The risk elicitation puzzle revisited: Across-methods (in)consistency?
Ontology highlight
ABSTRACT:
SUBMITTER: Holzmeister F
PROVIDER: S-EPMC8550567 | biostudies-literature | 2020 Sep
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:The question of how to measure and classify people's risk preferences is of substantial importance in the field of economics. Inspired by the multitude of ways used to elicit risk preferences, we conduct a holistic investigation of the most prevalent method, the multiple price list (MPL) and its derivations. In our experiment, we find that revealed preferences differ under various versions of MPLs as well as yield unstable results within a 30-minute time frame. We determine the most stable elicitation method with the highest forecast accuracy by using multiple measures of within-method consistency and by using behavior in two economically relevant games as benchmarks. A derivation of the well-known method by Holt and Laury (American Economic Review 92(5):1644-1655, 2002), where the highest payoff is varied instead of probabilities, emerges as the best MPL method in both dimensions. As we pinpoint each MPL characteristic's effect on the revealed preference and its consistency, our results have implications for preference elicitation procedures in general.
| S-EPMC5366177 | biostudies-literature
Project description:The Asymptotic Classification Theory of Cognitive Diagnosis (ACTCD) developed by Chiu, Douglas, and Li proved that for educational test data conforming to the Deterministic Input Noisy Output "AND" gate (DINA) model, the probability that hierarchical agglomerative cluster analysis (HACA) assigns examinees to their true proficiency classes approaches 1 as the number of test items increases. This article proves that the ACTCD also covers test data conforming to the Deterministic Input Noisy Output "OR" gate (DINO) model. It also demonstrates that an extension to the statistical framework of the ACTCD, originally developed for test data conforming to the Reduced Reparameterized Unified Model or the General Diagnostic Model (a) is valid also for both the DINA model and the DINO model and (b) substantially increases the accuracy of HACA in classifying examinees when the test data conform to either of these two models.
| S-EPMC5978609 | biostudies-literature
Project description:BackgroundThe evidence used to inform health care decision making (HCDM) is typically uncertain. In these situations, the experience of experts is essential to help decision makers reach a decision. Structured expert elicitation (referred to as elicitation) is a quantitative process to capture experts' beliefs. There is heterogeneity in the existing elicitation methodology used in HCDM, and it is not clear if existing guidelines are appropriate for use in this context. In this article, we seek to establish reference case methods for elicitation to inform HCDM.MethodsWe collated the methods available for elicitation using reviews and critique. In addition, we conducted controlled experiments to test the accuracy of alternative methods. We determined the suitability of the methods choices for use in HCDM according to a predefined set of principles for elicitation in HCDM, which we have also generated. We determined reference case methods for elicitation in HCDM for health technology assessment (HTA).ResultsIn almost all methods choices available for elicitation, we found a lack of empirical evidence supporting recommendations. Despite this, it is possible to define reference case methods for HTA. The reference methods include a focus on gathering experts with substantive knowledge of the quantities being elicited as opposed to those trained in probability and statistics, eliciting quantities that the expert might observe directly, and individual elicitation of beliefs, rather than solely consensus methods. It is likely that there are additional considerations for decision makers in health care outside of HTA.ConclusionsThe reference case developed here allows the use of different methods, depending on the decision-making setting. Further applied examples of elicitation methods would be useful. Experimental evidence comparing methods should be generated.
| S-EPMC8777312 | biostudies-literature
Project description:Desired fertility measures are routinely collected and used by researchers and policy makers, but their self-reported nature raises the possibility of reporting bias. In this paper, we test for the presence of such bias by comparing responses to direct survey questions with indirect questions offering a varying, randomized, degree of confidentiality to respondents in a socioeconomically diverse sample of Nigerian women ([Formula: see text]). We find that women report higher fertility preferences when asked indirectly, but only when their responses afford them complete confidentiality, not when their responses are simply blind to the enumerator. Our results suggest that there may be fewer unintended pregnancies than currently thought and that the effectiveness of family planning policy targeting may be weakened by the bias we uncover. We conclude with suggestions for future work on how to mitigate reporting bias.
| S-EPMC11348086 | biostudies-literature
Project description:Quantum-mechanical methods are used for understanding molecular interactions throughout the natural sciences. Quantum diffusion Monte Carlo (DMC) and coupled cluster with single, double, and perturbative triple excitations [CCSD(T)] are state-of-the-art trusted wavefunction methods that have been shown to yield accurate interaction energies for small organic molecules. These methods provide valuable reference information for widely-used semi-empirical and machine learning potentials, especially where experimental information is scarce. However, agreement for systems beyond small molecules is a crucial remaining milestone for cementing the benchmark accuracy of these methods. We show that CCSD(T) and DMC interaction energies are not consistent for a set of polarizable supramolecules. Whilst there is agreement for some of the complexes, in a few key systems disagreements of up to 8 kcal mol-1 remain. These findings thus indicate that more caution is required when aiming at reproducible non-covalent interactions between extended molecules.
| S-EPMC8225865 | biostudies-literature
Project description:Elicitation methods are known to influence second language speech production. For teachers and language assessors, awareness of such effects is essential to accurate interpretations of testing outcomes. For speech researchers, understanding why one method gives better performance than another may yield insights into how second-language phonological knowledge is acquired, stored, and retrieved. Given these concerns, this investigation compared L2 vowel intelligibility on two elicitation tasks and determined the degree to which differences generalized across vowels, vowels in context, lexical items, and individual speakers. The dependent variable was the intelligibility of Cantonese speakers' productions of English /i I u ℧/ in varying phonetic environments. In a picture-naming task, the speakers produced responses without an auditory prompt. In a second task-interrupted repetition-they heard exemplars of the same targets without pictures, and repeated each one after counting aloud to 10, a step intended to disrupt their short-term auditory store and therefore prevent simple mimicry. For target words with scores below 80% on picture naming, mean intelligibility was more than 10 points higher on interrupted repetition. However, that difference did not generalize across conditions or across speakers. Thus, although it is technically accurate to say that, on average, interrupted repetition yielded better vowel intelligibility than did picture naming, that observation requires a great deal of qualification, particularly because of individual speaker differences. The outcomes are interpreted in terms of their relevance to language assessment and phonetic learning.
| S-EPMC9326214 | biostudies-literature
Project description:Knowledge on the relationship between different biological modalities (RNA, chromatin, etc.) can help further our understanding of the processes through which biological components interact. The ready availability of multi-omics datasets has led to the development of numerous methods for identifying sources of common variation across biological modalities. However, evaluation of the performance of these methods, in terms of consistency, has been difficult because most methods are unsupervised. We present a comparison of sparse multiple canonical correlation analysis (Sparse mCCA), angle-based joint and individual variation explained (AJIVE) and multi-omics factor analysis (MOFA) using a cross-validation approach to assess overfitting and consistency. Both large and small-sample datasets were used to evaluate performance, and a permuted null dataset was used to identify overfitting through the application of our framework and approach. In the large-sample setting, we found that all methods demonstrated consistency and lack of overfitting; however, in the small-sample size setting, AJIVE provided the most stable results. We provide an R package so that our framework and approach can be applied to evaluate other methods and datasets.
| S-EPMC7373174 | biostudies-literature
Project description:Well-established methods of economic evaluation are used in many countries to inform decisions about the funding of new medical interventions. To guide such decisions, it is important to consider what health gains would be expected from the same level of investment elsewhere in the health care system. Recent research in the United Kingdom has evaluated the evidence available and the methods required to estimate the health effects of changes in health care expenditure within the National Health Service. Because of the absence of sufficiently broad-ranging data, assumptions were required in the previously mentioned work to estimate health effects in terms of a broader measure of health (quality-adjusted life-years), which is more relevant for policy. These assumptions constitute important sources of uncertainty. This work presents an application of the structured elicitation of the judgments of key individuals about these uncertain quantities. This article describes the design and conduct of the exercise, including the quantities elicited, the individual (rather than consensus) approach used, how uncertainty in knowledge was elicited (mode and bounds of an 80% credible interval), and methods to generate group estimates. It also reports on a successful application involving 28 clinical experts and 25 individuals with policy responsibilities. Although, as expected, most experts found replying to the questions challenging, they were able to express their beliefs quantitatively. Consistent across the uncertainties elicited, experts' judgments suggest that the quality-adjusted life-year (QALY) impacts of changes in expenditure from earlier work using assumptions are likely to have been underestimated and the "central" estimate of health opportunity cost from that work (£12,936 per QALY) to have been overestimated.
| S-EPMC7509606 | biostudies-literature
Project description:Despite the development of numerous gene regulatory network (GRN) inference methods in the last years, their application, usage and the biological significance of the resulting GRN remains unclear for our general understanding of large-scale gene expression data in routine practice. In our study, we conduct a structural and a functional analysis of B-cell lymphoma GRNs that were inferred using 3 mutual information-based GRN inference methods: C3Net, BC3Net and Aracne. From a comparative analysis on the global level, we find that the inferred B-cell lymphoma GRNs show major differences. However, on the edge-level and the functional-level-that are more important for our biological understanding-the B-cell lymphoma GRNs were highly similar among each other. Also, the ranks of the degree centrality values and major hub genes in the inferred networks are highly conserved as well. Interestingly, the major hub genes of all GRNs are associated with the G-protein-coupled receptor pathway, cell-cell signaling and cell cycle. This implies that hub genes of the GRNs can be highly consistently inferred with C3Net, BC3Net, and Aracne, representing prominent targets for signaling pathways. Finally, we describe the functional and structural relationship between C3Net, BC3Net and Aracne gene regulatory networks. Our study shows that these GRNs that are inferred from large-scale gene expression data are promising for the identification of novel candidate interactions and pathways that play a key role in the underlying mechanisms driving cancer hallmarks. Overall, our comparative analysis reveals that these GRNs inferred with considerably different inference methods contain large amounts of consistent, method independent, biological information.
| S-EPMC3864360 | biostudies-literature
Project description:Arabinoxylan-Oligosaccharide Intake changes the microbiota and softens stool consistency without changes in gut transit and metabolic health in healthy adults
| PRJEB32919 | ENA