The kappa coefficient and the chi square statistic are used as indices of agreement between two judges' ratings of a set of stimuli on a nominal scale. In this article, the logic of these indices is extended to weighted multi-way cases. (Author/JKS)

It is shown that the rank-biserial correlation coefficient is a linear function of the U-statistic (Mann and Whitney), so that a test of group mean difference is equivalent to a test of zero correlation for the rank-biserial coefficient. (RC)

Exact conditional tests of independence in cross-classification tables are formulated based on chi square and other statistics with stronger operational interpretations, such as some nominal and ordinal measures of association. Guidelines for table dimensions and sample sizes for which the tests are economically implemented on a computer are…

Presents two experiments concerning confirmatory bias in the way counselors collect data to test their hypotheses. Counselors were asked either to develop their own clinical hypothesis or were given a hypothesis to test. Confirmatory bias in hypothesis testing was not supported in either experiment. (JAC)

Handicapped students' scores on timed and untimed editions of the Scholastic Aptitude Test (SAT) were studied. Of the approximately 1800 students studied, 79 percent were learning disabled. Handicapped students' performance improved with extended time, the increase being greater than that for nonhandicapped students tested with extra time.…

Behavioral assessment is seen as a way of individualizing programing for special needs pupils by obtaining information (through such techniques as self-monitoring, direct observation, and permanent product review) and evaluating information (social validation and single-case research designs). (CL)

It is argued that the manual of the Test of Adolescent Language lacks precision even though the authors explicitly state that the test may be used for the purpose of identifying intraindividual differences. To alleviate this problem, a table containing statistically significant differences between subtest standard scores is presented. (Author/CL)

Performance of 286 normal children (grades K-9) on the De Renzi and Faglioni form of the Token Test and the De Renzi and Ferrari Reporter's Test were analyzed. Two different scoring conventions were compared: number correct versus weighted scores. Normative data are presented by grade level and age. Specific administration and scoring procedures…

Investigated the validity of a microcomputer-administered version of the Marital Adjustment Test (MAT) (N=100). Results showed no significant differences in the computer and the paper-pencil versions, regardless of order of presentation. (JAC)

A comparison of the Wechsler Intelligence Scale for Children-Revised (WISC-R) and the Kaufman Assessment Battery for Children (K-ABC) with 37 mentally retarded children revealed that the WISC-R Full Scale IQ resulted in scores significantly lower than the K-ABC Mental Processing Composite. (CL)

Discrepancy scores for four global scales and the Nonverbal Scale were examined for the Kaufman Assessment Battery for Children using the standardization sample as the data source. The range of subtest scores and the profile variability of subtest scores were determined for each sex, race, and age level. (Author/CL)

An exact probability test for use with certain Likert-type scales is presented. The procedure assumes equally-spaced points, independence of subjects' responses, and that each point has an equal likelihood of response for each subject. Tables for critical values are presented. (Author/JKS)

A procedure is specified for testing the significance of predicted trends in k independent correlations. An example is also provided for illustrative purposes. (Author)

Adaptive multistage testlet (MST) designs appear to be gaining popularity for many large-scale computer-based testing programs. These adaptive MST designs use a modularized configuration of preconstructed testlets and embedded score-routing schemes to prepackage different forms of an adaptive test. The conditional information targeting (CIT)…

Effects of two methods of test-taking behavior on performances were compared. Further study is in order of the hypothesis that students who take exams in small groups working for a common grade will perform better then when taking exams individually. (Author/JB)