Strategies for inference robustness in focused modelling

Advances in computation mean that it is now possible to fit a wide range of complex models to data, but there remains the problem of selecting a model on which to base reported inferences. Following an early suggestion of Box & Tiao, it seems reasonable to seek 'inference robustness' in reported models, so that alternative assumptions that are reasonably well supported would not lead to substantially different conclusions. We propose a four-stage modelling strategy in which we iteratively assess and elaborate an initial model, measure the support for each of the resulting family of models, assess the influence of adopting alternative models on the conclusions of primary interest, and identify whether an approximate model can be reported. The influence-support plot is then introduced as a tool to aid model comparison. The strategy is semi-formal, in that it could be embedded in a decision-theoretic framework but requires substantive input for any specific application. The one restriction of the strategy is that the quantity of interest, or 'focus', must retain its interpretation across all candidate models. It is, therefore, applicable to analyses whose goal is prediction, or where a set of common model parameters are of interest and candidate models make alternative distributional assumptions. The ideas are illustrated by two examples. Technical issues include the calibration of the Kullback-Leibler divergence between marginal distributions, and the use of alternative measures of support for the range of models fitted.     

A pooled time-series analysis of abortion demand

This study estimates the demand for abortion in the United States using state data pooled over years 1992 and 1982. The empirical results showed that the price elasticity of abortion demand ranged from -0.70 to -0.99 and an income elasticity between 0.27 and 0.35. The demand for abortion was found (1) not to be statistically related to a woman's educational level; (2) to be higher the greater a state's taste for abortion; (3) coincident with the business cycle; and (4) not to be related to the level of a state's welfare payment.     

Making Career Theories More Culturally Sensitive: Implications for Counseling

The primary question addressed in this article is whether and how career theories can be more culturally sensitive without losing value as conceptual explanations or their usefulness for counselors. Contextual action theory is identified as a means to develop culturally sensitive explanations. Six steps are proposed and illustrated, including using the naive observations and subjective reports and recognizing ongoing processes. The use of these steps in counseling is also addressed.     

Effects of Social Anxiety on Nonverbal Accuracy and Response Time I: Facial Expressions

Cognitive models of social anxiety provide a basis for predicting that the ability to process nonverbal information accurately and quickly should be impaired during the experience of state anxiety. To test this hypothesis, we assigned participants to threatening and non-threatening conditions and asked them to label the emotions expressed in a series of faces. It was predicted that social anxiety would be positively associated with errors and response times in threatening conditions, but not in a non-threatening condition. It was also predicted that high social anxiety would be associated with more errors and longer response times when identifying similar expressions such as sadness, anger, and fear. The results indicate that social anxiety was not associated with errors in identifying facial expressions of emotion, regardless of the level of state anxiety experienced. However, social anxiety scores were found to be significantly related to response times to identify facial expressions, but the relationship varied depending on the level of state anxiety experienced. Methodological and theoretical implications of using response time data when assessing nonverbal ability are discussed.     

Infants' Use of Constraints to Speed Information Processing and to Anticipate Events

Two experiments were conducted with 28‐week‐old infants using a modification of the Visual Expectation Paradigm. The first sought to determine whether speed of information processing (SIP) could be assessed in infants using a reaction time (RT) measure and approach that is widely used to measure SIP in adults. Infants saw a center fixation cue followed by a peripheral target that could appear in 1, 2, or 4 locations. There was a linear increase in RT of eye movements as the number of locations increased from 1 to 2 and to 4 targets, suggesting that the paradigm does measure SIP. The second experiment asked whether varying the number of cue‐target pairings would augment or impair infant's SIP in the trade‐off between the benefit of additional information and the liability of additional memory load. The findings showed that the presence of cue information can eliminate the difference in RT between the 1‐ and 2‐location conditions, whereas no benefit of cue was obtained for the 4‐location condition.