5%) had travel experience. After that, all participants were asked to answer “Instructional Manipulation Check” (IMC; Oppenheimer et al. 2009) in order to check the general involvement in the online survey.

Figure 3.

Choice sets in experiment 2. (A) Decoy choice set I: B^dBC. (B) Decoy choice set II: BCC^d. (C) Compromise choice set I: ABC. (D) Compromise choice set II: BCD.

Results

First, 15 participants were excluded in the main analysis as they failed on the IMC measure. In addition, no participant in the decoy condition (i.e., B^dBC vs. BCC^d) chose the dominated option in the choice task, whereas some participants chose the decoy option in the rejection task (n = 13). Participants chose option A or D in the compromise effect (n = 26). We excluded these participants from further analysis (new n = 250) to focus on the relative share of options B and C in order to calculate the decoy and compromise effects.

A binary logistic regression analysis found a significant three-way interaction effect between the two factors (B = –.262, Wald = 5.08, p < .05). Specifically, as shown in Table 2, the results supported the main predictions. Specifically, the decoy effect was significant only in the choice response mode (option B’s share under the B^dBC vs. BCC^d condition = 59.5% vs. 16.7%; decoy effect = +42.8%, χ²[1] = 12.61, p< .001), whereas the decoy effect was marginally significant in the rejection mode (option B’s share under the B^dBC vs. BCC^d condition = 34.2% vs. 15.8%; decoy effect = +18.4%, χ²[1] = 3.43, p = .064). Therefore, the decoy effect was reduced by 24.4% when the response mode was from choice to rejection, supporting hypothesis 1.

Table 2.

Results of Main Experiment 2: Decision Task and the Context Effect.

	Choice Response Mode				Rejection Response Mode
SetOptions	Decoy Set		Compromise Set		Decoy Set		Compromise Set
SetOptions	B^dBC	BCC^d	ABC	BCD	B^dBC	BCC^d	ABC	BCD
A or B^d	0.0% (0/37)	–	10.5% (4/38)	–	20.8% (10/48)	–	33.3% (10/30)	–
B	59.5% (22/37)	16.7% (5/30)	21.1% (8/38)	38.2% (13/34)	27.1% (13/48)	14.6% (6/41)	30.0% (9/30)	22.2% (10/45)
C	40.5% (15/37)	83.3% (25/30)	68.4% (26/38)	35.3% (12/34)	52.1% (25/48)	78.0% (32/41)	36.7% (11/30)	40.0% (18/45)
D or C^d	–	0.0% (0/35)	–	26.5% (9/34)	–	7.3% (3/41)	–	37.8% (17/45)
Context effect	Decoy effect +42.8%, p < .001 (22/37 vs. 5/30)		Compromise effect –28.5% p < .05 (8/34 vs. 13/25)		Decoy effect +18.4%, p = .064 (13/38 vs. 6/38)		Compromise effect +9.3%, p > .10 (9/20 vs. 10/28)

Note: Numbers in parentheses show the raw choice data and cell sizes.

Decoy effect = P{B; B, C}_{under the B}^d_{BC condition} – P{B; B, C}_{under the BCC}^d _condition.

Compromise effect = P{B; B, C}_{under the ABC condition} – P{B; B, C}_{under the BCD condition}.

In contrast, the compromise effect was significant, but negative in the choice response mode (option B’s share under the ABC vs. BCD condition = 23.5% vs. 52.0% compromise effect = −28.5%, χ²[1] = 5.09, p <.05). However, the compromise effect was positive, but not significant in the rejection response mode (option B’s share under the ABC vs. BCD condition = 45.0% vs. 35.7%; compromise effect = +9.3%, χ²[1] = .42, p >.10). Specifically, the compromise effect increased by 37.8% when the response mode was from choice to rejection, supporting hypothesis 2. In conclusion, this study replicated the results of experiment 1.

Decision Time Analysis

In this analysis, the total duration of decision time was measured to assess the context effect. On the basis of the previous literature (Fazio 1990; Sokolova and Krishna 2016), time data were log-transformed after excluding five participants whose decision time values were regarded as outliers, with data over two standard deviations from the mean (n = 313). Subsequently, a 2 (context effect types: decoy and compromise effects) × 2 (decision tasks: choice vs. rejection) analysis of variance was conducted.

The results indicated that the interaction effect (F[1, 309] = .02, p =.88) and the main effect of context effect type (F[1, 309] = 1.09, p =.30) were insignificant. The main effect of the decision task was significant (F[1, 309] = 9.69, p <.01) given that the decision time for the rejection task (M = 48.46 seconds, SD = 22.98) was longer than that for the choice task (M = 42.48 seconds, SD = 25.01). Although the respondents in the rejection task took approximately 6 seconds longer than the respondents in the choice task, the additional time was estimated based on the additional number of decisions. In particular, the results indicate that the choice task requires only one choice action, whereas the rejection task requires two choice actions. Therefore, we supposed that no significant difference occurred between choice and rejection tasks. The different deliberative processing levels failed to fully explain the results.

Methodology and Results of Main Experiment 3 (Replicating Previous Studies with Slight Modifications)

In this study, we replicated previous experiments with slight modifications. First, although price information was included in study 2, price levels were similar across different options. Price is one of the most important attributes for traveler decision making (Chung and Petrick 2013; S. Kim and Crompton 2002); thus, we used different price levels for each option. Second, a traveler can reject all options when making a decision. To reflect this rationale, a full rejection mode was considered and compared with our previous rejection modes (i.e., leaving one option in the final stage). Finally, a different method for manipulating the compromise and decoy effects was introduced. In particular, three-alternative choice sets for choice and rejection decision tasks were compared based on the previous literature (Larson and Billeter 2013).

Subjects, Design, and Procedure

A total of 166 adults from the United States participated in an online panel survey, that is, Amazon MTurk (50.0% female; average age = 36.64 years). Participants were randomly assigned to one of 2 (choice set: ABC [for the compromise effect] vs. B^dBC [for the decoy effect]) × 3 (decision task: choice vs. rejection I vs. rejection II⁴) between-subjects experimental conditions. The majority of participants (72.3%) had taken a vacation trip within the previous 24 months.

The main task was similar to that in experiment 2. First, participants were asked to imagine that they had planned a week-long vacation for spring break. Then, they were informed that they currently had three options, but they could no longer hold their reservations. Each vacation place differed in terms of four attributes as shown in Figure 4. We used the same instructions in experiment 2 for the choice and rejection I conditions. Therefore, participants under the rejection I condition should save one option in the final decision. By contrast, participants under the rejection II condition were asked to indicate their rejection options using the following instruction: “Which reservations would you decide to cancel? Please look carefully and reject two or more options! (You could cancel all options).”

Figure 4.

Choice sets in experiment 3. (A) Decoy choice set. (B) Compromise choice set. (C) Compromise choice set I: ABC. (D) Compromise choice set II: BCD.

Respondents were then asked to indicate the first and second options that they would cancel. Subsequently, they were told that they also had the choice to cancel the last option.

Results

First, the majority of participants under the rejection II condition did not want to cancel all their options (i.e., 86.4% [19/22] in the decoy effect set and 80.0% [16/20] in the compromise effect set). Furthermore, the rejection orders of the first and second cancellation options were similar under rejection I and rejection II conditions (all p’s > .10). Finally, the relative shares of option B in the decoy and compromise choice sets were similar under the two rejection conditions (all p’s > .10). In summary, the results indicated that the decision process of participants under the full rejection condition did not differ from that under the rejection while saving one option condition. Therefore, the two rejection conditions were combined to generate a total rejection condition for further analysis.⁵ In addition, the share of option B was calculated in the main analysis.

A binary logistic regression analysis indicated a significant two-way interaction effect between the two factors (B = 1.91, Wald = 8.35, p <.01). As shown in Table 3, the results supported the main predictions. In particular, the decoy effect (i.e., the share of option B under the B^dBC condition) was significantly higher under the choice condition than under the rejection condition (option B’s share under the choice vs. rejection condition = 78.0% vs. 52.4%; ∆ = 25.7%, χ²[1] = 6.01, p < .05). However, the compromise effect (i.e., share of option B under the ABC condition) was marginally higher in the rejection mode than in the choice mode (option B’s share under the choice vs. rejection condition = 45.2% vs. 63.4%; ∆ = 19.2%, χ²[1] = 2.76, p <.10). In conclusion, this study demonstrated a consensus between the results of experiments 1 and 2, even in the full rejection mode with different price attributes.

Table 3.

Results of Main Experiment 3: Decision Task and the Context Effect.

	Decoy Set { B^dBC }				Compromise Set {ABC}
Decision Mode	Choice Mode	Rejection Mode Total	Rejection Mode I	Rejection Mode II	Choice Mode	Rejection Mode Total	Rejection Mode I	Rejection Mode II
B^d / A	4.9% (2/41)	7.1% (3/42)	0.0% (0/20)	13.6% (3/22)	26.2% (11/42)	19.5% (8/41)	19.0% (4/21)	20.0% (4/20)
B	78.0% (32/41)	52.4% (22/42)	60.0% (12/20)	45.5% (10/22)	45.2% (19/42)	63.4% (26/41)	61.9% (13/21)	65.0% (13/20)
C	17.1% (7/41)	40.5% (17/42)	40.0% (8/20)	40.9% (9/22)	28.6% (12/42)	17.1% (7/41)	19.0% (4/21)	15.0% (3/20)
Context effect comparison	Decoy effect: Choice > Rejection ∆ = 25.7%, p < .05 (32/41 vs. 22/42)		Decoy effect: Rej. I = Rej. II ∆ = 14.5%, p = .35 (12/20 vs. 10/22)		Compromise effect: Choice < Rejection ∆ = 18.2%, p <.10 (19/42 vs. 26/41)		Compromise effect: Rej. I = Rej. II ∆ = 3.1%, p = .84 (13/21 vs. 13/20)

Note: Numbers in parentheses show the raw choice data and cell sizes.

Decoy effect comparison = P{B; B^d, B, C}_{under the choice condition} vs. P{B; B^d, B, C}_{under the rejection condition.}

Compromise effect comparison = P{B; A, B, C}_{under the choice condition} vs. P{B; A, B, C}_{under the rejection condition}.

Discussion and Implications

Based on the line of previous research in the preference construction paradigm (e.g., Bettman, Luce, and Payne 1998), we suggested that travelers’ decision outcomes could be influenced by two important factors (i.e., decision mode and the choice set—either in the compromise and decoy setting). We predicted that the decoy effect would be stronger in a choice (vs. rejection) task, whereas the compromise effect would be stronger in a rejection (vs. choice) task in the travel and tourism settings. Two main empirical studies provide supporting empirical evidence for the hypotheses.

In study 1, we show the predicted pattern for travel destination choice for the younger generation. Furthermore, the follow-up study provided procedural evidence of our assumption regarding the order of the rejected option in the rejection task. In study 2, we replicated the findings with a more complex and realistic travel options with the general public. Furthermore, we also tested and provided a significant mediating role of the importance of attributes in preference change. In sum, across two different experiments, we showed consistently that the decoy effect was stronger in the choice task, and that the compromise effect was stronger in the rejection task.

Theoretical Implications

The findings of this article have important theoretical implications. First, since Josiam and Hobson (1995) introduced the decoy effect in travel and tourism research, researchers have remained quiet about context effects. This is surprising in that the decoy and compromise effects have been widely investigated in other areas such as marketing (e.g., Huber, Payne, and Puto 1982; J. Park and Kim 2005; Simonson 1989), animal science (e.g., Scarpi 2011), economy (e.g., Barbos 2010), food choice (e.g., Carroll and Vallen 2014), gambling (e.g., Herne 1999), medical science (e.g., Bornstein and Emler 2001; J. A. Schwartz and Chapman 1999), political science (e.g., Herne 1997), and psychology (e.g., Wedell and Pettibone 1999). Therefore, showing the decoy and compromise effects in a travel and tourism setting could extend our understanding regarding travelers’ decision making. Moreover, the results of this study differ from those of previous studies, including those in other fields, by suggesting that different decision tasks can be regarded as a significant moderating factor for the decoy and compromise effects. This study is the first to introduce the moderating role of the decision task mode (i.e., choice vs. rejection) in the compromise and decoy effect setting. To our knowledge, this approach is novel across different fields, including psychology and consumer behavior. This study also attempted to test whether different decision procedures interact with various required decision tasks (i.e., choice vs. rejection decision modes).

Second, these results are consistent with the idea of framing the rejection task (i.e., individuals focus more on the negative features of options in a rejection task, whereas in a choice task, individuals focus more on the positive features of the options) (Shafir 1993). Specifically, the different context effects for choice versus rejection tasks could be due to the availability of different types of information. In the choice task, individuals place more weight on the positive attributes of the options. Therefore, compared to the compromise setting, individuals can easily choose the dominant option in the choice sets of the decoy setting. In the rejection task, individuals place more weight on the negative attributes of the option. Therefore, individuals are able to easily reject the extreme options in the compromise setting, whereas in the decoy setting, after individuals reject the inferior option, it becomes more difficult to reject a successive option, as both remaining options are of equal weight.

Third, this research could extend our understanding of two typical contexts effects: the decoy effect and compromise effect. Along with other research streams showing factors generating the opposite patterns for the two effects (e.g., Mourali, Böckenholt, and Laroche 2007; J. Kim 2017; Pocheptsova et al. 2009), this article also identifies a moderator (i.e., choice mode) distinguishing the direction of two effects. The significant moderating effect supports the different underlying mechanism of the decoy and compromise effects in that decoy effects are based on relatively simple decision-making processes, whereas compromise effects are based on relatively complex ones (Carroll and Vallen 2014; Herne 1999).

Fourth, the results of this research provide some empirical evidence of a “nudging” processing in the travel domains (e.g., Tan et al. 2018). The concept of nudge has been suggested by Thaler and Sunstein (2008), showing the importance of simple cues on behavioral change. For example, a simple change of display location can significantly change food consumption behavior (Keller, Markert, and Bucher 2015); preference for a healthy snack option can be higher when the healthy food option is placed in the middle (vs. on the edge) of a list. The results of this article suggested the decision mode and options characteristics could significantly influence the decision outcome.

Fifth, the results of our research are relevant to the study of Dhar and Simonson (2003) because both investigations examined the compromise and decoy effects. In particular, Dhar and Simonson (2003) reported that the compromise effect was significantly reduced, whereas the decoy effect increased when a no-choice option was available. The critical difference was found to be the key moderator for the context effect. The study of Dhar and Simonson (2003) addressed the importance of the no-choice option, whereas the current study emphasized the decision response mode. Furthermore, the underlying mechanisms for the effects differed given that Dhar and Simonson (2003) emphasized the role of the no-choice option to reduce preference uncertainty and discomfort. Therefore, a future study should integrate the two preceding research approaches to determine whether their results are different.

Finally, the results of this article provide empirical evidence for the preference construction paradigm (Bettman, Luce, and Payne 1998; Payne, Bettman, and Johnson 1993) in the travel domain. Specifically, travelers’ preference for a travel destination can be influenced by the structure of travel options available (Kozak, Kim, and Chon 2017). We are not denying that each traveler can have specific preferences for travel options. However, our study shows that the contexts of decision making also significantly influence the final choice outcome, resulting in extending our understanding regarding travelers’ decision and preference.

Managerial Implications

The managerial implications of these findings suggest that tourism marketers should consider which decision mode to use, depending on the option they would like to promote. Today, travel and hotel bookings are largely conducted in an online setting. Online search of multiple sources generates multiple options for consideration. First, many instances of search outcomes can be categorized as a compromise setting since the options will be different in terms of general price–quality attributes: higher price–quality options versus lower price–quality options. In this situation, tourism managers might use the results of our study to promote their travel option. For example, when a traveler has multiple reservations, she will be in a rejection mode for her final choice, resulting in her choosing the middle option among alternatives. Alternatively, when a traveler is in a choice mode, marketers of an extreme option (such as an expensive hotel) should introduce more extreme options to promote their target option.

Second, search outcomes can be related to the decoy effect, specially travelers are using price comparison websites (e.g., trivago.com or hotelscombined.com). The same travel/ hotel option can be presented at different price levels by different websites. In this situation, the lowest price option is expected to enjoy the comparative advantage based on the decoy effect.

Third, our article also finds the importance of decision mode: choice or rejection. We believe that the specific policy of travel industries could significantly influence travelers’ decision mode. For example, a transaction method such as no cancellation fee could encourage travelers to book multiple options in the reservation stage. Ultimately, travelers will face the rejection mode in the final decision. In this situation, they might finalize their choice by selecting the middle range option more frequently. Tourism managers might change their cancellation policy based on their characteristics of service and price level, in order to encourage the usage of their service.

Fourth, effective promotional information can convince potential customers into perceiving price fairness, reinforcing quality assurance, and thereby increasing demand (Chung and Petrick 2013). Previous studies have substantially evidenced the efficacy of promotional messaged in communicating with tourism customers (e.g., Chung et al. 2011; S. Kim and Crompton 2001; Steckenreuter and Wolf 2013). Thus, results of this study can facilitate development of promotional messages that can reduce customer resistance to price.

Fifth, it is better to provide more options for customers in a situation of lack of information which can occur due to information asymmetry between customers and tourism suppliers (Fleischer, Tchetchik, and Toledo 2015). According to attribution theory, customers in uncertain circumstances tend to use information to make inferences about a causal factor (Weiner 2000) or use peripheral cues to interpret price (Ferguson and Ellen 2013). Therefore, providing alternatives in the destination choice process can be conducive to a reduction in the likelihood of negative inferences related to the perception of price unfairness and alleviation of the likelihood of negative responses (Xia, Monroe, and Cox 2004). As a consequence, since it will help enhance customer satisfaction and trustworthiness, and fortify brand loyalty, tourism providers can benefit from such promotional strategies.

Conclusion and Future Studies

In this study, we found a significant decoy effect only in the choice task and a significant compromise effect only in the rejection task. One weakness of this study could be the number of attributes for choice options in that there were only two (in study 1) or relative few (in study 2) important attributes for each option. Since actual decision making involves multiple attributes, researchers should increase the number of attributes in subsequent studies in order to increase the external validity of the findings. Second, we employed a scenario-based experimental design. Even though the scenario-based method is well-established and frequently used in travel research (e.g., J. H. Kim and Jang 2014; J. Kim, Kim, and Kim 2018) as well as other area such psychology and marketing (e.g., Kahneman and Tversky 1979; Thaler 1985), external validity is limited. Future study should address this weakness by employing secondary data or a field experiment.

Our research findings suggest important directions for future studies. In this study, although we provided the theoretical explanation for the core argument, the evidence of the underlying mechanism was weak. Therefore, future studies could examine this effect by showing more direct evidence of the underlying mechanism. Second, we opted to use only one type of product (i.e., a vacation spot) where the purpose of travel is hedonic. It might be that compromise effects are more stronger in utilitarian versus hedonic consumption situations (S. A. Kim and Kim 2016). Therefore, future research needs to empirically test whether purpose of travel moderates the decoy and compromise effects. Finally, even though this study showed a consistent effect for two different products and two different populations (e.g., students and general public), future studies should extend this research across different countries in order to extend the external validity of the findings.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

ORCID iD

Jungkeun Kim

Author Biographies

Jungkeun Kim is a senior lecturer in the Department of Marketing, Auckland University of Technology. His main research interests are in consumer and travel decision-making and behaviors.

Peter Beomcheol Kim is an associate professor in the School of Hospitality and Tourism at the Auckland University of Technology (AUT), Auckland, New Zealand. His research interests include service management and marketing and strategic human resource management.

Jin-Soo Lee is an associate professor at the School of Hotel & Tourism Management, The Hong Kong Polytechnic University. His research interests include event management and hospitality marketing.

Seongseop (Sam) Kim is a professor of the School of Hotel and Tourism Management, The Hong Kong Polytechnic University, Hong Kong SAR, China. His main research interests include tourist behavior, convention/event management, destination marketing, and economic psychology.

Kenneth F. Hyde is an associate professor of Marketing at Auckland University of Technology. His research interests include tourist behaviour and decision-making.

References

Barbos

2010. “Context Effects: A Representation of Choices from Categories.” Journal of Economic Theory 145 (3): 1224–43.

Bettman

J. R.

Luce

M. F.

Payne

J. W.

1998. “Constructive Consumer Choice Processes.” Journal of Consumer Research 25 (3): 187–217.

Bornstein

B. H.

Emler

A. C.

2001. “Rationality in Medical Decision Making: A Review of the Literature on Doctors’ Decision-Making Biases.” Journal of Evaluation in Clinical Practice 7 (2): 97–107.

Carroll

Vallen

2014. “Compromise and Attraction Effects in Food Choice.” International Journal of Consumer Studies 38 (6): 636–41.

Chernev

2009. “Choosing versus Rejecting: The Impact of Goal-Task Compatibility on Decision Confidence.” Social Cognition 27 (2): 249–60.

Chung

J. Y.

Kyle

G. T.

Petrick

J. F.

Absher

J. D.

2011. “Fairness of Prices, User Fee Policy and Willingness to Pay among Visitors to a National Forest.” Tourism Management 32 (5): 1038–46.

Chung

J. Y.

Petrick

J. F.

2013. “Price Fairness of Airline Ancillary Fees: An Attributional Approach.” Journal of Travel Research 52 (2): 168–81.

Crompton

1992. “Structure of Vacation Destination Choice Sets.” Annals of Tourism Research 19 (3): 420–34.

Crompton

J. L.

Ankomah

P. K.

1993. “Choice Set Propositions in Destination Decisions.” Annals of Tourism Research 20 (3): 461–76.

10.

Dhar

Simonson

2003. “The Effect of Forced Choice on Choice.” Journal of Marketing Research 40 (2): 146–60.

11.

Ganzach

1995. “Attribute Scatter and Decision Outcome: Judgment versus Choice.” Organizational Behavior and Human Decision Processes 62 (1): 113–22.

12.

Gonzalez-Prieto

Sallan

J. M.

Simo

Carrion

2013. “Effects of the Addition of Simple and Double Decoys on the Purchasing Process of Airline Tickets,” Journal of Air Transport Management 29:39–45.

13.

Fazio

1990. “A Practical Guide to the Use of Response Latency in Social Psychological Research.” Research Methods in Personality and Social Psychology 11:74–97.

14.

Ferguson

Ellen

P. Scholder

. 2013. “Transparency in Pricing and Its Effect on Perceived Price Fairness.” Journal of Product & Brand Management 22 (5): 404–12.

15.

Fleischer

Tchetchik

Toledo

2015. “Does It Pay to Reveal Safety Information? The Effect of Safety Information on Flight Choice.” Transportation Research Part C: Emerging Technologies 56:210–20.

16.

Herne

1997. “Decoy Alternatives in Policy Choices: Asymmetric Domination and Compromise Effects.” European Journal of Political Economy 13 (3): 575–89.

17.

Herne

1999. “The Effects of Decoy Gambles on Individual Choice.” Experimental Economics 2 (1): 31–40.

18.

Ritchie

J. R. B.

1993. “Measuring Destination Attractiveness: A Contingency Approach.” Journal of Travel Research 32 (2): 25–34.

19.

Huber

Payne

J. W.

Puto

1982. “Adding Asymmetrically Dominated Alternatives: Violations of Regularity and the Similarity Hypothesis.” Journal of Consumer Research 9 (1): 90–8.

20.

Iyengar

S. S.

Lepper

M. R.

2000. “When Choice Is Demotivating: Can One Desire Too Much of a Good Thing?” Journal of Personality and Social Psychology 79 (6): 995–1006.

21.

Josiam

B. M.

Hobson

J. S.

1995. “Consumer Choice in Context: The Decoy Effect in Travel and Tourism.” Journal of Travel Research 34 (1): 45–50.

22.

Kahneman

Tversky

1979. “Prospect Theory: An Analysis of Decision under Risk.” Econometrica 47 (2): 263–92.

23.

Karl

Reintinger

Schmude

2015. “Reject or Select: Mapping Destination Choice.” Annals of Tourism Research 54:48–64.

24.

Keller

Markert

Bucher

2015. “Nudging Product Choices: The Effect of Position Change on Snack Bar Choice.” Food Quality and Preference 41:41–3.

25.

Khan

Zhu

Kalra

2011. “When Trade-offs Matter: The Effect of Choice Construal on Context Effects.” Journal of Marketing Research 48 (1): 62–71.

26.

Kim

2017. “The Influence of Graphical versus Numerical Information Representation Modes on the Compromise Effect.” Marketing Letters 28 (3): 397–409.

27.

Kim

P. B.

Kim

J. E.

2018. “Different or Similar Choices: The Effect of Decision Framing on Variety Seeking in Travel Bundle Packages.” Journal of Travel Research 57 (1): 99–105.

28.

Kim

Spence

M. T.

Marshall

2018. “The Color of Choice: The Influence of Presenting Product Information in Color on the Compromise Effect.” Journal of Retailing 94 (2): 167–85.

29.

Kim

J. H.

Jang

S. S.

2014. “A Scenario-Based Experiment and a Field Study: A Comparative Examination for Service Failure and Recovery.” International Journal of Hospitality Management 41:125–32.

30.

Kim

S. A.

Kim

2016. “The Influence of Hedonic versus Utilitarian Consumption Situations on the Compromise Effect.” Marketing Letters 27 (2): 387–401.

31.

Kim

Crompton

J. L.

2001. “The Effects of Different Types of Information Messages on Perceptions of Price and Stated Willingness-to-Pay.” Journal of Leisure Research 33 (3): 299–318.

32.

Kim

Crompton

J. L.

2002. “The Influence of Selected Behavioral and Economic Variables on Perceptions of Admission Price Levels.” Journal of Travel Research 41 (2): 144–52.

33.

Kozak

Kim

Chon

2017. “Competitiveness of Overseas Pleasure Destinations: A Comparison Study Based on Choice Sets.” International Journal of Tourism Research 19 (5): 569–83.

34.

Laran

Wilcox

2011. “Choice, Rejection, and Elaboration on Preference-Inconsistent Alternatives.” Journal of Consumer Research 38 (2): 229–41.

35.

Larson

J. S.

Billeter

D. M.

2013. “Consumer Behavior in ‘Equilibrium’: How Experiencing Physical Balance Increases Compromise Choice.” Journal of Marketing Research 50 (4): 535–47.

36.

Loewenstein

1999. “Is More Choice Always Better?” Social Security Brief 7:1–7.

37.

Luce

M. F.

Bettman

R. R.

Payne

J. W.

1997. “Choice Processing in Emotionally Difficult Decisions.” Journal of Experimental Psychology: Learning, Memory and Cognition 23 (2): 384–405.

38.

Meyer

Johnson

E. J.

1995. “Empirical Generalization in the Modeling of Consumer Choice.” Marketing Science 14 (3): G180–9.

39.

Mourali

Böckenholt

Laroche

2007. “Compromise and Attraction Effects under Prevention and Promotion Motivations.” Journal of Consumer Research 34 (2): 234–47.

40.

Mourali

Nagpal

2013. “The Powerful Select, the Powerless Reject: Power’s Influence in Decision Strategies.” Journal of Business Research 66 (7): 874–80.

41.

Murphy

H. C.

Chen

2016. “Online Information Sources Used in Hotel Bookings: Examining Relevance and Recall.” Journal of Travel Research 55 (4): 523–36.

42.

Nagpal

Lei

Khare

2015. “To Choose or to Reject: The Effect of Decision Frame on Food Customization Decisions.” Journal of Retailing 91 (3): 422–35.

43.

Oppenheimer

D. M.

Meyvis

Davidenko

2009. “Instructional Manipulation Checks: Detecting Satisficing to Increase Statistical Power. Journal of Experimental Social Psychology 45 (4): 867–72.

44.

Oppewal

Huybers

Crouch

G. I.

2015. “Tourist Destination and Experience Choice: A Choice Experimental Analysis of Decision Sequence Effects.” Tourism Management 48:467–76.

45.

Park

C. W.

Jun

S. Y.

MacInnis

D. J.

2000. “Choosing What I Want versus Rejecting What I Do Not Want: An Application of Decision Framing to Product Option Choice Decisions.” Journal of Marketing Research 37 (2): 187–202.

46.

Park

Kim

2005. “The Effects of Decoys on Preference Shifts: The Role of Attractiveness and Providing Justification.” Journal of Consumer Psychology 15 (2): 94–107.

47.

Park

J. Y.

Jang

2013. “Confused by Too Many Choices: Choice Overload in Tourism.” Tourism Management 35:1–12.

48.

Payne

J. W.

Bettman

J. R.

Johnson

E. J.

1993. The Adaptive Decision Maker. New York: Cambridge University Press.

49.

Perdue

R. R.

Meng

2006. “Understanding Choice and Rejection in Destination Consideration Sets.” Tourism Analysis 11 (6): 337–48.

50.

Pocheptsova

Amir

Dhar

Baumeister

R. F.

2009. “Deciding without Resources: Resource Depletion and Choice in Context.” Journal of Marketing Research 46 (3): 344–55.

51.

Scarpi

2011. “The Impact of Phantom Decoys on Choices in Cats.” Animal Cognition 14 (1): 127–36.

52.

Scheibehenne

Greifeneder

Todd

P. M.

2009. “What Moderates the Too-Much-Choice Effect.” Psychology and Marketing 26 (3): 229–53.

53.

Schoemaker

P. J. H.

1982. “The Expected Utility Model: Its Variants, Purposes, Evidence and Limitations.” Journal of Economic Literature 20 (2): 529–63.

54.

Schwartz

2000. “Self-Determination: The Tyranny of Freedom.” American Psychologist 55:79–88.

55.

Schwartz

J. A.

Chapman

G. B.

1999. “Are More Options Always Better? The Attraction Effect in Physicians’ Decisions about Medications.” Medical Decision Making 19 (3): 315–23.

56.

Shafir

1993. “Choosing versus Rejecting: Why Some Options Are Both Better and Worse Than Others.” Memory and Cognition 21 (4): 546–56.

57.

Shafir

Simonson

Tversky

1993. “Reason-Based Choice.” Cognition 49 (1/2): 11–36.

58.

Simonson

1989. “Choice Based on Reasons: The Case of Attraction and Compromise Effects.” Journal of Consumer Research 16 (2): 158–74.

59.

Sirakaya

Woodside

A. G.

2005. “Building and Testing Theories of Decision Making by Travellers.” Tourism Management 26:815–32.

60.

Smith

S. J.

Parsa

H. G.

Bujisic

van der Rest

J. P.

2015. “Hotel Cancelation Policies, Distributive and Procedural Fairness, and Consumer Patronage: A Study of the Lodging Industry.” Journal of Travel & Tourism Marketing 32 (7): 886–906.

61.

Sokolova

Krishna

2016. “Take It or Leave It: How Choosing versus Rejecting Alternatives Affects Information Processing.” Journal of Consumer Research 43 (4): 614–35.

62.

Steckenreuter

Wolf

I. D.

2013. “How to Use Persuasive Communication to Encourage Visitors to Pay Park User Fees.” Tourism Management 37:58–70.

63.

Tan

Liu

Gursoy

2018. “Evaluation Nudge: Effect of Evaluation Mode of Online Customer Reviews on Consumers’ Preferences.” Tourism Management 65:29–40.

64.

Thaler

1985. “Mental Accounting and Consumer Choice.” Marketing Science 4 (3): 199–214.

65.

Thaler

Sunstein

2008. Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven: Yale University Press.

66.

Crompton

J. L.

1990. “Attitude Determinants in Tourism Destination Choice.” Annals of Tourism Research 17 (3): 432–48.

67.

Van Middelkoop

Borgers

Timmermans

2003. “Inducing Heuristic Principles of Tourist Choice of Travel Mode: A Rule-Based Approach.” Journal of Travel Research 42 (1): 75–83.

68.

Wedell

D. H.

1997. “Another Look at Reasons for Choosing and Rejecting.” Memory & Cognition 25 (6): 873–87.

69.

Wedell

D. H.

Pettibone

J. C.

1999. “Preference and the Contextual Basis of Ideals in Judgment and Choice.” Journal of Experimental Psychology 128 (3): 346–61.

70.

Weiner

2000. “Intrapersonal and Interpersonal Theories of Motivation from an Attributional Perspective.” Educational Psychology Review 12 (1): 1–14.

71.

Woodside

A. G.

King

R. I.

2001. “An Updated Model of Travel and Tourism Purchase-Consumption Systems.” Journal of Travel & Tourism Marketing 10 (1): 3–27.

72.

Woodside

A. G.

Sherrel

1977. “Traveler Evoked, Inept, and Inert Sets of Vacation Destinations.” Journal of Travel Research 16 (1): 14–8.

73.

Xia

Monroe

K. B.

Cox

J. L.

2004. “The Price Is Unfair! A Conceptual Framework of Price Fairness Perceptions.” Journal of Marketing 68 (4): 1–15.