Confidence depends on level of aggregation

The credible intervals that people set around their point estimates are typically too narrow (cf. Lichtenstein, Fischhoff, & Phillips, 1982). That is, a set of many such intervals does not contain the actual values of the criterion variables as often as it should given the probability assigned to this event for each estimate. The typical interpretation of such data is that people are overconfident about the accuracy of their judgments. This paper presents data from two studies showing the typical levels of overconfidence for individual estimates of unknown quantities. However, data from the same subjects on a different measure of confidence for the same items, their own global assessment for the set of multiple estimates as a whole, showed significantly lower levels of confidence and overconfidence than their average individual assessment for items in the set. It is argued that the event and global assessments of judgment quality are fundamentally different and are affected by unique psychological processes. Finally, we discuss the implications of a difference between confidence in single and multiple estimates for confidence research and theory.     

A bayesian approach to confirmatory factor analysis

Confirmatory factor analysis is considered from a Bayesian viewpoint, in which prior information on parameter is incorporated in the analysis. An iterative algorithm is developed to obtain the Bayes estimates. A numerical example based on longitudinal data is presented. A simulation study is designed to compare the Bayesian approach with the maximum likelihood method.Computer facilities were provided by the Computer Services Center, The Chinese University of Hong Kong.     

Preferred Difficulty and Subjective Probability in Different Action Phases

Drawing on Gollwitzer's deliberative–implemental mindset distinction (P. M. Gollwitzer, 1990), it was predicted that people who are deliberating on different actions or goals would be more cautious or more realistic in their expectation of success in subsequent tasks than people who are going to implement a chosen action or goal. Participants were given a choice between different test-materials. They were interrupted before (deliberative) or immediately after decision-making (implemental). They then either had to choose between various levels of difficulty within one type of task (Experiment 1) or they had to predict their own future performance (Experiment 2). The results showed that deliberative participants preferred less difficult tasks and overestimated their probability of success less than implemental participants. In addition, deliberative participants referred more than implemental participants to their past performance when selecting levels of difficulty or predicting future performance; however, the two groups did not differ in actual performance. Taken together, the findings suggest that people are more realistic in a deliberative than in an implemental state of mind. The present studies extend prior research because for the first time they document mindset effects on peoples' estimates concerning their future performance in the achievement domain.     

Improved Test–retest Reliability Criteria as Applied to Probability Assessments

Test–retest reliability is a common indicator of response consistency. It is argued that using regression coefficients for detecting systematic response error is less appropriate than testing for shifts in the mean (median) and variance. This procedure is exemplified using probability response data. For this data, shifts in centrality were found to be about 2.5 times more likely than shifts in variability. The shifts in centrality did not favor any particular direction; however, variability tended to decrease over time in early sessions.     

A maximum likelihood method for latent class regression involving a censored dependent variable

The standard tobit or censored regression model is typically utilized for regression analysis when the dependent variable is censored. This model is generalized by developing a conditional mixture, maximum likelihood method for latent class censored regression. The proposed method simultaneously estimates separate regression functions and subject membership in K latent classes or groups given a censored dependent variable for a cross-section of subjects. Maximum likelihood estimates are obtained using an EM algorithm. The proposed method is illustrated via a consumer psychology application.     

An evaluation of the reliability of probability judgments across response modes and over time

Despite the importance of probability assessment methods in behavioral decision theory and decision analysis, little attention has been directed at evaluating their reliability and validity. In fact, no comprehensive study of reliability has been undertaken. Since reliability is a necessary condition for validity, this oversight is significant. The present study was motivated by that oversight. We investigated the reliability of probability measures derived from three response modes: numerical probabilities, pie diagrams, and odds. Unlike previous studies, the experiment was designed to distinguish systematic deviations in probability judgments, such as those due to experience or practice, from random deviations. It was found that subjects assessed probabilities reliably for all three assessment methods regardless of the reliability measures employed. However, a small but statistically significant decrease over time in the magnitudes of assessed probabilities was observed. This effect was linked to a decrease in subjects overconfidence during the course of the experiment.     

Subjective probability and delay.

Human subjects indicated their preference between a hypothetical $1,000 reward available with various probabilities or delays and a certain reward of variable amount available immediately. The function relating the amount of the certain-immediate reward subjectively equivalent to the delayed $1,000 reward had the same general shape (hyperbolic) as the function found by Mazur (1987) to describe pigeons' delay discounting. The function relating the certain-immediate amount of money subjectively equivalent to the probabilistic $1,000 reward was also hyperbolic, provided that the stated probability was transformed to odds against winning. In a second experiment, when human subjects chose between a delayed $1,000 reward and a probabilistic $1,000 reward, delay was proportional to the same odds-against transformation of the probability to which it was subjectively equivalent.     

On the sampling theory roundations of item response theory models

Item response theory (IT) models are now in common use for the analysis of dichotomous item responses. This paper examines the sampling theory foundations for statistical inference in these models. The discussion includes: some history on the stochastic subject versus the random sampling interpretations of the probability in IRT models; the relationship between three versions of maximum likelihood estimation for IRT models; estimating versus estimating -predictors; IRT models and loglinear models; the identifiability of IRT models; and the role of robustness and Bayesian statistics from the sampling theory perspective.A presidential address can serve many different functions. This one is a report of investigations I started at least ten years ago to understand what IRT was all about. It is a decidedly one-sided view, but I hope it stimulates controversy and further research. I have profited from discussions of this material with many people including: Brian Junker, Charles Lewis, Nicholas Longford, Robert Mislevy, Ivo Molenaar, Donald Rock, Donald Rubin, Lynne Steinberg, Martha Stocking, William Stout, Dorothy Thayer, David Thissen, Wim van der Linden, Howard Wainer, and Marilyn Wingersky. Of course, none of them is responsible for any errors or misstatements in this paper. The research was supported in part by the Cognitive Science Program, Office of Naval Research under Contract No. Nooo14-87-K-0730 and by the Program Statistics Research Project of Educational Testing Service.     

When is the Wilcoxon–Mann–Whitney procedure a test of location? Implications for effect-size measures

The Wilcoxon–Mann–Whitney procedure is invariant under monotone transformations but its use as a test of location or shift is said not to be so. It tests location only under the shift model, the assumption of parallel cumulative distribution functions (cdfs). We show that infinitely many monotone transformations of the measured variable produce parallel cdfs, so long as the original cdfs intersect nowhere or everywhere. Thus there are infinitely many effect sizes measured as shifts of medians, invalidating the notion that there is one true shift parameter and thereby rendering any single estimate dubious. Measuring effect size using the probability of superiority alleviates this difficulty.