Examined the reliability and validity of a career commitment measure using employees (N=266) of newspaper and insurance companies. Results showed career commitment could be reliably measured and was operationally distinct from job involvement and organizational commitment. Discusses findings in terms of meaning of career commitment. (Author/ABL)

Cohen's kappa statistic is frequently used to measure agreement between two observers using categorical polytomies. Cohen's statistic is: shown to be inherently multivariate in nature; expanded to analyze ordinal and interval data; and extended to over two observers. A non-asymptotic test of significance is provided for the generalized statistic.…

Gulliksen's matched random subtests method is a graphical method to split a test into parallel test halves, allowing maximization of coefficient alpha as a lower bound to the classical test reliability coefficient. This problem is formulated as a zero-one programing problem solvable by algorithms that already exist. (TJH)

Deals with issues of validity and reliability in qualitative research in education. Discusses philosophical assumptions underlying the concepts of internal validity, reliability, and external validity or generalizability. Presents strategies congruent with a qualitative research perspective for ensuring the rigor and trustworthiness of findings.…

Notes that conventional practice in social work research is to recommend level of reliability close to 0.80 as minimum standard. Contends that needed reliability varies by situation: that in situations in which important decisions about individuals are being made, 0.90 provides better standard; whereas in descriptive survey research with large…

Presents a series of four studies testing the hypothesis that people seek out and prefer information about causal mechanisms rather than information about covariation. Concludes that people attempt to seek out causal mechanisms in developing a causal explanation for a specific event. (DR)

An alternative formula is presented for the reliability of a difference score that contains the correlation between true scores instead of the correlation between observed scores. This approach provides more useful information and yields values that are not as anomalous as those usually obtained. (SLD)

The idea of test consistency is illustrated, with reference to two sets of test scores. A mathematical model is used to explain the relative consistency and relative inconsistency of measurements, and a means of indexing reliability is derived using the model. Practical aspects of estimating reliability are considered. (TJH)

Identifies and evaluates alternative methods for weighting tests. Presents formulas for composite reliability and validity as a function of component weights and suggests a rational process that identifies and considers trade-offs in determining weights. Discusses drawbacks to implicit weighting and explicit weighting and the difficulty of…

Developed an index for assessing interrater agreement with respect to a single target using a multi-item rating scale. The variance of rater mean scale scores is used as the numerator of the agreement index. Studied four variants of a disattenuated agreement index that vary in the random response term used as the denominator. (SLD)

Investigated how dissimilar in composition and variability samples inducting reliability coefficients from prior studies were from the cited prior samples from which coefficients were generalized. Results from 20 articles show that citing reliability coefficients from prior studies as the basis for concluding new scores are reliable is only…

Provides a sample of seven published studies in different disciplines that inappropriately generalized reliability coefficients involving several raters to scores generated by a single rater. Score reliability when only one rater is used for scoring is lower than the score reliability for which two raters are used. (SLD)

Describes the development of the Moral Justification Scale, an objective measure of justice and care orientations. The scale was administered to 100 college students. Results imply that the Moral Justification Scale shows promise as an easily administered, objectively scored measure of Giligan's constructs of care and justice. (Author/MKA)

The population discrepancy between Cronbach's Coefficient Alpha (L. Cronbach, 1951) and scale reliability with fixed congeneric measure, uncorrelated errors, and sampling of subjects was studied. The difference is expressed in terms of the individual component violations of the assumption of equal tau-equivalence that is necessary and sufficient…

A number of school-based programs address sexual violence by focusing on adolescents' attitudes about rape or acceptance of rape myths. However, really problems exist in the literature regarding measurement of rape myth acceptance, including issues of reliability and validity. This paper addresses measurement reliability issues and reviews…