nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2011‒08‒15
seventeen papers chosen by
Philip Yu
Hong Kong University

  1. Inferring Consideration Set from Scanner Data By Wirawan Dony Dahana; Nozomi Nakajima
  2. Upgrading or Downgrading? \ Framing Effects in Online Shopping Environments \ By Nozomi NAKAJIMA
  3. Effects of Mode Shares on Mode Choice By Carlos Carrion-Madera; Nebiyou Tilahun; David Levinson
  4. Accessibility, network structure, and consumers’ destination choice: a GIS analysis of GPS travel data. By Arthur Huang; David Levinson
  5. Manipulation of Choice Behavior By Manzini, Paola; Mariotti, Marco; Tyson, Christopher J.
  6. Importance of E-services for Cultural Tourism By Eveline van Leeuwen; Peter Nijkamp
  7. Farmersâ Willingness to Grow Switchgrass as a Cellulosic Bioenergy Crop: A Stated Choice Approach By Fewell, Jason E.; Bergtold, Jason; Williams, Jeffery
  8. Supply chain differentiation, contract agriculture, and farmersâ marketing preferences: the case of sweet pepper in Thailand By Schipmann, Christin; Qaim, Matin
  9. Investigating the Role of Poultry in Livelihoods and the Impact of HPAI on Livelihoods Outcomes in Africa: Evidence from Ethiopia, Ghana, Kenya and Nigeria By Birol, Ekin; Asare-Marfo, Dorene; Ayele, Gezahegne; Mensah-Bonsu, Akwasi; Ndirangu, Lydia K.; Okpukpara, Benjamin; Roy, Devesh; Yakhshilikov, Yorbol
  10. Measuring Market Potential for Fresh Organic Fruit and Vegetable in Ghana By Owusu, Victor; Owusu, Michael Anifori
  11. Determinants of the Speed of Adoption of Soil Fertility-Enhancing Technologies in Western Kenya By Odendo, Martins; Obare, Gideon A.; Salasya, Beatrice
  12. Using a Discrete Choice Experiment to Elicit Consumersâ WTP for Health Risk Reductions Achieved By Nanotechnology in the UK By Erdem, Seda; Rigby, Dan
  13. The effect of using labelled alternatives in stated choice experiments: an exploration focusing on farmland walking trails in Ireland By Doherty, Edel; Campbell, Danny; Hynes, Stephen; Van Rensberg, Tom
  14. Reclaimed Wastewater and the WTP to avoid Summer Water Restrictions: Incorporation Endogenous Free-riding Beliefs By Dupont, Diane P.
  15. Choosing âbuy noneâ in food choice analysis: the role of utility balance By Campbell, Danny; Doherty, Edel; Gibson, Vikki
  16. "Willingness to Pay for Electric Vehicles and their Attributes" By Michael K. Hidrue; George R. Parsons; Willett Kempton; Meryl Gardner
  17. Homeowners' Preferences for Adopting Residential Heating Systems: A Discrete Choice Analysis for Germany By Michelsen, Carl Christian; Madlener, Reinhard

  1. By: Wirawan Dony Dahana (Graduate School of Economics, Osaka University); Nozomi Nakajima (Graduate School of Economics, Osaka University)
    Abstract: Consideration set has been one of main research topics in marketing field for a long time. Using interview data, many studies have been conducted to investigate the nature of its content. However, only few studies tried to do so by using consumer purchase history. By modeling the process of consideration and choice set formation, in this research we try to elicit consumer consideration set from scanner data. Some managerial implications for marketing decisions derived from the information of consideration set are discussed.
    Keywords: Consideration Set, Choice Set, Choice Model
    JEL: M11 M31 M37
    Date: 2011–08
  2. By: Nozomi NAKAJIMA (Graduate School of Economics, Osaka University)
    Abstract: Recent development in behavioral decision theory reveals the important role of decision environment in the consumer's evaluation and choice processes. Often it is referred as "decision framing." Of particular interest is the online shopping environment, where buyers are usually forced to make their decisions under the sellers' (programmed) guidance on their web sites. How can the decision frames constructed in online shopping environment influence consumers' decision making? What should be done to exploit the characteristics of their framed decisions in the design of online shopping environments? In the present study, we considered an online PC shop as an example because it is one of the most popular and typical online shops and it will help us get insights into the consumers' online-framed decision characteristics. Buyers are usually led to specify the configurations of personal computers, i.e., CPU, memory and hard drive size, type of optical drives, etc., taking their preferences and budgets into account. In the course of specification processes, their decisions are framed in some ways and influenced by them. Among other things, the way the choice alternatives are presented (upgrading/ downgrading, etc.), from which buyers are expected to choose, is of special interest because it can be easily controlled by the sellers. Experimental studies were conducted to investigate the influence of some decision frames including the flow of selection process, the number of alternatives, the price intervals of the alternatives, and the default choice settings. The extremeness aversion, the shifts of the reference points, and the tradeoff between utility and economic loss aversion, are the examples of the involved effects. Above all, particular attention was paid to the default choice settings that provide the total prices as well as the reference points. Based on the results of the experiments, a set of theoretical conclusions and managerial implications of default choice settings are discussed.
    Keywords: online shopping, decision framing, pricing, choice model
    Date: 2011–08
  3. By: Carlos Carrion-Madera; Nebiyou Tilahun; David Levinson (Nexus (Networks, Economics, and Urban Systems) Research Group, Department of Civil Engineering, University of Minnesota)
    Abstract: This study considers the influence of the knowledge of existing mode shares on travelers mode choice. This contrasts with traditional mode choice models, where the main objective is to predict the overall mode shares as the aggregate of individual mode choices according to variables encompassing attributes of the modes, and characteristics of the travelers. In this study, a computer-administered adaptive stated preference survey is developed and applied to a sample of subjects selected from the University of Minnesota. The results indicate that the presence of mode shares in the mode choice model does influence the decision of travelers.
    Keywords: mode choice, mode shares, mixed logit, stated preference.
    JEL: R41 R48 R53
    Date: 2011
  4. By: Arthur Huang; David Levinson (Nexus (Networks, Economics, and Urban Systems) Research Group, Department of Civil Engineering, University of Minnesota)
    Abstract: Anecdotal and empirical evidence has shown that road networks, destination accessibility, and travelers' choice of destination are closely related. Nevertheless, there have not been systematic investigations linking individuals' travel behavior and retail clusters at the microscopic level. Based on GPS travel data in the Twin Cities, this paper analyzes the impacts of travelers' interactions with road network structure and clustering of services at the destination on travelers' destination choice. A multinomial logit model is adopted. The results reveal that higher accessibility and diversity of services in adjacent zones of a destination are associated with greater attractiveness of a destination. Further, the diversity and accessibility of establishments in an area are often highly correlated. In terms of network structure, a destination with a more circuitous or discontinuous route dampens its appeal. Answering where and why people choose to patronize certain places, our planning, our findings shed light on the design of road networks and clusters from a travel behavior perspective.
    Keywords: travel behavior, destination choice, accessibility, GIS analysis, GPS travel data, road networks
    JEL: R30 L22
    Date: 2011
  5. By: Manzini, Paola (University of St. Andrews); Mariotti, Marco (University of St. Andrews); Tyson, Christopher J. (Queen Mary, University of London)
    Abstract: We introduce and study the problem of manipulation of choice behavior. In a class of two-stage models of decision making, with the agent's choices determined by three "psychological variables," we imagine that a subset of these variables can be selected by a "manipulator." To what extent does this confer control of the agent's behavior? Within the specified framework, which overlaps with two existing models of choice under cognitive constraints, we provide a complete answer to this question.
    Keywords: attention, choice function, revealed preference, satisficing, threshold
    JEL: D01 D70
    Date: 2011–07
  6. By: Eveline van Leeuwen (VU University Amsterdam); Peter Nijkamp (VU University Amsterdam)
    Abstract: Despite a busy lifestyle – or perhaps as a result of a stressful lifestyle – more people than ever before make leisure trips, sometimes for a long time but in many cases just for short periods. Modern telecommunication technology brings attractive tourist destinations directly into the living rooms of potential travellers, also destinations that would otherwise have been difficult to find, such as certain cultural heritage objects. In this contribution, we will address the relevance of e-services in urban cultural tourism. Its aim is to map out the relative drivers of cultural visitors to cities with a particular view on the importance of modern e-services in the tourist sector. We focus on three case-study cities: Amsterdam, Leipzig and Genoa. We use discrete choice models and factor analysis to analyse the preferences of tourists for cultural heritage and e-services. Interestingly, in all three cities, the most important group of tourists, the cultural heritage enthusiasts, are often international tourists. This stresses the importance of multilingual e-services in order to maximise their impact on cultural heritage visitors and the tourism sector in general. In addition, it is also important to note that certain e-services become more important for tourists from further away, such as online booking systems.
    Keywords: Tourism; cultural heritage; e-services; ordered logit models; factor analysis
    JEL: L83 L86
    Date: 2011–08–09
  7. By: Fewell, Jason E.; Bergtold, Jason; Williams, Jeffery
    Abstract: Farmersâ Willingness to Grow Switchgrass as a Cellulosic Bioenergy Crop: A Stated Choice Approach Agricultureâs role as a source of feedstocks in a potential lignocellulosic-based biofuel industry is a critical economic issue. Several studies have assessed the technical feasibility of producing bioenergy crops on agricultural lands. However, few of these studies have assessed farmersâ willingness to produce or supply bioenergy crops or crop residues. Biomass markets for bioenergy crops do not exist, and developing these markets may take several years. Therefore, an important, yet unaddressed question is under what contractual or pricing arrangements farmers will grow biomass for bioenergy in these nascent markets. The purpose of this paper is to examine farmersâ willingness to produce switchgrass under alternative contractual, pricing, and harvesting arrangements. Contracts are likely to be the preferred method to bring together producers and processors of biomass for bioenergy. Contract design may vary across farmers and crop type, and may include attributes specific to annual crops, contract length, quantity or acreage requirements, quality specifications, payment dates, and other important features. A stated choice survey was administered in three, six-county areas of Kansas by Kansas State University and the USDA, National Agricultural Statistics Service from November 2010 to January 2011 to assess farmersâ willingness to produce cellulosic biomass under different contractual arrangements. This paper focuses on the switchgrass stated choice experiment from the survey. The stated choice experiment asked farmers to rank their preferred contractual arrangement from two contract options and one âdo not adoptâ option. Contractual attributes included percentage net returns above the next best alternative (e.g. CRP or hay production), contract length, a custom harvest option, insurance availability, and a seed-cost share option. Respondents then ranked their preferred contract option. The survey also collected data on farm characteristics, bioenergy crop preferences, socio-economic demographics, risk preferences, and marketing behavior. The survey used a stratified sample of farmers who farm more than 260 acres and grow corn. A total of 460 surveys were administered with a 65 percent completion rate. The underlying theoretical model uses the random utility model (RUM) approach to assess farmersâ willingness to grow switchgrass for bioenergy and determine the contractual attributes most likely to increase the likelihood of adoption. This framework allows us to define the âprice,â or farmersâ mean willingness to accept, for harvested biomass sold to an intermediate processor. The estimated choice models follow the approach of Boxall and Adamowicz (2002) to capture heterogeneity across farmers and geographic regions due to management differences, conservation practices, and risk preferences. Using the percentage net return above CRP or hay production allows prices to float to levels that will entice farmers to adopt switchgrass. This will help determine a market price for bioenergy crops based on current market and production conditions without specifying an exact monetary value for the biomass. In addition, the survey results will facilitate contract designs between biorefineries and farmers while informing policymakers and the biofuel industry about farmersâ willingness to supply biomass for bioenergy production. Reference: Boxall, P.C. and W.L. Adamowicz, âUnderstanding Heterogeneous Preferences in Random Utility Models: A Latent Class Approach,â Environmental and Resource Economics 23(2002): 421 â 446.
    Keywords: Biofuels, Cellulosic, Biomass, Switchgrass, Farmers, Willingness to Pay, Crop Production/Industries, Production Economics, Resource /Energy Economics and Policy,
    Date: 2011
  8. By: Schipmann, Christin; Qaim, Matin
    Abstract: There is an emerging body of literature analyzing how smallholder farmers in developing countries can be linked to modern supply chains. However, most of the available studies concentrate on farm and farmer characteristics, failing to capture details of institutional arrangements between farmers and traders. Moreover, farmersâ preferences have rarely been considered. Here, we address these gaps by analyzing different market channels for sweet pepper in Thailand. Using data from a survey and choice experiment with farmers, we find that there is a general preference for marketing options that do not involve a contract. Additional provision of inputs and credit can increase the attractiveness of contracts. Yet, the most important factor for farmers is to personally know the buyer they deal with, which may be related to issues of trust. Some policy implications are discussed.
    Keywords: Choice experiment, contract design, farmersâ stated preferences, modern agricultural supply chains, Thailand, Community/Rural/Urban Development, Institutional and Behavioral Economics, International Development, L14, O31, Q12, Q13,
    Date: 2011–04
  9. By: Birol, Ekin; Asare-Marfo, Dorene; Ayele, Gezahegne; Mensah-Bonsu, Akwasi; Ndirangu, Lydia K.; Okpukpara, Benjamin; Roy, Devesh; Yakhshilikov, Yorbol
    Abstract: In this paper we investigate the role of poultry in the livelihoods portfolios of households and the impact of supply and demand shocks that may be caused by Highly Pathogenic Avian Influenza (HPAI) on various livelihoods outcomes of households in four Sub-Saharan African (SSA) countries. The study countries include Ethiopia and Kenya in East Africa and Ghana and Nigeria in West Africa. These countries represent a spectrum of SSA countries in terms of disease status, role of poultry sector and means of disease spread. By using nationally representative household level secondary data and discrete choice methods (probit model and zero inflated negative binomial model) we profile the household, farm and regional characteristics of those households who are most likely to keep poultry, and those who are most likely to be engaged in intensive poultry production, i.e., keep larger household flocks. We estimate the impact of the disease outbreaks and scares/threats on livelihood outcomes by using matching methods (i.e., propensity score matching). The results of this study generate valuable information regarding the role of poultry in the livelihoods of small-scale poultry producing households and the livelihood impacts of HPAI induced demand and supply shocks. Such information is critical for the design of targeted and hence efficient and effective HPAI control and mitigation policies.
    Keywords: Highly Pathogenic Avian Influenza (HPAI), demand shock, supply shock, livelihoods, probit model, zero inflated negative binomial model, propensity score matching, Livestock Production/Industries,
    Date: 2010–09
  10. By: Owusu, Victor; Owusu, Michael Anifori
    Abstract: This paper examines the market potential for fresh organic lettuce and water melon with a recently collected data on consumers from Kumasi metropolis of Ghana. Using a doublebounded dichotomous choice contingent valuation technique, consumerâs willingness to pay is estimated with a Tobit model to address the zero willingness to pay responses in the sample data. As much as 71% of the consumers are willing to pay over 50% price premiums for organic vegetables and over 82% are willing to pay 1%â50% price premiums for organic fruits. The empirical results indicate that human capital, product attributes and consumer perception influence consumersâ willing to pay for organic food products. The estimated market potential for organic fruit is GH¢32,117,113 (US$ 26,453,433) per annum and that of organic vegetable is GH¢1,991,224 (US$1,640,083) per annum suggesting a huge market potential for organic fruits in Ghana.
    Keywords: Willingness to Pay, Price Premium, Organic Products, Consumer Perception, Market Potential, Africa, Crop Production/Industries,
    Date: 2010–09
  11. By: Odendo, Martins; Obare, Gideon A.; Salasya, Beatrice
    Abstract: Most adoption studies have employed cross-sectional data in a static discrete choice modelling framework to analyze why some farmers adopt at a certain point in time. The static approach does not consider the dynamic environment in which the adoption decision is made and thus does not incorporate the speed of adoption and the effect of time-dependent elements in explaining adoption. The adoption speed of an innovation is important in various aspects. Based on data from a survey of a random sample of 331 smallholder households in western Kenya, this study investigated determinants of time to adoption of mineral fertilizer, animal manure and compost using Duration analysis. Results revealed that factors that influenced timing of the adoption varied by the practices. Whilst education level of the household head, cattle ownership, location of the farm, access to extension services, and participation in land management programmes accelerated the adoption of different practices, age of household head, relative farming experience and market liberalization retarded the adoption. Gender of household head gave mixed results. To speed up adoption of the practices requires policies that promote farmersâ participation in land management programs, access to extension services and markets in addition to stratified targeting of different practices to specific locations and farmers.
    Keywords: Adoption, duration analysis, soil nutrients, Crop Production/Industries, Land Economics/Use,
    Date: 2010–09
  12. By: Erdem, Seda; Rigby, Dan
    Abstract: We present research findings on consumersâ willingness to pay (WTP) for reductions in the level of foodborne health risks. The research addresses how such valuations are affected by the means of which the risk reduction is delivered and the methods of risk presentations used in choice tasks. In this case, the research has two treatments. In the first treatment, the comparison is between risk reductions achieved by an improvement in the food system in general (e.g., more stringent regulations and inspection regimes) within the slaughter and meat processing stages of the food chain, as opposed to a risk reduction achieved via innovations in food packaging using nanotechnology, which is the use of nanosensors in packaging. If there is a contamination in packaging, nanosensors reveal a colour change on the packaging material. In the second treatment, the comparison is between valuations of risk reductions in which reductions in risks are presented via absolute values and grids and absolute values together. Both comparisons are achieved via split sample Discrete Choice Experiment surveys. The difference between consumersâ valuations of foodborne risk reductions provides an implicit value for nanotechnology (i.e., WTP to avoid) and the effect of risk grids on choices people make. General results show the existence of heterogeneity in British consumersâ preferences. The effects of nanosensors and risk grids on consumersâ choices are not strong across the models. The valuations of health risk reductions show some variations across the models in both treatment groups.
    Keywords: Discrete Choice Experiments, Nanotechnology, Nanosensors, Health Risks, Grids, UK, Health Economics and Policy,
    Date: 2011–04
  13. By: Doherty, Edel; Campbell, Danny; Hynes, Stephen; Van Rensberg, Tom
    Abstract: Discrete choice experiment data aimed at eliciting the demand for recreational walking trails on farmland in Ireland is used to explore whether some respondents reach their choices solely on the basis of the alternativeâs label. To investigate this type of processing strategy, the paper exploits a discrete mixtures approach which encompasses random parameters for the attributes. We find evidence that respondents employ different processing strategies for different alternatives and differences in processing emerge between rural and urban based respondents. Results highlight that model fit and policy conclusions are sensitive to assumptions related to processing strategies among respondents
    Keywords: Discrete choice experiments, processing strategies, discrete mixtures, rural and urban comparison, outdoor recreation, welfare estimates, Land Economics/Use,
    Date: 2011–04
  14. By: Dupont, Diane P.
    Abstract: Climate change may cause more frequent seasonal water shortages. Water-scarce countries already use reclaimed household wastewater for subsequent uses that do not require potable water. Views on the degree of acceptability of reusing wastewater and a distrust of oneâs water provider may deter countries like Canada and the United Kingdom from adopting these technologies. This paper reports on results from a 2009 Canadian Internet-based contingent valuation study. Two water management programs were presented: a program to reduce summer water use through water restrictions and a program to allow citizens to avoid summer water restrictions through the use of reclaimed household wastewater. The paper estimates the willingness-to-pay (WTP) for the second program and finds trust in the water utility, belief in future drought conditions, and age to be important factors, as is the belief that members of oneâs community will not voluntarily reduce water use by the required amount. The latter introduces a potential endogeneity bias in responses to discrete choice WTP questions. Joint estimation of the underlying WTP function with a model to explain a respondentâs belief in community free riding is undertaken. The paper contributes to the literature on valuing the avoidance of water restrictions, elicitation of free riding beliefs, and the use of endogenous regressors in discrete choice models
    Keywords: water shortages, reclaimed wastewater, WTP, free riding beliefs, endogeneity in discrete choice models, Environmental Economics and Policy, Q25, Q51, Q53,
    Date: 2011–04
  15. By: Campbell, Danny; Doherty, Edel; Gibson, Vikki
    Abstract: Stated choice analysis is now a widely used and accepted methodology for exploring food choice. In stated choice experiments respondents are asked to make a choice between two or more alternatives, one of which typically takes the form of a âbuy noneâ option. It is widely recognised that respondents often perceive this option differently from the other alternatives and various reasons for this have been offered. Nevertheless, the role that utility balance among the experimentally designed options plays on the propensity of respondentâs choosing âbuy noneâ has largely been overlooked. Using a non-linear representation of utility we show that the âbuy noneâ choices are sensitive to utility balance. We further show how accommodating this provides an additional insight into choice behaviour and has a bearing on welfare calculations
    Keywords: discrete choice experiments, utility balance, status-quo bias, food choice, Consumer/Household Economics,
    Date: 2011–04
  16. By: Michael K. Hidrue (Department of Agricultural Economics, Mississippi State University); George R. Parsons (Department of Economics, University of Delaware); Willett Kempton (Department of Electrical and Computer Engineering, University of Delaware); Meryl Gardner (Department of Business Administration, University of Delaware)
    Abstract: This article presents a stated preference study of electric vehicle choice using data from a national survey. We used a choice experiment wherein 3029 respondents were asked to choose between their preferred gasoline vehicle and two electric versions of that preferred vehicle. We estimated a latent class random utility model and used the results to estimate the willingness to pay for five electric vehicle attributes: driving range, charging time, fuel cost saving, pollution reduction, and performance. Driving range, fuel cost savings, and charging time led in importance to respondents. Individuals were willing to pay (wtp) from $35 to $75 for a mile of added driving range, with incremental wtp per mile decreasing at higher distances. They were willing to pay from $425 to $3250 per hour reduction in charging time (for a 50 mile charge). Respondents capitalized about 5 years of fuel saving into the purchase price of an electric vehicle. We simulated our model over a range of electric vehicle configurations and found that people with the highest values for electric vehicles were willing to pay a premium above their wtp for a gasoline vehicle that ranged from $6000 to $16,000 for electric vehicles with the most desirable attributes. At the same time, our results suggest that battery cost must drop significantly before electric vehicles will find a mass market without subsidy.
    Keywords: Electric Vehicles, Stated Preference, Discrete Choice
    JEL: Q42 Q51
    Date: 2011
  17. By: Michelsen, Carl Christian (E.ON Energy Research Center, Institute for Future Energy Consumer Needs and Behavior (FCN), RWTH Aachen University); Madlener, Reinhard (E.ON Energy Research Center, Institute for Future Energy Consumer Needs and Behavior (FCN), RWTH Aachen University)
    Abstract: Space heating accounts for a large fraction of the primary energy consumption and CO2 emissions of residential buildings. Besides targeting the insulation standard, residential heating systems (RHS) based on renewable energy sources offer the potential to reduce energy demand for space heating. Therefore, understanding the determinants of the homeowners’ adoption decisions in favor of RHS becomes increasingly important. In this paper, we analyze the influence of preferences about RHS-specific attributes on the adoption decision. Moreover, we control for the influence of socio-demographic, home and spatial characteristics. For this purpose, we specify the discrete appliance choice as a multinomial logit model. We apply the model to representative survey data for Germany. Our findings show that there are different drivers for the adoption of RHS in newly built and existing 1- and 2-family homes, and that the importance of key drivers also differs across groups of homeowners and RHS, respectively. First, we find that adopters of a gas- and oil-fired condensing boiler with solar thermal support have a strong preference for energy savings, while adopters of a heat pump or wood pellet-fired boiler prefer being more independent from fossil fuels. Second, we find preferences about RHS-specific attributes to be highly relevant for owners of newly built homes, while evidence for an influence of the control variables is scarce. Third, we find that socio-demographic, home and spatial characteristics have a higher and more differentiated impact in the case of existing homes. These variables predetermine the decision context and leave less leeway for the influence of preferences when replacing a RHS in an existing home.
    Keywords: Technology adoption; Consumer behavior; residential heating systems; space heating; discrete choice
    JEL: D12 O33 Q41 R22
    Date: 2011–05

This nep-dcm issue is ©2011 by Philip Yu. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.