We propose three procedures based on association rules (AR) learning and random forests (RF) to support the specification of a portfolio choice model applied in data from complex choice experiment data, specifically a Participatory Value Evaluation (PVE) choice experiment. In a PVE choice experiment, respondents choose a combination of alternatives, subject to a resource constraint. We combine a methodological-iterative (MI) procedure with AR learning and RF models to support the specification of parameters of a portfolio choice model. Additionally, we use RF model predictions to contrast the validity of the behavioural assumptions of different specifications of the portfolio choice model. We use data of a PVE choice experiment conducted to elicit the preferences of Dutch citizens for lifting COVID-19 measures. Our results show model fit and interpretation improvements in the portfolio choice model, compared with conventional model specifications. Additionally, we provide guidelines on the use of outcomes from AR learning and RF models from a choice modelling perspective.@en

Introduction Currently, it is not known what attributes of health care interventions citizens consider important in disinvestment decision-making (i.e. decisions to discontinue reimbursement). Therefore, this study aims to investigate the preferences of citizens of the Netherlands toward the relative importance of attributes of health care interventions in the context of disinvestment. Methods A participatory value evaluation (PVE) was conducted in April and May 2020. In this PVE, 1143 Dutch citizens were asked to save at least €100 million by selecting health care interventions for disinvestment from a list of eight unlabeled health care interventions, described solely with attributes. A portfolio choice model was used to analyze participants' choices. Results Participants preferred to disinvest health care interventions resulting in smaller gains in quality of life and life expectancy that are provided to older patient groups. Portfolios (i.e. combinations of health care interventions) resulting in smaller savings were preferred for disinvestment over portfolios with larger savings. Conclusion The disinvestment of health care interventions resulting in smaller health gains and that are targeted at older patient groups is likely to receive most public support. By incorporating this information in the selection of candidate interventions for disinvestment and the communication on disinvestment decisions, policymakers may increase public support for disinvestment.@en

Background: The COVID-19 outbreak early 2020 was followed by an unprecedented package of measures. The relative calmness of the pandemic early 2022 provides a momentum to prepare for various scenarios. Objectives: As acceptance of COVID-19 measures is key for public support we investigated citizens’ preferences towards imposing measures in four scenarios: 1) spring/summer scenario with few hospitalizations; 2) autumn/winter scenario with many hospitalizations; 3) a new contagious variant, the impact on hospitalizations is unclear; 4) a new contagious variant, hospitalizations will substantially increase. Methods: Study 1 comprised a Participatory Value Evaluation (PVE) in which 2011 respondents advised their government on which measures to impose in the four scenarios. Respondents received information regarding the impact of each measure on the risk that the health system would be overloaded. To triangulate the results, 2958 respondents in Study 2 evaluated the acceptability of the measures in each scenario. Results: Measures were ranked similarly by respondents in Study 1 and 2: 1) the majority of respondents thought that hygiene measures should be upheld, even in the spring/summer; 2) the majority supported booster vaccination, working from home, encouraging self-testing, and mandatory face masks from scenario 2 onwards; 3) even in scenario 4, lockdown measures were not supported by the majority. Young respondents were willing to accept more risks for the health system than older respondents. Conclusion: The results suggest that policies that focus on prevention (through advising low-impact hygiene measures) and early response to moderate threats (by scaling up to moderately restrictive measures and boostering) can count on substantial support. There is low support for lockdown measures even under high-risk conditions, which further emphasizes the importance of prevention and a timely response to new threats. Our results imply that young citizens’ concerns, in particular, should be addressed when restrictive COVID-19 measures are to be implemented.@en

Since its origins in the 1970s, choice modelling has become an important field of study in diverse areas, including transportation, health economics, environmental economics and marketing. Choice modellers have developed several methods to collect and model individual choices. Researchers and policymakers use such methods to understand individual preferences in diverse contexts, derive economic values or predict behaviour. Over the years, the field of choice modelling has been developed in two key areas. Firstly, choice modellers have developed new data collection tools to account for more realistic forms of decision-making. While discrete choice experiments (DCEs) are still popular and highly customisable, they force respondents to choose among mutually-exclusive alternatives, which may not reflect how individuals choose in real life. In response, new SC experiments have been proposed to incorporate more realistic forms of decision-making, such as Participatory Value Evaluation (PVE). In a PVE experiment, respondents select a combination of alternatives without surpassing resource constraints. Secondly, while theory-driven models based on utility theory, e.g., random utility maximisation (RUM) or Kuhn-Tucker, are still the norm to model choice behaviour, there is a broader recognition that individual' behaviour is ultimately unknown from the analyst perspective, data-driven methods can help to uncover such behaviour. Despite the latter, to the author’s knowledge, three methodological and practical challenges are still unresolved in the literature. Firstly, no research has been done to explore the potential of data-driven methods to analyse data from SC experiments outside DCEs, and in particular for PVE experiments, either as complements to improve the specification of choice models or as standalone data analysis methods. Secondly, while data-driven methods for discrete choices (and DCEs) are available in the literature, such methods either sacrifice their flexibility to learn from the data to satisfy consistency assumptions or vice versa. Thus, a method that balances flexibility and consistency assumptions is lacking. Thirdly, there is a lack of software tools to estimate and compare data-driven methods easily and conveniently, hindering their widespread use. Considering these challenges, this thesis further investigates how data-driven methods can be used for analysing individual choice behaviour from SC experiments, either to complement theory-driven choice models or alternatives to theory-driven choice models; and to develop methodological tools for such purposes, i.e., new models and software. This thesis scopes its research to two specific SC experiments: PVE and DCEs. To reach the goals of this thesis, five novel studies are proposed. The first study (Chapter 2) introduces the reader to how PVE experiments are conducted in real-life and how they are conventionally analysed with theory-driven choice models. The second study (Chapter 3) proposes three procedures based on association rules (AR) learning and random forests (RF) to assist the specification and test the validity of the assumptions of theory-driven choice models for PVE experiments. The third study (Chapter 4) shows how XGBoost and SHAP -a machine learning model and explainable artificial intelligence method, respectively- can be used to analyse PVE experiments data as an alternative to theory-driven analysis. The fourth study (Chapter 5) proposes a new discrete choice model based on artificial neural networks that balances flexibility to learn utility functions from the data while satisfying consistency with RUM and economic theory. The fifth study (Chapter 6) introduces NP4VTT, a new software tool that provides five nonparametric models to uncover the VTT distribution from two-attribute-two-alternative DCEs. Together, these studies provide further evidence that supports the use of data-driven methods to analyse individual choice behaviour and specific methodological tools were provided for such purposes. This thesis concludes by highlighting that while the primary research goal and sub-goals were achieved, the relevance of the findings and conclusions of this thesis shall be put into perspective. Firstly, using data-driven methods, either to assist choice models or as an alternative to them, lead to “moderate-to-modest” model fit improvements. Consequently, researchers or policymakers interested in using the methods proposed in this thesis for prediction should not expect considerable differences compared with conventional choice models. Secondly, the methods proposed in this thesis provide a considerable number of new insights of behavioural interest. Choice modellers could benefit from thesis insights to contrast or further assist the development of choice models, while policymakers have a wide range of new information for targeting decisions to specific policies or individuals. However, researchers should consider how to synthesise all these new insights effectively. Thirdly, this thesis made efforts to make more data-driven methods available by, for instance, publishing the studies in open-access journals and, when possible, making code and data publicly available for the general public. Nevertheless, there are still conceptual challenges to make these methods more amicable to researchers accustomed to the concepts and structure of the choice modelling community. As a final reflection, while having the potential to help choice modellers to increase their understanding of individual choice behaviour, data-driven methods still require more development (and being easily accessible) to serve as a real alternative to choice models. @en

Two-attribute-two-alternative stated choice experiments are widely used to infer the Value-of-Travel-Time (VTT) distribution. Two-attribute-two-alternative stated choice experiments have the advantage that their data can be analysed using nonparametric models, which allow for the inference of the VTT distribution without having to impose assumptions on its shape. However, a software package that enables researchers to estimate nonparametric models promptly is currently lacking. As a result, nonparametric models are underused. This paper aims to fill this software void. It presents NP4VTT, a Python package that enables researchers to estimate and compare nonparametric models in a fast and convenient way. It comprises five nonparametric models for estimating the VTT distribution from data coming from two-attribute-two-alternative stated choice experiments. We illustrate the use of NP4VTT by applying it to the Norwegian 2009 VTT data. We hope this software package will help researchers studying the VTT make more informed decisions concerning the shape of the VTT distribution and encourages the use and development of nonparametric models for choice behaviour analyses.@en

Overweight and obesity are a growing problem, especially among people with a low income. Policymakers aspire to alleviate this problem by implementing publicly funded projects. This study has three aims: 1) to explore citizen preferences regarding the public funding of projects promoting a healthy body weight among people with a low income, 2) to identify whether such preferences differ between citizens with a low income and those with a higher income, and 3) to identify the reasons underlying these preferences. We conducted a Participatory Value Evaluation (PVE) among 1053 Dutch citizens to achieve these aims. In an online choice experiment, respondents were asked to advise on the implementation of eight different projects that encourage a healthy body weight among citizens with a low income, with a total resource constraint of 100,000 euros. The projects were 1) lifestyle coaching including sports, 2) lifestyle coaching without sports, 3) local sports coach, 4) fruit and vegetable boxes, 5) bariatric surgery, 6) improving the living environment, 7) courses on healthy lifestyles, and 8) sports vouchers. We used the “Multiple Discrete-Continuous Extreme Value” model to estimate the preferences of respondents towards these eight projects. Fruit and vegetable boxes and sports vouchers were the most popular projects, while bariatric surgery was least popular. Respondents with a low income tended to spend less of the budget than respondents with a higher income. Respondent arguments for the choices they made were qualitatively analysed using inductive content analysis. They often mentioned the value judgements ‘importance’, ‘healthiness’ and ‘usefulness’, as well as project costs and efficacy, as reasons for their decisions. Policymakers could use the results to ensure their decisions on the allocation of public funding to projects that encourage a healthy weight among people with a low income are aligned with citizen preferences.@en

Following the outbreak of COVID-19, governments took unprecedented measures to curb the spread of the virus. Public participation in decisions regarding (the relaxation of) these measures has been notably absent, despite being recommended in the literature. Here, as one of the exceptions, we report the results of 30, 000 citizens advising the government on eight different possibilities for relaxing lockdown measures in the Netherlands. By making use of the novel method Participatory Value Evaluation (PVE), participants were asked to recommend which out of the eight options they prefer to be relaxed. Participants received information regarding the societal impacts of each relaxation option, such as the impact of the option on the healthcare system. The results of the PVE informed policymakers about people's preferences regarding (the impacts of) the relaxation options. For instance, we established that participants assign an equal value to a reduction of 100 deaths among citizens younger than 70 years and a reduction of 168 deaths among citizens older than 70 years. We show how these preferences can be used to rank options in terms of desirability. Citizens advised to relax lockdown measures, but not to the point at which the healthcare system becomes heavily overloaded. We found wide support for prioritising the re-opening of contact professions. Conversely, participants disfavoured options to relax restrictions for specific groups of citizens as they found it important that decisions lead to "unity"and not to "division". 80% of the participants state that PVE is a good method to let citizens participate in government decision-making on relaxing lockdown measures. Participants felt that they could express a nuanced opinion, communicate arguments, and appreciated the opportunity to evaluate relaxation options in comparison to each other while being informed about the consequences of each option. This increased their awareness of the dilemmas the government faces.@en