C.G. Chorus
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
94 records found
1
Journal article
(2026)
-
Patrick J. Thoral, Jesse de Metz, Stella Mulia, Annebel ten Broeke, Nicolaas Heyning, Birkitt L. ten Tusscher, Rolf K. Gigengack, Hilde M. Feijen, Caspar G. Chorus, More Authors
Background
Withholding and withdrawing life-sustaining therapy (LST) is common in European ICUs but significant variations exist. Behaviour artificial intelligence technology (BAIT) may help standardize the ethical dilemma to continue or withdraw LST for patients already admitted to the ICU.
Methods
Several sessions with intensivists of an academic medical centre and a large urban teaching hospital were held to determine the criteria influencing the process. A discrete choice experiment was conducted during which 25 hypothetical cases were presented to the participants. For each case the participants had to decide whether they would continue, continue with a time limited trial of one week, or withdraw LST. The results of the experiment were used to develop a multinomial logistic regression model that was incorporated in a web-based decision-support system.
Results
Thirty-six participants (intensivists and fellows in intensive care medicine) completed the experiment. The estimated model consisted of twelve covariates and showed good model fit (McFadden’s ρ2 0.25). The most important covariates were age, patient values, expected cardiovascular and pulmonary impairment after ICU discharge and frailty at admission. The BAIT system lets intensivists view expected decisions based on documented criteria and uses color-coding to show the magnitude of the effect and its direction (i.e. to continue or withdraw LST).
Conclusions
We developed a BAIT system that may support clinicians facing the dilemma of continuing or withdrawing LST by elucidating the key criteria involved in assessing medical futility. ...
Withholding and withdrawing life-sustaining therapy (LST) is common in European ICUs but significant variations exist. Behaviour artificial intelligence technology (BAIT) may help standardize the ethical dilemma to continue or withdraw LST for patients already admitted to the ICU.
Methods
Several sessions with intensivists of an academic medical centre and a large urban teaching hospital were held to determine the criteria influencing the process. A discrete choice experiment was conducted during which 25 hypothetical cases were presented to the participants. For each case the participants had to decide whether they would continue, continue with a time limited trial of one week, or withdraw LST. The results of the experiment were used to develop a multinomial logistic regression model that was incorporated in a web-based decision-support system.
Results
Thirty-six participants (intensivists and fellows in intensive care medicine) completed the experiment. The estimated model consisted of twelve covariates and showed good model fit (McFadden’s ρ2 0.25). The most important covariates were age, patient values, expected cardiovascular and pulmonary impairment after ICU discharge and frailty at admission. The BAIT system lets intensivists view expected decisions based on documented criteria and uses color-coding to show the magnitude of the effect and its direction (i.e. to continue or withdraw LST).
Conclusions
We developed a BAIT system that may support clinicians facing the dilemma of continuing or withdrawing LST by elucidating the key criteria involved in assessing medical futility. ...
Background
Withholding and withdrawing life-sustaining therapy (LST) is common in European ICUs but significant variations exist. Behaviour artificial intelligence technology (BAIT) may help standardize the ethical dilemma to continue or withdraw LST for patients already admitted to the ICU.
Methods
Several sessions with intensivists of an academic medical centre and a large urban teaching hospital were held to determine the criteria influencing the process. A discrete choice experiment was conducted during which 25 hypothetical cases were presented to the participants. For each case the participants had to decide whether they would continue, continue with a time limited trial of one week, or withdraw LST. The results of the experiment were used to develop a multinomial logistic regression model that was incorporated in a web-based decision-support system.
Results
Thirty-six participants (intensivists and fellows in intensive care medicine) completed the experiment. The estimated model consisted of twelve covariates and showed good model fit (McFadden’s ρ2 0.25). The most important covariates were age, patient values, expected cardiovascular and pulmonary impairment after ICU discharge and frailty at admission. The BAIT system lets intensivists view expected decisions based on documented criteria and uses color-coding to show the magnitude of the effect and its direction (i.e. to continue or withdraw LST).
Conclusions
We developed a BAIT system that may support clinicians facing the dilemma of continuing or withdrawing LST by elucidating the key criteria involved in assessing medical futility.
Withholding and withdrawing life-sustaining therapy (LST) is common in European ICUs but significant variations exist. Behaviour artificial intelligence technology (BAIT) may help standardize the ethical dilemma to continue or withdraw LST for patients already admitted to the ICU.
Methods
Several sessions with intensivists of an academic medical centre and a large urban teaching hospital were held to determine the criteria influencing the process. A discrete choice experiment was conducted during which 25 hypothetical cases were presented to the participants. For each case the participants had to decide whether they would continue, continue with a time limited trial of one week, or withdraw LST. The results of the experiment were used to develop a multinomial logistic regression model that was incorporated in a web-based decision-support system.
Results
Thirty-six participants (intensivists and fellows in intensive care medicine) completed the experiment. The estimated model consisted of twelve covariates and showed good model fit (McFadden’s ρ2 0.25). The most important covariates were age, patient values, expected cardiovascular and pulmonary impairment after ICU discharge and frailty at admission. The BAIT system lets intensivists view expected decisions based on documented criteria and uses color-coding to show the magnitude of the effect and its direction (i.e. to continue or withdraw LST).
Conclusions
We developed a BAIT system that may support clinicians facing the dilemma of continuing or withdrawing LST by elucidating the key criteria involved in assessing medical futility.
Bert van Wee, professor in Transport Policy at Delft University of Technology, the Netherlands, faculty Technology, Policy and Management, is retiring. Given his large contributions to EJTIR as Editor-in-Chief, editorial board member, author and reviewer, this Editorial Note is dedicated to his work for EJTIR. Over 25 years, van Wee published 18 papers and 1 book review in EJTIR covering a wide range of topics from road pricing to urban rail transport, vehicle automation and port throughput. What his studies have in common is that they explore how transport policies affect land-use and travel behaviour, as well as the economic and wider societal impacts of those policies. Bert van Wee’s generalist view on the transport system is rare, but, given the rising complexity of the system, increasingly needed to indeed be able to address the future challenges.
...
Bert van Wee, professor in Transport Policy at Delft University of Technology, the Netherlands, faculty Technology, Policy and Management, is retiring. Given his large contributions to EJTIR as Editor-in-Chief, editorial board member, author and reviewer, this Editorial Note is dedicated to his work for EJTIR. Over 25 years, van Wee published 18 papers and 1 book review in EJTIR covering a wide range of topics from road pricing to urban rail transport, vehicle automation and port throughput. What his studies have in common is that they explore how transport policies affect land-use and travel behaviour, as well as the economic and wider societal impacts of those policies. Bert van Wee’s generalist view on the transport system is rare, but, given the rising complexity of the system, increasingly needed to indeed be able to address the future challenges.
Integral system safety for machine learning in the public sector
An empirical account
Journal article
(2024)
-
J. Delfos, A. M.G. Zuiderwijk, S. van Cranenburgh, C. G. Chorus, R. I.J. Dobbe
This paper introduces systems theory and system safety concepts to ongoing academic debates about the safety of Machine Learning (ML) systems in the public sector. In particular, we analyze the risk factors of ML systems and their respective institutional context, which impact the ability to control such systems. We use interview data to abductively show what risk factors of such systems are present in public professionals' perceptions and what factors are expected based on systems theory but are missing. Based on the hypothesis that ML systems are best addressed with a systems theory lens, we argue that the missing factors deserve greater attention in ongoing efforts to address ML systems safety. These factors include the explication of safety goals and constraints, the inclusion of systemic factors in system design, the development of safety control structures, and the tendency of ML systems to migrate towards higher risk. Our observations support the hypothesis that ML systems can be best regarded through a systems theory lens. Therefore, we conclude that system safety concepts can be useful aids for policymakers who aim to improve ML system safety.
...
This paper introduces systems theory and system safety concepts to ongoing academic debates about the safety of Machine Learning (ML) systems in the public sector. In particular, we analyze the risk factors of ML systems and their respective institutional context, which impact the ability to control such systems. We use interview data to abductively show what risk factors of such systems are present in public professionals' perceptions and what factors are expected based on systems theory but are missing. Based on the hypothesis that ML systems are best addressed with a systems theory lens, we argue that the missing factors deserve greater attention in ongoing efforts to address ML systems safety. These factors include the explication of safety goals and constraints, the inclusion of systemic factors in system design, the development of safety control structures, and the tendency of ML systems to migrate towards higher risk. Our observations support the hypothesis that ML systems can be best regarded through a systems theory lens. Therefore, we conclude that system safety concepts can be useful aids for policymakers who aim to improve ML system safety.
Gender violence encompasses a multitude of morally problematic psychological, physical, and sexual behaviors that, in most countries, constitute criminal offenses. In this study, we investigate the association between moral foundations (Care, Fairness, Loyalty, Authority, and Sanctity) and punitive responses to gender violence offenses. Our case study focuses on gender violence in Portugal, a country in which these offenses are a prevalent social problem. We collected data on gender violence legal cases decided in Portuguese courts between 2002 and 2022, and we used a latent class cluster analysis model to identify the complex patterns in the data and reduce such patterns to a distinct number of clusters. Four main clusters unravel latent relations between the foundations mapped in the legal narratives and corresponding punitive responses: (i) Affirmative with suspended prison time (moral rhetoric rooted in Authority); (ii) Mixed outcomes but no prison time (moral rhetoric rooted in Sanctity); (iii) Affirmative with lengthy prison time large compensation (moral rhetoric rooted in Loyalty and Care); and (iv) Affirmative with court fines (moral rhetoric rooted in Fairness). The moral foundations provide a valuable lens to understand the problem of gender violence, but further research is needed to establish the causal mechanisms between morality and punitive responses to gender violence.
...
Gender violence encompasses a multitude of morally problematic psychological, physical, and sexual behaviors that, in most countries, constitute criminal offenses. In this study, we investigate the association between moral foundations (Care, Fairness, Loyalty, Authority, and Sanctity) and punitive responses to gender violence offenses. Our case study focuses on gender violence in Portugal, a country in which these offenses are a prevalent social problem. We collected data on gender violence legal cases decided in Portuguese courts between 2002 and 2022, and we used a latent class cluster analysis model to identify the complex patterns in the data and reduce such patterns to a distinct number of clusters. Four main clusters unravel latent relations between the foundations mapped in the legal narratives and corresponding punitive responses: (i) Affirmative with suspended prison time (moral rhetoric rooted in Authority); (ii) Mixed outcomes but no prison time (moral rhetoric rooted in Sanctity); (iii) Affirmative with lengthy prison time large compensation (moral rhetoric rooted in Loyalty and Care); and (iv) Affirmative with court fines (moral rhetoric rooted in Fairness). The moral foundations provide a valuable lens to understand the problem of gender violence, but further research is needed to establish the causal mechanisms between morality and punitive responses to gender violence.
Journal article
(2023)
-
Mathijs de Haas, Maarten Kroesen, Caspar Chorus, Sascha Hoogendoorn-Lanser, Serge Hoogendoorn
In mobility panels, respondents may use a strategy of soft-refusal to lower their response burden, e.g. by claiming they did not leave their house even though they actually did. Soft-refusal leads to poor data quality and may complicate research, e.g. focused on people with actual low mobility. In this study we develop three methods to detect the presence of soft-refusal in mobility panels, based on respectively (observed and predicted) out-of-home activity, straightlining and speeding. For each indicator, we explore the relation with reported immobility and panel attrition. The results show that speeding and straightlining in a questionnaire is strongly related to reported immobility in a (self-reported) travel diary. Using a binary logit model, respondents who are predicted to leave their home but report no trips are identified as possible soft refusers. To reveal different patterns of soft-refusal and assess how these patterns influence the probability to drop out of the panel, a latent transition model is estimated. The results show four behavioral patterns with respect to soft-refusal ranging from a large class of reliable respondents who score positive on all three soft-refusal indicators, to a small ‘high-risk’ class of respondents who score poorly on all indicators. This ‘high-risk’ group also reports the highest immobility and has the highest attrition rate. The model also shows that respondents who do not drop out of the panel, tend to stay in the same behavioral pattern over time. The amount of soft-refusal expressed by a respondent therefore seems to be a stable behavioral trait.
...
In mobility panels, respondents may use a strategy of soft-refusal to lower their response burden, e.g. by claiming they did not leave their house even though they actually did. Soft-refusal leads to poor data quality and may complicate research, e.g. focused on people with actual low mobility. In this study we develop three methods to detect the presence of soft-refusal in mobility panels, based on respectively (observed and predicted) out-of-home activity, straightlining and speeding. For each indicator, we explore the relation with reported immobility and panel attrition. The results show that speeding and straightlining in a questionnaire is strongly related to reported immobility in a (self-reported) travel diary. Using a binary logit model, respondents who are predicted to leave their home but report no trips are identified as possible soft refusers. To reveal different patterns of soft-refusal and assess how these patterns influence the probability to drop out of the panel, a latent transition model is estimated. The results show four behavioral patterns with respect to soft-refusal ranging from a large class of reliable respondents who score positive on all three soft-refusal indicators, to a small ‘high-risk’ class of respondents who score poorly on all indicators. This ‘high-risk’ group also reports the highest immobility and has the highest attrition rate. The model also shows that respondents who do not drop out of the panel, tend to stay in the same behavioral pattern over time. The amount of soft-refusal expressed by a respondent therefore seems to be a stable behavioral trait.
Surgery or comfort care for neonates with surgical necrotizing enterocolitis
Lessons learned from behavioral artificial intelligence technology
Journal article
(2023)
-
Otis C. van Varsseveld, Annebel ten Broeke, Caspar G. Chorus, Nicolaas Heyning, Elisabeth M.W. Kooi, Jan B.F. Hulscher
Background: Critical decision making in surgical necrotizing enterocolitis (NEC) is highly complex and hard to capture in decision rules due to case-specificity and high mortality risk. In this choice experiment, we aimed to identify the implicit weight of decision factors towards future decision support, and to assess potential differences between specialties or centers. Methods: Thirty-five hypothetical surgical NEC scenarios with different factor levels were evaluated by neonatal care experts of all Dutch neonatal care centers in an online environment, where a recommendation for surgery or comfort care was requested. We conducted choice analysis by constructing a binary logistic regression model according to behavioral artificial intelligence technology (BAIT). Results: Out of 109 invited neonatal care experts, 62 (57%) participated, including 45 neonatologists, 16 pediatric surgeons and one neonatology physician assistant. Cerebral ultrasound (Relative importance = 20%, OR = 4.06, 95% CI = 3.39–4.86) was the most important factor in the decision surgery versus comfort care in surgical NEC, nationwide and for all specialties and centers. Pediatric surgeons more often recommended surgery compared to neonatologists (62% vs. 57%, p = 0.03). For all centers, cerebral ultrasound, congenital comorbidity, hemodynamics and parental preferences were significant decision factors (p < 0.05). Sex (p = 0.14), growth since birth (p = 0.25), and estimated parental capacities (p = 0.06) had no significance in nationwide nor subgroup analyses. Conclusion: We demonstrated how BAIT can analyze the implicit weight of factors in the complex and critical decision for surgery or comfort care for (surgical) NEC. The findings reflect Dutch expertise, but the technique can be expanded internationally. After validation, our choice model/BAIT may function as decision aid.
...
Background: Critical decision making in surgical necrotizing enterocolitis (NEC) is highly complex and hard to capture in decision rules due to case-specificity and high mortality risk. In this choice experiment, we aimed to identify the implicit weight of decision factors towards future decision support, and to assess potential differences between specialties or centers. Methods: Thirty-five hypothetical surgical NEC scenarios with different factor levels were evaluated by neonatal care experts of all Dutch neonatal care centers in an online environment, where a recommendation for surgery or comfort care was requested. We conducted choice analysis by constructing a binary logistic regression model according to behavioral artificial intelligence technology (BAIT). Results: Out of 109 invited neonatal care experts, 62 (57%) participated, including 45 neonatologists, 16 pediatric surgeons and one neonatology physician assistant. Cerebral ultrasound (Relative importance = 20%, OR = 4.06, 95% CI = 3.39–4.86) was the most important factor in the decision surgery versus comfort care in surgical NEC, nationwide and for all specialties and centers. Pediatric surgeons more often recommended surgery compared to neonatologists (62% vs. 57%, p = 0.03). For all centers, cerebral ultrasound, congenital comorbidity, hemodynamics and parental preferences were significant decision factors (p < 0.05). Sex (p = 0.14), growth since birth (p = 0.25), and estimated parental capacities (p = 0.06) had no significance in nationwide nor subgroup analyses. Conclusion: We demonstrated how BAIT can analyze the implicit weight of factors in the complex and critical decision for surgery or comfort care for (surgical) NEC. The findings reflect Dutch expertise, but the technique can be expanded internationally. After validation, our choice model/BAIT may function as decision aid.
Moral rhetoric in discrete choice models
A Natural Language Processing approach
This paper proposes a new method to combine choice- and text data to infer moral motivations from people’s actions. To do this, we rely on moral rhetoric, in other words, extracting moral values from verbal expressions with Natural Language Processing techniques. We use moral rhetoric based on a well-established moral, psychological theory called Moral Foundations Theory. We use moral rhetoric as input in Discrete Choice Models to gain insights into moral behaviour based on people’s words and actions. We test our method in a case study of voting and party defection in the European Parliament. Our results indicate that moral rhetoric have significant explanatory power in modelling voting behaviour. We interpret the results in the light of political science literature and propose ways for future investigations.
...
This paper proposes a new method to combine choice- and text data to infer moral motivations from people’s actions. To do this, we rely on moral rhetoric, in other words, extracting moral values from verbal expressions with Natural Language Processing techniques. We use moral rhetoric based on a well-established moral, psychological theory called Moral Foundations Theory. We use moral rhetoric as input in Discrete Choice Models to gain insights into moral behaviour based on people’s words and actions. We test our method in a case study of voting and party defection in the European Parliament. Our results indicate that moral rhetoric have significant explanatory power in modelling voting behaviour. We interpret the results in the light of political science literature and propose ways for future investigations.
Give and take
Moral aspects of travelers' intentions to participate in a hypothetical established social routing scheme
Journal article
(2023)
-
Teodora Szep, Tom van den Berg, Nicolas Cointe, Aemiro Melkamu Daniel, Andreia Martinho, Tanzhe Tang, Caspar Chorus
Social routing schemes are widely regarded as promising tools to reduce traffic congestion in urban networks. We contribute to the growing literature on such schemes and their effect on travel behavior, by exploring the interaction between the characteristics and framing of the scheme on the one hand, and travelers' moral personality and moral motivations on the other hand. Our method uses a two-wave stated intention experiment eliciting preferences in a hypothetical context where a social routing scheme is presumed to have been established already. This is followed by a morality survey. We hypothesize and then confirm the following: when a social routing scheme is framed and designed as an altruistic effort requesting personal sacrifices for the benefit of other travelers, people who strongly adhere to care related notions of morality are attracted to such a scheme. On the contrary, a scheme that is designed and framed as a collective endeavour which would also benefit participating travelers attracts those who strongly adhere to moral notions related to fairness. We derive tentative policy recommendations from our findings, suggesting that a collective good scheme, albeit more difficult to implement, is likely to be more viable in the long run.
...
Social routing schemes are widely regarded as promising tools to reduce traffic congestion in urban networks. We contribute to the growing literature on such schemes and their effect on travel behavior, by exploring the interaction between the characteristics and framing of the scheme on the one hand, and travelers' moral personality and moral motivations on the other hand. Our method uses a two-wave stated intention experiment eliciting preferences in a hypothetical context where a social routing scheme is presumed to have been established already. This is followed by a morality survey. We hypothesize and then confirm the following: when a social routing scheme is framed and designed as an altruistic effort requesting personal sacrifices for the benefit of other travelers, people who strongly adhere to care related notions of morality are attracted to such a scheme. On the contrary, a scheme that is designed and framed as a collective endeavour which would also benefit participating travelers attracts those who strongly adhere to moral notions related to fairness. We derive tentative policy recommendations from our findings, suggesting that a collective good scheme, albeit more difficult to implement, is likely to be more viable in the long run.
The long road to automated trucking
Insights from driver focus groups
Work towards making automated driving systems a reality is well underway. In this study, we look at what is likely to be one of the first widespread implementations of a form of automated driving on public roads, i.e., truck platooning, where virtually connected trucks drive at short headways to save fuel and associated emissions. With progressing technology, we may see platoons with drivers resting while being in the truck, or even platoons in which not all trucks require drivers. Hence, platooning technology has a significant impact on the jobs of truck drivers especially when driver involvement reduces. Driver acceptance of this emerging technology is therefore an important factor in the implementation of platooning and, consequently, automated driving in general. In this study, we explore the range of perspectives that exist among drivers by conducting focus groups in The Netherlands. These discussions indicate that drivers foresee that platooning will eventually become a reality but believe it will have a negative impact on the quality of their work and their job satisfaction.
...
Work towards making automated driving systems a reality is well underway. In this study, we look at what is likely to be one of the first widespread implementations of a form of automated driving on public roads, i.e., truck platooning, where virtually connected trucks drive at short headways to save fuel and associated emissions. With progressing technology, we may see platoons with drivers resting while being in the truck, or even platoons in which not all trucks require drivers. Hence, platooning technology has a significant impact on the jobs of truck drivers especially when driver involvement reduces. Driver acceptance of this emerging technology is therefore an important factor in the implementation of platooning and, consequently, automated driving in general. In this study, we explore the range of perspectives that exist among drivers by conducting focus groups in The Netherlands. These discussions indicate that drivers foresee that platooning will eventually become a reality but believe it will have a negative impact on the quality of their work and their job satisfaction.
Efficiently allocating scarce healthcare resources requires nuanced understanding of individual and collective interests as well as relative concerns, which may overlap or conflict. This paper is the first to empirically investigate whether and to what extent self-interest (SI), positional concerns (PC) and distributional considerations (DC) simultaneously explain individual decision making related to access to healthcare services. Our investigation is based on a stated choice experiment conducted in two countries with different healthcare systems, the United States (US) and the United Kingdom (UK). The choice experiment is on allocation of medical treatment waiting times for a hypothetical disease. We carry out the investigation under two different perspectives: (i) in a socially inclusive personal perspective decision makers were asked to choose between waiting time distributions for themselves and (ii) in a social perspective decision makers were asked to make similar choices for a close relative or friend of opposite gender. The results obtained by estimating a variety of advanced choice models indicate that DC, SI and PC, in this order of importance, are significant drivers of choice behaviour in our empirical context. These findings are consistent regardless of the choice perspective and the country where decision makers live. Comparing the results from different choice perspectives, we find that US respondents who chose for their close relative or friend attach significantly larger weight to their close relative’s or friend’s waiting times as well as to the overall distribution of waiting times than US respondents who chose for themselves. Looking at differences between countries, our results show that UK respondents who made choices for themselves placed significantly larger weight on SI and DC than US respondents, while US respondents, in turn, displayed relatively stronger but not significantly different positional concerns than UK respondents. In addition, we observe that UK respondents who chose for their close relative or friend put a larger weight on DC than their US counterparts. We conclude that the methodological (data collection and analysis) approach allows for disentangling the relative importance of the three motivations and discusses the potential implications of these findings for healthcare decision making.
...
Efficiently allocating scarce healthcare resources requires nuanced understanding of individual and collective interests as well as relative concerns, which may overlap or conflict. This paper is the first to empirically investigate whether and to what extent self-interest (SI), positional concerns (PC) and distributional considerations (DC) simultaneously explain individual decision making related to access to healthcare services. Our investigation is based on a stated choice experiment conducted in two countries with different healthcare systems, the United States (US) and the United Kingdom (UK). The choice experiment is on allocation of medical treatment waiting times for a hypothetical disease. We carry out the investigation under two different perspectives: (i) in a socially inclusive personal perspective decision makers were asked to choose between waiting time distributions for themselves and (ii) in a social perspective decision makers were asked to make similar choices for a close relative or friend of opposite gender. The results obtained by estimating a variety of advanced choice models indicate that DC, SI and PC, in this order of importance, are significant drivers of choice behaviour in our empirical context. These findings are consistent regardless of the choice perspective and the country where decision makers live. Comparing the results from different choice perspectives, we find that US respondents who chose for their close relative or friend attach significantly larger weight to their close relative’s or friend’s waiting times as well as to the overall distribution of waiting times than US respondents who chose for themselves. Looking at differences between countries, our results show that UK respondents who made choices for themselves placed significantly larger weight on SI and DC than US respondents, while US respondents, in turn, displayed relatively stronger but not significantly different positional concerns than UK respondents. In addition, we observe that UK respondents who chose for their close relative or friend put a larger weight on DC than their US counterparts. We conclude that the methodological (data collection and analysis) approach allows for disentangling the relative importance of the three motivations and discusses the potential implications of these findings for healthcare decision making.
Towards machine learning for moral choice analysis in health economics
A literature review and research agenda
Review
(2023)
-
Nicholas V.R. Smeele, Caspar G. Chorus, Maartje H.N. Schermer, Esther W. de Bekker-Grob
Background: Discrete choice models (DCMs) for moral choice analysis will likely lead to erroneous model outcomes and misguided policy recommendations, as only some characteristics of moral decision-making are considered. Machine learning (ML) is recently gaining interest in the field of discrete choice modelling. This paper explores the potential of combining DCMs and ML to study moral decision-making more accurately and better inform policy decisions in healthcare. Methods: An interdisciplinary literature search across four databases – PubMed, Scopus, Web of Science, and Arxiv – was conducted to gather papers. Based on the Preferred Reporting Items for Systematic and Meta-analyses (PRISMA) guideline, studies were screened for eligibility on inclusion criteria and extracted attributes from eligible papers. Of the 6285 articles, we included 277 studies. Results: DCMs have shortcomings in studying moral decision-making. Whilst the DCMs' mathematical elegance and behavioural appeal hold clear interpretations, the models do not account for the ‘moral’ cost and benefit in an individual's utility calculation. The literature showed that ML obtains higher predictive power, model flexibility, and ability to handle large and unstructured datasets. Combining the strengths of ML methods with DCMs has the potential for studying moral decision-making. Conclusions: By providing a research agenda, this paper highlights that ML has clear potential to i) find and deepen the utility specification of DCMs, and ii) enrich the insights extracted from DCMs by considering the intrapersonal determinants of moral decision-making.
...
Background: Discrete choice models (DCMs) for moral choice analysis will likely lead to erroneous model outcomes and misguided policy recommendations, as only some characteristics of moral decision-making are considered. Machine learning (ML) is recently gaining interest in the field of discrete choice modelling. This paper explores the potential of combining DCMs and ML to study moral decision-making more accurately and better inform policy decisions in healthcare. Methods: An interdisciplinary literature search across four databases – PubMed, Scopus, Web of Science, and Arxiv – was conducted to gather papers. Based on the Preferred Reporting Items for Systematic and Meta-analyses (PRISMA) guideline, studies were screened for eligibility on inclusion criteria and extracted attributes from eligible papers. Of the 6285 articles, we included 277 studies. Results: DCMs have shortcomings in studying moral decision-making. Whilst the DCMs' mathematical elegance and behavioural appeal hold clear interpretations, the models do not account for the ‘moral’ cost and benefit in an individual's utility calculation. The literature showed that ML obtains higher predictive power, model flexibility, and ability to handle large and unstructured datasets. Combining the strengths of ML methods with DCMs has the potential for studying moral decision-making. Conclusions: By providing a research agenda, this paper highlights that ML has clear potential to i) find and deepen the utility specification of DCMs, and ii) enrich the insights extracted from DCMs by considering the intrapersonal determinants of moral decision-making.
Data-driven assisted model specification for complex choice experiments data
Association rules learning and random forests for Participatory Value Evaluation experiments
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.
...
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.
Empirical studies on individual behaviour often, implicitly or explicitly, assume a single type of decision rule. Other studies do not specify behavioural assumptions at all. We advance sociological research by introducing (random) regret minimization, which is related to loss aversion, into the sociological literature and by testing it against (random) utility maximization, which is the most prominent decision rule in sociological research on individual behaviour. With an application to neighbourhood choice, in a sample of four European cities, we combine stated choice experiment data and discrete choice modelling techniques and find a considerable degree of decision rule-heterogeneity, with a strong prevalence of regret minimization and hence loss aversion. We also provide indicative evidence that decision rules can affect expected neighbourhood demand at the macro level. Our approach allows identifying heterogeneity in decision rules, that is, the degree of regret/loss aversion, at the level of choice attributes such as the share of foreigners when comparing neighbourhoods, and can improve sociological practice related to linking theories and social research on decision-making.
...
Empirical studies on individual behaviour often, implicitly or explicitly, assume a single type of decision rule. Other studies do not specify behavioural assumptions at all. We advance sociological research by introducing (random) regret minimization, which is related to loss aversion, into the sociological literature and by testing it against (random) utility maximization, which is the most prominent decision rule in sociological research on individual behaviour. With an application to neighbourhood choice, in a sample of four European cities, we combine stated choice experiment data and discrete choice modelling techniques and find a considerable degree of decision rule-heterogeneity, with a strong prevalence of regret minimization and hence loss aversion. We also provide indicative evidence that decision rules can affect expected neighbourhood demand at the macro level. Our approach allows identifying heterogeneity in decision rules, that is, the degree of regret/loss aversion, at the level of choice attributes such as the share of foreigners when comparing neighbourhoods, and can improve sociological practice related to linking theories and social research on decision-making.
Perceived challenges and opportunities of machine learning applications in governmental organisations
An interview-based exploration in the Netherlands
As the application of machine learning (ML) algorithms becomes more widespread, governmental organisations try to benefit from this technology. While ML has the potential to support public services, its application also introduces challenges. Several scholars have described the possible opportunities and challenges of ML applications in the public sector conceptually. However, it is not yet investigated if and how these concepts materialise and are perceived by end-users in the public sector when ML is applied. Therefore, it is neither clear whether these concepts are valid, nor what regulation could be introduced to address them effectively. This empirical study's objective is to shed light on how challenges and opportunities of governmental use of ML algorithms are perceived by Dutch professionals in the public sector. We attain our objective by conducting interviews with twelve professionals from Dutch executive and supervisory organisations in the public sector that respectively use ML and supervise the use of ML. Results show that ML is used primarily for improvements in the accuracy and speed of public task execution. Furthermore, interviewed professionals experience several barriers for ML implementation as well as risks following from the use of ML. The implications of these findings for practice are discussed, as well as opportunities for further research.
...
As the application of machine learning (ML) algorithms becomes more widespread, governmental organisations try to benefit from this technology. While ML has the potential to support public services, its application also introduces challenges. Several scholars have described the possible opportunities and challenges of ML applications in the public sector conceptually. However, it is not yet investigated if and how these concepts materialise and are perceived by end-users in the public sector when ML is applied. Therefore, it is neither clear whether these concepts are valid, nor what regulation could be introduced to address them effectively. This empirical study's objective is to shed light on how challenges and opportunities of governmental use of ML algorithms are perceived by Dutch professionals in the public sector. We attain our objective by conducting interviews with twelve professionals from Dutch executive and supervisory organisations in the public sector that respectively use ML and supervise the use of ML. Results show that ML is used primarily for improvements in the accuracy and speed of public task execution. Furthermore, interviewed professionals experience several barriers for ML implementation as well as risks following from the use of ML. The implications of these findings for practice are discussed, as well as opportunities for further research.
Objectives: Research efforts evaluating the role of altruistic motivations behind health policy support are usually based on direct preference elicitation procedures, which may be biased. We propose an indirect measurement approach to approximate self-protection–related and altruistic motivations underlying preferences for public health policies. Methods: Our new approach relies on associations between on the one hand decision makers’ perceived health risk for themselves and for close relatives and on the other hand their observed preferences for health policies that reduce such risks. The approach allows to make a rough distinction between health-related self-protection and local altruistic motives behind preferences for health policies. We illustrate our approach using data obtained from a discrete choice experiment in the context of policies to relax coronavirus-related lockdown measures in The Netherlands. Results: Our results show that the approach is able to uncover that (1) people who think they have a high chance of experiencing health risks from a COVID-19 infection are more willing to accept a societal or personal sacrifice, (2) people with a higher health risk perception for their relatives have a higher willingness to accept sacrifices than people with a higher health risk perception for themselves, and (3) people who perceive that they have a high risk of dying of COVID-19 have a higher willingness to accept sacrifices than those anticipating less severe consequences of COVID-19. Conclusions: Our method offers a useful proxy metric to distinguish health-related self-protection and local altruism as drivers of citizens’ responses to healthcare policies.
...
Objectives: Research efforts evaluating the role of altruistic motivations behind health policy support are usually based on direct preference elicitation procedures, which may be biased. We propose an indirect measurement approach to approximate self-protection–related and altruistic motivations underlying preferences for public health policies. Methods: Our new approach relies on associations between on the one hand decision makers’ perceived health risk for themselves and for close relatives and on the other hand their observed preferences for health policies that reduce such risks. The approach allows to make a rough distinction between health-related self-protection and local altruistic motives behind preferences for health policies. We illustrate our approach using data obtained from a discrete choice experiment in the context of policies to relax coronavirus-related lockdown measures in The Netherlands. Results: Our results show that the approach is able to uncover that (1) people who think they have a high chance of experiencing health risks from a COVID-19 infection are more willing to accept a societal or personal sacrifice, (2) people with a higher health risk perception for their relatives have a higher willingness to accept sacrifices than people with a higher health risk perception for themselves, and (3) people who perceive that they have a high risk of dying of COVID-19 have a higher willingness to accept sacrifices than those anticipating less severe consequences of COVID-19. Conclusions: Our method offers a useful proxy metric to distinguish health-related self-protection and local altruism as drivers of citizens’ responses to healthcare policies.
Decision Field Theory
Equivalence with probit models and guidance for identifiability
We examine identifiability and distinguishability in Decision Field Theory (DFT) models and highlight pitfalls and how to avoid them. In the past literature, the models’ parameters have been put forward as being able to capture the psychological processes in a decision maker's mind during deliberation. DFT models have been widely used to analyse human decision making behaviour, and many empirical applications in the choice modelling domain rely solely on data concerning the observed final choice. This raises the question if such data are rich enough to allow for the identification of the model's parameters. Insight into identifiability and distinguishability is crucial as it allows the researcher to determine which behavioural and psychological conclusions can or cannot be drawn from the estimated DFT model and how a DFT model can be specified in such a way that resulting parameters have meaningful interpretations. In this paper, we address this issue. To do this, we first show which specifications of DFT are equivalent to conventional probit models. Then, building on this equivalence result, we apply established analytical methods to highlight and explain the identification and distinguishability issues that arise when estimating DFT models on conventional choice data. We find evidence that some of the DFT models’ special cases suffer from identifiability issues. Our results warrant caution when DFT models are used to infer psychological processes and human behaviour from conventional choice data, and they help researchers choose the correct specification of DFT models.
...
We examine identifiability and distinguishability in Decision Field Theory (DFT) models and highlight pitfalls and how to avoid them. In the past literature, the models’ parameters have been put forward as being able to capture the psychological processes in a decision maker's mind during deliberation. DFT models have been widely used to analyse human decision making behaviour, and many empirical applications in the choice modelling domain rely solely on data concerning the observed final choice. This raises the question if such data are rich enough to allow for the identification of the model's parameters. Insight into identifiability and distinguishability is crucial as it allows the researcher to determine which behavioural and psychological conclusions can or cannot be drawn from the estimated DFT model and how a DFT model can be specified in such a way that resulting parameters have meaningful interpretations. In this paper, we address this issue. To do this, we first show which specifications of DFT are equivalent to conventional probit models. Then, building on this equivalence result, we apply established analytical methods to highlight and explain the identification and distinguishability issues that arise when estimating DFT models on conventional choice data. We find evidence that some of the DFT models’ special cases suffer from identifiability issues. Our results warrant caution when DFT models are used to infer psychological processes and human behaviour from conventional choice data, and they help researchers choose the correct specification of DFT models.
Within moral psychology, theories focusing on the conceptualization and empirical measurement of people’s morality in terms of general moral values –such as Moral Foundation Theory- (implicitly) assume general moral values to be relevant concepts for the explanation and prediction of behavior in everyday life. However, a solid theoretical and empirical foundation for this idea remains work in progress. In this study we explore this relationship between general moral values and daily life behavior through a conceptual analysis and an empirical study. Our conceptual analysis of the moral value-moral behavior relationship suggests that the effect of a generally endorsed moral value on moral behavior is highly context dependent. It requires the manifestation of several phases of moral decision-making, each influenced by many contextual factors. We expect that this renders the empirical relationship between generic moral values and people’s concrete moral behavior indeterminate. Subsequently, we empirically investigate this relationship in three different studies. We relate two different measures of general moral values -the Moral Foundation Questionnaire and the Morality As Cooperation Questionnaire- to a broad set of self-reported morally relevant daily life behaviors (including adherence to COVID-19 measures and participation in voluntary work). Our empirical results are in line with the expectations derived from our conceptual analysis: the considered general moral values are poor predictors of the selected daily life behaviors. Furthermore, moral values that were tailored to the specific context of the behavior showed to be somewhat stronger predictors. Together with the insights derived from our conceptual analysis, this indicates the relevance of the contextual nature of moral decision-making as a possible explanation for the poor predictive value of general moral values. Our findings suggest that the investigation of morality’s influence on behavior by expressing and measuring it in terms of general moral values may need revision.
...
Within moral psychology, theories focusing on the conceptualization and empirical measurement of people’s morality in terms of general moral values –such as Moral Foundation Theory- (implicitly) assume general moral values to be relevant concepts for the explanation and prediction of behavior in everyday life. However, a solid theoretical and empirical foundation for this idea remains work in progress. In this study we explore this relationship between general moral values and daily life behavior through a conceptual analysis and an empirical study. Our conceptual analysis of the moral value-moral behavior relationship suggests that the effect of a generally endorsed moral value on moral behavior is highly context dependent. It requires the manifestation of several phases of moral decision-making, each influenced by many contextual factors. We expect that this renders the empirical relationship between generic moral values and people’s concrete moral behavior indeterminate. Subsequently, we empirically investigate this relationship in three different studies. We relate two different measures of general moral values -the Moral Foundation Questionnaire and the Morality As Cooperation Questionnaire- to a broad set of self-reported morally relevant daily life behaviors (including adherence to COVID-19 measures and participation in voluntary work). Our empirical results are in line with the expectations derived from our conceptual analysis: the considered general moral values are poor predictors of the selected daily life behaviors. Furthermore, moral values that were tailored to the specific context of the behavior showed to be somewhat stronger predictors. Together with the insights derived from our conceptual analysis, this indicates the relevance of the contextual nature of moral decision-making as a possible explanation for the poor predictive value of general moral values. Our findings suggest that the investigation of morality’s influence on behavior by expressing and measuring it in terms of general moral values may need revision.
Economic theory is built on the assumption that people are omniscient utility maximizers. In reality, this is unlikely to be true and often people lack information about all alternatives that are available to them; either because the information is unavailable or that the cost of searching for and evaluating that information is high. In this paper, we develop a simple and tractable model that captures satisficing behavior. We show that the model can retrieve consistent parameters under a large range of experimental conditions. We test our model on synthetic data and present an empirical application. We discuss the implications of our results for the use of satisficing choice models in explaining choice.
...
Economic theory is built on the assumption that people are omniscient utility maximizers. In reality, this is unlikely to be true and often people lack information about all alternatives that are available to them; either because the information is unavailable or that the cost of searching for and evaluating that information is high. In this paper, we develop a simple and tractable model that captures satisficing behavior. We show that the model can retrieve consistent parameters under a large range of experimental conditions. We test our model on synthetic data and present an empirical application. We discuss the implications of our results for the use of satisficing choice models in explaining choice.
E-bike user groups and substitution effects
Evidence from longitudinal travel data in the Netherlands
Journal article
(2021)
-
Mathijs de Haas, Maarten Kroesen, Caspar Chorus, Sascha Hoogendoorn-Lanser, Serge Hoogendoorn
In recent years, the e-bike has become increasingly popular in many European countries. With higher speeds and less effort needed, the e-bike is a promising mode of transport to many, and it is considered a good alternative for certain car trips by policy-makers and planners. A major limitation of many studies that investigate such substitution effects of the e-bike, is their reliance on cross-sectional data which do not allow an assessment of within-person travel mode changes. As a consequence, there is currently no consensus about the e-bike’s potential to replace car trips. Furthermore, there has been little research focusing on heterogeneity among e-bike users. In this respect, it is likely that different groups exist that use the e-bike for different reasons (e.g. leisure vs commute travel), something which will also influence possible substitution patterns. This paper contributes to the literature in two ways: (1) it presents a statistical analysis to assess the extent to which e-bike trips are substituting trips by other travel modes based on longitudinal data; (2) it reveals different user groups among the e-bike population. A Random Intercept Cross-Lagged Panel Model is estimated using five waves of data from the Netherlands Mobility Panel. Furthermore, a Latent Class Analysis is performed using data from the Dutch national travel survey. Results show that, when using longitudinal data, the substitution effects between e-bike and the competing travel modes of car and public transport are not as significant as reported in earlier research. In general, e-bike trips only significantly reduce conventional bicycle trips in the Netherlands, which can be regarded an unwanted effect from a policy-viewpoint. For commuting, the e-bike also substitutes car trips. Furthermore, results show that there are five different user groups with their own distinct behaviour patterns and socio-demographic characteristics. They also show that groups that use the e-bike primarily for commuting or education are growing at a much higher rate than groups that mainly use the e-bike for leisure and shopping purposes.
...
In recent years, the e-bike has become increasingly popular in many European countries. With higher speeds and less effort needed, the e-bike is a promising mode of transport to many, and it is considered a good alternative for certain car trips by policy-makers and planners. A major limitation of many studies that investigate such substitution effects of the e-bike, is their reliance on cross-sectional data which do not allow an assessment of within-person travel mode changes. As a consequence, there is currently no consensus about the e-bike’s potential to replace car trips. Furthermore, there has been little research focusing on heterogeneity among e-bike users. In this respect, it is likely that different groups exist that use the e-bike for different reasons (e.g. leisure vs commute travel), something which will also influence possible substitution patterns. This paper contributes to the literature in two ways: (1) it presents a statistical analysis to assess the extent to which e-bike trips are substituting trips by other travel modes based on longitudinal data; (2) it reveals different user groups among the e-bike population. A Random Intercept Cross-Lagged Panel Model is estimated using five waves of data from the Netherlands Mobility Panel. Furthermore, a Latent Class Analysis is performed using data from the Dutch national travel survey. Results show that, when using longitudinal data, the substitution effects between e-bike and the competing travel modes of car and public transport are not as significant as reported in earlier research. In general, e-bike trips only significantly reduce conventional bicycle trips in the Netherlands, which can be regarded an unwanted effect from a policy-viewpoint. For commuting, the e-bike also substitutes car trips. Furthermore, results show that there are five different user groups with their own distinct behaviour patterns and socio-demographic characteristics. They also show that groups that use the e-bike primarily for commuting or education are growing at a much higher rate than groups that mainly use the e-bike for leisure and shopping purposes.