Some developments in multivariate generalizability

This article is concerned with estimation of components of maximum generalizability in multifacet experimental designs involving multiple dependent measures. Within a Type II multivariate analysis of variance framework, components of maximum generalizability are defined as those composites of the dependent measures that maximize universe score variance for persons relative to observed score variance. The coefficient of maximum generalizability, expressed as a function of variance component matrices, is shown to equal the squared canonical correlation between true and observed scores. Emphasis is placed on estimation of variance component matrices, on the distinction between generalizability- and decision-studies, and on extension to multifacet designs involving crossed and nested facets. An example of a two-facet partially nested design is provided.Appreciation is expressed to the Office of Research in Medical Education, University of Texas Medical Branch, for permitting use of their data.     

Conflicting codes: Professional, ethical, and legal obligations in archaeology

Archaeologists employed in governmental positions often deal with issues that produce conflicts between their professional duties to their employer, their ethical responsibilities to the resource, and their obligations as established by legislation. The paper examines some of the conflicts imposed on governmental archaeologists by each of these systems but focuses on the conflicts imposed by federal legislation and regulations on governmental archaeologists, using “Kennewick Man” as an example. This is a revised edition of a paper written for the symposium: “Ethics in Science: Special Problems in Anthropology and Archaeology,” organized by Merrilee Salmon; 1998 Annual Meeting of the American Association for the Advancement of Science (Philadelphia, February 15, 1998). The original paper was in a more conversational style as befitting a discussion atmosphere. Revisions, comments, and opinions are entirely the author’s. Joe Watkins is the Anadarko Agency Archaeologist at the Bureau of Indian Affairs and is a Research Associate in the Department of Anthropology, Indiana University.     

THE IMPACT OF AUDITION ON THE DEVELOPMENT OF VISUAL ATTENTION

Interactions between audition and vision were investigated in two experiments In the first experiment, school-age hearing children, deaf children with cochlear implants, and deaf children without implants participated in a task in which they were to respond to some visual signals and not others This task did not involve sound at all Deaf children without implants performed much more poorly than hearing children Deaf children with cochlear implants performed considerably better than deaf children without implants The second experiment employed a longitudinal design and showed that the rate of development in visual selective attention was faster for deaf children with cochlear implants than deaf children without implants Moreover, the gams were rapid–occurring within 2 years post-implant surgery The results suggest that a history of experience with sounds matters in the development of visual attention The results are discussed in terms of multimodal developmental processes     

Depression and decision-making among intravenous drug users

Two short scales, the TCU Depression Scale and the TCU Decision-making Scale, were psychometrically evaluated in a sample of 145 intravenous drug users. Coefficient alpha reliabilities were .78 for the 6-item TCU Depression Scale and .77 for the 9-item TCU Decision-making Scale. Concurrent validity of the former scale was assessed by correlating scores with those on the Beck Depression Inventory, r = .75. Based on the Beck Depression Inventory Clinical cutoff scores, 83% of the sample showed some depression, with 23% severely depressed, 39% moderately depressed, and 21% mildly depressed. Individuals scoring higher on depression on both tests tended to score lower on decision-making. Significant demographic associations of age, gender, education, and race-ethnicity were found for the depression and decision-making scales. More depression was noted for women, those younger, white, and having less education. Older and more educated intravenous drug users tended to score higher on decision-making. Validity for the depression and decision-making scales was assessed by examining correlations with behaviors. Significant positive correlations were found between depression scores and intravenous use of cocaine only, heroin and cocaine combined, and heroin only. Also, intravenous use of cocaine only and of cocaine and heroin combined were negatively related to decision-making. AIDS sex-risky behavior was positively correlated with depression and negatively correlated with decision-making.     

COVID-19, Virtual Engagement and the Psychoid Imagination1

The impact of the COVID-19 pandemic on analytic training and the delivery of educational content is explored here. The proliferation of Zoom therapy and teaching is creating a post-human platform to which nearly everyone in contemporary society has had to adapt. Looking at the possible meanings of the pandemic, a psychoid factor (the virus) engaging the imagination has come to the fore as a response to climate change. The striking similarity to the H1N1 viral pandemic (“Spanish flu”) is noted, especially in the context of C. G. Jung having had a case in 1919 during which he experienced a number of visions and dreams. The imagery produced can be seen as an implicit attempt at “re-enchanting the world” found in The red book. Finally, a reconsideration of pedagogy in response to the pandemic is discussed with an eye to the archetypal aspects of internet communications.     

Order‐constrained linear optimization

Despite the fact that data and theories in the social, behavioural, and health sciences are often represented on an ordinal scale, there has been relatively little emphasis on modelling ordinal properties. The most common analytic framework used in psychological science is the general linear model, whose variants include ANOVA, MANOVA, and ordinary linear regression. While these methods are designed to provide the best fit to the metric properties of the data, they are not designed to maximally model ordinal properties. In this paper, we develop an order‐constrained linear least‐squares (OCLO) optimization algorithm that maximizes the linear least‐squares fit to the data conditional on maximizing the ordinal fit based on Kendall's τ. The algorithm builds on the maximum rank correlation estimator (Han, 1987, Journal of Econometrics, 35, 303) and the general monotone model (Dougherty & Thomas, 2012, Psychological Review, 119, 321). Analyses of simulated data indicate that when modelling data that adhere to the assumptions of ordinary least squares, OCLO shows minimal bias, little increase in variance, and almost no loss in out‐of‐sample predictive accuracy. In contrast, under conditions in which data include a small number of extreme scores (fat‐tailed distributions), OCLO shows less bias and variance, and substantially better out‐of‐sample predictive accuracy, even when the outliers are removed. We show that the advantages of OCLO over ordinary least squares in predicting new observations hold across a variety of scenarios in which researchers must decide to retain or eliminate extreme scores when fitting data.     

TACIT: An open-source text analysis,crawling, and interpretation tool

As human activity and interaction increasingly take place online, the digital residues of these activities provide a valuable window into a range of psychological and social processes. A great deal of progress has been made toward utilizing these opportunities; however, the complexity of managing and analyzing the quantities of data currently available has limited both the types of analysis used and the number of researchers able to make use of these data. Although fields such as computer science have developed a range of techniques and methods for handling these difficulties, making use of those tools has often required specialized knowledge and programming experience. The Text Analysis, Crawling, and Interpretation Tool (TACIT) is designed to bridge this gap by providing an intuitive tool and interface for making use of state-of-the-art methods in text analysis and large-scale data management. Furthermore, TACIT is implemented as an open, extensible, plugin-driven architecture, which will allow other researchers to extend and expand these capabilities as new methods become available.