FIXED EFFECTS ANALYSIS OF VARIANCE BY REGRESSION ANALYSIS

This paper is concerned with the solution of typical analysis of variance problems using general purpose multiple regression computer programs. Specific models, restrictions on the parameters for hypothesis testing, and computational aspects are discussed. It is argued that this approach has many pedagogical advantages over traditional procedures.     

TESTING FOR FAIRNESS WITH A MODERATED MULTIPLE REGRESSION STRATEGY: AN ALTERNATIVE TO DIFFERENTIAL ANALYSIS1

It is argued that analyses of subgroup differences utilizing a bivariate correlation strategy do not provide an adequate examination of test fairness. An analysis of differential prediction, which involves slopes and intercepts of regression lines results in more complete coverage of the test fairness issue, since the overall regression line determines the way in which a test is used for prediction. While subgroup correlation coefficients yield information concerning the slopes and intercepts, means and standard deviations must also be examined. A moderated multiple regression strategy is recommended as an alternative to separate analyses by subgroups. An ordered step-up regression procedure is presented which is more encompassing than the bivariate strategies, while avoiding inherent problems associated with subgroup coding in multiple regression.     

ISSUES CONCERNING THE USE OF REGRESSION ANALYSIS IN SALARY DISCRIMINATION CASES

Twenty-nine salary discrimination court cases were examined to determine ways an organization can refute a regression analysis that leads to the inference of discrimination in compensation. It was found to be equally effective for the organization to introduce its own regression, other statistics, or no statistics. The plaintiff won all of the cases when discrimination was proven in promotion. The defendant won all cases when discrimination was not proven in promotion. Defendants also won all cases when plaintiffs inappropriately treated jobs as fungible or when they failed to include important variables. Market variables may be included in regressions if defendants can show they were applied consistently to determine salaries. The implications of the courts' acceptance of some regressions for comparable worth is presented, and a list of recommendations is made for organizations that may be faced with results from regression analysis in court.     

ANALYSIS OF A SIMPLIFIED TREATMENT FOR STUTTERING IN CHILDREN

We investigated the effectiveness of a simplified program for the treatment of stuttering in children. The simplified treatment included awareness training, in which the subjects learned to detect every occurrence of stuttering; training a response incompatible with stuttering, which involved relaxation and regulation of air flow over the larynx when speaking; and social support, which involved parent-delivered prompts and praise of children's use of the techniques in everyday environments. Eight children were treated in their homes with the simplified treatment, in a multiple baseline across subjects design, and all reached the criterion level of less than 3% words stuttered. In addition, the reduction in stuttering generalized to the school setting and was maintained at posttreatment (10 to 13 months). The subjects' rates of speech remained stable throughout baseline and treatment. Pretreatment and posttreatment ratings by the parents showed that they found treatment to be both acceptable and credible. Finally, social validity measures revealed a noticeable improvement in the subjects' speech to parents and speech pathologists.     

A FADED BEDTIME WITH RESPONSE COST PROTOCOL FOR TREATMENT OF MULTIPLE SLEEP PROBLEMS IN CHILDREN

The sleep-wake cycles of 4 developmentally delayed individuals with longstanding severe sleep disturbances were regulated using a faded bedtime procedure with response cost. Bedtimes were systematically delayed for each individual, thus increasing the probability of short latency to sleep onset. The response cost component, consisting of removing the individual from bed for 1 hour, was implemented when an individual did not experience short latency to sleep onset. A fading procedure was then applied successfully to advance the bedtimes and to gradually increase durations of sleep. Specifically, all 4 individuals had decreased amounts of nighttime sleep that increased following treatment. Two of the 4 individuals showed excessive daytime sleep that decreased following treatment. Three of the 4 individuals experienced decreases in night wakings following treatment. Both environmental and biological manipulations of the sleep-wake cycle are hypothesized as mechanisms of treatment. The relative advantages of this procedure over other procedures for the treatment of pediatric sleep disorders are discussed, as are directions for future research.     

MULTIPLE FRACTURES IN TIME: REFLECTIONS ON A CAR CRASH

This essay offers a story of a traumatic event experienced by the author on September 6, 1999. Presented here in autoethnographic and multivoiced form, the author narrates a fraction of time in her life that has left her forever changed: a near-death car crash. She uses literary mode in the crash story and recollected dialogue in the rescue story. In the effort to engage readers into the particulars of such a horrific experience, she seeks to connect with trauma survivors and perhaps offer illumination for those who (for their own reasons) seek further understanding in what will, hopefully for them, never be a first-order experience.     

A METHOD FOR SCALING VERBAL DISQUALIFICATION

A disqualification is a message that says something “without really saying it”—one that is evasive, indirect, or effectively ambiguous in some other way. This article describes the rationale for, and development of, an objective method for measuring the degree of disqualification in brief written messages: nonexpert judges independently assess four basic elements of the message (sender, content, receiver, context) for ambiguity, by use of a physical scale of length. The raw scale values are standardized for each judge, and the reliability of these values is quite high. Research uses of the method, as well as its limitations and possible modifications, are discussed.     

COURT STANDARDS FOR JOB ANALYSIS IN TEST VALIDATION

Selected Federal court are reviewed and analyzed to determine the criteria used by the courts in their assessment of job analyses in the development and validation of selection tests. A set of standards which delineates the components and characteristics of a job analysis necessary to withstand legal scrutiny is presented. Implications are discussed.     

A SCALING METHOD FOR MEASURING COLOR CONSTANCY

Recently developed types of indirect psychophysical scaling method served as starting point in finding a handy procedure for measuring differences in the degrees of 'color constancy'. A design for carrying out categorical comparisons with color paper proofs was constructed. The raw scores, given as percentage distributions, prove transformable into standard scale values. These demonstrate the shifted positions of the colors in question when presented in alternating conditions of illumination.     

PROMOTING LAW ENFORCEMENT FOR CHILD PROTECTION: A COMMUNITY ANALYSIS

The Colorado Occupant Protection Project (COPP) intervention provided police with brief instruction concerning the importance of citations for drivers' failure to use child safety seats and special coupons to accompany citations. Coupons were exchangeable by drivers for a safety seat and brief training in its use, plus a waiver of the $50 citation fine. Over 4.5 years of archival records were employed, using an ABA design and a comparison community to evaluate the program. Few tickets were issued for nonuse of safety seats during the 3-year baseline in either community. Citations for nonuse of safety seats increased to over 50 per month during the intervention period at the test site, whereas rates remained essentially zero at the comparison site. After the COPP intervention was removed at the intervention site, citation rates for nonuse of safety seats decreased to about 15 per month. Differences between intervention conditions and settings were statistically significant. During the intervention, officers were 44 times more likely to write citations than were controls. Results suggested that a behavioral program can increase police citation writing for child protection purposes.     

SOCIAL-SKILLS TRAINING FOR UNASSERTIVE CHILDREN: A MULTIPLE-BASELINE ANALYSIS1

The effects of social-skills training consisting of instructions, feedback, behavior rehearsal, and modelling were examined in a multiple-baseline analysis in four unassertive children. The treatment was effective in that the behaviors selected for modification changed markedly. The effects of treatment generalized from trained to untrained items (interpersonal situations requiring assertive responding) and gains were maintained at the two- and four-week followup probe sessions. In addition, overall assertiveness in all subjects increased from baseline assessment to the conclusion of treatment and into followup.     

CONTEMPORARY EVOLUTIONARY THEORY AS A NEW HEURISTIC MODEL FOR THE SOCIOSCIENTIFIC METHOD IN BIBLICAL STUDIES

Abstract. Notions of uniform and gradual evolution have been replaced in some circles by biological and paleontological models that postulate that periods of rapid change punctuate long periods of evolutionary stasis. This new theory, called punctuated equilibria (or PE for short), may have implications for paradigms in scholarly disciplines other than the sciences. Whereas old evolutionary models exerted great influence upon historians, sociologists, anthropologists, and students of religion for more than a century, the new model may provide heuristic paradigms for research that correlate more adequately with the current observations of scholars. We therefore provide suggestions for deployment of this new scientific paradigm in history and anthropology. In particular, this model can explain the rise of the Israelite state and the religious ethos in the Hebrew Bible, two major concerns of today's socioscientific study of biblical materials. Thus the possibility of an overarching paradigm for the social sciences is entertained.     

A ROTATIONAL PROCEDURE FOR SEPARATION OF TRAIT,METHOD AND INTERACTION FACTORS IN MULTTTRAIT-MULTIMETHOD MATRICES

Cluster recovery indices are more important than ever, because of the necessity for comparing the large number of clustering procedures available today. Of the cluster recovery indices prominent in contemporary literature, the Hubert and Arabie (1985) adjustment to the Rand index (1971) has been demonstrated to have the most desirable properties (Milligan &; Cooper, 1986). However, use of the Hubert and Arabie adjustment to the Rand index is limited to cluster solutions involving non-overlapping, or disjoint, clusters. The present paper introduces a generalization of the Hubert and Arabie adjusted Rand index. This generalization, called the Omega index, can be applied to situations where both, one, or neither of the solutions being compared is non-disjoint. In the special case where both solutions are disjoint, the Omega index is equivalent to the Hubert and Arabie adjusted Rand index.     

INCREASING INDICES OF HAPPINESS AMONG PEOPLE WITH PROFOUND MULTIPLE DISABILITIES: A PROGRAM REPLICATION AND COMPONENT ANALYSIS

We replicated a program to increase indices of happiness among people with profound multiple disabilities and conducted a component analysis of the program. The program involved presentation and contingent withdrawal of stimuli that had been identified as preferred based on preference assessments and staff opinion. The program was implemented with 3 adult students and was accompanied by increased happiness indices for each student. Subsequent implementation of the two types of stimuli indicated that preferred stimuli based on preference assessments were more consistently accompanied by increased happiness indices than were preferred stimuli based on staff opinion. Social validity measures supported the definition of happiness, in that raters' subjective opinions of the students' happiness coincided with the observed happiness indices. Results are discussed regarding the importance and practical implications of using preference assessments for determining stimuli for increasing happiness indices. Future research areas are suggested, focusing on increasing happiness and other quality-of-life indicators during the daily routines of people with profound disabilities.     

USE OF COMPONENT ANALYSES TO IDENTIFY ACTIVE VARIABLES IN TREATMENT PACKAGES FOR CHILDREN WITH FEEDING DISORDERS

We evaluated the separate components in treatment packages for food refusal of 4 young children. First, treatment packages were implemented until food acceptance improved. Next, a component analysis was conducted within a multielement or reversal design to identify the active components that facilitated food acceptance. The results indicated that escape extinction was always identified as an active variable when assessed; however, other variables, including positive reinforcement and noncontingent play, were also identified as active variables for 2 of the children. The results suggest that the component analysis was useful for identifying variables that affected food acceptance.     

A RAPID METHOD FOR EVALUATING THE NECESSITY FOR BOTH A BEHAVIORAL INTERVENTION AND METHYLPHENIDATE

We examined the effects of a behavioral intervention and methylphenidate (MPH) on inappropriate behavior and sleep disturbance displayed by a 6-year-old boy who had been diagnosed with attention deficit hyperactivity disorder. Results showed that the behavioral intervention was effective in reducing inappropriate behaviors to near-zero levels regardless of the presence or absence of MPH.     

A COMPREHENSIVE EVALUATION OF REINFORCER IDENTIFICATION PROCESSES FOR PERSONS WITH PROFOUND MULTIPLE HANDICAPS

We evaluated comprehensively a preference assessment for identifying reinforcers for persons with profound multiple handicaps. Four experiments were conducted involving 18 individuals. Results of Experiment 1 replicated previous findings in that the assessment identified student preferences for respective stimuli, and caregiver opinion of preferences did not coincide with the systematic assessment. Results of Experiment 2 indicated highly preferred stimuli were likely to function as reinforcers in training programs, whereas stimuli not highly preferred did not function as reinforcers. Results of Experiment 3 suggested the 12 stimuli used in the assessment represented a comprehensive stimulus set for identifying preferences, although the utility of the set sometimes could be enhanced by caregiver opinion. Results of Experiment 4 indicated the assessment identified preferences likely to be maintained over time. Overall, results are discussed in terms of identifying limits and alternatives to a behavioral teaching technology when applied to persons with profound multiple handicaps.