Error production in children's reproductions of the Rey-Osterrieth Complex Figure (R-OCF) was examined in this study. Subjects were a control group of normal children and two clinical groups--a language-based learning disorder group (LD) and a group of children with learning disorders secondary to tumors in the posterior fossa (infra-tentorial) region of the brain. The children were between the ages of 9 and 12 years. The LD groups consisted of 7 subjects with learning disorders primarily referenced to left frontal systems (LD-LF) and 18 subjects with learning disorders primarily referenced to left hemisphere systems in general (LD-LH). The tumor group consisted of 13 subjects. Subjects in the clinical groups were matched for age, sex, and handedness with the control subjects. A scoring system was devised to evaluate the production of ten error types across copy, immediate recall, and delayed recall administrations of the R-OCF. Statistical analyses were conducted to determine whether there were significant differences in error production between the control group and the two clinical groups across the copy and immediate recall conditions and between the two clinical groups across the copy, immediate recall, and delayed recall conditions. Problems of statistical power and sample size resulted in the elimination of the LD-LF group from the analyses. Consequently, the question of differences in error production between the LD-LF and LD-LH groups could not be answered and awaits research with a larger sample. Overall, results indicate that error analysis can reliably differentiate normal children from children in clinical groups. Children in the LD-LH and tumor groups produced significantly higher frequencies of errors in reproductions of the figure than children in the control group under copy and immediate recall conditions. Data on specific error types differentiating the control and the clinical groups is presented. The results also indicate that error analysis can reliably differentiate children in the LD-LH group from children in the tumor group. Data on specific error types differentiating the two clinical groups across the copy and immediate recall conditions is presented. No significant differences emerged between these two groups in error production for the delayed recall condition. Suggestions for conducting further research on error production in children's reproductions of the R-OCf are discussed.

This study consisted of a comprehensive review of the consequential aspects of validity of a grade 10 mathematics assessment. This test is part of a larger state-mandated assessment system in which the studied test is one of two assessments that a student must pass in order to graduate from high school in the state of Massachusetts. Validity evidence was collected using three rigorous measurement methods. Qualitative and quantitative procedures were used to ensure a more complete collection and analyses of validity evidence. A survey was developed and administered to all participating high school mathematics teachers and key education personnel. Fifty-six percent of the surveys were completed and analyzed. In addition, focus group and one-on-one interviews were conducted within each participating school district. The results indicated that the Massachusetts' education reform initiative had created significant changes in high school mathematics curriculum and instruction. In addition, many positive and negative intended and unintended consequences connected to this assessment system were identified. The results were discussed based on a classification system in which a representative sample of school districts was selected from the state population. In this study, a comprehensive analysis of a few specific consequential validity questions was addressed using sound quantitative and qualitative research methods. This type of research, examining the consequential aspects of validity of a state mandated test as a component of a larger assessment system, represents a huge undertaking. The social, politic, and educational implications involved in any reform effort are complex and difficult to document. As education reform affects more and more students across this nation, answers to the outlined questions may assist key administrators in the state of Massachusetts, perhaps even other states in the middle of similar reform efforts, in making important mid-course corrections, and/or merely provide needed validity evidence regarding intended and unintended consequences of the program using solid, data-driven information.

Curriculum Based Measurement (CBM) has become an increasingly popular instrument/methodology for reading assessment. In part, its popularity derives from promises of formative assessment (i.e., progress monitoring). However, a review of the literature suggests CBM formative assessment applications may lack the requisite reliability evidence. Furthermore, available research provides support and direction to improve the accuracy and stability of formative assessment outcomes. The primary purpose of this research was to evaluate and compare the effects of a controlled set of reading passages on student performance. Researchers developed a controlled set of Curriculum Like Measurement (CLM) reading passages from a sample of unfamiliar grade-specific reading curriculums. Each grade-specific passage-set was controlled for passage-difficulty using the Spache and Dale-Chall readability formulas. Analysis compared CBM and CLM formative assessment outcomes. A second purpose of this study was to compare short-term (10-week) assessment outcomes with the negatively accelerating developmental trends that have been documented with long-term assessment (i.e., 36-week). Analysis tested for differences in stability of growth-estimates [SE(b)], accuracy of predictions (SEE), and observed growth-rates/slope (b). 99 students in grades second to fifth participated over 10 weeks. Results suggest CLM progress monitoring outcomes are more stable and accurate than CBM. Results did not demonstrate the negatively accelerating curvilinear relationship between grades. Results and implications are discussed.

No description available.

No description available.

Emerging research has suggested that the nature of cognitive deficits in Amnestic-MCI (a-MCI) may extend beyond memory impairments and can include deficits in attention (Gualtieri & Johnson, 2005; 2007). Recent studies have found significant differences between subjects with a-MCI and cognitively healthy individuals on measures of processing speed (Gorus, De Raedt, Lambert, Lemper, & Mets, 2008; Gualtieri & Johnson, 2005; Levinoff, Saumier, & Chertkow, 2005; Silveri et al., 2007). The current study sought to add to the limited research currently available on processing speed in a-MCI by comparing cognitively healthy individuals (M age = 64.57; SD = 6.61) and an age-matched a-MCI group (M age = 64.19; SD = 5.79) on a measure of processing speed, an index of attention. Second, the study evaluated the utility of the “25 Numbers Test” in differentiating individuals with a-MCI from a cognitively healthy group of participants. As expected, the participants with a-MCI performed more poorly (t(72) = -5.96, p < .01) on the 25 Numbers Test and demonstrated greater intra-individual variability in comparison to the cognitively healthy group (t(72) = -3.009, p < .01). The 25 Numbers Test effectively discriminated cognitively healthy individuals from those with a-MCI (AUC = 0.85, p < .01). Results will add to the limited research on processing speed in a-MCI and provide a basis for the importance of evaluating processing speed as part of routine screening for a-MCI.

The advantages that computer adaptive testing offers over linear tests have been well documented. The Computer Adaptive Test (CAT) design is more efficient than the Linear test design as fewer items are needed to estimate an examinee's proficiency to a desired level of precision. In the ideal situation, a CAT will result in examinees answering different number of items according to the stopping rule employed. Unfortunately, the realities of testing conditions have necessitated the imposition of time and minimum test length limits on CATs. Such constraints might place a burden on the CAT test taker resulting in aberrant response behaviors by some examinees. Occurrence of such response patterns results in inaccurate estimation of examinee proficiency levels. This study examined the effects of test lengths, time limits and the interaction of these factors with the examinee proficiency levels on the occurrence of aberrant response patterns. The focus of the study was on the aberrant behaviors caused by rushed guessing due to restrictive time limits. Four different testing scenarios were examined; fixed length performance tests with and without content constraints, fixed length mastery tests and variable length mastery tests without content constraints. For each of these testing scenarios, the effect of two test lengths, five different timing conditions and the interaction between these factors with three ability levels on ability estimation were examined. For fixed and variable length mastery tests, decision accuracy was also looked at in addition to the estimation accuracy. Several indices were used to evaluate the estimation and decision accuracy for different testing conditions. The results showed that changing time limits had a significant impact on the occurrence of aberrant response patterns conditional on ability. Increasing test length had negligible if not negative effect on ability estimation when rushed guessing occured. In case of performance testing high ability examinees while in classification testing middle ability examinees suffered the most. The decision accuracy was considerably affected in case of variable length classification tests.

Pica is a potentially life-threatening behavior problem displayed by many individuals with developmental disabilities that can result in serious medical complications. The purpose of this study was to compare findings from a functional analysis (FA), which is the state of the art method of behavioral assessment, with two indirect assessment measures: the Motivational Assessment Scale (MAS) and the Questions About Behavioral Function (QABF). For three participants, results from the indirect measures paralleled those of the FA in identifying the function of pica. The indirect measures were more cost and time efficient. Use of these indirect methods of assessment not only identified the functional relations that served to maintain pica, but also served as a simpler, less time consuming and accurate alternative to an FA. When an FA is time or cost prohibitive, use of the MAS and QABF may be helpful in suggesting the function of pica.

Body dysmorphic disorder (BDD) is characterized by an intense preoccupation with an imagined or real defect of an individual's physical appearance (Sobanski & Schmidt, 2000). This present study was an extension of the work of Mayville(1998, 1999), who studied BDD symptoms in a non-clinical multi-ethnic adolescent population. The results of Mayville's (1998, 1999) study indicated that African Americans have a more positive body image compared to any other ethnic group, and females demonstrated less satisfaction with their physical appearance in comparison to males. In addition to replicating Mayville's study (1998, 1999), this study focused on the component of social anxiety symptoms, and its relationship to body dysmorphic symptoms. It is imperative to study the relationship between BDD and social anxiety in adolescents, as there may be diminished functioning in terms of social and academic performance for those who are diagnosed with BDD (Albertini & Phillips, 1999). In this study scores were examined on two different scales: Social Physique Anxiety Scale (SPAS, Hart, Leary, & Rejeski, 1989) and the Fear of Negative Evaluations Scale (FNE, Watson & Friend, 1969) with their respective relationships to scores on the Body Image Rating Scale (BIRS, Mayville, Gipson, and Katz, 1997). As expected the SPAS contributed significantly and substantially to the prediction of scores on the BIRS in comparison to the FNE. The large squared semi-partial correlation coefficients in this study indicate that there is some overlap between the constructs of body dysmorphic disorder symptoms and the two measures of social anxiety symptoms. In this study, however, the Sex variable and Ethnicity variable did not contribute substantially to predicting scores on the BIRS.

No description available.