This article extends multivariate generalizability theory (MGT) to tests with different random-effects designs for each level of a fixed facet. There are numerous situations in which the design of a test and the resulting data structure are not definable by a single design. One example is mixed-format tests that are composed of multiple-choice and…

An information-correction method for testlet-based tests is introduced. This method takes advantage of both generalizability theory (GT) and item response theory (IRT). The measurement error for the examinee proficiency parameter is often underestimated when a unidimensional conditional-independence IRT model is specified for a testlet dataset. By…

The "Common European Framework of Reference" (CEFR; Council of Europe, 2001) provides a competency model that is increasingly used as a point of reference to compare language examinations. Nevertheless, aligning examinations to the CEFR proficiency levels remains a challenge. In this article, we propose a new, level-centered approach to…

A paradoxical phenomenon of decreases in reliability as the number of elements averaged over increases is shown to be possible in multifacet reliability procedures (intraclass correlations or generalizability coefficients). Conditions governing this phenomenon are presented along with implications and cautions. (Author)

Items previously described as "negatively worded" are redefined as "connotatively inconsistent" because this term has a broader base for generalization. Using generalizability theory with a sample of 102 graduate students, the study showed that connotatively consistent and reversed connotatively inconsistent items were not…

This paper describes a four-step approach to constructing diagnostic test profiles that provide precise but practical information on students' instructional needs. A test of pronoun use was constructed to represent 32 categories of usage defined by different combinations of five factors in a domain. (Author/LMO)

This study applied generalizability theory to investigate the contributions of persons, items, sections, and language backgrounds to the score dependability of the Test of English for International Communication (TOEIC). I replicated and extended Brown's (1999) study of the Test of English as a Foreign Language (TOEFL), using data from two…

This paper reports the results of generalizability theory (G) analyses done for new writing and speaking tasks for the Test of English as a Foreign Language (TOEFL). For writing, a special focus was placed on evaluating the impact on the reliability of the number of raters (or ratings) per essay (one or two) and the number of tasks (one, two, or…

Tables of specifications are used to guide test developers in sampling items and maintaining consistency from form to form. This paper is a generalizability study of the American College Testing Program (ACT) Achievement Program Mathematics Test (AAP), with the content areas of the table of specifications representing multiple dependent variables.…

Pronunciation is regarded as a valuable subskill in foreign language teaching and testing. Its quality is commonly assessed in a global way by having examinees read aloud. An atomistic test is a more systematic and explicit approach. Such a test would consist of about 40 items, use recorded performances, and draw on an inventory of pronunciation…

The concepts of universe of admissible observation and universe of generalization from the generalizability theory were applied to calculate the intraclass correlation coefficient of a licensure test. The internal consistency coefficient of a dichotomously scored test is identical to the intraclass correlation coefficient of a two-facet design.…