The Effect of Attribute Ranges on Weights in Multiattribute Utility Measurements

Nitzsch, Rüdiger von; Weber, Martin

doi:10.1287/mnsc.39.8.937

Cited by 151 publications

(82 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In SMART (von Winterfeldt and Edwards, 1986), ratio (Edwards, 1977) and point allocation methods the decision maker directly assigns weights to objectives. These methods do not explicitly incorporate ranges when weight judgments are derived (von Nitzsch andBorcherding, 1993). In AHP pairwise comparisons typically do not include the impact ranges.…”

Section: Multi-criteria Decision Analysis (Mcda)mentioning

confidence: 99%

“…The range insensitivity bias refers to the phenomenon that participants do not sufficiently adjust their weights if the range of attributes is changed (von Nitzsch and Weber, 1993). For instance, if the difference between the lowest and highest costs increases from 100,000 € to 500,000 € after the addition of a new alternative and all other ranges remain the same, then the weight of the cost objective should increase.…”

Section: Biases Related To Weighting Proceduresmentioning

confidence: 99%

See 1 more Smart Citation

Are objectives hierarchy related biases observed in practice? A meta-analysis of environmental and energy applications of Multi-Criteria Decision Analysis

Marttunen

Belton

Lienert

2018

European Journal of Operational Research

View full text Add to dashboard Cite

Procedural and behavioural biases have received little attention in recent Multi-Criteria Decision Analysis (MCDA) research. Our literature review shows that most research on biases was done 15-30 years ago.This study focuses on biases that are introduced at an early stage of MCDA when building objectives hierarchies and their effect on the weights. The main objective is to investigate whether prior findings regarding such biases, which were mostly based on laboratory experiments, can be found in real-world applications. We conducted a meta-analysis of the objectives hierarchies and weight elicitation procedures in 61 environmental and energy MCDA cases. Relationships between the structural characteristics of the objectives hierarchy and assigned objectives' weights were analysed with statistical tests. Our main research questions were: (i) how does hierarchy size and structure affect the objectives' weights? (ii) how are weights distributed across economic, social and environmental objectives? (iii) is there support for the equalising bias? Our findings are mostly aligned with earlier research and suggest that the hierarchy structure and content can substantially influence weight distributions. For example, hierarchical weighting seems to be sensitive to the asymmetry bias, which can occur when a hierarchy has branches that differ in the number of sub-objectives. We found no evidence for the equalising bias. We highlight issues deserving more attention when developing objectives hierarchies and eliciting weights.The research demonstrates the potential to use meta-analysis, which has not previously been used in this way in the MCDA field, to learn from a collection of applications.

show abstract

Section: Multi-criteria Decision Analysis (Mcda)mentioning

confidence: 99%

Section: Biases Related To Weighting Proceduresmentioning

confidence: 99%

Are objectives hierarchy related biases observed in practice? A meta-analysis of environmental and energy applications of Multi-Criteria Decision Analysis

Marttunen

Belton

Lienert

2018

European Journal of Operational Research

View full text Add to dashboard Cite

show abstract

“…Borcherding et al (1991) reported that in their experiment decision makers found it difficult to give consistent TRADEOFF statements. Regardless of this, TRADEOFF weights have been used to validate the weights derived with other methods (von Nitzsch and Weber, 1993). We think that the dependency between inconsistencies of the statements and the number of attributes compared simultaneously needs further attention (Hypothesis 6, Exhibit 1).…”

Section: H6mentioning

confidence: 99%

On the convergence of multiattribute weighting methods

Pöyhönen¹,

Hämäläinen²

2001

European Journal of Operational Research

244

106

View full text Add to dashboard Cite

Five multiattribute weighting methods were compared in an Internet experiment. This is the first experiment where the subjects created the alternatives and attributes themselves. Each subject used five methods to assess attribute weights, one version of the Analytic Hierarchy Process (AHP), DIRECT weighting, Simple Multiattribute Rating Technique (SMART), SWING weighting and TRADEOFF weighting. They can all be used following the principles of multiattribute value theory. Furthermore, each of them asks decision makers to give numerical estimates of weight ratios although the elicitation questions are different. In earlier studies, however, these methods have yielded different weights. Our results suggest that weights are different because the methods explicitly or implicitly lead the decision makers to choose their responses from a limited set of numbers. The other consequences are that the spread of weights and the inconsistency between the preference statements depend on the number of attributes that a decision maker considers simultaneously.

show abstract

“…Contrary to this prediction, several studies have provided evidence that weight increases with increase in range (Fischer, 1995;Goldstein, 1990;von Nitzsch & Weber, 1993). In these studies the researchers have observed this relationship while using a variety of different methods (direct importance ratings, swing weights, regression weights, or weights inferred from value judgments).…”

mentioning

confidence: 93%

Testing models of trade-off contrast in pairwise choice.

Wedell

1998

Journal of Experimental Psychology: Human Perception and Perfor

View full text Add to dashboard Cite

In three experiments, students completed sets of three pairwise choices in each of several domains. The first two pairs in each set were contextual pairs, and the third was the target pair. Context was manipulated by widening the range of values on one dimension and narrowing it on the second dimension. Consistent with previous research, participants on target trials more often chose the alternative whose poorer value was on the widened dimension. Four models explained this effect as (a) decreased weighting of the wide dimension, (b) contextual shift in values, (c) a tendency to equalize responses, or (d) a tendency to add value based on dominance relationships. Path models provided evidence against weight change and response equalization, and manipulation of dominance did not reduce the effect. Results supported the value-shift explanation of trade-off contrast.When choosing between multiattribute alternatives, one is faced with the problem of having to make trade-offs between values on different attributes. For example, imagine a choice between two cars that differ primarily on the dimensions of gas mileage and safety features. Would you choose the car that gets high gas mileage but has few safety features or the one that gets low gas mileage but has many safety features? Ultimately the choice depends on whether the increase in gas mileage is worth the decrease in safety: a trade-off.How are such trade-offs resolved? One approach to the problem is given by multiattribute utility theory, in which preference values are assigned to each level on a dimension and importance weights are assigned to the different dimensions (Von Winterfeldt & Edwards, 1986). The overall utility of an alternative is determined by combining weights and values, typically using a weighted additive model:( 1) where Vy is the preference value of alternative i on attribute/ Wj is the weight assigned to that attribute, and m indexes the set of relevant attributes. The choice rule is then to select the alternative with the highest overall attractiveness or utility, a difference operation. If one considers a dimensionwise processor of information in a pairwise choice situation, the weighted additive model can give rise to the weighted difference model. When Equation 1 is rearranged, the difference in utilities for two alternatives can be expressed as

show abstract

The Effect of Attribute Ranges on Weights in Multiattribute Utility Measurements

Cited by 151 publications

References 11 publications

Are objectives hierarchy related biases observed in practice? A meta-analysis of environmental and energy applications of Multi-Criteria Decision Analysis

Are objectives hierarchy related biases observed in practice? A meta-analysis of environmental and energy applications of Multi-Criteria Decision Analysis

On the convergence of multiattribute weighting methods

Testing models of trade-off contrast in pairwise choice.

Contact Info

Product

Resources

About