Conjoint Data Research Articles

How European consumers value wine credence attributes: a cross-country comparison of France, Greece and Italy

Several attributes can be used to differentiate wine products to meet consumer interest and thus increase producer visibility, attractiveness and revenues. The perception towards the same attribute may depend on various factors that characterise the subjects, such as individual, behavioural and situational characteristics, like their country of origin. This study aims to identify which credence attributes and related levels motivate consumers the most to buying wine, by comparing the results obtained in three different European countries: France, Greece and Italy. A conjoint experiment based on linear assumption was administered using price, production method, geographical indication and wine origin as product attributes. The conjoint data were analysed in three steps: performing a model with the whole sample; performing three models using national data to compare results between the countries; performing a cluster analysis using the Ward method to associate consumer characteristics with product attributes. Results show that origin is the most valued attribute for choosing wine, followed by the production method. Cross-country evaluation reveals several significant differences among the attributes of the production method, geographical indication and origin. The cluster analysis identified three groups named: High price and national wine seekers; Certification seekers and Price-sensitive seekers. This paper provides several implications for both academicians and enterprises. Indeed, it is the first evaluation comparing the role given by consumers to biodynamic certification in a cross-country evaluation. Several indications are also provided for producers who can help differentiate better wine production by earning a higher income.

Price Optimization Combining Conjoint Data and Purchase History: A Causal Modeling Approach

Abstract: Pricing decisions of companies require an understanding of the causal effect of a price change on the demand. When real-life pricing experiments are infeasible, data-driven decision-making must be based on alternative data sources such as purchase history (sales data) and conjoint studies where a group of customers is asked to make imaginary purchases in an artificial setup. We present an approach for price optimization that combines population statistics, purchase history, and conjoint data in a systematic way. We build on the recent advances in causal inference to identify and quantify the effect of price on the purchase probability at the customer level. The identification task is a transportability problem whose solution requires a parametric assumption on the differences between the conjoint study and real purchases. The causal effect is estimated using Bayesian methods that take into account the uncertainty of the data sources. The pricing decision is made by comparing the estimated posterior distributions of gross profit for different prices. The approach is demonstrated with simulated data resembling the features of real-world data.

Test-Retest Reliability in Metric Conjoint Experiments: A New Workflow to Evaluate Confidence in Model Results

Metric conjoint studies are a popular research design in the entrepreneurship domain. For these studies, test-retest reliabilities of ρ > .70 or higher are an often-cited reliability criterion. Despite their widespread use, however, there is little rigorous analysis of whether test-retest reliability in metric conjoint studies relates to model efficacy. Informed by a systematic literature review, we conducted two Monte Carlo simulations to evaluate the effects of various determinants of test-retest reliability in conjoint experiments. We then illustrate a workflow for entrepreneurship researchers employing conjoint designs to better evaluate—and communicate—confidence in statistical models estimated from conjoint data.

Consumer preferences for circular outdoor sporting goods: An Adaptive Choice-Based Conjoint analysis among residents of European outdoor markets

Since 2015, the concept of a Circular Economy (CE) has been gaining attention within the field of corporate sustainability including the textile sector. Further, CE related literature often mentions single actors from the Outdoor Sporting Goods Industry (OSGI) as forerunners. However, an analysis of users’ acceptance of circular outdoor products is so far missing in that literature. Thus, this study aimed to analyse if outdoor sporting goods users value circular product attributes using an Adaptive Choice-Based Conjoint analysis. The conjoint data from 1012 respondents, resident in important European outdoor markets, show that functional product attributes as well as social sustainability are more important for the purchase choice than circular product attributes while product durability and the share of recycled material used in the product are the most relevant circular attributes. Further, environmental awareness (strong influence), gender and age (weak influence) moderate the preference for functionality rather than circularity: environmentally aware, young, or female users are comparatively receptive to circular product features. As the study relates these findings to previous research and discusses practical implications, it adds an OSGI-specific perspective to the literature on consumer acceptance of circular products and thereby gives action recommendations to industry managers.

Changes in Traveler Stated Preference for Bus and Car Modes Due to Real-Time Schedule Information: A Conjoint Analysis

This paper reports a conjoint analysis that explored potential impacts of real-time transit schedule information on mode preference. Conjoint analysis is a stated-preference approach to choice modeling in which respondents are asked to rate hypothetical products or services described by a single level of each of a number of attributes. Respondent ratings are decomposed into “part-worths “ describing preferences for each attribute level. Subjects for the study were 500 randomly-sampled employees on the University of Michigan Medical Campus.The conjoint data indicate potential significance of real-time transit schedule information for circumstances under which modal choice decisions are made on a day-today basis. Stated mode preference is not, however, significantly affected by availability of such information when decisions are made on a month-by-month basis. These results should further motivate transit system designers to provide such information, with particular attention paid to developing a highly accessible method of information dissemination.

S290 Is Fecal Immunochemical Test (FIT) the Answer to Increasing Colorectal Cancer Screening (CRC) Uptake in the Filipino Community? Results From a Conjoint Analysis Survey

Introduction: Asian Americans are the fastest growing major racial/ethnic group in the US, with Filipinos comprising the 3rd largest group. While >80% of Filipinos are proficient in English, and have insurance rates, education levels, and incomes that exceed the general US population, they have lower CRC screening rates and worse CRC outcomes vs. non-Hispanic Whites. To begin to address this disparity, we used conjoint analysis to understand Filipinos’ preferences for the different CRC screening test options. Methods: To quantify Filipinos’ preferences for CRC screening tests, we conducted a choice-based conjoint analysis survey for individuals ≥40yo at average risk for CRC who had not undergone prior screening. From 4/29-11/7/21, we recruited Filipinos at an academic medical center and through a national survey research firm (Cint). Using the conjoint data, we performed simulations to determine each individual’s preferred screening test; for this analysis, we focused on the proportion of people who preferred annual FIT or colonoscopy every 10 years as both are tier 1 tests according to the US Multi-Society Task Force (MSTF) on CRC. We then performed logistic regression to explore whether demographics predicted decision making on FIT vs colonoscopy; variables with p< .20 from bivariate analyses were included as covariates in the regression model. Results: Overall, 105 participants completed the survey; most respondents were female (74.3%) and aged 40-49y (84.8%). Moreover, 64.8% of participants stated they planned to get screened for CRC and they reported high self-perceived benefits of CRC screening (median 4.4, IQR 3.8-4.8; 1-5 scale, higher=more beneficial). When performing simulations using the conjoint data for the US MSTF tier 1 tests, we found that 66.7% of respondents preferred an annual FIT while 33.3% preferred a colonoscopy every 10 years (Figure). In a regression analysis that accounted for sex, marital status, household income, employment status, and geographic region, no variables were significantly associated with individual's preference for FIT over colonoscopy (Table). Conclusion: We found that 2 in 3 Filipinos prefer annual FIT over colonoscopy for their CRC screening and that demographics poorly predict individual decision making. To improve CRC screening uptake in the Filipino community, our data suggest that community-based interventions should either focus primarily on FIT or employ a choice-based approach (ie, FIT or colonoscopy).Figure 1.: Data from simulations using conjoint analysis data assessing the proportion of respondents who would prefer each MSTF tier 1-recommended test (N=105) Table 1. - Regression analysis on preferring FIT every year over colonoscopy every 10 years for CRC screening; screening test preferences were determined through simulations from conjoint analysis-derived data (N=105) Variable Prefers FIT every year for CRC screening n (% of row) aOR [95% CI] Sex: Male 21 (77.8%) Reference Female 49 (62.8%) 0.61 (0.20, 1.86) Marital status: Married or living with a partner 53 (62.4%) Reference Not married 17 (85.0%) 2.65 (0.61, 11.44) Total household income: ≤$100,000 34 (79.1%) reference >$100,000 31 (62.0%) 0.72 (0.24, 2.20) Prefer not to say 5 (41.7%) 0.23 (0.05, 1.12) Employment status: Unemployed, on disability, on leave of absence from work, retired, or a homemaker 19 (82.6%) reference Employed or student 51 (62.2%) 0.36 (0.10, 1.31) US region: Northeast/South/ Midwest 15 (83.3%) reference West 55 (63.2%) 0.70 (0.15, 3.21) Has non-first degree relative or friend diagnosed with CRC 11 (44.0%) 0.42 (0.15, 1.16)

Using Conjoint Experiments to Analyze Election Outcomes: The Essential Role of the Average Marginal Component Effect

AbstractPolitical scientists have increasingly deployed conjoint survey experiments to understand multidimensional choices in various settings. In this paper, we show that the average marginal component effect (AMCE) constitutes an aggregation of individual-level preferences that is meaningful both theoretically and empirically. First, extending previous results to allow for arbitrary randomization distributions, we show how the AMCE represents a summary of voters’ multidimensional preferences that combines directionality and intensity according to a probabilistic generalization of the Borda rule. We demonstrate why incorporating both the directionality and intensity of multi-attribute preferences is essential for analyzing real-world elections, in which ceteris paribus comparisons almost never occur. Second, and in further empirical support of this point, we show how this aggregation translates directly into a primary quantity of interest to election scholars: the effect of a change in an attribute on a candidate’s or party’s expected vote share. These properties hold irrespective of the heterogeneity, strength, or interactivity of voters’ preferences and regardless of how votes are aggregated into seats. Finally, we propose, formalize, and evaluate the feasibility of using conjoint data to estimate alternative quantities of interest to electoral studies, including the effect of an attribute on the probability of winning.

Robust Product Line Design by Protecting the Downside While Minding the Upside

Many decisions must be made before relevant uncertainties are revealed; hence robustness approaches to decision making are often considered. There are two major approaches to robustness in the literature: (i) maximizing the value of the worst outcome and (ii) minimizing the maximum ex post regret of the decision. While the first approach seeks the best worst‐case outcome, it may result in substantial opportunity losses relative to the ex post optimal decision; this is particularly critical in new product design decisions as it leaves the leading firm exposed to a late entrant who designs its products after uncertainties are resolved. The second approach seeks to minimize these ex post opportunity losses, but it may expose the decision maker to greater downside risk, that is, lower outcome in the worst case. Most studies have examined one of these approaches in isolation; in contrast, in this study, we explore the trade‐offs between protecting upside regret and downside risk with a focus on a product line design (PLD) problem, which relies greatly on conjoint analysis data. The PLD problem may suffer from significant uncertainty in the partworth utilities which define the attribute‐level consumer preferences on a product for different population segments. Our numerical experiments, based on a real conjoint data set, show that neither robustness approaches dominates the other and hence our proposed approach uses both. We construct a Min‐Value vs. Max‐Relative‐Regret efficient frontier to narrow down the PLD choice and propose a metric to quantify the trade‐offs and finalize the selection.

Public Perspectives on Decisions About Emergency Care Seeking for Care Unrelated to COVID-19 During the COVID-19 Pandemic

Delaying critical care for treatable conditions owing to fear of contracting COVID-19 in the emergency department (ED) is associated with avoidable morbidity and mortality. To assess and quantify how people decided whether to present to the ED during the COVID-19 pandemic for care unrelated to COVID-19 using conjoint analysis, a form of trade-off analysis that examines how individuals make complex decisions. This cross-sectional survey study was conducted using a nationwide sample from June 1, 2020, during the initial peak of the COVID-19 pandemic. Included participants were adults aged 18 years or older in the US who self-reported that they had not tested positive for COVID-19. Data were analyzed from July 2020 through May 2021. Participants completed a self-administered online survey. Using a choice-based conjoint analysis survey, the relative importance was assessed for the following attributes for individuals in deciding whether to seek ED care for symptoms consistent with myocardial infarction or appendicitis: reduction in chance of dying because of ED treatment, crowdedness of ED with other patients, and chance of contracting COVID-19 in the ED. We also performed latent class analyses using conjoint data to identify distinct segments of the respondent population with similar choice patterns. Logistic regression was then used to explore whether patient sociodemographics and political affiliations were factors associated with decision-making. Among 1981 individuals invited to participate, 933 respondents (47.1%) completed the survey; participants' mean (SD) age was 40.1 (13.0) years, and 491 (52.6%) were women. In latent class analyses, 158 individuals (16.9%) with symptoms of myocardial infarction and 238 individuals (25.5%) with symptoms of appendicitis prioritized avoidance of COVID-19 exposure in the ED (ie, chance of contracting COVID-19 in the ED or crowdedness of ED with other patients) over seeking appropriate care for symptoms. Having a usual source of care was a factor associated with lower odds of prioritizing avoidance of COVID-19 exposure (myocardial infarction scenario: adjusted odds ratio, 0.49 [95% CI, 0.32-0.76]; P = .001; appendicitis scenario: adjusted odds ratio, 0.57 [95% CI, 0.40-0.82]; P = .003), but most sociodemographic factors and political affiliations were not factors associated with decision-making. This study found that up to one-quarter of individuals were willing to forgo potentially life-saving ED care to avoid exposure to COVID-19. These findings suggest that health care systems and public health organizations should develop effective communications for patients and the community at large that reassure and encourage timely health care for critical needs during the ongoing COVID-19 pandemic and other scenarios.

What inhibits consumers in emerging countries from engaging in status consumption? A latent class conjoint analysis approach

PurposeThis paper aims to address how the status consumption tendency of consumers in emerging markets is negatively influenced by five individual traits: self-control, self-actualization, religiosity, future orientation and self-efficacy.Design/methodology/approachA conjoint experiment measured the importance of certain smartphone product attributes. A latent class regression analysis was then employed to estimate segment-level part-worths using conjoint data collected from 500 Bangladeshi consumers.FindingsThe results revealed three segments with members that differ in how they evaluate smartphone product attributes. Those susceptible to a product's brand name (i.e. status seekers) appear to have low self-control, are less religious and are more myopic.Research limitations/implicationsAn issue may exist with generalizability, as the analysis was conducted based on data collected in one country and for one product category. However, this study's framework provides direction for future researchers to better understand status consumption in emerging countries.Practical implicationsThe findings are useful for marketers selling status products to improve market segmentation and target their offerings more efficiently.Originality/valueThe originality of this paper is twofold. First, it investigates the influencing factors of status consumption that have not been addressed in the extant literature. Second, it is the first to use experimental data to measure segment-level status consumption accurately.

Clonal selection algorithms for optimal product line design: A comparative study

Product design constitutes a critical process for a firm to stay competitive. Whilst the biologically inspired Clonal Selection Algorithms (CSA) have been applied to efficiently solve several combinatorial optimization problems, they have not yet been tested for optimal product lines. By adopting a previous comparative analysis with real and simulated conjoint data, we adapt and compare in this context 23 CSA variants. Our comparison demonstrates the efficiency of specific cloning, selection and somatic hypermutation operators against other optimization algorithms, such as Simulated Annealing and Genetic Algorithm. To further investigate the robustness of each method to combinatorial size, we extend the previous paradigm to larger product lines and different optimization objectives. The consequent performance variation elucidates how each operator shifts the search focus of CSAs. Collectively, our study demonstrates the importance of a fine balance between global and local search in such combinatorial problems, and the ability of CSAs to achieve it.

Using Hierarchical Bayes draws for improving shares of choice predictions in conjoint simulations: A study based on conjoint choice data

The use of Hierarchical Bayes (HB) estimation techniques for choice-based conjoint (CBC) data offers the opportunity to directly use HB draws for preference simulations. This paper analyzes the use of HB draws for shares of choice predictions. Five different choice rules are compared: the first choice rule applied to HB draws, the logit choice rule applied to HB draws, the randomized first choice rule, the traditional first choice rule and the traditional logit choice rule. Each two different holdout choice scenarios are constructed containing one time two extremely similar and the other time very unique alternatives to assess how well the choice rules tolerate the IIA property in predicting choice shares. We present a Monte Carlo study to systematically explore shares of choice predictions based on the different choice rules and further verify whether our findings hold in empirical settings. The key finding of our Monte Carlo study is that using HB draws either combined with the first choice rule or the logit choice rule substantially improves choice share predictions when compared to the other choice rules, regardless of the type of holdout choice scenario. While the logit choice rule applied to HB draws performs a touch better for simulated data, the first choice rule applied to HB draws provides the best choice share predictions for each of the five empirical data sets. Using HB draws does not only provide the best predictive validity but, more importantly, it is theoretically correct when applying a Bayesian estimation approach to CBC data.

Exact Logit-Based Product Design

The share-of-choice product design (SOCPD) problem is to find the product, as defined by its attributes, that maximizes market share arising from a collection of customer types or segments. When customers follow a logit model of choice, the market share is given by a weighted sum of logistic probabilities, leading to the logit-based share-of-choice product design problem. At first glance, this problem appears hopeless: one must optimize an objective function that is neither convex nor concave, over an exponentially-sized discrete set of attribute combinations. In this paper, we develop an exact methodology for solving this problem based on modern integer, convex and conic optimization. At the heart of our methodology is the surprising result that the logit-based SOCPD problem can be exactly reformulated as a mixed-integer convex program. Using both synthetic problem instances and instances derived from real conjoint data sets, we show that our methodology can solve large instances to provable optimality or near optimality in operationally feasible time frames.

Home charging and electric vehicle diffusion: Agent-based simulation using choice-based conjoint data

Electric vehicles offer the potential of climate-friendly driving and have received more and more attention in diffusion research recently. However, the effect of the availability of home charging options on the market potential of electric vehicles has not been targeted in detail in the diffusion process so far, although it represents a beneficial differentiation to conventional vehicles. Therefore, we present an agent-based simulation to address the buying behavior of German consumers regarding electric and plug-in hybrid electric vehicles using empirical data primarily derived from a choice-based conjoint study. We explicitly consider individual possibilities of home charging in our model, which appears to have an important influence on the diffusion of both, electric and plug-in hybrid electric vehicles. However, its importance decreases with faster charging times at public charging stations. Furthermore, our findings reveal that in the longer term, technological progress in charging time, range, and charging station density of electric vehicles is presumably cannibalizing plug-in hybrid electric vehicles’ market shares more than that of conventional vehicles. However, with still lower technological capabilities of electric vehicles, a governmental subsidy can initially promote plug-in hybrid electric vehicles, but electric vehicles will benefit later from that promotional effect.

A Low-Dimension Shrinkage Approach to Choice-Based Conjoint Estimation

Estimating consumers' heterogeneous preferences using choice-based conjoint (CBC) data poses a considerable modeling challenge, as the amount of information elicited from each consumer is often limited. Given the lack of individual-level information, effective information pooling across consumers becomes critical for accurate CBC estimation. In this paper, we propose an innovative low-dimension shrinkage approach to pooling information and modeling preference heterogeneity, in which we learn a low-dimensional affine subspace approximation of the heterogeneity distribution and shrink the individual-level part-worth estimates toward this affine subspace. Drawing on recent modeling techniques for low-rank matrix recovery, we develop a computationally tractable machine learning model for implementing this low-dimension shrinkage and apply it to CBC estimation. We use an extensive simulation experiment and a field data set to demonstrate the superior performance of our low-dimension shrinkage approach as compared to alternative benchmark models.

A Bayesian Dynamic Model for Incomplete Preferences with No-Choice Options in Conjoint Analysis

In the analysis of ranking data such as ranking based conjoint analysis, it is common that all preference ranks are obtained in each trial. However, if there is an no-choice option in alternatives such as conjoint profiles, the ranking ends there, and partial ranking data with a structure depending on the individuals and the trials is observed. We propose a rank-ordered logit model and Bayesian estimation method for partial ranking data with no-choice options that have different ranking structures for each individual and trial. In the model, we consider the latent variables that vary according to each individual and time and decompose them into individual and temporal heterogeneity. Specifically, we consider individuals' heterogeneities using a hierarchical Bayesian model, individuals' learning and evolution of preference using a dynamic linear model, and estimate parameters via the Markov chain Monte Carlo method. For empirical analysis, we apply the proposed model to the analysis of the ranking data obtained by the conjoint measurement method. From the results of conjoint data analysis for smartphones, it is confirmed that the preference or relative importance of the attributes change among trials.

Using Conjoint Experiments to Analyze Elections: The Essential Role of the Average Marginal Component Effect (AMCE)

Political scientists have increasingly deployed conjoint survey experiments to understand multi-dimensional choices in various settings. We demonstrate that the Average Marginal Component Effect (AMCE) constitutes an aggregation of individual-level preferences that translates into a primary quantity of interest to empirical election scholars: the effect of a change in an attribute on a candidate or party's expected vote share. This property holds irrespective of the heterogeneity, strength, or interactivity of voters' preferences and regardless of how votes are aggregated into seats. Overall, our results indicate the AMCE's central role in understanding elections, a conclusion buttressed by a corresponding literature review. We also provide practical advice on interpreting AMCEs. Finally, we propose, formalize, and evaluate the feasibility of using conjoint data to estimate alternative quantities of interest to electoral studies, including the effect of an attribute on the probability of winning.

Personal preferences for Personalised Trials among patients with chronic diseases: an empirical Bayesian analysis of a conjoint survey.

ObjectiveTo describe individual patient preferences for Personalised Trials and to identify factors and conditions associated with patient preferences.DesignEach participant was presented with 18 conjoint questions via an online survey. Each question provided two choices of Personalised Trials that were defined by up to eight attributes, including treatment types, clinician involvement, study logistics and trial burden on a patient.SettingOnline survey of adults with at least two common chronic conditions in the USA.ParticipantsA nationally representative sample of 501 individuals were recruited from the Chronic Illness Panel by Harris Poll Online. Participants were recruited from several sources, including emails, social media and telephone recruitment of the target population.Main outcome measuresThe choice of Personalised Trial design that the participant preferred with each conjoint question.ResultsThere was large variability in participants’ preferences for the design of Personalised Trials. On average, they preferred certain attributes, such as a short time commitment and no cost. Notably, a population-level analysis correctly predicted 62% of the conjoint responses. An empirical Bayesian analysis of the conjoint data, which supported the estimation of individual-level preferences, improved the accuracy to 86%. Based on estimates of individual-level preferences, patients with chronic pain preferred a long study duration (p≤0.001). Asthma patients were less averse to participation burden in terms of data-collection frequency than patients with other conditions (p=0.002). Patients with hypertension were more cost-sensitive (p<0.001).ConclusionThese analyses provide a framework for elucidating individual-level preferences when implementing novel patient-centred interventions. The data showed that patient preference in Personalised Trials is highly variable, suggesting that individual differences must be accounted for when marketing Personalised Trials. These results have implications for advancing precise interventions in Personalised Trials by indicating when rigorous scientific principles, such as frequent monitoring, is feasible in a substantial subset of patients.

A practical approach to revenue management in passenger train services: A case study of the Indonesian railways Argo Parahyangan

We proposed a practical approach to incorporating demand function in pricing optimization and its corresponding capacity allocation problem. Based on the willingness-to-pay approach, demand function was estimated from choice-based conjoint (CBC) data using a combination of hierarchical Bayes estimation, randomized first choice simulation, and cubic spline interpolation. The approach was implemented in the largest intercity passenger train service in Indonesia, assuming two customer segments and four fare classes. Solution to the mixed-integer nonlinear pricing optimization problem was obtained using enumeration. Subsequently, the expected marginal seat revenue heuristic was used in the seat inventory allocation problem.

Product Launches with New Attributes: A Hybrid Conjoint–Consumer Panel Technique for Estimating Demand

The authors propose and empirically evaluate a new hybrid estimation approach that integrates choice-based conjoint with repeated purchase data for a dense consumer panel, and they show that it increases the accuracy of conjoint predictions for actual purchases observed months later. The key innovation lies in combining conjoint data with a long and detailed panel of actual choices for a random sample of the target population. By linking the actual purchase and conjoint data, researchers can estimate preferences for attributes not yet present in the marketplace, while also addressing many of the key limitations of conjoint analysis, including sample selection and contextual differences. Counterfactual product and pricing exercises illustrate the managerial relevance of the approach.