Using network of species interactions to value biodiversity conservation in a megadiverse country: a comparison of latent class and mixed logit models

Typical convergent validity tests of benefit transfer based on stated preference data assume that willingness to pay (WTP) estimates have been accurately measured, and that differences in WTP arise from differences in observable and unobservable characteristics between the study and the policy sites. In this paper, we conduct a convergent validity test assuming equality of underlying preferences, but allow for the possibility that transfer errors arise from differences in the way that respondents process information in the preference elicitation tasks. Using data from an identical survey instrument applied to the population of two river basins in Spain, we obtain marginal and total WTP estimates for ecological improvements of water bodies and the corresponding transfer errors across sites. Results of equality constrained latent class (ECLC) models that infer attribute non-attendance (AN-A) are compared to results from mixed logit (MXL) models in WTP space. We find large absolute and relative differences in marginal and total WTP between sites for the MXL models, and significantly reduced transfer errors for the ECLC models. This paper therefore provides further evidence that AN-A can significantly affect environmental values derived from attribute-based stated preference methods and is the first to investigate the implications for benefit transfer.

A choice to use a seat belt is largely dependent on the psychology of the vehicles’ occupants, and thus those decisions are expected to be characterized by preference heterogeneity. Despite the importance of seat belt use on the safety of the roadways, the majority of existing studies ignored the heterogeneity in the data and used a very standard statistical or descriptive method to identify the factors of using a seatbelt. Application of the right statistical method is of crucial importance to unlock the underlying factors of the choice being made by vehicles’ occupants. Thus, this study was conducted to identify the contributory factors to the front-seat passengers’ choice of seat belt usage, while accounting for the choice preference heterogeneity. The latent class model has been offered to replace the mixed logit model by replacing a continuous distribution with a discrete one. However, one of the shortcomings of the latent class model is that the homogeneity is assumed across a same class. A further extension is to relax the assumption of homogeneity by allowing some parameters to vary across the same group. The model could still be extended to overlay some attributes by considering attributes non-attendance (ANA), and aggregation of common-metric attributes (ACMA). Thus, this study was conducted to make a comparison across goodness of fit of the discussed models. Beside a comparison based on goodness of fit, the share of individuals in each class was used to see how it changes based on various model specifications. In summary, the results indicated that adding another layer to account for the heterogeneity within the same class of the latent class (LC) model, and accounting for ANA and ACMA would improve the model fit. It has been discussed in the content of the manuscript that accounting for ANA, ACMA and an extra layer of heterogeneity does not just improve the model goodness of fit, but largely impacts the share of class allocation of the models.

Understanding urban consumers’ preferences for rice attributes is crucial for rice breeders, producers, and retailers to meet diverse and evolving market demands. Based on the sample data of 629 rice consumers in Shanghai, China, obtained through the choice experiment (CE) approach, this study uses the mixed logit (ML) model to analyze consumers’ preferences and willingness to pay (WTP) for food safety labels, brands, nutritional quality, and taste quality. Furthermore, the latent class (LC) model examines the heterogeneity in consumer group preferences. The research findings highlight that consumers prioritize taste quality as the most crucial attribute, followed by nutritional quality, food safety labels, and brand attributes. The premium rates for superior taste quality, organic certification labels, and green certification labels exceeded 100%. Interestingly, while combining organic certification with well-known international or domestic brands does not uniformly boost consumer preferences, incorporating green certification alongside well-known international or domestic brands significantly elevates those preference levels. Factors such as the external environment, consumption habits, and personal characteristics significantly influence individuals’ preferences for rice attributes. Based on these insights, the study puts forth policy recommendations for rice breeders, producers, and retailers.

ObjectivesThe global coronavirus disease 2019 (COVID‐19) pandemic has not been well controlled, and vaccination could be an effective way to prevent this pandemic. By accommodating attribute nonattendance (ANA) in a discrete choice experiment (DCE), this paper aimed to examine Chinese public preferences and willingness to pay (WTP) for COVID‐19 vaccine attributes, especially the influence of ANA on the estimated results.MethodsA DCE was designed with four attributes: effectiveness, protection period, adverse reactions and price. A random parameter logit model with an error component (RPL‐EC) was used to analyse the heterogeneity of respondents' preferences for COVID‐19 vaccine attributes. Two equality constraint latent class (ECLC) models were used to consider the influence of ANA on the estimated results in which the ECLC‐homogeneity model considered only ANA and the ECLC‐heterogeneity model considered both ANA and preference heterogeneity.ResultsData from 1,576 samples were included in the analyses. Effectiveness had the highest relative importance, followed by adverse reactions and protection period, which were determined by the attributes and levels presented in this study. The ECLC‐heterogeneity model improved the goodness of fit of the model and obtained a lower probability of ANA. In the ECLC‐heterogeneity model, only a small number of respondents (29.09%) considered all attributes, and price was the most easily ignored attribute (64.23%). Compared with the RPL‐EC model, the ECLC‐homogeneity model obtained lower WTPs for COVID‐19 vaccine attributes, and the ECLC‐heterogeneity model obtained mixed WTP results. In the ECLC‐heterogeneity model, preference group 1 obtained higher WTPs, and preference groups 2 and 3 obtained lower WTPs.ConclusionsThe RPL‐EC, ECLC‐homogeneity and ECLC‐heterogeneity models obtained inconsistent WTPs for COVID‐19 vaccine attributes. The study found that the results of the ECLC‐heterogeneity model considering both ANA and preference heterogeneity may be more plausible because ANA and low preference may be confused in the ECLC‐homogeneity model and the RPL‐EC model. The results showed that the probability of ANA was still high in the ECLC‐heterogeneity model, although it was lower than that in the ECLC‐homogeneity model. Therefore, in future research on DCE (such as the field of vaccines), ANA should be considered as an essential issue.Public ContributionChinese adults from 31 provinces in mainland China participated in the study. All participants completed the COVID‐19 vaccine choice questions generated through the DCE design.

Assessment of urban transportation pricing policies with incorporation of unobserved heterogeneity

Purpose This study examines consumer preferences for chicken meat produced using circular economy strategies. It assesses willingness to pay (WTP) for different feeding practices (insect meal, food by-products) and packaging materials (chicken feather-based packaging) across Spain, Denmark, the United Kingdom, and Poland. It identifies key socioeconomic and attitudinal factors influencing consumer acceptance. Design/methodology/approach A discrete choice experiment (DCE) was conducted with 1,967 consumers. Participants evaluated chicken breast options varying by feed type, packaging material, and price. Mixed logit models in WTP space and latent class models estimated heterogeneity in preferences. Interactions with demographic and attitudinal factors were also analysed to identify key drivers of acceptance. Findings Results reveal cross-country differences in WTP. Danish consumers exhibit a positive WTP for chicken feather-based packaging, while Spanish consumers show negative WTP. UK and Polish consumers have no significant preference. Insect meal as feed is largely rejected, whereas food by-products receive mixed responses. Familiarity with circular economy concepts, openness to food technologies, and political ideology influence acceptance. Consumer segmentation identifies clusters, including strong resistors and partial adopters. Originality/value This study provides novel insights into consumer acceptance of circular strategies in meat production. It highlights cultural and socioeconomic factors shaping sustainability preferences and underscores the importance of creating policy mechanisms to close the gap between consumer WTP and industrial feasibility.

Latent class models offer an alternative perspective to the popular mixed logit form, replacing the continuous distribution with a discrete distribution in which preference heterogeneity is captured by membership of distinct classes of utility description. Within each class, preference homogeneity is usually assumed, although interactions with observed contextual effects are permissible. A natural extension of the fixed parameter latent class model is a random parameter latent class model which allows for another layer of preference heterogeneity within each class. A further extension is to overlay attribute processing rules such as attribute non-attendance (ANA) and aggregation of common-metric attributes (ACMA). This paper sets out the random parameter latent class model with ANA and ACMA, and illustrates its application using a stated choice data set in the context of car commuters and non-commuters choosing amongst alternative packages of travel times and costs pivoted around a recent trip in Australia. What we find is that for the particular data set analysed, in the presence of attribute processing together with the discrete distributions defined by latent classes, that adding an additional layer of heterogeneity through random parameters within a latent class only very marginally improves on the statistical contribution of the model. Nearly all of the additional fit over the fixed parameter latent class model is added by the account for attribute processing. This is an important finding that might suggest the role that attribute processing rules play in accommodating attribute heterogeneity, and that random parameters within class are essentially a potentially confounding effect. An interesting finding, however, is that the introduction of random parameters increases the probability of membership to full attribute attendance classes, which may suggest that some individuals assign a very low marginal disutility (but not zero) to specific attributes or that there are very small differences in the marginal disutility of common-metric attributes, and this is being accommodated by random parameters, but not observed under a fixed parameter latent class model.

The fundamental objective of discrete Choice Experiments (CEs) model or Choice Modelling (CM) is to understand the behavioural processes among individuals which drive the choice decisions based on selected attributes and choice options. Preferences may differ among individuals triggered by their specific characteristics such as socio-demographics, constraints and attitudes. Preferences could also vary among groups and within a particular group by forming different segments of groups or subgroups. The Latent Class (LC) model is a distinctive approach which can accommodate preference heterogeneity where preferences are assumed to be relatively homogenous within the segments, but substantially different between the segments. This LC model was applied to account for preference heterogeneity among aquaculturists in the Setiu Wetlands, Terengganu. Currently, fish cage culture is the main socioeconomic activity which imposes considerable impacts on the wetland ecosystem and thus affect its ability to deliver ecosystem service outcomes to other stakeholder groups. This research quantifies the aquaculturists' preferences heterogeneity with regard to the ecosystem impacts under different management scenarios. The existence of subdivisions of preferences within the stakeholder subgroups was tested. This study revealed three latent classes or segments which show relatively distinct sets of preferences. Segment 1 shows a strong preference for higher harvest rates, a moderately strong preference for higher fisheries income and a moderately strong preference for lower shellfish collections. Segment 2 shows a moderately strong linear preference for higher harvest rates and a very strong preference for retaining the status quo. Segment 3 shows a perplexing set of significant preferences for increasing harvest rates and a modest preference for higher fisheries income. In direct contrast to Segment 2, Segment 3 shows a very strong aversion to retaining the status quo. The results of aquaculturists' preferences for delivery of different ecosystem services in Setiu Wetlands suggest that Latent Class Model (LCM) could be applied successfully in a Malaysian setting. The success of the LC model is evidenced by the high level of fit obtained from the best fitting models. The main finding of this research suggests that a good understanding of the main research objective, familiarity with the research area and carefully designed choice set, as well as employing appropriately trained enumerators are the main factors that particularly contribute to a successful application of the LC model in a developing country setting.

PurposeThe purpose of this paper is to examine smallholders’ preferences and willingness to pay for microcredit products with varying attribute combinations, in order to contribute to the debate on the optimal design of rural microcredit.Design/methodology/approachData used in this study are based on a discrete choice experiment from 552 randomly selected respondents. Mixed logit and latent class models are estimated to examine the choice probability and sources of preference heterogeneity. Endogenous attribute attendance models are applied to account for attribute non-attendance (ANA) phenomenon, focusing on separate non-attendance probability as well as joint non-attendance probability.FindingsThe results demonstrate that preference heterogeneity and ANA exist in the smallholder farmers’ microcredit choices. Averagely, smallholder farmers prefer longer credit period, smaller credit size, lower transaction costs and lower interest rate. Guarantor collateral method and installment repayment positively affect their preferences as well. Moreover, respondents are found to be willing to pay more for the attributes they consider important. The microcredit providers are able to attract new customers under the current interest rates, if the combination of attributes is appropriately adjusted.Originality/valueThis study contributes to the debate by assessing the preference trade-off of different microcredit attributes more comprehensively than in previous analyses, by taking preference heterogeneity and ANA into account.

This paper investigates freight market segmentation to seek the potential market of modal shift from road to rail-based intermodal (only pre- and end- haulage are transported by road), using discrete choice analysis. The Defficient design approach was used to generate SP scenarios. The data, including 2160 SP observations of 135 companies, was collected in China's Yangtze River Delta region. The multinomial logit (MNL) model, the mixed logit (ML) model, and the latent class (LC) model were constructed successively. Freight mode attributes (cost, time and reliability), shipment characteristics (distance, time threshold), cargo characteristics (value and density) were considered. In addition, the logistics elements such as the stages of production process and psychological characteristics such as choice inertia were also introduced in this paper. The results show that the LC model with segment variables performs better than the counterpart without segment variables, the MNL model and the ML model. Using the LC model, shippers are classified into four market segments. There are significant differences in shippers' preference among different classes. Based on the class membership function, the characteristic of each class can be identified by a combination of different attributes. The difficulty of achieving the modal shift from road to rail-based intermodal is ranked according to the values of attributes, further determining the priority of the potential market. This study helps to understand the shippers' freight mode choice better. Rail service suppliers can design transportation products that match the shippers' demand characteristics for different classes to improve the market share.

This paper investigates methodological issues related to attribute non‐attendance (ANA) in choice experiments (CEs). First evaluates the variation in the prevalence of ANA across various non‐hypothetical choice experiments (NHCEs) elicitation formats compared to hypothetical choice experiments (HCEs). Second, it investigates whether eliciting respondents’ self‐reported ANA after each choice set (choice task (CT)) or at the end of the choice task (seria task (ST)yields comparable results. Finally, it examines how incorporating self‐reported ANA information affects respondents’ willingness to pay (WTP) and the external predictive powers of the estimated choice models. To answer these research questions, four treatments were conducted: HCE, NHCE, non‐hypothetical rank‐order discrete choice experiment (NHROCE), and non‐hypothetical best worst scaling (NHBWS). The results indicate that accounting for ANA information significantly improves the goodness‐of‐fit of the estimated choice models, especially when full ranking information (NHROCE and NHBWS) is used. In terms of marginal WTP estimates, the results show that modeling ANA, independently of the elicitation approach of ANA (ST or CT), significantly influences consumers’ WTP values. However, the results suggest that incorporating ANA information does not substantially enhance the predictive power of the estimated choice model.

In the current issue of Environmental and Resource Economics, Morey et al. (2006) discuss a new approach to using attitudinal data in latent class modeling. We compare this approach with the one taken in Boxall and Adamowicz (2002), in the context of a discrete choice, random utility framework with heterogeneous preferences. We derive the respective likelihood functions of the two approaches to show that they are structurally similar, and discuss their implications for the use of attitudinal data. We conclude with a discussion comparing the relative merits of latent class and random parameters (mixed logit) modeling, offering the view that as a practical matter, choosing between them depends on the analyst’s judgment about the correlation of preference parameters.

Despite high coverage of China's Family Doctor Contract Services (FDCS), substantive utilization among hypertensive populations remains suboptimal. By decomposing service packages into granular clinical components, this study addresses the limitations of prior research focused on generic primary care attributes. We aim to quantify patient preferences and identify heterogeneity to align service delivery with specialized management needs, thereby facilitating the transition from nominal enrollment to substantive engagement. A Discrete Choice Experiment (DCE) was conducted to community hypertension patients in Nanjing, China. Five key attributes were identified through literature review, qualitative interviews, and expert consultation. A Mixed Logit Model (MLM) and Latent Class Model (LCM) were employed to estimate attribute importance, willingness to pay (WTP), and preference heterogeneity. Analysis of 638 responses, with 596 participants passing the internal consistency check. The Mixed Logit Model demonstrated that all five attributes exerted a statistically significant influence on patient choices. In terms of relative importance, medication type was the primary driver, followed by the scope of services, payment method, appointment scheduling, and the annual contract fee. WTP estimates indicated positive valuations for original-brand medications, integrated clinical service bundles, and multi-source payment structures. Furthermore, the Latent Class Model identified two distinct subgroups reflecting preference heterogeneity within the sample. Therapeutic certainty significantly outweighs economic considerations for community hypertension patients, with the pronounced preference for original-brand medications serving as a critical proxy for clinical safety. Policy should encompass state-led support for original-drug development while simultaneously enhancing institutional trust in generic alternatives through transparent quality evidence. Transitioning toward stratified, patient-centered management is essential to address preference heterogeneity and improve the substantive effectiveness of the family doctor system in China.

Designing biodiversity conservation policies requires not only estimating social benefits but also confirming that results exhibit sensitivity to scope — a necessary methodological condition for validity. This means that willingness to pay (WTP) should increase monotonically with the size or quantity of the goods. This study evaluates that effect in the economic valuation of biodiversity conservation in Manu National Park, Peru, based on specific attributes. A total of 2,240 choice experiments were conducted with household heads in Lima through face-to-face surveys, and logit models were applied to estimate marginal WTP. Results show that the most valued attributes are reducing endangered plant species (PEN 5.08/month) and reducing deforestation (PEN 4.69/month); however, no attribute showed sensitivity to range. Latent class analysis identified two heterogeneous preference groups regarding biodiversity conservation in this protected area. The “pro-conservation” group (78.3%) assigned positive and significant values to all attributes except the reduction from 24 to 8 endangered fauna species. This lack of scope sensitivity may stem from Peru’s status as a megadiverse country, where the inherent non-use value of biodiversity alone could justify in situ conservation policies involving modest costs for relatively few species or specific charismatic habitats.

While a successful survey requires engaged and attentive respondents, careless survey completion remains a great concern in online market research. In this article, we test metrics of engagement in an online willingness-to-pay (WTP) study for fresh blueberry attributes using a major U.S. panel company and evaluate the impact that poorly behaving respondents have on subsequent data quality. In doing so, we investigate in detail the complex joint relationship between attribute non-attendance (ANA) and measures of respondent engagement in web surveying. Using fixed latent classes, an approach known as the Equality Constrained Latent Class procedure, we export individual probabilistic class assignment of all levels of attribute attendance to cross reference with respondents who fail measures of engagement and fraudulence, and analyze their composition and impact on latent classes, indicating non-attendance of individual and combinations of attributes. We also analyze engagement impacts on the tau variance parameter in the scaled mixed logit model and find strong links to unnecessarily increased heterogeneity when not properly filtering poorly behaving respondents. While WTP estimates between respondents passing and failing engagement metrics are similar with the ECLC model, filtering failing respondents in the scaled mixed logit model reduces overall WTP estimates. Results have implications for both WTP researchers and general online market researchers.