Best-worst Scaling Experiment Research Articles

Assessing good governance principles of renewable energy megaprojects

Australia has a net zero emission target by 2050. The transition from fossil fuel to renewable energy sources is critical to meeting this target and the construction of large-scale renewable energy projects is booming. These developments bring economic benefits, but their size means they can also have environmental and social consequences. Rural landscapes are particularly vulnerable because they are expected to be at the forefront of new large-scale renewable energy developments. Such projects are complex and, although they might be socially acceptable, need to be managed and governed well if they are to proceed smoothly. Good governance incorporates fairness, trust and transparency, aspects that foster public acceptance for renewable energy projects. While social acceptance has attracted substantial research, little is known about the public's views on the governance of renewable energy megaprojects. In Australia, a continent with abundant solar radiation and space, large projects are novel and our study aimed to assess which good governance principles the public considers to be most important if megaprojects are to be approved for construction in remote Australia. This insight can complement state and investors views on governance. We carried out an Australia-wide online survey that included a best-worst scaling (BWS) experiment and received 2223 valid responses, using an ambitious 12,000-ha solar megafarm planned for remote northern Australia as a case study. The most important governance principal for respondents was taking responsibility for environmental risk, followed by benefits such as cheap energy and jobs for the local community - governance principles referred to as responsiveness and participation. Accountability if something goes wrong was also perceived as important, but mainly by those respondents who lived in the jurisdiction that is the proposed host of the solar megafarm, and less so by people living in other places of Australia. Perhaps surprisingly, the governance principles of transparency, fairness and the rule of law were considered to be less important. Adhering to national regulations was also affected by the location of respondents, with those living closer to the proposed megafarm being least concerned about adherence.

Prioritization of ethical concerns regarding HIV molecular epidemiology by public health practitioners and researchers

BackgroundHIV molecular epidemiology (HIV ME) can support the early detection of emerging clusters of new HIV infections by combining HIV sequence data routinely obtained during the clinical treatment of people living with HIV with behavioral, geographic, and sociodemographic information. While information about emerging clusters promises to facilitate HIV prevention and treatment efforts, the use of this data also raises several ethical concerns. We sought to assess how those working on the frontlines of HIV ME, specifically public health practitioners (PHPs) and researchers, prioritized these issues.MethodsEthical issues were identified through literature review, qualitative in-depth interviews, and stakeholder engagement. PHPs and researchers using HIV ME prioritized the issues using best–worst scaling (BWS). A balanced incomplete block design was used to generate 11 choice tasks each consisting of a sub-set of 5 ethical concerns. In each task, respondents were asked to assess the most and least concerning issue. Data were analyzed using conditional logit, with a Swait-Louviere test of poolability. Latent class analysis was then used to explore preference heterogeneity.ResultsIn total, 57 respondents completed the BWS experiment May–June 2023 with the Swait-Louviere test indicating that researchers and PHPs could be pooled (p = 0.512). Latent class analysis identified two classes, those highlighting “Harms” (n = 29) (prioritizing concerns about potential risk of legal prosecution, individual harm, and group stigma) and those highlighting “Utility” (n = 28) (prioritizing concerns about limited evidence, resource allocation, non-disclosure of data use for HIV ME, and the potential to infer the directionality of HIV transmission). There were no differences in the characteristics of members across classes.ConclusionsThe ethical issues of HIV ME vary in importance among stakeholders, reflecting different perspectives on the potential impact and usefulness of the data. Knowing these differences exist can directly inform the focus of future deliberations about the policies and practices of HIV ME in the United States.

Assessing patient information needs for new antidiabetic medications to inform shared decision-making: A best-worst scaling experiment in China.

Shared decision-making (SDM) is a patient-centred approach to improve the quality of care. An essential requirement for the SDM process is to be fully aware of patient information needs. Our study aimed to assess patient information needs for new antidiabetic medications using the best-worst scaling (BWS) experiment. BWS tasks were developed according to a literature review and the focus group discussion. We used a balanced incomplete block design and blocking techniques to generate choice sets. The final BWS contains 11 attributes, with 6-choice scenarios in each block. The one-to-one, face-to-face BWS survey was conducted among type 2 diabetic patients in Jiangsu Province. Results were analyzed using count-based analysis and modelling approaches. We also conducted a subgroup analysis to observe preference heterogeneity. Data from 539 patients were available for analysis. The most desired information domain was the comparative effectiveness of new antidiabetic medications. It consists of the incidence of macrovascular complications, the length of extended life years, changes in health-related quality of life, the incidence of microvascular complications, and the control of glycated haemoglobin. Of all the attributes, the incidence of macrovascular complications was the primary concern. Patients' glycemic control and whether they had diabetes complications exerted a significant influence on their information needs. Information on health benefits is of critical significance for diabetic patients. Patients have different information needs as their disease progresses. Personalized patient decision aids that integrate patient information needs and provide evidence of new antidiabetic medications are worthy of being established. Before data collection, a pilot survey was carried out among diabetic patients to provide feedback on the acceptability and intelligibility of the attributes.

Eliciting citizens’ priorities for active travel infrastructure investments: A qualitative analysis of best-worst scaling experiments

IntroductionThe built environment plays an important role in individuals’ propensity to walk and cycle and local authorities increasingly invest financial resources towards its development. Organisations responsible for the built environment have developed auditing tools as guidelines to inspect routes and identify improvements to support active travel. MethodsUsing these auditing tools as a starting point, this study developed 21 walking and 25 cycling investment-relevant factors that were embedded into two choice-based survey instruments, respectively. The study used cognitive interview pre-testing to internally validate a preference-based elicitation approach known as Best-Worst Scaling (BWS), which aimed to capture pedestrian and cyclist preferences. We report findings from cognitive interviews (data analysed thematically) with 20 participants (10 pedestrians and 10 cyclists). ResultsIn both sets of interviews, four themes emerged regarding how the participants approached the BWS task and five themes related to the understanding of the factors. The BWS choice tasks required refinement regarding the ‘frame of reference’, ‘travel context’, the ‘decision-making strategy’, and the ‘concrete thinking’ (finding some factors easier to interpret). Additionally, issues with understanding the factors, the wording, ‘overlapping’, negatively phrased factors, and technical jargon all pointed towards the need to refine auditing tools if these were to be introduced in a preference elicitation context. ConclusionsThis study helps to empirically uncover how citizens interpret infrastructure related aspects of walking and cycling by pointing to nuanced aspects that auditing tools may miss. The findings also helped develop an internally consistent preference elicitation survey-instrument that any local authority can implement to determine which walking and cycling infrastructure investments are a priority in their area.

Preferences for Monitoring Comprehensive Heart Failure Care: A Latent Class Analysis.

To measure preference heterogeneity for monitoring systems among patients with a chronic heart failure. A best-worst scaling experiment (BWS case 3) was conducted among patients with chronic heart failure to assess preferences for hypothetical monitoring care scenarios. These were characterized by the attributes mobility, risk of death, risk of hospitalization, type and frequency of monitoring, risk of medical device, and system-relevant complications. A latent class analysis (LCA) was used to analyze and interpret the data. In addition, a market simulator was used to examine which treatment configurations participants in the latent classes preferred. Data from 278 respondents were analyzed. The LCA identified four heterogeneous classes. For class 1, the most decisive factor was mobility with a longer distance covered being most important. Class 2 respondents directed their attention toward attribute "monitoring," with a preferred monitoring frequency of nine times per year. The attribute risk of hospitalization was most important for respondents of class 3, closely followed by risk of death. For class 4, however, risk of death was most important. A market simulation showed that, even with high frequency of monitoring, most classes preferred therapy with high improvement in mobility, mortality, and hospitalization. Using LCA, variations in preferences among different groups of patients with chronic heart failure were examined. This allows treatment alternatives to be adapted to individual needs of patients and patient groups. The findings of the study enhance clinical and allocative decision-making while elevating the quality of clinical data interpretation.

Apples to advocacy: Evaluating consumer preferences for hard cider policies

AbstractHard cider is a sector of a maturing craft beverage industry that continues to experience growth in the United States. Cider is also experiencing challenges, however, such as competition from other alcohol markets, changing consumer preferences, the supply chain, and inflationary pressures. National policy changes may help promote more optimal outcomes for this sector, but public support is important to policy formation. This study uses survey data from a best-worst scaling experiment of consumers in four leading cider-producing states (Michigan, Washington, Wisconsin, and Vermont) to understand preferences toward ten broad cider policy initiatives. The results of multinomial logistic modeling reveal that consumers prefer policies mandating ingredients, nutrition facts, and allergen labeling across all ciders. The least preferred policy initiatives include allowing producers to use vintage on labeling and funding regional cider development. These results have important implications for stakeholders across the industry, including the benefits of labeling disclosures in marketing and the need to improve public awareness of barriers to cider industry development.

Best-worst scaling in studying the impact of the coronavirus pandemic on health professionals in Ghana

In this study, we utilized a best-worst scaling experiment design to assess the potential factors associated with depression, anxiety, and stress among health professionals following the experience of the COVID-19 pandemic. The maximum difference model was performed to analyze the potential risk factors associated with depression, anxiety, and stress. As a case study, a total of 300 health professionals in Ghana were included in the survey. The majority, 112 (68.7%) male health professionals and 97 (70.8%) female health professionals reported that they had encountered suspected COVID-19 patients. 83 (50.9%) of the male health professionals and 76 (55.5%) of the female health professionals reported that they had encountered confirmed COVID-19 patients. A considerable proportion of the males 59 (36.2%) and females 57 (41.6%) health professionals reported coming into direct contact with COVID-19 lab specimens. The findings indicated that a high proportion of health professionals encountered suspected or confirmed COVID-19 patients, while a considerable proportion had direct contact with COVID-19 lab specimens leading to psychological problems. Risk factors such as contact with confirmed COVID-19 patients, the relentless spread of the coronavirus, death of patients and colleagues, shortage of medical protective equipment, direct contact with COVID-19 lab specimens, and the permanent threat of being infected should be given special attention, and necessary psychological intervention provided for health professionals endorsing these risk factors. Improving the supply of medical protective equipment to meet occupational protection practices, sufficient rest, and improving the vaccination of the population might help safeguard health professionals from depression, anxiety, and stress. Our results provide insight into policy discussions on the mental health of health professionals and interventions that are essential to enhance psychological resilience.

Aligning values to labels: A best-worst analysis of food labels

AbstractConsumer misperception and misinterpretation of food labels can lead to consumers not buying a product or purchasing products that do not align with their environmental or sustainability interests. Consumer purchasing behavior can be explained by looking at consumer food values or food quality attributes. This study aimed to (a) determine the effect label information has on consumer preference shares for selected sustainability-related food labels and (b) if correlations exist between food labels and food values. To the best of our knowledge, this is the first study to examine the comprehension of 12 different labels and identify how food labels relate to food value preferences. Responses from the best-worst scaling experiment of food value and environmental food label choice sets were analyzed using the random parameter logit model. Results reveal preference shares changed for each label as more information was provided to the respondents about the various labels included in the study. These findings should support food policy efforts requiring strict, clear label standards. Food labels should represent the food’s core food values to increase consumer preference for the product. These findings also further support the need for efforts to increase consumer knowledge and understanding of the labels on food packaging.

Older adult patient preferences for the content and format of prescription medication labels – A best-worst scaling and discrete choice experiment study

BackgroundPatient preferences for the content and format of prescription medication labels (PMLs, i.e., sticker labels placed on medication bottles/packets at dispensing) have been extensively studied. However, accommodating all preferences on PMLs is impractical due to space limitations. Understanding how patients prioritise the content and format attributes of PMLs can inform improvements while working within PML space constraints. ObjectivesWe aimed to (1) identify a ranking of medication-related content attributes to be prioritised on PMLs using best-worst scaling (BWS), and (2) determine the relative importance of format attributes when incorporated onto PMLs using discrete choice experiment (DCE), from the perspective of older adult patients in Singapore. MethodsAttributes were informed by our prior qualitative study and PML best practice guidelines. For the BWS component, the assessed content attributes were indication, precautions, interaction or paired medicines, food instructions, side effects, expiry date, and missed dose action, all of which are currently not legally mandated on PMLs in Singapore. A BWS object case was used to rank the content attributes. For the DCE component, in a series of questions, participants were asked to choose between two PML options each time, that varied in the presentation of dosage-frequency instructions, font size, presentation of dosage, presentation of precautions, and font colour of precautions. A mixed logit model estimated the relative utilities of format attribute levels, enabling the calculation of importance scores of the format attributes. ResultsThe study recruited 280 participants (mean age: 68.8 ± 5.4 years). The three most-preferred content attributes were indication, precautions and interaction or paired medicines. The top three format preferences were tabular style presentation of dosage-frequency instructions, large font size and precautions in red colour. ConclusionsHealthcare institutions should consider improving their PMLs based on the leading content and format preferences voiced by older adult patients. The methodology adopted in the study can also be used for aligning the content and format of other patient education materials with patient preferences.

The role of information heterogeneity in blockchain-based traceability systems: evidence from fresh fruits buyers in China

Blockchain technology is now being piloted to agri-food traceability systems to restore consumers’ confidence for food quality and safety. It is important for the industry to understand what information to be recorded and tracked in blockchain-based fresh produce traceability systems to meet consumers’ preferences for information. Yet little research has focused specifically on consumers’ preferences concerning information attributes traced by this new blockchain technology. This study conducts a best-worst scaling experiment with fresh fruit buyers in China to investigate consumers’ preference and perceived value regarding sixteen information attributes about blockchain-based fresh fruit traceability systems. The results from the analysis of a random parameter logit model reveal that consumers consistently rank testing information as the first-most valuable attribute, followed by production inputs (pesticides and fertilizers), quality certification and grades information attributes, while supplier and logistics information are considered to be the least valuable traceability one. Furthermore, there exist significant heterogeneity in relative value placed on traceable information attributes. The findings identify four different consumer segments by using a latent class modelling approach: (1) sensitivity for authoritative information, (2) preferences for comprehensive information, (3) information preferences equally, and (4) preferences for production inputs information. Preference heterogeneity is mainly explained by risk attitude, risk perception, information concern, traceability cognition, gender and other factors. The findings from this study can provide stakeholders and policymakers with certain insights as well as strategies on information provision and disclosure for fresh produce blockchain-based traceability.

Signaling Origin: Consumer Perceptions of Product of Canada and Made in Canada Food Labels

ABSTRACT A Product of Canada label can be used on food if all major ingredients are grown or raised in Canada and the product is processed in a Canadian facility. A Made in Canada label is permissible when the final processing step occurs in Canada. Using survey data, we examine whether Canadian consumers understand the difference between these labels, their effectiveness as quality cues, and the relative importance of country-of-origin labels. A best-worst scaling experiment reveals a positive response to origin labels in general, and a Product of Canada label in particular. Nevertheless, few consumers understand the meaning of the labels or the difference between them. Probit regression models find little evidence of a relationship between valuing the source of ingredients or location of processing and choosing Product of Canada or Made in Canada labels. Consumers may be responding to the “Canadianness” of the labels, rather than what the labels signify.

Using MaxDiff Analysis to Elicit Patients' Treatment Preferences for Distal Radius Fractures in Patients Aged 60 Years and Older

The objective of our study was to determine how the attributes of surgical and nonsurgical distal radius fracture (DRF) treatments affect patient treatment preferences. Two hundred fifty patients aged 60 years and older were contacted from a single-hand surgeon's practice, and 172 chose to participate. We built a series of best-worst scaling experiments for the MaxDiff analysis to determine the relative importance of treatment attributes. Hierarchical Bayes analysis was used to generate individual-level item scores (ISs) for each attribute that together have a total sum of100. One hundred general hand clinic patients without a history of a DRF and 43 patients with a history of a DRF completed the survey. For the general hand clinic patients, the most important attributes to avoid when choosing a DRF treatment (in descending order) were the longer time to full recovery (IS, 24.9; 95% confidence interval [CI]: 23.4-26.3), longer time spent in a cast (IS, 22.8; 95% CI, 21.5-24.2), and higher complication rates (IS, 18.4; 95% CI, 16.9-19.8). Meanwhile, for patients with a history of a DRF, the most important attributes to avoid (in descending order) were a longer time to full recovery (IS, 25.6; 95% CI, 23.3-27.9), longer time spent in a cast (IS, 22.8; 95% CI, 19.9-25.7), and abnormal alignment of the radius on x-ray (IS, 18.3; 95% CI, 15.4-21.3). For both the groups, the least concerning attributes based on the IS were appearance-scar, appearance-bump, and the need for anesthesia. Eliciting patient preferences is a vital component of shared decision-making and advancing patient-centered care. As conceptualized in this MaxDiff analysis, when choosing a DRF treatment, patients mostly want to avoid a longer time to full recovery and a longer time in a cast, whereas patients have the least concern about appearance and need for anesthesia. Eliciting patient preferences is a vital component of shared decision-making. Our results may provide guidance to surgeons in discussions on the relative benefits of surgical and nonsurgical DRF treatments, by quantifying the most and least important factors to patients.

Effective Return Prevention Measures in the Post-purchase Stage: A Best-Worst Scaling Approach

Within the rather nascent literature stream examining product returns, the authors identify that the vast majority of studies enlightens return avoidance measures (before ordering), whereas return averting measures (after purchasing) showed to increase customer satisfaction the most. Moreover, extant literature is characterized by methodological paucity, which prevents a more diverse view on the problem of product returns. As a result, the authors enrich the literature by conducting a Best-Worst Scaling (BWS) experiment dealing with various product averting measures (including different types of measures with different levels of immediacy). Focusing on the most frequent online shopping segment (Generation Y) and the second most frequently occurring reason for product returns (i.e., consumers did not like the product), the authors exploratively evince that monetary/utilitarian measures are preferred most and are effective, as well as funding of social and sustainability projects. However, substantial differences were detected based on consumers’ green consumption values. Apart from that, the study contributes to the BWS literature by outlining the necessity to include anchor scaling and quality assessment criteria.

Environmental sustainability perception toward obvious recovered waste content in paper-based packaging: An online and in-person survey best-worst scaling experiment

This study explores consumers’ visual sustainability impressions of paper-based packaging that has incorporated obvious waste content. Two research questions were addressed concerning (i) the environmental sustainability perception of noticeable waste content in packaging and (ii) the impact of the presentation format (i.e., online versus in-person surveys) when studying these perceptions. Best-worst scaling experiments were conducted, which made respondents choose the ‘most’ and ‘least’ environmentally friendly package. Packages were designed using paperboard substrates blending either brown linerboard or white hardwood pulp with different recovered waste materials. The results showed that consumers perceive obvious waste-containing packaging as more environmentally friendly than classical packaging (with no visual waste). Samples with a brown base and agricultural waste were perceived as more sustainable compared to white packaging and the use of paper waste. In addition, the presentation format changed respondents’ perception, and should therefore be carefully considered when designing surveys.

Shades of fake news: how fallacies influence consumers’ perception

ABSTRACT So far, fake news has been mostly associated with fabricated content that intends to manipulate or shape opinions. In this manuscript, we aim to establish that the perception of information as fake news is influenced by not only fabricated content but also by the rhetorical device used (i.e., how news authors phrase the message). Based on argumentation theory, we advance that fallacies – a subset of well-known deceptive rhetorical devices – share a conceptual overlap with fake news and are therefore suitable for shedding light on the issue’s grey areas. In a first two-by-two, between-subject, best-worst scaling experiment (case 1), we empirically test whether fallacies are related to the perception of information as fake news and to what extent a reader can identify them. In a second two-by-two experiment, we presume that a reader believes that some of a sender’s messages contain fake news and investigate recipients’ subsequent reactions. We find that users distinguish nuances based on the applied fallacies; however, they will not immediately recognise some fallacies as fake news while overemphasising others. Regarding users’ reactions, we observe a more severe reaction when the message identified as fake news comes from a company instead of an acquaintance.

Does the relative importance of the OxCAP-MH’s capability items differ according to mental ill-health experience?

BackgroundSome capability dimensions may be more important than others in determining someone’s well-being, and these preferences might be dependent on ill-health experience. This study aimed to explore the relative preference weights of the 16 items of the German language version of the OxCAP-MH (Oxford Capability questionnaire-Mental Health) capability instrument and their differences across cohorts with alternative levels of mental ill-health experience.MethodsA Best–Worst-Scaling (BWS) survey was conducted in Austria among 1) psychiatric patients (direct mental ill-health experience), 2) (mental) healthcare experts (indirect mental ill-health experience), and 3) primary care patients with no mental ill-health experience. Relative importance scores for each item of the German OxCAP-MH instrument were calculated using Hierarchical Bayes estimation. Rank analysis and multivariable linear regression analysis with robust standard errors were used to explore the relative importance of the OxCAP-MH items across the three cohorts.ResultsThe study included 158 participants with complete cases and acceptable fit statistic. The relative importance scores for the full cohort ranged from 0.76 to 15.72. Findings of the BWS experiment indicated that the items Self-determination and Limitation in daily activities were regarded as the most important for all three cohorts. Freedom of expression was rated significantly less important by psychiatric patients than by the other two cohorts, while Having suitable accommodation appeared significantly less important by the expert cohort. There were no further significant differences in the relative preference weights of OxCAP-MH items between the cohorts or according to gender.ConclusionsOur study indicates significant between-item but limited mental ill-health related heterogeneity in the relative preference weights of the different capability items within the OxCAP-MH. The findings support the future development of preference-based value sets elicited from the general population for comparative economic evaluation purposes.

Consumers' preferences and willingness to pay for improved environmental standards: insights from cane sugar in the Great Barrier Reef region*

Reducing nutrient runoff from sugarcane production into the Great Barrier Reef (GBR) has become a major policy focus for the Queensland and Australian Governments. This study explores consumer willingness to pay (WTP) to achieve higher environmental standards for sugar originating from the GBR catchments, through the use of a GBR‐safe ecolabel. A Best‐Worst Scaling (BWS) and a Contingent Valuation (CV) experiment are conducted on a random sample of 1,100 Australian residents. The BWS experiment reveals that personal health considerations are more important than sustainability and environmental factors, including impacts on the GBR. Results of the CV experiment show that respondents are more likely to pay a premium to support Reef‐friendly sugar if they are living in urban areas, plan to visit the GBR in the future, think that the GBR condition has declined, and are generally concerned about keeping a healthy diet. We estimate that the average WTP is $24.5/year/household, which only represents 0.34 per cent of the average weekly grocery bill of Australian households. This small contribution through increased sugar prices could conservatively raise $46.9M/year in support of sugar producers to improve water quality in the GBR. Based on these results, we recommend policy‐makers consider instruments that further involve sugar consumers.

Best–worst scaling methodology to evaluate constructs of the Consolidated Framework for Implementation Research: application to the implementation of pharmacogenetic testing for antidepressant therapy

BackgroundDespite the increased demand for pharmacogenetic (PGx) testing to guide antidepressant use, little is known about how to implement testing in clinical practice. Best–worst scaling (BWS) is a stated preferences technique for determining the relative importance of alternative scenarios and is increasingly being used as a healthcare assessment tool, with potential applications in implementation research. We conducted a BWS experiment to evaluate the relative importance of implementation factors for PGx testing to guide antidepressant use.MethodsWe surveyed 17 healthcare organizations that either had implemented or were in the process of implementing PGx testing for antidepressants. The survey included a BWS experiment to evaluate the relative importance of Consolidated Framework for Implementation Research (CFIR) constructs from the perspective of implementing sites.ResultsParticipating sites varied on their PGx testing platform and methods for returning recommendations to providers and patients, but they were consistent in ranking several CFIR constructs as most important for implementation: patient needs/resources, leadership engagement, intervention knowledge/beliefs, evidence strength and quality, and identification of champions.ConclusionsThis study demonstrates the feasibility of using choice experiments to systematically evaluate the relative importance of implementation determinants from the perspective of implementing organizations. BWS findings can inform other organizations interested in implementing PGx testing for mental health. Further, this study demonstrates the application of BWS to PGx, the findings of which may be used by other organizations to inform implementation of PGx testing for mental health disorders.

Governance of forest resource use in western Nepal: Current state and community preferences.

Improved governance of natural resource use is critical to the sustainability and maintenance of environmental quality. In western Nepal, unsustainable resource extraction is seen by the local community as a major threat to forest sustainability. While most respondents to a survey of 243 households inside and outside a protected area (PA) thought the laws for managing resource use were adequate and appropriate, a far smaller proportion thought they were achieving their objectives. Disenchantment with the existing governance regime was strongest outside the PA, probably because there was greater investment in community engagement within the PA. The most likely reason for this failure is the deeply embedded corruption within the forest governance system. Devolution of power to local communities by increasing governance participation is one of the most likely means of containing corruption. It was therefore not surprising that governance participation was rated as the most important governance principle by respondents in a best-worst scaling experiment. Respondents also regarded effectiveness, accountability and transparency as important governance principle to improve management of forest resource extraction from red panda habitat.

Valuing informal carers\u2019 quality of life using best-worst scaling\u2014Finnish preference weights for the Adult \ufeffSocial Care Outcomes Toolkit for carers (ASCOT-Carer)

This study developed Finnish preference weights for the seven-attribute Adult Social Care Outcomes Toolkit for carers (ASCOT-Carer) and investigated survey fatigue and learning in best-worst scaling (BWS) experiments. An online survey that included a BWS experiment using the ASCOT-Carer was completed by a sample from the general population in Finland. A block of eight BWS profiles describing different states from the ASCOT-Carer were randomly assigned to each respondent, who consecutively made four choices (best, worst, second best and second worst) per profile. The analysis panel data had 32,160 choices made by 1005 respondents. A scale multinomial logit (S-MNL) model was used to estimate preference weights for 28 ASCOT-Carer attribute levels. Fatigue and learning effects were examined as scale heterogeneity. Several specifications of the generalised MNL model were employed to ensure the stability of the preference estimates. The most and least-valued states were the top and bottom levels of the control over daily life attribute. The preference weights were not on a cardinal scale. We observed the position effect of the attributes on preferences associated with the best or second-best choices. A learning effect was found. The established preference weights can be used in evaluations of the effects of long-term care services and interventions on the quality of life of service users and caregivers. The learning effect implies a need to develop study designs that ensure equal consideration to all profiles (choice tasks) in a sequential choice experiment.