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.     

A stochastic model for test-retest correlations

A simple stochastic model is formulated in order to determine the optimal time between the first test and the second test when the test-retest method of assessing reliability is used. A forgetting process and a change in true score process are postulated. The optimal time between tests is derived by maximizing the probability that the respondent has not remembered the response on the first test and has not had a change in true score. The resulting test-retest correlation is then found to be a linear function of the true reliability of the test, where the slope of this function is the key probability of not remembering and having no change in true score. Some numerical examples and suggestions for using the results in empirical studies are given. Specific recommendations are presented for improved design and analysis of intentions data.This research was made possible by a grant from the Center for Food Policy Research, Graduate School of Business, Columbia University, New York, New York, 10027.     

Reaction times of younger and older men and temporal contingencies of reinforcement

Influences of extended training and temporal contingencies on reaction time were studied in relation to developmental differences. Older and younger men were trained on a chained schedule in which completion of a variable interval produced a terminal link in which reaction time was measured. The reaction-time procedure involved a conditional discrimination with matching to sample in one component and oddity matching in the other. During baseline training, no time limit was placed on the response to the discrimination choice stimuli. Subsequently, increasingly severe time limits were imposed over a series of sessions. Older and younger men showed increased speeds (decreased reaction times) when temporal contingencies were imposed, and these changes were maintained during post-training baseline sessions when there was unlimited time in which to respond. The younger men generally responded faster than the older ones, and age differences were not appreciably reduced during the course of the experiment. The results indicated the feasibility of studying reaction time in human subjects using operant procedures analogous to those developed for the study of nonverbal organisms.     

Concurrent schedules: Effects of time- and response-allocation constraints

Five pigeons were trained on concurrent variable-interval schedules arranged on two keys. In Part 1 of the experiment, the subjects responded under no constraints, and the ratios of reinforcers obtainable were varied over five levels. In Part 2, the conditions of the experiment were changed such that the time spent responding on the left key before a subsequent changeover to the right key determined the minimum time that must be spent responding on the right key before a changeover to the left key could occur. When the left key provided a higher reinforcer rate than the right key, this procedure ensured that the time allocated to the two keys was approximately equal. The data showed that such a time-allocation constraint only marginally constrained response allocation. In Part 3, the numbers of responses emitted on the left key before a changeover to the right key determined the minimum number of responses that had to be emitted on the right key before a changeover to the left key could occur. This response constraint completely constrained time allocation. These data are consistent with the view that response allocation is a fundamental process (and time allocation a derivative process), or that response and time allocation are independently controlled, in concurrent-schedule performance.     

A stochastic multidimensional scaling procedure for the empirical determination of convex indifference curves for preference/choice analysis

The vast majority of existing multidimensional scaling (MDS) procedures devised for the analysis of paired comparison preference/choice judgments are typically based on either scalar product (i.e., vector) or unfolding (i.e., ideal-point) models. Such methods tend to ignore many of the essential components of microeconomic theory including convex indifference curves, constrained utility maximization, demand functions, et cetera. This paper presents a new stochastic MDS procedure called MICROSCALE that attempts to operationalize many of these traditional microeconomic concepts. First, we briefly review several existing MDS models that operate on paired comparisons data, noting the particular nature of the utility functions implied by each class of models. These utility assumptions are then directly contrasted to those of microeconomic theory. The new maximum likelihood based procedure, MICROSCALE, is presented, as well as the technical details of the estimation procedure. The results of a Monte Carlo analysis investigating the performance of the algorithm as a number of model, data, and error factors are experimentally manipulated are provided. Finally, an illustration in consumer psychology concerning a convenience sample of thirty consumers providing paired comparisons judgments for some fourteen brands of over-the-counter analgesics is discussed.     

Essential independence and likelihood-based ability estimation for polytomous items

A definition ofessential independence is proposed for sequences of polytomous items. For items satisfying the reasonable assumption that the expected amount of credit awarded increases with examinee ability, we develop a theory ofessential unidimensionality which closely parallels that of Stout. Essentially unidimensional item sequences can be shown to have a unique (up to change-of-scale) dominant underlying trait, which can be consistently estimated by a monotone transformation of the sum of the item scores. In more general polytomous-response latent trait models (with or without ordered responses), anM-estimator based upon maximum likelihood may be shown to be consistent for under essentially unidimensional violations of local independence and a variety of monotonicity/identifiability conditions. A rigorous proof of this fact is given, and the standard error of the estimator is explored. These results suggest that ability estimation methods that rely on the summation form of the log likelihood under local independence should generally be robust under essential independence, but standard errors may vary greatly from what is usually expected, depending on the degree of departure from local independence. An index of departure from local independence is also proposed.This work was supported in part by Office of Naval Research Grant N00014-87-K-0277 and National Science Foundation Grant NSF-DMS-88-02556. The author is grateful to William F. Stout for many helpful comments, and to an anonymous reviewer for raising the questions addressed in section 2. A preliminary version of section 6 appeared in the author's Ph.D. thesis.     

Sensory superstition on multiple interval schedules.

Pigeons were exposed to multiple schedules in which an irregular repeating sequence of five stimulus components was correlated with the same reinforcement schedule throughout. Stable, idiosyncratic, response-rate differences developed across components. Components were rank-ordered by response rate; an approximately linear relation was found between rank order and the deviation of mean response rate from the overall mean rate. Nonzero slopes of this line were found for multiple fixed-interval and variable-time schedules and for multiple variable-interval schedules both when number of reinforcements was the same in all components and when it varied. The steepest function slopes were found in the variable schedules with relatively long interfood intervals and relatively short component durations. When just one stimulus was correlated with all components of a multiple variable-interval schedule, the slope of the line was close to zero. The results suggest that food-rate differences may be induced initially by different reactions to the stimuli and subsequently maintained by food.     

A multi-element analysis of the effect of teacher aides in an "open"-style classroom

Open classrooms with few rules, individualized instruction, and informal class organization present new problems for the application of behavior principles. The effects of three types of teacher aides on student achievement and on-task behavior were studied. Each was compared with a standard no-aide condition. Subjects were 54 third graders in two "open"-style classrooms. The three types of aide, helping adult, disciplinary adult, and helping fifth-grade aide, were compared in a multi-element design with a no-aide control. The helping-adult aide significantly affected the academic output of the class, when compared with the no-aide condition. All aide conditions produced more academic work and on-task behavior than did the standard no-aide condition.     

Choice, time allocation, and response rate during stimulus generalization

Six pigeons were trained to discriminate between two noise intensities using a procedure that assessed choice, time allocation, and response rate simultaneously and independently. Responses on the left or right key (R1 or R2) were respectively correct in the presence of two different intensities, S1 and S2. After a correct response, reinforcement became available for pecks on the center key. Reinforcement density for R1¦S1 relative to R2¦S2 was varied across experimental conditions. Generalization tests followed extensive training at each condition. As a function of stimulus intensity, proportions of initial choices of R2, of time spent in R2-initiated components, and of center-key responses emitted in R2-initiated components all yielded sigmoidal gradients of similar slope, which shifted slightly in location when relative reinforcement density changed. Changeovers were maximal where initial choice proportions approximated 0.5. Gradients relating the absolute number of center-key responses to stimulus intensity were also roughly sigmoidal, but were more sensitive to changes in reinforcement density. Gradients of momentary response rate also depended on reinforcement density. During training, large but transitory shifts in choice responding occurred when reinforcement density changed, while differences in momentary response rate developed slowly, suggesting separate control of choice and response rate by the contingencies of reinforcement.     

A comparison of frequency, interval, and time-sampling methods of data collection

Data representing high, medium, and low response rates in constant and nonconstant patterns were generated by electromechanical equipment to determine whether the same data collected by time-sampling, interval recording, and frequency recording would be represented similarly by each method. Results indicated: (1) that time-sampling provided an extremely inaccurate estimate of responding, and (2) that interval recording accurately represented responding of low and medium rates, but grossly underestimated high-rate responding.