Different or Similar Choices: The Effect of Decision Framing on Variety Seeking in Travel Bundle Packages

Abstract

This research investigates whether different ways of framing decisions influence travelers’ variety-seeking tendencies in choosing bundled product options. Based on the literatures of bundling, variety-seeking, and the decision-framing effect, we empirically test whether travelers show higher variety-seeking in travel package decisions when the bundle package is selected from a combined decision rather than from two single decisions. We also examine whether this different variety-seeking tendency is influenced by travelers’ preference for consistency. The results of six experimental studies support our main proposition, and the substantive theoretical and managerial implications of the findings are discussed.

Keywords

travel bundle variety-seeking decision framing preference for consistency

Introduction

According to U.S. Department of Commerce data published in 2015, approximately 30% of all passengers who planned leisure/holiday travel used Internet booking and obtained their information from an online travel agency source. When travelers search for information or decide to buy travel services, they often have opportunities to buy bundle package deals from travel service providers. With the popularity of online travel services, bundle packages are more easily and commonly used promotional tools for most travel service agencies. For example, many of the highly recognized top travel websites such as expedia.com, travelocity.com, and orbitz.com provide package options for consumers to choose multiple elements of travel service, including flights, hotels, car rental, and travel insurance. The travel bundle package refers to a combination of related travel services in a single offer (Morrison 2010). Bundle packages can benefit consumers because of their convenience or competitive pricing; moreover, they benefit travel service providers because they increase the attractiveness of travel options and overall customer satisfaction (e.g., Noone and Mattila 2009; Tanford, Baloglu, and Erdem 2012).

According to World Best Enterprises (2004), website design and content are considered to be essential factors that contribute the most to the success of online business. The decision of how to offer different elements of travel services and goods as bundle packages is very important for travel service providers. Nonetheless, there is a lack of research in the tourism literature that addresses effective ways of combining, framing, and presenting these options in travel website design and content. Consumers can seek different levels of variety in their decision outcomes/processes of selecting multiple options in terms of the price level, attribute level, and/or brand level. For example, when a traveler buys a bundle package consisting of a flight ticket and rental car, he or she could choose a relatively cheap economy-class flight ticket and an inexpensive small rental car—a combination considered to be low variety-seeking in terms of price level. On the other hand, his or her bundle choice would be high in variety-seeking if he or she chooses a cheap economy-class airline ticket, but an expensive large-sized rental car. Variety is also generated from non-price-based considerations, such as brand. The bundle choice could be low in variety when the same name of a franchised hotel is chosen in different locations; however, it could be high in variety if two different names of franchised hotels are chosen in the destinations.

The primary purpose of this article is to investigate how decision framing in travel package options and consumers’ tendency to seek variety in their choices are related. Specifically, we aim to examine the influence of the two different ways in which bundle options are combined:

travel websites allow consumers to select elements a la carte by clicking on each element, one by one; or

travelers are offered several pre-determined sets of bundle packages and are asked to choose a bundle package option on the website.

This investigation is of particular importance in the travel industry, given that these two package options are being practiced in most online travel agencies, yet little is known regarding consumers’ views on these two different ways of product presentation. Currently, travelers can select each element one by one at expedia.com or orbitz.com and, alternatively, select bundle packages all at once by choosing a combined package option at freetravelguides.com or travelinformation.com. This study can shed light on understanding the influence of different ways of combining bundle packages on consumers’ decisions in the tourism context, thus extending previous research on the processes of decision making when choosing a variety of options (McAlister 1982; Ratner, Kahn, and Kahneman 1999). In the next section, we review bundling studies in the literatures of tourism, variety-seeking, and the decision-framing effect, followed by a series of experimental studies.

Literature Review and Prediction

Bundling in the Travel Industry

Bundling, a commonplace marketing method, involves selling two or more products/services in a package, oftentimes for a special price (Adams and Yellen 1976; Guiltinan 1987). Examples of bundled products in the hospitality and tourism industries include combined restaurant meals, hotel amenities, and travel packages (Kim, Bojanic, and Warnick 2009; Kwon and Jang 2011; Repetti, Roe, and Gregory 2015). Bundled travel products such as accommodation and transportation are particularly prevalent in online travel intermediaries (OTAs; Kim, Bojanic, and Warnick 2009; Tanford, Baloglu, and Erdem 2012). Purchasing bundled travel products is favored by frequent travelers, given that by booking a flight and a hotel together, they can save a substantial amount of money on a trip (Kohl 2016). It is no wonder that previous bundling research has mainly focused on the economic value perceived by consumers of bundled products (e.g., Soman and Gourville 2001; Xia, Monroe, and Cox 2004; Noone and Mattila 2009). Another benefit of purchasing bundled products is the reduced search costs and greater convenience for consumers’ who need to make a series of purchase decisions that are interrelated to one another (Guiltinan 1987; Tanford, Baloglu, and Erdem 2012).

Bundling is also favored by suppliers, as selling products in bundles can save in distribution and transaction costs; furthermore, it can be sales-effective by maximizing sales volume (Guiltinan 1987; Stremersch and Tellis 2002). Bundled products are often sold at a discounted rate, as suppliers save various costs associated with promotion, distribution, and reservations (Stremersch and Tellis 2002; Chiam, Soutar, and Yeo 2009). Given the “perishability” of products in the hospitality and tourism industries, selling multiple products in a package is an attractive sales option for firms striving to avoid untaken restaurant tables, unoccupied hotel rooms, and empty flight seats. Suppliers, therefore, are increasingly offering their products and services jointly at OTAs to boost the sales of their products and services in combinations such as airlines and hotels, and airlines and rental cars (Kim, Bojanic, and Warnick 2009; Tanford, Baloglu, and Erdem 2012).

A related concept called dynamic packaging enables travelers to build and reserve service packages by bundling travel components (Fesenmaier et al. 2003; Cardoso and Lange 2007). In the websites of OTAs, customers are informed of available and attractive package options created by recommender systems (Ricci and Werthner 2002; Garcia, Torre, and Linaza 2013). Although the range of travel service options to be bundled varies, the objective of dynamic packaging technology is to enable consumers to choose travel components by assembling a package that most suits their preferences. Furthermore, owing to the development of technology that allows marketers to access consumers’ previous purchase experiences (cf. market basket analysis), it is possible for OTAs and suppliers to maximize their yield by customizing service options that appeal to consumers (Pan, Lee, and Tsai 2014; Solnet, Boztug, and Dolnicar 2016).

For bundling strategies, there are two core dimensions for marketers to take into account: price and product (Stremersch and Tellis 2002). Marketers are required to choose how many different products are included, and how much of a discount is offered for a bundled product. Most bundled travel products are sold at a discounted rate in a package, while no discount is applied when each travel product is purchased separately. Most OTAs make both options available: (1) purchasing each travel product separately and (2) purchasing multiple travel products in a package. It has become imperative for marketers to effectively design bundled products by understanding how consumers make purchase decisions. Based on theories of variety-seeking and the decision-framing effect, this research examines how different presentations of bundled travel products influence travelers’ decisions using a series of experimental studies. The following sections entail the literature of variety-seeking theories and the decision-framing effect, along with the research propositions.

Variety-Seeking Literature

Many researchers have shown that people often tend to seek variety in terms of the decision outcome (Menon and Kahn 1995; Kahn, Ratner, and Kahneman 1997; Chintagunta 1998; Ratner, Kahn, and Kahneman 1999; Inman 2001; Ratner and Kahn 2002). For example, Simonson (1990) showed that the number of choices made at one time can influence people’s variety-seeking tendency. He argues that people tend to choose more diverse alternatives when they simultaneously choose multiple items than when they sequentially make the same number of choices. Specifically, the participants in his study were asked to choose three out of six snacks during three consecutive weeks. Half of the participants were asked to choose all three snacks in the first week (i.e., the simultaneous-choice condition), although they did not receive their snacks until the appointed times. The other participants were asked to choose one snack for each week (i.e., the sequential-choice condition). The results indicated that the participants in the simultaneous-choice condition were more likely to select a variety of items than those in the sequential-choice condition. In a follow-up study, Simonson and Winer (1992) replicated the previous findings using actual supermarket scanner data.

Table 1.

Summary of Studies 1–4.

Study	Independent Variable	Dependent Variable	Target Product	N (Sample)
1	Decision framing of the bundle	Choice	Travel bundle packages consisting of an airline ticket ($300, $400) and hotel choice (3-star, 4-star)	47 (students)
2A	Decision framing of the bundle in ascending order of price	Choice	Travel bundle packages consisting of an airline ticket (economy, premium economy) and hotel choice (3-star, 4-star)	321 (adults)
2B	Decision framing of the bundle in descending order of price	Choice	Travel bundle packages consisting of an airline ticket (economy, premium economy) and hotel choice (3-star, 4-star)	300 (adults)
2C	Decision framing of the bundle, controlling for price transparency and income	Choice	Travel bundle packages consisting of an airline ticket (economy, premium economy) and hotel choice (3-star, 4-star)	186 (adults)
3	Decision framing of the bundle, hotel brands	Choice	Travel bundle package of existing hotel brands (a low-price/quality hotel brand, a high-price/quality hotel brand)	272 (adults)
4	Decision framing of the bundle, preference for consistency	Choice	Travel bundle packages consisting of an airline ticket (economy, premium economy) and a rental car (standard, full size)	239 (adults)

Similar phenomena were proposed by Read and Loewenstein (1995). They (also Read et al. 2001) introduced the concept of diversification bias. This type of bias concerns people’s tendency to adopt more variety when choosing multiple alternatives at the same time, as opposed to choosing a single alternative sequentially (Read and Loewenstein 1995). They showed that people included more varied options when choosing multiple options at the same time versus choosing a single option sequentially. To explain this diversification bias, Read, Loewenstein, and Rabin (1999) use the concept of (narrow and broad) choice bracketing, which designates the grouping of individual choices together into sets. Narrow choice bracketing involves a small set of choices, containing one or very few choices, whereas broad choice bracketing involves a larger set of choices. One example to consider could be the decision to smoke or abstain. Narrow choice bracketing would entail the consumer’s decision to smoke one cigarette at a time. However, broad choice bracketing would be more likely to entail the consumer’s decision to smoke 7,300 cigarettes over a one-year period. Read, Loewenstein, and Rabin (1999) suggest that broad choice bracketing usually yields higher utility because it allows people to take into account all of the consequences of their actions. However, people often fail to bracket broadly when it would be feasible for them to do so.

This variety-seeking tendency is a powerful motivation for consumers, in that they sometimes choose the less-preferred items for the sake of variety. For example, Ratner, Kahn, and Kahneman (1999) provide empirical evidence of people choosing less-preferred options. Participants were asked to rate the experience of listening to multiple songs. Half of them were exposed only to their favorite song (i.e., the repeated-favorite condition), whereas the other half of the participants were exposed to both their favorite and not favorite songs (i.e., the mixed-sequence condition). The results indicate that the participants’ reported listening experience declined more steeply in the repeated-favorite condition than in the mixed-sequence condition. This finding is related to the antecedents of variety seeking, in that variety seeking happens as a result of the information-gathering function from the less-preferred option (Kahn and Lehmann 1991), or in order to reduce satiation (McAlister 1982). Therefore, the variety-seeking tendency in decision making is a common phenomenon in consumer choice behavior.

On the other hand, the variety-seeking tendency could be related to the decision process. In other words, variety seeking can occur in the manner with which consumers choose items, as well as in the items they choose. For example, Drolet (2002) provides evidence that variety seeking can be extended to decision processes. Specifically, the participants in her study were asked to make three sets of choices between a high-quality/high-price option and a low-quality/low-price option sequentially. In addition, the choice set was different. The results indicate that consumers who selected the high-quality/high-price option (low-quality/low-price option) in the first two choices were likely to select the low-quality/low-price option (high-quality/high-price option) for the third choice. These results suggest that consumers have a tendency to vary their use of decision rules. In a follow-up study, Kim and Drolet (2003) replicated the previous findings and additionally suggested that these decision rules governing variety seeking are stronger for people in individualistic cultures than for those in collectivistic cultures. In sum, people seek variety through the decision process, and the variety of the decision process also influences the decision outcome.

In the tourism literature, the concept of variety seeking has been examined as the consumer’s propensity or trait in switching or revisiting destinations/service providers (e.g., Niininen, Szivas, and Riley 2004; Castro, Armario, and Ruiz 2007; Jang and Feng 2007; Jung and Yoon 2012; Assaker and Hallak 2013). For example, Castro, Armario, and Ruiz (2007) and Jang and Feng (2007) investigated the impact of travelers’ variety-seeking tendencies on destination revisit intentions, while Assaker and Hallak (2013) and Jung and Yoon (2012) examined consumers’ variety-seeking tendency as a moderator between customers’ satisfaction and behavioral intentions. Given that few efforts have been made to understand what influences consumers’ variety-seeking tendency in the tourism literature, we examine decision framing as a determinant of travelers’ variety-seeking propensity.

Finally, most of the previous literatures regarding variety-seeking have been conducted in the same and single product category. However, we will investigate variety-seeking mostly from a different, but related product category in travel packages. This investigation is related to the research stream suggesting that the variety-seeking tendency could occur across different, but related, product categories (see Lattin and McAlister 1985; Kim and Drolet 2003, 374).

Decision-Framing Effect

Decision framing can differ in a variety of ways, and may involve:

the number of choices, e.g., choosing one versus two (Simonson 1990; Read and Loewenstein 1995; Kim, Kim, and Marshall 2014; Mittelman et al. 2014);

the way the choice is expressed, e.g., choosing versus rejecting (Huber, Neale, and Northcraft 1987; Shafir, Simonson, and Tversky 1993; Chernev 2009; Laran and Wilcox 2011); or

the decision mode, e.g., choice versus judgment/ratings (Einhorn, Kleinmuntz, and Kleinmuntz 1979; Einhorn and Hogarth 1981; Johnson and Russo 1984; Billings and Scherer 1988).

It should be noted that the number of choices significantly influences the variety-seeking tendency (i.e., Simonson 1990; Read and Loewenstein 1995), as mentioned earlier. Likewise, decision framing may also be related to the way bundle options are created, whether in a sequential or nonsequential manner.

Recently, Mittelman et al. (2014) have provided empirical evidence that people show higher levels of variety seeking when they are asked to make two single choices sequentially than when they are asked to choose from a combined offering. For example, participants in their study were asked to choose between two cans of soft drinks: either Coke or Sprite. Participants given two-single-choice offerings frequently selected different soft drink brands, compared to those in the one-combined-choice condition. They termed this phenomenon the “offer-framing effect.” The authors have suggested that the offer-framing effect could be explained by different decision objectives. In detail, people in the two-single-choice offering conditions tend to ask themselves a typical question when they make the first choice: “What is my favorite brand?” However, when people make the second (the same choice set) choice in a row, they already own the product they prefer the most; hence, people are less likely to ask themselves the same question. Rather, people ask themselves a different question: “Should I make the same or a different choice?” As a consequence, people are more likely to seek variety in the second single-offering condition. This implies that people’s inherent preference is reflected in the one-combined-choice condition; however, only their first choice (not the second choice) in the single-offering condition reflects their inherent preferences. Taken together, Mittelman et al. (2014) suggested that the change in the question or approach toward the sequential second choice in the single-offering condition could result in high levels of variety seeking overall. The single-offering choice could enhance people’s variety-seeking tendency, ultimately hurting the brand loyalty of their favorite brands.

Proposition Development

In contrast to Mittelman et al.’s (2014) study, which suggested a higher variety-seeking tendency in the two single (vs. one combined) choice condition, we propose the reverse effect of Mittelman et al.’s (2014) findings in travel bundle package decisions. Based on the preference of consistency theory (e.g., Cialdini, Trost, and Newsom 1995), the balancing/highlighting strategy (Dhar and Simonson 1999), and the different characteristics of travel packages (versus consumption goods), we predict that the variety-seeking tendency will be higher in the one-combined-choice than in the two-single-choice condition in travel bundle decisions.

The main differences between Mittelman et al.’s (2014) research and this study are the types of elements constituting the bundle packages: products (e.g., Coke and Sprite in Mittelman et al.’s study) and travel services (e.g., flight tickets and hotels in this study). Mittelman et al. (2014) argued that people engage in two processes of probing: “What is my favorite brand?” in the first choice, then “Should I make the same or a different choice?” in the second choice. However, in travel bundle situations, travelers in the two-single-choice condition are less likely to engage in the same probing process. Rather, they are more likely to repeat the first question: “What is my favorite hotel?” in the second choice, since the choice involving each travel service is regarded as independent and different. Kim and Park (2016) also found lower variety-seeking under two-single-choice situations when two decisions were framed differently (e.g., a choice task for the first task, and a rejection task for the second task). They suggested the moderating role of decision similarity for the offer-framing effect. Because of the different characteristics of the bundle package target (travel service vs. consumption goods), we expect a lower variety-seeking tendency in the two-single-choice situations.

This argument is also highly related to the preference for consistency theory. People are assumed to have a high level of consistency (Epstein 1979; Cialdini, Trost, and Newsom 1995; Cialdini 2009), in that people want to show their consistent behavior and opinions across different situations. For example, Epstein (1979) argued that the stability of personality could easily be found in repeated events. In the domain of decision making, people also have “preference consistency,” which is related to the tendency to adhere to the same decision rule, especially under multiple decision task situations. For example, if people show their consistency or intransitivity in multiple binary choices, they are likely to have high preference consistency (Lee, Amir, and Ariely 2009). Brand loyalty (e.g., Nam, Ekinci, and Whyatt 2011) or customer loyalty (e.g., Kandampully, Zhang, and Bilgihan 2015) are also highly related to consistent preferences in the domains of hospitality and tourism. High brand loyalty indicates the consistent repetition of choosing specific brand options against competing brands. Nam, Ekinci, and Whyatt (2011) suggested that previous satisfaction increases brand loyalty. Furthermore, the concept of variety seeking is contrary to the concept of consistency motivation. For example, Fishbach, Ratner, and Zhang (2011) have suggested that people have two opposite motivations for consistency seeking and variety seeking; additionally, they provided empirical evidence that activating one construct results in a reduction in the other construct. The consistency motivation has also been investigated across different situations.

In sum, the consistency motivation could influence travel bundle package decisions, in that people in the two-single-choice condition are more likely to show consistent decisions, as compared to those in the one-combined-choice condition. Specifically, we assume that travelers who made the first decision in the two-single-choice condition tend to show a similar decision rule or decision outcome in the second decision. We expect lower variety-seeking in the two-single (one-combined)-choice condition, mainly because of the consistency motivation from decision makers.

The concept of consistency in multiple decision situations is also closely related to the balancing/highlighting strategy in subsequent choice situations (Dhar and Simonson 1999). Dhar and Simonson (1999) investigated the influence of a previous choice (e.g., entrée) on a subsequent one (e.g., dessert) and argued that if a choice involves a trade-off between two goals (e.g., pleasure and good health), consumers tend to prefer the balancing strategy. This is because focusing on a single goal can generate satiation or boredom. For instance, the choice of an unhealthy pizza dish was more preferable to participants after a workout versus after watching television. This result indicates that people try to balance two different goals, such as health and enjoyment. On the other hand, if choice involves a trade-off between a goal (e.g., pleasure) and a resource (e.g., money), consumers tend to prefer the highlighting strategy. In this case, consumers do not want to be adversely affected by the less enjoyable component (such as an inexpensive dessert ruining the taste of a previously expensive and delicious entrée) by adopting the balancing option. For example, the researchers show that compared to people who order an inexpensive main course, those who order an expensive main course are also likely to choose an expensive dessert (Dhar and Simonson 1999). The selection of the highlighting strategy over the balancing strategy is directly related to the consistency motivation.

In this article, we assume that most travel packages involve a trade-off between a goal (e.g., pleasure) and a resource (e.g., money), and that the conflict or trade-off is more salient in the two-single-choice rather than the one-combined-choice condition. Therefore, people under the two-single-choice condition will exhibit the highlighting strategy, which is related to low variety seeking. In conclusion, based on the three reasons above, we expect that when people are exposed to the two sequential single-offering condition (vs. the one-combined-offering condition), they are less likely to seek variety in travel bundle packages. More formally, we state this as follows:

Overall Proposition: People will show lower variety seeking in travel package decisions when the bundle package is selected by two single decisions rather than by a combined decision.

In the next section, we address this proposition by varying the types of decision framing across five experiments. Study 1 presents the initial confirmation of the prediction, above, in the context of a vacation trip consisting of an airline ticket and a hotel as a bundle package. In studies 2A–2C, we replicate the results of Study 1 with a general public sample and exclude multiple alternative explanations (i.e., the order of the information presentation or the pricing effect). Study 3 shows the same effect with the use of real hotel brands. Finally, Study 4 demonstrates a boundary condition for the above effect through measuring preference for consistency. All six empirical studies suggest that the various types of variety-seeking could be reduced when the decision involves a two-single-choice condition rather than one combined condition for travel package decisions.

Study 1

The main purpose of Study 1 was to test the effect of decision framing on the variety-seeking tendency in the choice of travel bundle packages consisting of an airline ticket and hotel choice. We expected that the variety-seeking tendency would be lower in the two-single-choice condition than under the combined-choice condition.

Method

Subjects and design

Forty-seven undergraduate and graduate students (72.3% female; average age = 23.5) at a large university in New Zealand participated in the study. Participants were assigned to one of two decision-framing conditions—selecting a bundle through two single choices versus one combined choice—in a between-subjects experimental design.

Procedure

Participants first were asked to read the study’s general instructions and then to imagine that they were planning to take a vacation in Melbourne, Australia. They were also asked to select an airline ticket and hotel as a bundle package. We manipulated the decision framing by changing the type of decision. Specifically, participants in the two-single-choice condition were first asked to select an airline ticket from two options. One was a regular economy-class ticket priced at $300 with limited baggage and meal service, and the other option was a relatively expensive premium economy-class ticket priced at $400 with better baggage and meal services. Then, they were asked to select a hotel from two options, consisting of a relatively inexpensive three-star hotel and an expensive four-star hotel. On the other hand, participants in the one-combined-choice condition were first exposed to the combined-choice options, consisting of four combinations of airline tickets and hotels, as shown in Figure 1, and were asked to select one option. Finally, participants completed all sections on a survey, which included questions about their gender and age. Then, they were thanked and debriefed.

Figure 1.

Stimuli of Study 1.

Results

In this analysis, we assume that the combination of a cheaper airline ticket and hotel or the combination of an expensive airline ticket and hotel is low variety seeking. In contrast, the choice of a mix with a cheaper airline ticket and an expensive four-star hotel or with a higher airline ticket and an inexpensive three-star hotel is considered to be high variety seeking.

The results confirmed our expectation: participants in the one-combined-choice condition showed higher variety seeking (52.4% [11/21]) than those in the two-single-choice condition (23.1% [6/26], χ²(1) = 4.32; p < .05). The detailed decision outcome is described in Table 2.

Table 2.

Results of Studies 1 and 2.

Study	Choice Condition	Option
Study	Choice Condition	Economy Ticket and 3-Star Hotel	Economy Ticket and 4-Star Hotel	Premium Ticket and 3-Star Hotel	Premium Ticket and 4-Star Hotel
1: Trip to Melbourne	Two single	34.6% = 9/26	7.7% = 2/26	15.4% = 4/26	42.3% = 11/26
1: Trip to Melbourne	One combined	9.5% = 2/21	47.6% = 10/21	4.8% = 1/21	38.1% = 8/21
2A: Ascending order	Two single	39.8% = 64/161	37.9% = 61/161	3.1% = 5/161	19.3% = 31/161
2A: Ascending order	One combined	21.3% = 34/160	50.0% = 80/160	10.0% = 16/160	18.8% = 30/160
2B: Descending order	Two single	38.2% = 58/152	30.9% = 47/152	8.6% = 13/152	22.4% = 34/152
2B: Descending order	One combined	19.6% = 29/148	45.3% = 67/148	6.8% = 10/148	28.4% = 42/148
2C: Price discounting information	Two single	32.3% = 30/93	33.3% = 31/93	7.5% = 7/93	26.9% = 25/93
2C: Price discounting information	One combined	19.4% = 18/93	43.0% = 40/93	14.0% = 13/93	23.7% = 22/93

Note: The first and fourth options represent low variety seeking (i.e., a combination of cheaper options or a combination of expensive options), whereas the second and third options represent high variety seeking (i.e., combination of a cheaper option and an expensive option).

Discussion

The results of this study generally supported our core prediction. In this study, we found a significant impact of decision framing on the variety-seeking tendency regarding bundle package decisions. One possible explanation for this finding could be a price-based explanation. In detail, the price pressure in the two-single-choice decision condition could generate the avoidance of selecting the two expensive options sequentially. Put differently, in the two-single-choice condition, participants who had already selected the expensive airline ticket might feel monetary pressure and choose the less expensive option in the second element choice. However, as can be seen in Table 2, the data did not support this alternative explanation. Specifically, the choice ratio for selecting two expensive options was not significantly different from the two-single-choice condition (42.3% [11/26]) or the one-combined-choice condition (38.1% [8/21], χ²(1) = 0.09; p =.77).

Studies 2A and 2B

The main purpose of Studies 2A and 2B was to replicate the results of Study 1 using different samples (i.e., the general public from an online panel). Specifically, we asked adults from the general public rather than students to participate. Furthermore, we asked about participants’ prior travel experience in order to increase the external validity of our finding. In addition, to avoid the price effect, we did not provide price information. Finally, in the previous study, the options were displayed from an inexpensive to an expensive option. The price/quality display pattern (e.g., ascending and descending format) could influence consumers’ decisions and judgments by changing the reference price (e.g., Danziger and Segev 2006). Therefore, in this second study, we conducted two experiments: one in ascending order (Study 2A); and the other in descending order (Study 2B).

Method

Subjects and design

Three hundred twenty-one U.S. adults (Study 2A: average age = 33.9, 41.1% female) and 300 US adults (Study 2B: average age = 32.1, 37.7% female) from an online panel (i.e., Amazon Mechanical Turk)¹ were recruited for nominal compensation. Participants in this study were assigned to one of two conditions (decision framing of the bundle: two single choice vs. one combined choice) in a between-subjects experimental design.

Procedure

The procedures were similar to those employed in Study 1, with the exception of several modifications. Participants were asked to imagine that they were planning to take a vacation soon, and that they needed to select an airline ticket and hotel as a bundle package. Rather than providing the full information, we merely provided a general description of each option, such as “economy” versus “premium economy” and a “three-star hotel” versus a “four-star hotel.” The manipulation of the decision framing was similar to that of Study 1. Specifically, participants in the two-single-choice condition were first asked to select an airline ticket from two options. One option was described as an economy ticket, and the other one was described as a premium economy ticket. Then, they were asked to select a hotel from two options between a three-star hotel and a four-star hotel. We did not provide the price or any other quality information. On the other hand, participants in the one-combined-choice condition were first exposed to the combined-choice options consisting of four combinations of an airline ticket and hotel and were asked to select one. We also manipulated the display order, in that the participants in Study 2A were exposed to options in ascending price order, from the least expensive to the most expensive option. In contrast, the participants in Study 2B were exposed to the opposite order (from the most expensive to the least expensive option). Finally, participants completed all sections of a survey, which asked their gender and age as well as their personal travel experience involving airlines and hotels in the last two years. Then, participants were thanked.

Results

Study 2A

The results of Study A confirmed our expectation. Specifically, participants in the one-combined-choice condition showed higher variety-seeking (60.0% [96/160]) than those in the two-single-choice condition (41.0% [66/161], χ²(1) = 11.60; p < .01). The detailed decision outcome is described in Table 2. In addition, the choice ratio for selecting two expensive options was not significantly different from the two experimental conditions (two single choice = 19.3% [31/161] vs. one combined choice = 18.8% [30/160], χ²(1) = .01; p =.91). Therefore, price- or budget-related concerns could not explain this result.

In order to investigate the impact of previous travel experience, we excluded 82 participants who did not have any experience traveling with airlines or hotels (n = 239, average age = 33.4, 42.7% female). The result was quite similar to that of all participants. Specifically, participants in the one-combined-choice condition (60.2% [74/123]) showed higher variety-seeking than those in the two-single-choice condition (44.0% [51/116], χ² (1) = 6.28; p < .05).

Study 2B

In this study, the options were displayed in descending order of price/quality. The results were similar to those of Study 2A. Specifically, participants in the one-combined-choice condition (52.0% [77/148]) showed higher variety-seeking than those in the two-single-choice condition (39.5% [60/152], χ² (1) = 4.76; p < .05). The detailed decision outcome is described in Table 2. In addition, the choice ratio for selecting two expensive options did not significantly differ between the two experimental conditions (two single choice = 22.4% [34/152] vs. one combined choice = 28.4% [42/148], χ²(1) = 1.43; p =.23). Therefore, price- or budget-related concerns could not explain this result, either.

We also excluded 76 participants who had not traveled using airlines or hotels in the past two years (n = 224, average age = 31.6, 37.1% female). The result was similar to that of all participants. Specifically, participants in the one-combined-choice condition (52.2% [59/113]) showed marginally higher variety seeking than those in the two-single-choice condition (40.5% [45/111], χ²(1) = 3.07; p <.10).²

Discussion

The results of this study also suggest the significant effect of decision framing on the variety-seeking tendency in bundle combinations. Furthermore, the results replicated the options shown without price information. In addition, the higher variety-seeking tendency under the bundle choice (vs. the two-single-choice) condition was found when options were presented in both ascending (Study 2A) and descending (Study 2B) order of price/quality. Finally, the results were consistent for the participants having recent travel experience and for all participants. In Study 2C, we replicate the finding in Study 2A in order to exclude alternative explanations for these findings.

Study 2C

The main purpose of Study 2C was to exclude several alternative explanations for the results from the previous studies. First, the role that price plays in a bundle choice could be different from the role it plays in two single choices. Specifically, the price of a prepackaged bundle is determined before any choice is made, whereas the total price of two single choices is determined only after choosing both elements. In order to exclude this different role of price, we provided the price information of bundle expectation for two conditions.

Second, related to the first point, the way pricing is revealed in the two experimental conditions generates different levels of transparency (i.e., providing the itemized price for each component or not; Tanford, Erdem, and Baloglu 2011; Tanford, Baloglu, and Erdem 2012) for the bundle and single-choice conditions. In this experiment, we provided the same price information across the two conditions and we also measured the perceived transparency across the two experimental conditions. If the transparency is a key underlying mechanism for the different choice effect from the bundle versus two-single-choice condition, the perceived transparency should differ between the two experimental conditions. In contrast, if our original theorizing is correct, we should have replicated the previous results even though individual component price information was provided and the perceived price transparency was the same across the two experimental conditions.

Finally, the socioeconomic characteristics of Amazon MTurk users could have influenced the choice pattern, in that they might have preferred the cheaper options. In order to control the personal wealth effect, we measured income and used a control variable for this study.

Method

Subjects and design

One hundred eighty-six U.S. adults (average age = 33.2, 52.2% female) from the same online panel (i.e., Amazon Mechanical Turk) were recruited for nominal compensation. Participants in this study were assigned to one of two conditions (decision framing of the bundle: two single choice vs. one combined choice) in a between-subjects experimental design.

Procedure

The procedures were similar to those employed in Study 2A, with a few modifications. First, before making a decision, all participants were informed of the following additional information: “By purchasing the airline ticket and hotel booking together, you will save 15% off the regular price on each component.” Therefore, all participants knew that their choice was a bundle outcome. In addition, they were informed about the same price discounting information in order to control for the possible effect of price transparency. Then, participants were first asked to select the option(s) based on two experimental conditions. After choosing their travel package, they were also asked to rate their agreement with three statements about the perceived price transparency of the options (i.e., “The price of the travel package above is transparent/I understand the pricing of the travel package above/ The price of the travel package above is clear”) along a 7-point scale (1 = completely disagree, 9 = completely agree, Cronbach’s alpha = .91). Finally, participants completed all sections of a survey, which included questions about their gender and age, as well as their yearly family income along an 11-point scale (1=$0–$10,000, 11 = $100,001 or above). Then, participants were thanked.

Results

Choice pattern

The results of Study 2C replicated those of Study 2A. Specifically, participants in the one-combined-choice condition (57.0% [53/93]) showed higher variety seeking than those in the two-single-choice condition (40.9% [38/93], χ²(1) = 4.84; p < .05). The detailed decision outcome is described in Table 2. In addition, the choice ratio for selecting two expensive options was not significantly different from the two decision-framing conditions (two single choice = 26.9% [25/93] vs. one combined choice = 23.7% [22/93], χ²(1) = .26; p = .61). Therefore, price- or budget-related concerns could also not explain this result.

Additional analysis

In order to test the price transparency alternative explanation, we compared the perceived transparency across the two experimental conditions. The results indicated that the perceived transparency was not significantly different across the two experimental conditions (M_single = 4.19 vs. M_bundle = 3.85, F(1, 184) = 1.51, p > .10). Furthermore, in order to control for the role of income, we conducted a binary logistic analysis based on Preacher and Hayes (2008). We used Preacher and Hayes’s SPSS macro modules (model 1 with 5,000 bootstrapped samples). The independent variable is decision framing. The dependent variable is choice, and the moderator is income level. The results indicated that the main effect of income level was not significant (z = .46, p = .65, 95% bootstrap confidence interval [CI]: –0.077, 0.125). More importantly, the interaction effect between the two factors was not significant either (z = −0.20, p = .84, 95% CI: −0.222, 0.182), whereas the main effect of decision framing was still significant (z = 2.21, p = .027, 95% CI: 0.074, 0.124). In detail, the choice pattern for a different choice was consistent, regardless of income level.³

Discussion

Through this additional study, we were able to exclude several alternative explanations, such as the different role of price, price transparency, and income effect. We also replicated the results very similarly. One weakness of Studies 1 and 2 is that the stimuli are hypothetical. Thus, we conducted Study 3 using real hotel brands existing on the market.

Study 3

The main purpose of Study 3 is to replicate Studies 1 and 2 using actual brands as choices. In addition, even though we did not provide price information in Study 2, the options could have been different in terms of the information on quality. The descriptions of the quality of the options could have implicitly suggested the price level; thus, the budget-based explanation could explain the results. In order to exclude this alternative explanation, we did not provide any price/quality information but still manipulated the level of service and quality by using existing hotel brands (i.e., a low-price/quality hotel brand, a high-price/quality hotel brand). Additionally, we mainly used only a single product category (i.e., hotels) in each bundle package in this study. Finally, we used a pictorial format for providing option information, in order to produce results that were robust across different information presentation modes (e.g., Kim et al. 2016).

Method

Subjects and design

Two hundred seventy-two U.S. adults (average age = 33.0, 40.8% female) from an online panel (i.e., Amazon Mechanical Turk) were recruited for nominal compensation. Participants in this study were assigned to one of two (decision framing of the bundle: two single choice vs. one combined choice) × two (hotel brands: similar vs. different in price/quality level) conditions in a between-subjects experimental design.

Procedure

The procedures were similar to those employed in previous studies, with the exception of several modifications. Participants were asked to imagine that they were planning to take a vacation trip to two cities in the near future and to select two hotels for each destination of cities A and B. The participants were offered real hotel brand logos. For the similar price/quality condition, “Courtyard by Marriott” and “Holiday Inn” were chosen because these brands are considered as being at a similar level in terms of price and quality of service. On the other hand, we used the brands of “Courtyard by Marriott” and “JW Marriott” for a condition with options of differing price/quality, because the latter hotel is considered as having a higher price and better quality. Participants in the two-single-choice condition were first asked to select one hotel from the two options of hotels for city A. Then, they were asked to select another hotel for city B. On the other hand, participants in the one-combined-choice condition were first exposed to the combined-choice options, consisting of four combinations of hotels, and were asked to select one option. Finally, participants completed all sections of a survey, which included asking their gender and age, as well as their personal travel experience involving airlines and hotels in the last two years. Then, participants were thanked.

Results

In this analysis, we assume that selecting the same brand of hotel for two cities indicates low variety seeking, whereas selecting different brands signifies high variety seeking. To investigate the effect of the type of hotel in terms of price/quality, we first conducted a binary logistic analysis with the two experimental conditions. The main effect and interaction effect regarding the type of hotel in terms of price/quality was not significant (all p’s >.28). However, the main effect of decision framing was significant (Wald = 10.93, p <.01). Overall, the result confirmed our prediction. Specifically, participants in the one-combined-choice condition (36.1% [48/133]) showed higher variety seeking than those in the two-single-choice condition (18.0% [25/139], χ²(1) = 11.35; p < .01). We conducted a separate analysis for the same and different brand type in terms of price/quality level. For the different price/quality condition, the selection of two different hotel brands was higher in the one-combined-choice condition (33.3% [22/66]) than in the two-single-choice condition (15.3% [11/72], χ²(1) = 6.17; p < .05). For the same price/quality condition, the result was quite similar (selecting two different hotel brands: the one-combined-choice condition = 38.8% [26/67] vs. the two-single-choice condition = 20.9% [14/67], χ²(1) = 5.13; p < .05). The detailed decision outcome is described in Table 3.

Table 3.

Results of Studies 3 and 4.

Study	Choice Condition	Option
Study	Choice Condition	Hotel A and Hotel A	Hotel A and Hotel B	Hotel B and Hotel A	Hotel B and Hotel B
3: Same-quality Hotel	Two single	50.7% = 34/67	14.9% = 10/67	6.0% = 4/67	28.4% = 19/67
3: Same-quality Hotel	One combined	44.8% = 30/67	17.9% = 12/67	20.9% = 14/67	16.4% = 11/67
3: Different-quality Hotel	Two single	51.4% = 37/72	12.5% = 9/72	2.8% = 2/72	33.3% = 24/72
3: Different-quality Hotel	One combined	37.9% = 25/66	25.8% = 17/66	7.6% = 5/66	28.8% = 19/66
		Economy ticket and standard car	Economy ticket and full-size car	Premium ticket and standard car	Premium ticket and full-size car
4: Low PFC	Two single	49.0% = 24/49	14.3% = 7/49	26.5% = 13/49	10.2% = 5/49
4: Low PFC	One combined	45.6% = 26/57	15.8% = 9/57	29.8% = 17/57	8.8% = 5/57
4: High PFC	Two single	60.4% = 32/53	15.1% = 8/53	9.4% = 5/53	15.1% = 8/53
4: High PFC	One combined	43.8% = 21/48	16.7% = 8/48	29.2% = 14/48	10.4% = 5/48

Note: The first and fourth options represent low variety seeking, whereas the second and third options represent high variety seeking.

In order to investigate the impact of previous travel experience, we excluded 68 participants who did not have any experience staying at hotels in the past two years (n = 204, average age = 33.5, 39.7% female). The results were similar to that of all participants. A binary logistic analysis indicated that only the main effect, of decision framing, was significant (Wald = 8.03, p < .01). Specifically, participants in the one-combined-choice condition (34.6% [34/104]) showed higher variety seeking than those in the two-single-choice condition (17.0% [17/100], χ²(1) = 8.23; p < .01). This pattern was similar for the different price/quality condition (selecting two different hotel brands: the one-combined-choice condition = 32.7% [18/55] vs. the two-single-choice condition = 16.0% [8/50], χ²(1) = 3.93; p < .05), as well as for the same price/quality level (selecting two different hotel brands: the one-combined-choice condition = 36.7% [18/49] vs. the two-single-choice condition = 18.0% [9/50], χ² (1) = 4.38; p < .05)

Discussion

In this study, we found that choosing different hotels in the subsequent cities to be visited (i.e., variety seeking) was higher when the participants were in the combined-choice condition rather than in the two-single-choice condition. The results extend the generalizability of previous findings, in that the significant effect was replicated in the single service bundle package situation with an actual brand using the pictorial information presentation mode. In addition, the effect was also replicated in the same price/quality decision, as well. Therefore, we could exclude price-based or budget-based alternative explanations.

Study 4

In previous studies, we primarily showed the main effect of decision framing on the variety-seeking tendency in various travel decisions. The main purpose of Study 4 was to provide empirical evidence of the underlying mechanism of the decision-framing effect. In our theorizing, we consider variety-seeking and consistency behaviors as opposite constructs in the preference of travel packages. In addition, we expected the consistency motivation, especially in the two-single-choice condition, to generate a significant effect. In order to verify this argument, we measured individuals’ difference of preference for consistency (Cialdini, Trost, and Newsom 1995). If our theorizing is strong and correct, we should find a significant effect of preference for consistency (hereafter PFC) in the variety seeking for the two-single-choice condition. Specifically, the variety-seeking tendency should be lower when the participants have higher PFC for the two-single-choice situation. This is because the consistency motivation is more salient when people need to express their preference with multiple actions. In contrast, PFC will not influence the one-combined-choice situation when decision-makers need to make a single decision. This action-based consistency was supported by the recent research of Ritov and Baron (1992), suggesting that the status quo effect (i.e., the tendency to choose the same option across different times and situations) was mainly driven by the choice action.

In addition, in this study, we attempted to increase the external validity of the findings by providing more detailed information for each option and by presenting this information in a format similar to travel websites. On many websites (e.g., expedia.com), consumers can build up their bundle choices through multiple webpages; thus, in this study, two-single-choice options were given in two different sequences on the screen. Specifically, the second bundle option was not available until the first bundle option was chosen by the participants. Finally, we used a different product combination for the bundle package design (i.e., an airline ticket plus rental car).

Method

Subjects and design

Two hundred thirty-nine U.S. adults (average age = 36.2, 43.1% female) from an online panel (i.e., Amazon Mechanical Turk) were recruited for nominal compensation. Participants in this study were assigned to one of three (decision framing of the bundle: one-page two single choice vs. two-page two single choice vs. one combined choice)⁴ × two (PFC: high vs. low PFC based on individual measurement) conditions in a between-subjects experimental design.

Procedure

The procedures were quite similar to those employed in the previous study, with the exception of several modifications. Participants were asked to imagine that they were planning to take a vacation soon and to select an airline ticket and a rental car as a bundle package. The decision-framing manipulation was quite the same as in the previous studies, as shown in Figure 2. The one-page two-single-choice condition was the same as in the previous study. The participants in the two-page two-single-choice condition were asked to select their airline choice first and to go to the next page, which contained the second bundle of elements choice (i.e., rental car). After finishing their bundle choice across the different experimental conditions, they were asked to conduct another task that was not related to the main study. Afterward, participants were asked to answer 18 items from the PFC scales, based on Cialdini, Trost, and Newsom (1995) (e.g., “I make an effort to appear consistent to others”) along a 9-point scale (1 = strongly disagree, 9 = strongly agree). In order to check the participants’ involvement in the relatively long survey, we also included an item from the Instrumental Manipulation Check⁵ (IMC; Oppenheimer, Meyvis, and Davidenko 2009). Then, participants were thanked.

Figure 2.

Stimuli of Study 4.

Results

We excluded 32 participants who answered the Instrumental Manipulation Check and/or who did not answer all of the survey materials (new n = 207, average age = 36.1, 44.0% female). The reliability of the PFC scale was high (Cronbach’s alpha = .92). The page manipulation for the two-single-choice condition was not significantly different from each other for variety seeking (p = .36). Therefore, we collapsed the two experimental conditions as a one “two-single-choice” condition.

Main effect

The overall results for all participants confirmed our main prediction. Specifically, participants in the one-combined-choice condition (45.7% [48/105]) showed higher variety seeking than those in the two-single-choice condition (32.4% [33/102], χ²(1) = 3.88; p < .05). The detailed decision outcome is described in Table 3.

The role of preference for consistency

In order to investigate the moderating role of an individual’s PFC on the impact of decision framing on variety seeking, we conducted a binary logistic analysis. The results indicated that the interaction between the PFC score and decision framing was significant (Wald = 3.99, p < .05). To illustrate the interaction effect, we used a median split (median of PFC = 6.22) and categorized 106 participants as low PFC and 101 participants as high PFC. For the one-combined-choice condition, the variety-seeking tendency was quite the same across low-PFC (variety-seeking tendency = 45.6% [26/57]) and high-PFC participants (45.8% [22/48], χ²(1) = .01; p =.98). On the other hand, for the two-single-choice condition, the variety-seeking tendency was different between the low- and high-PFC participants. Specifically, the participants’ variety-seeking tendency in the travel bundle package was lower for high-PFC participants (variety-seeking tendency = 24.5% [13/53]) than for low-PFC participants (40.8% [20/49], χ²(1) = 3.09; p < .10),⁶ as shown as Figure 3.

Figure 3.

Results of Study 4.

Discussion

In this study, we replicate the previous findings in the different bundle combinations (i.e., an airline ticket and rental car) while providing more detailed information. More importantly, by measuring individual PFC, we can provide empirical evidence of the consistency mechanism, especially for the two-single-choice condition.

Discussion and Conclusions

The main proposition of this research was supported in all of the empirical studies. Decision framing revealed a significant impact on the variety-seeking tendency for travel products—travelers with combined options are more likely to engage in variety seeking than those with separate single options. This study contributes to the body of knowledge as one of the few research endeavors that has applied the theories of decision framing and variety seeking in the context of the tourism industry, where bundled products are commonplace.

It should be noted that a series of empirical studies were deliberately designed in this research to refute alternative explanations of the findings, as well as to test the main proposition. For example, although decision framing had a significant impact on the variety-seeking tendency in Study 1, participants may have been influenced by a product’s price rather than the way the products were presented (i.e., decision framing). We removed the price information in Studies 2A and 2B, and the same result was obtained for participants who were not given any price information about the products. Furthermore, Study 3 was implemented using hotel brands without providing any price/quality information to examine whether a product’s quality (cf. implicit price) could have influenced the findings of Studies 2A and 2B. Again, we found that participants with bundled products are more likely to engage in variety seeking than those with single separate products. Lastly, we found empirical evidence of our theoretical reasoning (i.e., the consistency mechanism) by detecting the effect of PFC, particularly in the two-single-choice condition.

Theoretical and Managerial Implications

This research has several theoretical and managerial implications. First, this research can contribute to our understanding regarding travel bundle consumption. In contrast to the existing literature showing the benefits of bundling (e.g., Noone and Mattila 2009) or preferences between the bundle and nonbundle options (e.g., Kim, Bojanic, and Warnick 2009), we focus on the impact of how a bundle option is constructed on decision outcomes. In addition, our study is different from previous research, which has investigated the antecedents for the preference of bundled packages in the travel industry (e.g., Tanford, Baloglu, and Erdem 2012).

Second, the results of Study 4 could exclude choice bracketing theory as an alternative explanation. Specifically, choice bracketing theory predicts a significant difference in variety seeking involving one-page single versus two-page single-choice conditions. However, our empirical findings indicate no difference in the variety-seeking tendency between the two conditions. In addition, the empirical settings of the two-single-choice and the one-combined-choice conditions are very similar in terms of the broadness of choice bracketing (i.e., narrow vs. broad choice bracketing). Therefore, the systematic difference from the two types of decision framing of the two bundle conditions can extend our understanding regarding choice bracketing theory.

Third, in this study, we investigated the variety-seeking tendency in a mostly different, but related, product category. Traditionally, the variety seeking has been investigated in the same product category. However, recent literature has suggested that the variety-seeking tendency could occur across different product categories that are related (Lattin and McAlister 1985; Kim and Drolet 2003, 374). Therefore, this article is important theoretically in providing the conditional effect of different levels of variety seeking across different, but related, product categories.

Fourth, the results of our Study 4 can provide a boundary condition for the effect of preference for consistency. PFC influenced the variety-seeking tendency only for the two-single-choice condition, and not for the one-combined-choice condition. Based on the results, we can infer that multiple actions themselves are a very important factor for preference regarding the consistency motivation. This implication is quite similar to the current understanding of PFC, given that typical PFC-based phenomena (e.g., foot-in-the-door effect; Freedman and Fraser 1966) require multiple actions.

Fifth, the higher preference for the middle-range option in the bundle condition could be explained by the compromise effect (Simonson 1989), which involves the share of a product that is enhanced when it is in the middle (versus extreme) option of the choice set. If the price-based preference for the middle option is the only mechanism, we cannot explain the result of Study 3, where the prices of all available options were the same. Furthermore, Kim and Kim (2016) have recently provided empirical evidence that the compromise effect was higher in utilitarian products, such as computers or TVs, than in hedonic products, such as travel or chocolate. The key products of this article could be categorized as hedonic in nature. Therefore, we believe that the compromise effect could not explain all of our results.

Sixth, the results of this study could provide unique theoretical contributions for the literature, extending previous studies. Rewtrakunphaiboon and Oppewal (2008) manipulated the order of travel package information, such as the travel destination or price. However, they failed to find any significance of this order effect in the preference for travel destination. In contrast, we found a significant effect of decision framing on the decision outcome in travel package decisions. As mentioned earlier, price transparency (e.g., Tanford, Erdem, and Baloglu 2011; Tanford, Baloglu, and Erdem 2012) was not the key driving force for our results. Finally, price- or budget-related factors (e.g., Harlam et al. 1995; Xu 2009) could not explain our results, either. In sum, we believe that our approach can provide a new perspective regarding travel package decisions.

In addition, it should be noted that there are differences in the two high variety-seeking options (i.e., the expensive category followed by the cheap category, vs. the cheap category followed by the expensive category). Overall, our findings show that consumers are more likely to engage in variety seeking when a cheap product category is presented first than when an expensive product category is presented first among the two high variety-seeking options. This finding is consistent with the study of Jin, He, and Song (2012), in that consumers are likely to prefer upgrading options (i.e., starting from an economic package and customizing upwards) rather than downgrading options (i.e., beginning with a luxury package and customizing downwards).

Finally, practitioners in the tourism industry may find the results of this study useful for their marketing endeavors. To maximize their yield, it can be recommended that marketers develop and properly display a diverse range of options on their websites. Our findings indicate that consumers are more likely to engage in variety-seeking when a combined form of product information is displayed. Understanding end-users’ responses to different decision-framing contexts can enable service providers to design effective marketing activities. Marketers, for example, are encouraged to present their promotional contents of bundle/package products as a combined form rather than as a form with multiple separate sections. The three common travel products are flights, hotels, and rental cars, and most major OTAs provide consumers with both bundled products and separate product options. In designing bundle/package products, it would be more effective to present a succinct promotional message in a window, including brief information about the three different products, linked to additional detailed information regarding each product.

Traditionally, separate individual choice options in practice are usually underpinned by dynamic packaging informed by a recommender system equipped with technology, while combined-choice options are usually prepackaged prior to distribution rather than being dynamically generated packages at the point of sale. However, combined-choice options can also be dynamically created by a recommender system at the point of sale (e.g., showing combined accommodations and rental car options after a flight option is selected), supported by technology. The findings of this study also imply that although attractive options are recommended for subsequent products/services, a consumer who chooses a certain category (e.g., cheap) of a product is unlikely to change to another type of subsequent product (e.g., expensive) in selecting bundle items in one sitting, because of the preference for consistency. In conjunction with the use of market basket analysis, which enables marketers to foresee consumers’ purchase behavior, our findings can be used as a reference for practitioners in designing their package products effectively.

In addition, when marketers design bundle/package products, they need to keep in mind that consumers with bundled (combined) package options are more likely to engage in variety-seeking than those with separate product options. Marketers can be advised to create diverse options in developing bundle/package products with a combination of target (newly introduced) and existing (popular) products. For example, when a hotel restaurant launches a new set menu, a hotel manager often develops a package (e.g., the new set menu + room). When displaying this package online, it would be prudent to provide multiple package options rather than limited options in order to motivate consumers who anticipate a variety of package options. Our findings also suggest that OTAs may need to keep attracting new partners across industries in order to provide diverse options for their consumers. It would be a competitive advantage for tourism companies to establish and maintain a wide range of partners (e.g., airlines, hotels, and rental car companies) that are keen to be a part of these bundle/package products. Although we aim to understand how decision framing influences the variety-seeking tendency in online traction, preference for consistency can be discussed from the perspective of customer loyalty. For example, given that a consumer who used an OTA for his or her travel would return to the OTA expecting a similar purchase experience, marketers need to note the consumer’s past experience for effective product design accordingly (cf. market basket analysis).

Limitations and Future Research

This study has several limitations that highlight the need for future research. First, in this article, variety-seeking was relatively and narrowly conceptualized and operationalized—a choice outcome from two options. Future research needs to extend the impact of different types of decision framing on various variety-seeking aspects, such as a different number of options or the degree of difference among the options. It should be noted that in this study the elements of the bundled products were quite similar or were in a complementary relationship (e.g., an airline ticket and hotel, or an airline ticket and rental car; Yan and Bandyopadhyay 2011). Future research is warranted to investigate the different combinations of bundled products, including various relationships among bundled products’ components (e.g., rental cars and restaurants).

Furthermore, we designed the experimental studies to hold other factors (e.g., price) constant so as to examine the pure impact of decision framing on the variety-seeking tendency in travel contexts. We acknowledge that the process of constructing these two packages is somewhat different in terms of price. In practice, unlike combined-package bundles that are usually created at a discounted rate, a discount rate of separate choice-package options is usually calculated once the consumer has selected all components of the package. However, we believe that our findings are valid, owing to a series of experimental studies designed to offset any alternative explanations. Lastly, this research used a scenario-based experimental method, which is strong in theory testing (internal validity) but weak in generalizability (external validity) (Kim and Jang 2014). It is, therefore, recommended for future researchers to examine our proposition using real business data, such as actual transaction records at an OTA.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Research Foundation of Korea Grant funded by the Korean Government (NRF-2014S1A5A2A03065829). This work was also supported by the Research Project Grant funded by the Faculty of Business and Law, Auckland University of Technology (R10043.18).

1.

For the survey quality of MTurk, refer to Casler, Bickel, and Hackett (2013) and Buhrmester, Kwang, and Gosling (2011).

2.

Even though this is marginally significant due to a relatively small sample size, the one-tailed result approaches .05 in the p value (i.e., p = .053).

3.

The different choice-seeking between the single vs. bundle conditions was 35.4% vs. 56.1% for −1 standard deviation [SD] of income, 40.7% vs. 57.0% for 0 SD of income, and 43.1% vs. 58.0% for +1 SD of income.

4.

We do not expect a significant difference between the one-page two-single (n = 61) and two-page two-single (n = 59) conditions. This significant result is expected because the time gap between two choices is very short, and all participants in this condition were expected to make multiple decisions. Therefore, we increase the same size for the one bundle-choice (n = 119) condition to match the number of subjects.

5.

For this IMC task, participants were given instructions, based on a modification of Oppenheimer, Meyvis, and Davidenko (2009): “For this study, and for most of the research we conduct, it is very important for us to know whether you tend to read the instructions you are given. If you are reading these instructions, please just skip the following questions and proceed to the next page. What is your favorite color?” If participants answered this question, it showed a low level of engagement regarding the survey. We believe that we could exclude the participants who did not read the basic instructions of this study by administering the IMC measurement.

6.

Even though this is marginally significant because of a relatively small sample size, the one-tailed result approaches .06 in the p value (i.e., p = .061).

Author Biographies

Jungkeun Kim, is a senior lecturer of Marketing in the Faculty of Business and Law at the Auckland University of Technology, New Zealand. His main research interests are in consumer decision-making and consumer behavior. He currently serves as an associate editor of Australasian Marketing Journal.

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. He serves as an associate director of New Zealand Tourism Research Institute (NZTRI).

Jae-Eun Kim, is a senior lecturer in the School of Communication, Journalism and Marketing at the Massey University, New Zealand. Her research interests lie in the area of consumer behavior related to retailing, design, and tourism. She serves as an associate director for Center for Advanced Retail Studies (CARS).

References

Adams

W. J.

Yellen

J. L.

1976. “Commodity Bundling and the Burden of Monopoly.” Quarterly Journal of Economics 90 (3): 475–98.

Assaker

Hallak

2013. “Moderating Effects of Tourists’ Novelty-Seeking Tendencies on Destination Image, Visitor Satisfaction, and Short- and Long-Term Revisit Intentions.” Journal of Travel Research 52 (5): 600–13.

Billings

R. S.

Scherer

L. L.

1988. “The Effects of Response Mode and Importance on Decision-Making Strategies: Judgment versus Choice.” Organizational Behavior and Human Decision Processes 41 (1): 1–19.

Buhrmester

Kwang

Gosling

S. D.

2011. “Amazon’s Mechanical Turk: A New Source of Inexpensive, Yet High-Quality, Data?” Perspectives on Psychological Science 6 (1): 3–5.

Cardoso

Lange

2007. “A Framework for Assessing Strategies and Technologies for Dynamic Packaging Applications in E-tourism.” Information Technology & Tourism 9 (1): 27–44.

Casler

Bickel

Hackett

2013. “Separate but Equal? A Comparison of Participants and Data Gathered via Amazon’s MTurk, Social Media, and Face-to-Face Behavioral Testing.” Computers in Human Behavior 29 (6): 2156–60.

Castro

C. B.

Armario

E. M.

Ruiz

D. M.

2007. “The Influence of Market Heterogeneity on the Relationship between a Destination’s Image and Tourists’ Future Behaviour.” Tourism Management 28 (1): 175–87.

Chernev

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

Chiam

Soutar

Yeo

2009. “Online and Off-Line Travel Packages Preferences: A Conjoint Analysis.” International Journal of Tourism Research 11 (1): 31–40.

10.

Chintagunta

P. K.

1998. “Inertia and Variety Seeking in a Model of Brand-Purchase Timing.” Marketing Science 17 (3): 253–70.

11.

Cialdini

R. B.

2009. Influence: Science and Practice, Vol. 4. Boston: Pearson Education.

12.

Cialdini

R. B.

Trost

M. R.

Newsom

J. T.

1995. “Preference for Consistency: The Development of a Valid Measure and The Discovery of Surprising Behavioral Implications.” Journal of Personality and Social Psychology 69 (2): 318–28.

13.

Danziger

Segev

2006. “The Effects of Informative and Non-informative Price Patterns on Consumer Price Judgments.” Psychology and Marketing 23 (6): 535–53.

14.

Dhar

Simonson

1999. “Making Complementary Choices in Consumption Episodes: Highlighting versus Balancing.” Journal of Marketing Research 36 (1): 29–44.

15.

Drolet

2002. “Inherent Rule Variability in Consumer Choice: Changing Rules for Change’s Sake.” Journal of Consumer Research 29 (3): 293–305.

16.

Einhorn

H. J.

Hogarth

R. M.

1981. “Behavioral Decision Theory: Processes of Judgment and Choice.” Journal of Accounting Research 19 (1): 1–31.

17.

Einhorn

H. J.

Kleinmuntz

D. N.

Kleinmuntz

1979. “Linear Regression and Process-Tracing Models of Judgment.” Psychological Review 86 (5): 465–85.

18.

Epstein

1979. “The Stability of Behavior: I. On Predicting Most of the People Much of the Time.” Journal of Personality and Social Psychology 37 (7): 1097–1126.

19.

Freedman

J. L.

Fraser

S. C.

1966. “Compliance without Pressure: The Foot-in-the-Door Technique.” Journal of Personality and Social Psychology 4 (2): 195–202.

20.

Fesenmaier

D. R.

Ricci

Schaumlechner

Wöber

Zanellai

2003. “DIETORECS: Travel Advisory for Multiple Decision Styles.” In Information and Communication Technologies in Tourism, 29–31. Helsinki, Finland: Springer.

21.

Fishbach

Ratner

R. K.

Zhang

2011. “Inherently Loyal or Easily Bored?: Nonconscious Activation of Consistency versus Variety-Seeking Behavior.” Journal of Consumer Psychology 21 (1): 38–48.

22.

Garcia

Torre

Linaza

M. T.

2013. “Mobile Social Travel Recommender System.” In Information and Communication Technologies in Tourism, edited by Xiang

Tussyadiah

, 3–16. Cham, Germany: Springer.

23.

Guiltinan

J. P.

1987. “The Price Bundling of Services: A Normative Framework.” Journal of Marketing 51 (2): 74–85.

24.

Harlam

B. A.

Krishna

Lehmann

D. R.

Mela

1995. “Impact of Bundle Type, Price Framing and Familiarity on Purchase Intention for the Bundle.” Journal of Business Research 33 (1): 57–66.

25.

Huber

V. L.

Neale

M. A.

Northcraft

G. B.

1987. “Decision Bias and Personnel Selection Strategies.” Organizational Behavior and Human Decision Processes 40 (1): 136–47.

26.

Inman

J. J.

2001. “The Role of Sensory-Specific Satiety in Attribute-Level Variety Seeking.” Journal of Consumer Research 28 (1): 105–20.

27.

Jang

S. S.

Feng

2007. “Temporal Destination Revisit Intention: The Effects of Novelty Seeking and Satisfaction.” Tourism Management 28 (2): 580–90.

28.

Jin

Song

2012. “Service Customization: To Upgrade or to Downgrade? An Investigation of How Option Framing Affects Tourists’ Choice of Package-Tour Services.” Tourism Management 33 (2): 266–75.

29.

Johnson

E. J.

Russo

J. E.

1984. “Product Familiarity and Learning New Information.” Journal of Consumer Research 11 (1): 542–50.

30.

Jung

H. S.

Yoon

H. H.

2012. “Why Do Satisfied Customers Switch? Focus on the Restaurant Patron Variety-Seeking Orientation and Purchase Decision Involvement.” International Journal of Hospitality Management 31 (3): 875–84.

31.

Kahn

Lehmann

D. R.

1991. “Modeling Choice Assortments.” Journal of Retailing 67 (3): 274-299.

32.

Kahn

Ratner

Kahneman

1997. “Patterns of Hedonic Consumption over Time.” Marketing Letters 8 (1): 85–96.

33.

Kandampully

Zhang

Bilgihan

2015. “Customer Loyalty: A Review and Future Directions with a Special Focus on the Hospitality Industry.” International Journal of Contemporary Hospitality Management 27 (3): 379–414.

34.

Kim

H. S.

Drolet

2003. “Choice and Self-Expression: A Cultural Analysis of Variety-Seeking.” Journal of Personality and Social Psychology 85 (2): 373–82.

35.

Kim

Bojanic

D. C.

Warnick

R. B.

2009. “Price Bundling and Travel Product Pricing Practices Used by Online Channels of Distribution.” Journal of Travel Research 47 (4): 403–12.

36.

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.

37.

Kim

J. E.

Marshall

2014. “Search for the Underlying Mechanism of Framing Effects in Multi-alternative and Multi-attribute Decision Situations.” Journal of Business Research 67 (3): 378–85.

38.

Kim

P. B.

Kim

J. E.

Magnini

V. P.

2016. “Application of Construal-Level Theory to Promotional Strategies in the Hotel Industry.” Journal of Travel Research 55 (3): 340–52.

39.

Kim

Park

2016. “An Extension and Refinement of Offer Framing Effects.” Working paper, Auckland University of Technology.

40.

Kim

S. A.

Kim

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

41.

Kohl

2016. “These 5 Tips Can Save You Money on Travel This Year.” March 10. http://blog.apastyle.org/apastyle/2010/11/how-to-cite-something-you-found-on-a-website-in-apa-style.html (accessed 10 March, 2016).

42.

Kwon

Jang

S. S.

2011. “Price Bundling Presentation and Consumer’s Bundle Choice: The Role of Quality Certainty.” International Journal of Hospitality Management 30 (2): 337–44.

43.

Laran

Wilcox

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

44.

Lattin

J. M.

McAlister

1985. “Using a Variety-Seeking Model to Identify Substitute and Complementary Relationships among Competing Products.” Journal of Marketing Research 22 (3): 330–39.

45.

Lee

Amir

Ariely

2009. “In Search of Homo Economicus: Cognitive Noise and the Role of Emotion in Preference Consistency.” Journal of Consumer Research 36 (2): 173–87.

46.

McAlister

1982. “A Dynamic Attribute Satiation Model of Variety-Seeking Behavior.” Journal of Consumer Research 9 (2): 141–50.

47.

Menon

Kahn

B. E.

1995. “The Impact of Context on Variety Seeking in Product Choices.” Journal of Consumer Research 22 (3): 285–95.

48.

Mittelman

Andrade

E. B.

Chattopadhyay

Brendl

C. M.

2014. “The Offer Framing Effect: Choosing Single versus Bundled Offerings Affects Variety Seeking.” Journal of Consumer Research 41 (4): 953–64.

49.

Morrison

A. E.

2010. Hospitality and Travel Marketing, 4th ed. New York: Delmar.

50.

Nam

Ekinci

Whyatt

2011. “Brand Equity, Brand Loyalty and Consumer Satisfaction.” Annals of Tourism Research 38 (3): 1009–30.

51.

Niininen

Szivas

Riley

2004. “Destination Loyalty and Repeat Behaviour: An Application of Optimum Stimulation Measurement.” International Journal of Tourism Research 6 (6): 439–47.

52.

Noone

B. M.

Mattila

A. S.

2009. “Hotel Revenue Management and the Internet: The Effect of Price Presentation Strategies on Customers’ Willingness to Book.” International Journal of Hospitality Management 28 (2): 272–79.

53.

Oppenheimer

D. M.

Meyvis

Davidenko

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

54.

Pan

Lee

Tsai

2014. “Travel Photos: Motivations, Image Dimensions, and Affective Qualities of Places.” Tourism Management 40:59–69.

55.

Preacher

K. J.

Hayes

A. F.

2008. “Asymptotic and Resampling Strategies for Assessing and Comparing Indirect Effects in Multiple Mediator Models.” Behavior Research Methods 40 (3): 879–91.

56.

Ratner

R. K.

Kahn

B. E.

2002. “The Impact of Private versus Public Consumption on Variety-Seeking Behavior.” Journal of Consumer Research 29 (2): 246–57.

57.

Ratner

R. K.

Kahn

B. E.

Kahneman

1999. “Choosing Less-Preferred Experiences for the Sake of Variety.” Journal of Consumer Research 26 (1): 1–15.

58.

Read

Loewenstein

1995. “Diversification Bias: Explaining the Discrepancy in Variety Seeking between Combined and Separated Choices.” Journal of Experimental Psychology: Applied 1 (1): 34–49.

59.

Read

Antonides

Van den Ouden

Trienekens

2001. “Which Is Better: Simultaneous or Sequential Choice?” Organizational Behavior and Human Decision Processes 84 (1): 54–70.

60.

Read

Loewenstein

Rabin

1999. “Choice Bracketing.” Journal of Risk and Uncertainty 19 (1-3): 171–97.

61.

Repetti

Roe

Gregory

2015. “Pricing Strategies for Resort Fees: Consumer Preferences Favor Simplicity.” International Journal of Contemporary Hospitality Management 27 (5): 790–809.

62.

Rewtrakunphaiboon

Oppewal

2008. “Effects of Package Holiday Information Presentation on Destination Choice.” Journal of Travel Research 47 (2): 127–36.

63.

Ricci

Werthner

2002. “Case Based Reasoning for Travel Planning Recommendation.” Information Technology and Tourism 4 (3/4): 215–26.

64.

Ritov

Baron

1992. “Status-quo and Omission Biases.” Journal of Risk and Uncertainty 5 (1): 49–61.

65.

Shafir

Simonson

Tversky

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

66.

Simonson

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

67.

Simonson

1990. “The Effect of Purchase Quantity and Timing on Variety-Seeking Behavior.” Journal of Marketing Research 27 (2): 150–62.

68.

Simonson

Winer

R. S.

1992. “The Influence of Purchase Quantity and Display Format on Consumer Preference for Variety.” Journal of Consumer Research 19 (1): 133–38.

69.

Solnet

Boztug

Dolnicar

2016. “An Untapped Gold Mine? Exploring the Potential of Market Basket Analysis to Grow Hotel Revenue.” International Journal of Hospitality Management 56:119–25.

70.

Soman

Gourville

J. T.

2001. “Transaction Decoupling: How Price Bundling Affects the Decision to Consume.” Journal of Marketing Research 38 (1): 30–44.

71.

Stremersch

Tellis

G. J.

2002. “Strategic Bundling of Products and Prices: A New Synthesis for Marketing.” Journal of Marketing 66 (1): 55–72.

72.

Tanford

Baloglu

Erdem

2012. “Travel Packaging on the Internet: The Impact of Pricing Information and Perceived Value on Consumer Choice.” Journal of Travel Research 51 (1): 68–80.

73.

Tanford

Erdem

Baloglu

2011. “Price Transparency of Bundled Vacation Packages.” Journal of Hospitality & Tourism Research 35 (2): 213–34.

74.

U.S. Department of Commerce. 2015. “U.S. Department of Commerce 2014 U.S. Travel and Tourism Statistics (U.S. Resident Outbound).” International Trade Administration (ITA) report. http://travel.trade.gov/outreachpages (accessed April 12, 2016).

75.

World Best Enterprises. 2004. “World Best Enterprises Quality Criteria for Website Excellence.” http://www.worldbest.com/criteria (accessed October 9, 2016).

76.

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.

77.

2009. “Examining the Effects of Bundling Strategies on Travelers’ Value Perception and Purchase Intention of a Vacation Package.” PhD Diss., Virginia Tech University, Blacksburg.

78.

Yan

Bandyopadhyay

2011. “The Profit Benefits of Bundle Pricing of Complementary Products.” Journal of Retailing and Consumer Services 18 (4): 355–61.