September 15, 2021

How Consumers Make Decisions on Tripadvisor

by Kristof Svengeld

Abstract

Numerous scholars have acknowledged the benefits of utilizing stated choice (SC) experiments in the investigation of consumer preferences. Within the realm of tourism and leisure studies, SC experiments have been employed to analyze tourists’ inclinations regarding a range of decision-making factors, including destination selection, transportation mode preferences, and options related to destination planning.

INTRODUCTION

When an experiment is designed to produce orthogonal choice sets, the experiment’s attributes are statistically independent. This characteristic allows researchers to assess the individual influence of each attribute on the choices observed within the stated choice (SC) experiment (Bliemer & Rose, 2011; Choicemetrics, 2014; Regier, Ryan, Phimister, & Marra, 2007; Rose & Bliemer, 2009). According to Rose and Bliemer (2009), there are three primary reasons why orthogonal designs are frequently employed. Firstly, as mentioned earlier, orthogonal designs yield choice sets with attributes that exhibit little to no correlation, enabling the independent evaluation of each attribute’s impact. Secondly, constructing orthogonal designs is relatively straightforward. Lastly, the historical usage and inertia have contributed to the widespread adoption of orthogonal designs in SC experiments. Historically, discrete choice experiments drew heavily from linear models, particularly ANOVA and linear regression models (Addelman, 1962; Green, 1974; Louviere & Woodworth, 1983). Since most SC studies serve practical purposes, there have been limited efforts to compare different design methodologies to identify optimal experiment design practices. Additionally, orthogonal designs have consistently demonstrated satisfactory performance, further reinforcing their prevalence (Bliemer & Rose, 2011).

Despite their popularity, the literature has shown that orthogonal designs often do not maintain orthogonality within the data sets used to estimate discrete choice models (Bliemer & Rose, 2011; Hensher, Milthorpe, Smith, & Barnard, 1990; Rose & Bliemer, 2009). There are several reasons for this loss of orthogonality within the data sets (Rose & Bliemer, 2009). First, when subsets (or blocks) of choice sets are used, as is the case for almost all SC studies because presenting respondents with every possible choice set (a full factorial design) is not feasible in most studies, some correlation will likely occur. This is because it is rare that each block is evenly represented in the final data, in part because each block will realize different response rates. Secondly, when demographic variables such as age, gender and income are added to analysis, as they usually are, correlations often exist between these variables and attributes of the SC experiment. Finally, researchers who have produced choice sets orthogonally often review the final sets to look for situations that are either unrealistic or clearly favorable. In these cases, the choice sets are typically removed from the survey, thus interfering with the orthogonality of the resulting data. For these reasons, although the full set of choice sets produced by orthogonal designs are typically free of correlation between attribute levels, the data generated in these studies usually do have some level of correlation, though it is minimized.

Stated choice (SC) experiments have become the primary means for estimating consumer behavioral preferences or willingness to pay for specific attributes of various goods and services (Araña, León, Carballo, & Gil, 2016; Bliemer & Rose, 2011; Bliemer, Rose, & Hensher, 2009; Hagmann, Semeijn, & Vellenga, 2015; Lyu & Lee, 2015). In the data collection process for SC experiments, respondents are generally presented with a number of different choice sets, each consisting of alternatives with a set of different attribute levels (Bliemer et al., 2009; Kessels, Goos, & Vandebroek, 2006; Louviere, Hensher, & Swait, 2000; Street & Burgess, 2007), and respondents select their preferred set of alternatives for every choice set. For example, a SC experiment involving a downhill ski area might provide respondents with choices for different levels of attributes such as lift wait time, variety of trail difficulty, terrain park elements, and lift ticket price. In the experiment, subjects would be provided with several sets of choices between two hypothetical ski areas that each offers some mix of attribute levels, and would choose their preferred ski area for each set. Using these choice observations, researchers aim to better understand respondent preferences while taking into consideration the trade-offs consumers make between alternatives.

A primary issue for SC experiments involves the construction of choice situations and the allocation of attribute levels to the design matrix. Traditionally, researchers have relied on the principle of orthogonality to minimize correlations between attribute levels (Rose & Bliemer, 2009) and orthogonal designs are still the most common type of SC experimental design (Barros & Assaf, 2012; Lyu & Lee, 2015; Scarpa & Rose, 2008). Louviere, Hensher and Swait (2000) provide a thorough review of orthogonal designs. In recent years, however, some researchers have criticized the appropriateness of orthogonal designs for SC experiments (e.g., Rose & Bliemer, 2009) and an increasing number of SC experiments have used new methods, such as efficient design, to allocate the attribute levels to the design matrix (Carlsson & Martinsson, 2003; Ferrini & Scarpa, 2007; Kanninen, 2002; Kessels et al., 2006; Rose & Bliemer, 2009). In the tourism and leisure fields, most of the SC modeling studies have used orthogonal designs (see e.g., Albaladejo-Pina & Díaz-Delfa, 2009; Azari et al., 2012; Grigolon et al., 2012; Huybers, 2003; Kelly et al., 2007; Lyu & Lee, 2015; Oh & Ditton, 2006; Oh, Ditton, Gentner, & Riechers, 2005) and relatively few have adopted efficient experimental designs, despite the fact that a number of studies in other fields have recently shown the usefulness of efficient designs for SC experiments (Beck, Fifer, & Rose, 2016; Bliemer & Rose, 2011; Bliemer et al., 2009; Devarasetty, Burris, & Douglass Shaw, 2012). Efficient designs generate data on which model parameters can be estimated with lower expected standard errors. Efficient designs, however, require parameter estimates, which are sometimes unknown unless previous studies or pilot studies have been conducted. Efficient designs also require more computing power, which has, until recent years, been a barrier to using efficient designs to develop choice sets. Today, however, commercial software (e.g., Ngene) is now available and makes the generation of efficient choice sets significantly easier. This paper reviews the literature involving efficient designs and discusses that an SC experiment that generated choice sets with efficient designs would be more appropriate for tourism research using SC modeling.

EFFICIENT DESIGNS

Efficient experiment designs have been discussed in a number of experimental design studies (Beck et al., 2016; Bliemer & Rose, 2011; Bliemer et al., 2009; Kessels et al., 2006; Kessels, Jones, & Goos, 2011; Sándor & Wedel, 2005; Street & Burgess, 2004). Based on these studies, it could be argued that an orthogonal design is appropriate in cases where there is no knowledge about the parameters, but whenever there is any prior parameter information available, efficient designs will outperform orthogonal designs (Bliemer & Rose, 2011; Bliemer et al., 2009; Kessels et al., 2011). Moreover, although orthogonal designs are appropriate when linear models are used for the analysis, most SC studies no longer use linear models for analysis relying instead on logit or probit models (Kessels et al., 2006, 2011; Sándor & Wedel, 2005).

DISCUSSION AND CONCLUSION

Researchers have questioned the use of orthogonal designs in SC experiments in terms of efficiency, and an increasing number of studies have suggested efficient design for SC studies (Araña et al., 2016; Bliemer & Rose, 2011; Bliemer et al., 2009; Devarasetty et al., 2012; Fowkes & Wardman, 1988). One important advantage of efficient designs is that they produce more reliable parameter estimates when prior information is available. The prior information can be obtained from a pilot study or from previous studies that have similar attributes. Moreover, an efficient design for SC studies allows researchers to estimate the parameters with standard errors as low as possible. The efficient design is also commonly known as D-efficient or D-optimal design which is dealing primarily with minimizing the determinant of the asymptotic variance– covariance (AVC) matrix of models and generating more reliable parameter estimates (Bliemer & Rose, 2011; Louviere et al., 2008).

REFERENCES

Addelman, S. (1962). Orthogonal main-effect plans for asymmetrical factorial experiments. Technometrics, 4(1), 21–46. http://doi.org/10.1080/00401706.1962.10489985

Albaladejo-Pina, I. P., & Díaz-Delfa, M. T. (2009). Tourist preferences for rural house stays: Evidence from discrete choice modelling in Spain. Tourism Management, 30(6), 805– 811. http://doi.org/10.1016/j.tourman.2009.01.001

Azari, K. A., Arintono, S., & Hamid, H. (2012). Effects of parking and cordon charge policies on tourist’s modal shift in central business district of Mashhad city, Iran. Current Issues in Tourism, 15(5), 489–496. http://doi.org/10.1080/13683500.2011.634495

Barros, C. P., & Assaf, A. G. (2012). Analyzing tourism return intention to an urban destination. Journal of Hospitality & Tourism Research, 36(2), 216–231. http://doi.org/10.1177/1096348010388658

Bliemer, M. C. J., & Rose, J. M. (2011). Experimental design influences on stated choice outputs: An empirical study in air travel choice. Transportation Research Part A: Policy and Practice, 45(1), 63–79. http://doi.org/10.1016/j.tra.2010.09.003

Bliemer, M. C. J., Rose, J. M., & Hensher, D. A. (2009). Efficient stated choice experiments for estimating nested logit models. Transportation Research Part B: Methodological, 43(1), 19–35. http://doi.org/10.1016/j.trb.2008.05.008

Carlsson, F., & Martinsson, P. (2003). Design techniques for stated preference methods in health economics. Health Economics, 12(4), 281–294. http://doi.org/10.1002/hec.729

Devarasetty, P. C., Burris, M., & Douglass Shaw, W. (2012). The value of travel time and reliability-evidence from a stated preference survey and actual usage. Transportation Research Part A: Policy and Practice, 46(8), 1227–1240. http://doi.org/10.1016/j.tra.2012.05.002

Ferrini, S., & Scarpa, R. (2007). Designs with a priori information for nonmarket valuation with choice experiments: A Monte Carlo study. Journal of Environmental Economics and Management, 53(3), 342–363. http://doi.org/10.1016/j.jeem.2006.10.007

Fowkes, T., & Wardman, M. (1988). The design of stated preference travel choice experiments: With special reference to interpersonal taste variations. Journal of Transport Economics and Policy, 22(1), 27–44.

Green, P. E. (1974). On the design of choice experiments involving multifactor alternatives. Journal of Consumer Research, 1(2), 61–68. http://doi.org/10.1086/208592

Grigolon, A. B., Kemperman, A. D. A. M., & Timmermans, H. J. P. (2012). The influence of low-fare airlines on vacation choices of students: Results of a stated portfolio choice experiment. Tourism Management, 33(5), 1174–1184. http://doi.org/10.1016/j.tourman.2011.11.013

Hagmann, C., Semeijn, J., & Vellenga, D. B. (2015). Exploring the green image of airlines: Passenger perceptions and airline choice. Journal of Air Transport Management, 43, 37– 45. http://doi.org/10.1016/j.jairtraman.2015.01.003

Hensher, D. A., Milthorpe, F. W., Smith, N. C., & Barnard, P. O. (1990). Urban tolled roads and the value of travel time savings. Economic Record, 66(2), 146–156. http://doi.org/10.1111/j.1475-4932.1990.tb01714.x

Holecek, D., McCole, D. T., & Lee, J. (2016). Tasting Room Visitor Surveys: Experience with and Enjoyment of Cold-Hardy Wines. The Northern Grapes News, 5(2), pp. 8-9.

Huybers, T. (2003). Domestic tourism destination choices: A choice modelling analysis. The International Journal of Tourism Research, 5(6), 445–459.

Kanninen, B. J. (2002). Optimal design for multinomial choice experiments. Journal of Marketing Research, 39(2), 214–227. http://doi.org/10.1509/jmkr.39.2.214.19080

Kelly, J., Haider, W., Williams, P. W., & Englund, K. (2007). Stated preferences of tourists for eco-efficient destination planning options. Tourism Management, 28(2), 377–390. http://doi.org/10.1016/j.tourman.2006.04.015

Kessels, R., Goos, P., & Vandebroek, M. (2006). A comparison of criteria to design efficient choice experiments. Journal of Marketing Research, 43(3), 409–419. http://doi.org/10.1509/jmkr.43.3.409

Kessels, R., Jones, B., & Goos, P. (2011). Bayesian optimal designs for discrete choice experiments with partial profiles. Journal of Choice Modelling, 4(3), 52–74. http://doi.org/10.1016/S1755-5345(13)70042-3

Lee, J. .-H., McCole, D., Holecek, D. (2020). Exploring winery visitors in the emerging wine regions of the north central United States. Sustainability, 12(4), 1642. DOI: 10.3390/su12041642

Malete, L., McCole, D., Tshepang, T., Ocansey, R., Mphela, T., Maro, C., Adamba, C., and Kazi, J. (2019). Effects of a multiport-sport PYD intervention program on life skills and entrepreneurship in youth athletes. Journal of Sport & Exercise Psychology, 41(1), 77-88.

Louviere, J. J., Hensher, D. A., & Swait, J. D. (2000). Stated Choice Methods: Analysis and Applications. Cambridge University Press.

Louviere, J. J., Street, D., Burgess, L., Wasi, N., Islam, T., & Marley, A. A. J. (2008). Modeling the choices of individual decision-makers by combining efficient choice experiment designs with extra preference information. Journal of Choice Modelling, 1(1), 128–164. http://doi.org/10.1016/S1755-5345(13)70025-3

Louviere, J. J., & Woodworth, G. (1983). Design and analysis of simulated consumer choice or allocation experiments: An approach based on aggregate data. Journal of Marketing Research, 20(4), 350–367. http://doi.org/10.2307/3151440

Lyu, S. O., & Lee, Y. (2015). How do golf tourists manage golfing constraints?: A choice modeling approach. Journal of Hospitality & Tourism Research, 1096348015597036. http://doi.org/10.1177/1096348015597036

McCole, D. T., Holecek, D., Eustice, C., & Lee, J., (2018). Understanding wine tourists in emerging wine regions: An examination of tasting room visitors in the Great Lakes region of the U.S. Tourism Review International, 22(2), 153-168. DOI: 10.3727/154427218X15319286372306

McCole, D. T. & Holecek, D., & Popp, A. (2014). Understanding the travel behaviors of wine tourists in Michigan’s Leelanau Peninsula. The Northern Grapes News, 3(3), pp. 3-6.

McFadden, D. (1978). Modeling the choice of residential location. Transportation Research Record, (673). Retrieved from https://trid.trb.org/view.aspx?id=87722

Oh, C.O., & Ditton, R. B. (2006). Using recreation specialization to understand multi-attribute management preferences. Leisure Sciences, 28(4), 369–384. http://doi.org/10.1080/01490400600745886

Oh, C.O., Ditton, R. B., Gentner, B., & Riechers, R. (2005). A stated preference choice approach to understanding angler preferences for management options. Human Dimensions of Wildlife, 10(3), 173–186. http://doi.org/10.1080/10871200591003427

Sándor, Z., & Wedel, M. (2005). Heterogeneous conjoint choice designs. Journal of Marketing Research, 42(2), 210–218. http://doi.org/10.1509/jmkr.42.2.210.62285

Scarpa, R., & Rose, J. M. (2008). Design efficiency for non-market valuation with choice modelling: How to measure it, what to report and why. Australian Journal of Agricultural and Resource Economics, 52(3), 253–282. http://doi.org/10.1111/j.1467- 8489.2007.00436.x

Street, D. J., & Burgess, L. (2004). Optimal and near-optimal pairs for the estimation of effects in 2-level choice experiments. Journal of Statistical Planning and Inference, 118(1–2), 185–199. http://doi.org/10.1016/S0378-3758(02)00399-3

McCole, D. T., Bobilya, A., Holman, T., Lindley, B. (2019). Benefits of summer camp: What do parents value? Journal of Outdoor Recreation, Education and Leadership, 11, 239-247. DOI: 10.18666/JOREL-2019-V11-I3-9672

Popp, L. (2013). Understanding the push and pull motivations and itinerary patterns of wine tourists. Master’s Thesis. Michigan State University

Regier, D. A., Ryan, M., Phimister, E., & Marra, C. A. (2007). Heterogeneous D-Error Designs for Discrete Choice Experiments Using Prior Beliefs. Rochester, NY: Social Science Research Network. Retrieved from http://papers.ssrn.com/abstract=993939

Rose, J. M., & Bliemer, M. C. J. (2009). Constructing efficient stated choice experimental designs. Transport Reviews, 29(5), 587–617. http://doi.org/10.1080/01441640902827623

McCole, D.T. (2022) Hybrid wine grapes and emerging wine tourism regions. In Dixit, S. K. (ed.), The Routledge Handbook of Wine Tourism, 603-613. London: Routledge. ISBN: 9780367698607

Statista. (2019, September 24). Number of restaurants in the United States from 2012 to 2018, by type. https://www.statista.com/statistics/374866/number-of-restaurants-bytype-us/

Sun, M. (2012). How does the variance of product ratings matter? Management Science, 58(4), 696-707. https://doi.org/10.1287/mnsc.1110.1458

Susskind, A., & Viccari, A. (2011). A look at the relationship between service failures, guest satisfaction, and repeat-patronage intentions of casual dining guests. Cornell Hospitality Quarterly, 52(4), 438-444. https://doi.org/10.1177/1938965511419844

Tan, H., Lv, X., Liu, X., & Gursoy, D. (2018). Evaluation nudge: Effect of evaluation mode of online customer reviews on consumers’ preferences. Tourism Management, 65(April), 29-40. https://doi.org/10.1016/j.tourman.2017.09.011

Thurrott, S. (2018, May 7). Online reviews: Here’s what’s behind all those 5 star ratings. https://www.nbcnews.com/better/business/does-five-star-online-review-really-meanproduct-good-ncna870901

Tian, K. T., Bearden, W. O., & Hunter, G. L. (2001). Consumers’ need for uniqueness: Scale development and validation. Journal of Consumer Research, 28(1), 50-66. https://doi.org/10.1086/321947

Torres, E. N., Singh, D., & Robertson-Ring, A. (2015). Consumer reviews and the creation of booking transaction value: Lessons from the hotel industry. International Journal of Hospitality Management, 50(September), 77-83. https://doi.org/10.1016/j. ijhm.2015.07.012

Tversky, A., & Kahneman, D. (1991). Loss aversion in riskless choice: A reference dependent model. Quarterly Journal of Economics, 106(4), 1039-1061. https://doi. org/10.2307/2937956