The requirement of parallel parts has long been the cornerstone of classic reliability theory. By recasting reliability in a structural equation framework, items, raters, or judges no longer need to be treated as equivalent entities. Instead, unique reliability estimates can be determined for each and collectively used to assess the maximal reliability of a weighted composite, with the composite reliability submitted to inferential test. Procedures are shown to generalize from single to multifactor applications. Ramifications of a structural approach to reliability determination are probed, and the dilemma posed by possible falsification of the true score hypothesis presented for individual researcher consideration.

The difference between mental health and mental ability measurement hinges on a single concept-zero. Dysfunctional mental health is manifested by symptoms defined as self-reported feelings of unpleasantness due to pathological causes. Symptoms can be meaningfully reported as present or absent whereas mental abilities are generally considered to be ever present in some positive amount. Absence of symptoms creates a population zero class with unknown membership and proportion. Inadvertent mixture of zero-and non-zero classes, as often occurs in community samples, biases symptom estimates of means, variances, and covariance for the non-zero class, resulting in what is herein referred to as the zero-problem.Two-part modeling is proposed as a means of circumventing the zero-problem. In Part I, zero-class sample members are identified and deleted. Part II provides users a symptoms research paradigm based on a multiplicative measurement model. Data are logarithmically transformed, and the log-normal distribution assumed. The hypothesis that symptom statements are unidimensional is tested by confirmatory factor analysis (CFA). If accepted, statements are combined into a weighted pathology score. Pathology scores can be correlated, corrected for attenuation, and used as input to multivariate statistical applications. Computer routines are provided as a user service.

Classical test theory has been ascribed the status of weak theory. Early development reflected a native realism, with variables being operationally defined. Subsequent developments in mathematics, statistics, and structural modeling, however, have rendered classical test theory problematic. Yet the basic framework is sound. Decomposition of an observed score into the sum of a true and an error component is inherently meaningful. What is required is a modernization that addresses past exclusions and deficiencies while retaining the classical framework. Consequently, a neo-classical test theory formalized as fourteen tenets is presented. Each tenet is accompanied by an elaborative discussion. Benefits accruing from and costs associated with theory modernization are discussed. Educational and developmental implications for three user types are identified.

A unifying theory of subject-centered scalability is offered that is grounded in structural true score modeling, is conceptually distinct from internal consistency and homogeneity as determined by item correlations, and is empirically confirmable. Scalability holds when item true scores are perfectly correlated but differ in their individual scale metric. The condition of scalability imposes constraints that allow individual item reliability to be estimated independently of scalability. Scalability is shown to imply unit rank and to be testable by a single-factor confirmatory factor analysis reinterpreted as a test of unit rank. High item correlations are shown, contrary to intuition, to be an insufficient condition for scalability. Conversely, low item correlations do not necessarily imply lack of scalability. A stepped decision-oriented procedure is offered as a guideline in summated rating scale construction.

No abstract