On computing ELECTRE's credibility indices under partial information

The ELECTRE family of decision aid methods is a well‐known approach to help decision makers (DMs) advance in a decision process. Among the most recent methods in the family, ELECTRE IS, III, and TRI compute fuzzy indices for the credibility of a given action outranking some other. We consider the case when the DMs are unsure which values each parameter should take, which may result from insufficient, imprecise or contradictory information, as well as from different preferences among a group of DMs. In the framework of a robustness analysis approach, where DMs provide only partial information on the parameter values (through constraints on acceptable combinations), we study how to find whether an outranking among two actions in an ELECTRE method is robust. In this context, we study the resulting nonlinear problems of optimizing a credibility index (under the type of constraints that we consider to appear in practice) and present some examples. Copyright © 1999 John Wiley & Sons, Ltd.     

The Influence of Arbitrary Breakpoints on Judgments of Maximum Output

Consumers often wonder about the product's maximum output: the highest rotation speed of a blender or the best printing quality of a printer. We examine how the number of levels (e.g., a blender with 3 vs. 7 speeds) influences judgments of maximum product output. Objectively speaking, the number of levels is no more than a set of breakpoints in an already predetermined continuum from the product's minimum to maximum output. Nevertheless, because of the ubiquitous association between number of breakpoints and quantity in daily life, consumers do not simply view more levels as a signal of greater precision (i.e., giving consumers more control over the possible outputs). They also incorrectly believe that the product has greater power (i.e., a higher maximum output), even when such an inference is in conflict with diagnostic attribute information (e.g., watts). A series of five studies documents the phenomenon, its asymmetric nature, and its boundary conditions. Reliance on the inaccurate “more levels, more power” lay theory weakens when participants consider a reduction rather than an increase in number of levels, and it disappears when the consumer is presented with an explicit relationship between each level and its corresponding output value (e.g., level 4:400 W).     

The proper sequence for correcting correlation coefficients for range restriction and unreliability

Corrections of correlations for range restriction (i.e., selection) and unreliability are common in psychometric work. The current rule of thumb for determining the order in which to apply these corrections looks to the nature of the reliability estimate (i.e., restricted or unrestricted). While intuitive, this rule of thumb is untenable when the correction includes the variable upon which selection is made, as is generally the case. Using classical test theory, we show that it is the nature of the range restriction, not the nature of the available reliability coefficient, that determines the sequence for applying corrections for range restriction and unreliability.We would like to thank Malcolm James Ree for his encouragement and helpful comments as well as those of the editors, associate editor, and reviewers.     

Response of the rat brain beta-endorphin system to novelty: importance of the fornix connection

In control rats, a step-down inhibitory avoidance training trial using a 0.8 mA footshock, or simple exposure to the training apparatus without footshock, was followed by a decrease of beta-endorphin-like immunoreactivity measured in the hypothalamus and ventral thalamus. The effect of inhibitory avoidance training was also measured in rats submitted to a brain sham operation, to bilateral transection of the dorsal fornix, to anterior or to posterior hypothalamic deafferentation, to adrenal medullectomy, to an adrenal sham operation, to 16 daily ip injections of 0.2 mg/kg dexamethasone, or to 16 daily ip injections of 1 ml/kg saline. The diencephalic beta-endorphin-like immunoreactivity response to training was abolished by fornix transection and was unaffected by all other treatments. This suggests that the response is not mediated by anterior or posterior neural afferents to the hypothalamus, or by a hypersecretion of epinephrine by the adrenal medullae, or of ACTH by the pituitary gland. The response, instead, appears to require the integrity of the pathway that sends projections from the septo-hippocampal system to the hypothalamus. Previous evidence had suggested that the diencephalic beta-endorphin-like immunoreactivity response to training is a result of novelty, and the septo-hippocampal system has been postulated to play a role in the registration of novelty.     

A significance test for multisample sphericity

The paper obtains a maximum likelihood criterion test for multisample sphericity. The test contains Mauchly's sphericity test as a special case.I would like to thank Professor R. E. Bargmann for his most valuable help.     

Sex differences in aggression: Are current notions misleading?

Studied experimentally the influence of norms and sex of subjects on aggressive behaviour in same sex dyads. Two hypotheses were tested: (1) subjects will react aggressively to an unpleasant state of affairs, if they interpret it as being the result of violation of a norm on the part of another; (2) female subjects will display more aggression than male subjects under conditions of repeated provocation while male subjects will be more aggressive under conditions of infrequent or no provocation. In a 2 × 2 × 2 complete factorial design (norm violation versus norm enforcement; male versus female; low versus high reward for performance) 20 same sex pairs of students performed alternatively a sensory-motor task (victim) and a shock delivery task (aggressor). As predicted, subjects who consider other's behaviour to be a norm violation aggress more often (p <.0001). A significant interaction between sex of subject and norm violation is found in support for the second hypothesis (p <.05). It is concluded that positions grounded on the S-R paradigm are misleading for the understanding of sex differences in aggression.     

A critical reappraisal of the evidence for unconscious abstraction of deterministic rules in complex experimental situations

In a recent experiment (Lewicki et al., 1988) subjects were submitted to a four-choice RT paradigm for 3600 trials. On each of the successive logical blocks of five trials, the first two locations of the target were randomly distributed, and the last three locations were determined by complex rules. Although subjects were unable to verbalize the actual nature of the manipulation, performance on the last trials of each block improved at a faster rate and was better overall than performance on the first trials. In addition, subsequent rules changes on 480 additional trials only affected performance on the last three trials of each block. The present paper demonstrates that contrary to Lewicki et al's assertions this performance pattern requires neither acquisition of tacit knowledge of the composition rules, nor partitioning by the subjects of the sequence into logical blocks of five trials. Rather, the results can be accounted for by the relative frequency of a few simple sequences of target locations. Moreover, this alternative explanation alone correctly anticipates some striking features of fine-grained performance (Lewicki et al., 1988). The discussion focuses on methodological implications of these findings for investigation of unconscious learning, and speculates on what and how people learn when they encounter a complex and structured situation.     

Comparison of outcomes across low-intensity psychological interventions for depression and anxiety within a stepped-care setting: A naturalistic cohort study using propensity score modelling

Low-intensity interventions for common mental disorders (CMD) address issues such as clinician shortages and barriers to accessing care. However, there is a lack of research into their comparative effectiveness in routine care. We aimed to compare treatment effects of three such interventions, utilizing four years' worth of routine clinical data. Users completing a course of guided self-help bibliotherapy (GSH), internet-delivered cognitive behavioural therapy (iCBT) or psychoeducational group therapy (PGT) from a stepped-care service within the NHS in England were included. Propensity score models (stratification and weighting) were used to control for allocation bias and determine average treatment effect (ATE) between the interventions. 21,215 users comprised the study sample (GSH = 12,896, iCBT = 6862, PGT = 1457). Adherence-to-treatment rates were higher in iCBT. All interventions showed significant improvements in depression (PHQ-9), anxiety (GAD-7) and functioning (WSAS) scores, with largest effect sizes for iCBT. Both propensity score models showed a significant ATE in favour of iCBT versus GSH and PGT, and in favour of GSH versus PGT. Discernible differences in effectiveness were seen for iCBT in comparison with GSH and PGT. Given variance in delivery mode and human resources between different low-intensity interventions, building on these findings would be valuable for future service provision and policy decision making.     

The biasing effects of selection and attrition on estimating the mean

Organizational and validation researchers often work with data that has been subjected to selection on the predictor and attrition on the criterion. These researchers often use the data observed under these conditions to estimate either the predictor or criterion's restricted population means. We show that the restricted means due to direct or indirect selection are a function of the population means plus the selection ratios. Thus, any difference between selected mean groups reflects the population difference plus the selection ratio difference. When there is also attrition on the criterion, the estimation of group differences becomes even more complicated. The effect of selection and attrition induces measurement bias when estimating the restricted population mean of either the predictor or criterion. A sample mean observed under selection and attrition does not estimate either the population mean or the restricted population mean. We propose several procedures under normality that yield unbiased estimates of the mean. The procedures focus on correcting the effects of selection and attrition. Each procedure was evaluated with a Monte Carlo simulation to ascertain its strengths and weaknesses. Given appropriate sample size and conditions, we show that these procedures yield unbiased estimators of the restricted and unrestricted population means for both predictor and criterion. We also show how our findings have implications for replicating selected group differences.