Regression analyses of repeated measures data in cognitive research

Repeated measures designs involving nonorthogonal variables are being used with increasing frequency in cognitive psychology. Researchers usually analyze the data from such designs inappropriately, probably because the designs are not discussed in standard textbooks on regression. Two commonly used approaches to analyzing repeated measures designs are considered in this article. It is argued that both approaches use inappropriate error terms for testing the effects of independent variables. A more appropriate analysis is presented, and two alternative computational procedures for the analysis are illustrated.     

Men's heterosocial skill and attitudes toward women as predictors of verbal sexual coercion and forceful rape

This study assessed whether men's heterosocial skill and attitudes toward women are related to verbal and physical sexual coercion in a nonincarcerated population. We predicted that heterosocially skilled men would be more likely than unskilled men to have engaged in verbal sexual coercion, whereas unskilled men would be more likely than skilled men to have engaged in physical sexual coercion (i.e., forceful rape). We expected an interaction, with this pattern holding only for men who accepted traditional gender roles or male sexual dominance; men who did not accept these attitudes were expected to be unlikely to have engaged in either verbal or physical coercion. To identify sexually coercive men, we administered questionnaires to 1152 male introductory psychology students. They completed anonymous self-report measures of (a) heterosocial skill, (b) attitudes toward women, and (c) involvement with verbal and physical sexual coercion. Results showed that heterosocially skilled men were more likely than unskilled men to have engaged in verbal sexual coercion, but heterosocial skill was unrelated to forceful rape. Men who accepted traditional gender roles or male sexual dominance were more likely than other men to have engaged in both verbal sexual coercion and forceful rape. There were no interactions. Theoretical issues and treatment implications are discussed.The authors wish to thank Gayla Goolsby, Trudi Zaplac, Diane Collinge, Shelley Davis, Gina Benedicto, and DiAnne Poehl. This paper is based on the second author's Senior Honors Thesis, done as part of the Texas A&M University Undergraduate Fellows Program, which provided funding.     

Ms. and the manager: A tale of two stereotypes

Recent research on the Ms. stereotype suggests a potentially important exception to the strong link between gender and managerial stereotypes observed in the 1970s. Specifically, we hypothesized that a woman who prefers “Ms.” as her title of address would be seen by perceivers of both sexes as more similar in the “requisite” traits of a successful, middle manager than would a woman who prefers a traditional title of address for herself. These hypotheses were tested by having respondents rate a “successful, middle manager” or other stimulus persons varying in titles of address, sex, and marital status. Correlational analyses of respondents' ratings of stimulus persons yielded strong support for the hypotheses. That is, a woman preferring Ms. as her title of address was seen by adult perceivers of both sexes as possessing the “requisite” personality characteristics of a successful, middle manager; whereas a traditionally titled woman was not. Implications of these findings are discussed.     

Contributions of the care-giving role to women's development

It was hypothesized that older widows (N=309) who had been care givers to their husbands would have undertaken a greater number of household responsibilities than had women who had not been care givers. Findings indicated that, prior to the husband's death, more traditionally male and more traditionally female tasks were performed by the care-giving wives (n=158). This was particularly true in the case of Alzheimer's care givers (n=21). More importantly, widows' present feelings of competence were related to undertaking household responsibilities, with those who had been care givers expressing significantly higher levels of competence.     

Predicting treatment outcome and recidivism among patients in a maximum security token economy

The relationship between program performance and outcome for patients in maximum security token economy programs has not been previously studied. We examined variables related to success in a token economy program and to postrelease failure for 92 men in a maximum security psychiatric hospital. Point earnings later in the program were best predicted by earlier point earnings. Variables that best predicted postrelease failure, defined in various ways, were similar to those found in other studies; youthfulness, having spent more time in institutions, a diagnosis of personality disorder or retardation rather than psychosis, not having been found not guilty by reason of insanity or unfit to stand trial, and having been referred from another psychiatric hospital. That program variables were generally unrelated to later outcome suggests that treatment programs for patients in maximum security settings must emphasize skills that are relevant to postrelease success.     

Overt responding in computer-based training

This experiment compared the extent to which students learned facts included in computer-based-training frames that required an overt response to those that did not. Frames included two types of facts: Some facts had one word missing which had to be supplied by the student (active facts). Other facts had no missing words and required no overt response (passive facts). Each student completed four 14-fact modules in random order. Two of the modules contained frames with all active facts. Each frame in the other two modules contained one active and six passive facts. Paper pretests and posttests showed that students learned twice as many facts when all facts in the module were active than when only one in seven were active. The modules that included passive facts required one-fifth the time to complete than those with only active facts. Students learned more facts per minute of training in one of the two modules that included passive facts than in the two modules with only active facts.     

The electronic teaching theater: Interactive hypermedia and mental models of the classroom

The introduction of hypermedia into the classroom presents both an opportunity to expand the power of teaching through electronic facilitation of the media and a challenge to redesign the classroom and instructional environment to exploit the enhanced features of hypermedia. This paper briefly surveys the current state of computers in the classroom and then argues that a more integrated approach is required. To this end, the concept of an electronic teaching theater is proposed which takes advantage of hypermedia and collaborative work environments. The aim of the electronic teaching theater is not to totally redesign the classroom and instructional interaction, but rather to facilitate it using well understood metaphors to lay out the dynamics of classroom interaction on the electronic media and to use the results from research in human/computer interaction to design the interface. To facilitate the application of hypermedia, a model of instructional interaction among the students, the instructors, the course material, and the products of instruction is presented. A number of component models pertaining to course preparation, lectures, note taking, and other instructional activities are then discussed that should prove useful in designing the electronic classroom. The origin of these models is the mental representation of the interaction as perceived by the teacher and the student. The idea is to start with such models as a base metaphor, to instantiate the metaphors in the electronic classroom, and then to explore innovations in the technology that go beyond the strict application of the metaphor. This work was supported in part by a grant from AT&T to the University of Maryland during the author’s visit to the Medical Research Council—Applied Psychological Unit, Cambridge, England. Appreciation is expressed to Patricia Wright and Richard M. Young for their helpful comments on this paper.     

Feigned insanity in nineteenth-century America: Tactics,trials, and truth

Feigned insanity in nineteenth-century America is appraised through a review of the medical and legal literature. The authors focus on the explanations for feigning, procedures used in uncovering feigning, and the role of feigning in the courtroom. This discussion of feigned insanity demonstrates the remarkable consistency of approach to this form of malingering over the past 200 years.     

Motivational Systems and Stress-Related Cardiovascular Reactivity

This study examined the relation of motivational systems to patterns of autonomic responses to stress. Specifically, we examined patterns of physiological response resulting from differential activation of motivational systems for behavioral approach and behavioral inhibition. We also examined the relation of these motivational systems to threat and challenge responses to potential stress. Self-report, cardiac measures (i.e., pre-injection period, or PEP, and heart rate, or HR), and blood pressure (i.e., systolic blood pressure, or SBP, and diastolic blood pressure, or DBP) measures reliably distinguished between an experimental condition facilitating behavioral approach and an experimental condition facilitating coactivation of behavioral approach and behavioral inhibition systems. Vascular (e.g., total peripheral resistance, or TPR) and electrodermal (e.g., electrodermal response, or EDR) measures, however, did not vary by condition. We discuss the results in relation to other research on motivational factors as they relate to stress and cardiovascular reactivity.     

Notes on the Tenacity of Therapeutic Presuppositions in Process Research: Examining the Artfulness of Blamings in Family Therapy

In their recent contribution to the Journal of Family Therapy , Frosh et al . apply discourse analytic methods to the study of 'process' in family therapy. While we welcome this contribution, we suggest that the authors' reading of the data has been affected by a number of fundamental presuppositions, which have led them to neglect the skilful rhetorical work taking place in the session. In this paper, we have undertaken an alternative analysis of the data, examining its rhetorical features and suspending judgement on the efficacy of the therapeutic project and on the truth status of statements made by participants in the session. While acknowledging that our microanalytic method has some shortcomings, we contend that attention to the situated and strategic nature of 'therapy talk' has exposed the primacy of notions of responsibility and blame in family work, and has shown how these issues are artfully negotiated by therapist and family in the cut and thrust of the therapeutic interview.