John Balla

Goodness of fit in confirmatory factor analysis: The effects of sample size and model parsimony

The purpose of the present investigation is to examine the influence of sample size (N) and model parsimony on a set of 22 goodness-of-fit indices including those typically used in confirmatory factor analysis and some recently developed indices. For sample data simulated from two known population data structures, values for 6 of 22 fit indices were reasonably independent ofN and were not significantly affected by estimating parameters known to have zero values in the population: two indices based on noncentrality described by McDonald (1989; McDonald and Marsh, 1990), a relative (incremental) index based on noncentrality (Bentler, 1990; McDonald & Marsh, 1990), unbiased estimates of LISRELs GFI and AGFI (Joreskog & Sorbom, 1981) presented by Steiger (1989, 1990) that are based on noncentrality, and the widely known relative index developed by Tucker and Lewis (1973). Penalties for model complexity designed to control for sampling fluctuations and to address the inevitable compromise between goodness of fit and model parsimony were evaluated.

Directions for Integration in Australia: Overview of a Report to the Commonwealth Schools Commission. Part II.

Abstract Drawing upon information from a national survey, this paper examines the extent to which integration is being achieved for children with disabilities in Australian schools. Technical and organizational issues leading to identification of major barriers to integration together with possible strategies to overcome these are discussed. A need for policy‐oriented research in this field is identified, particularly the provision of accurate prevalence figures and other data to help monitor the impact of policy. Overall, there is evidence of a national movement towards integration but typically this is occurring within a context of inadequate policy formulation, organizational development and resource provision.

Computers in the teaching process and software development: Co-operation the key

Micro-computer technology has been a part of our education environment since the late 1970s but it is evident that successful incorporation of this technology into the teaching process has been somewhat sporadic. A number of reasons could be advanced to explain this, including: lack of suitable hardware; lack of relevant and/or quality software; inadequate teacher training; and insufficient teacher time to design the changes in curriculum and modes of presentation required. It could be argued that some of these areas are being addressed currently via technological advances and lower pricing structures. However, curriculum applications in the form of units of instruction incorporating computer technology, and effective models of software development have not been systematically supported nor researched. This void is addressed by detailing a co-operative project between tertiary institutions and schools involving the creation, implementation and evaluation of units of instruction incorporating computer technology. In addition, this project addressed the issue of training of teachers and teacher trainees through the development of courseware, and also highlights the need for a systematic model for software development which is also described.

The Use of Verbal Self-Instruction Training to Promote Indirect Generalization

A battery of tests was administered to forty-eight institutionalized moderately and severely intellectually disabled men and twenty non-institutionalized moderately disabled men and women, providing data on language, conceptual tempo and manual dexterity. The groups were then randomly subdivided with half receiving training in the cognitive strategy of verbal self- instruction and the others being trained in the more traditional procedure of modelling and imitation. The primary tasks taught through these methods were life skills of high ecological validity for these subjects (i.e vacuuming, sandwich making and collating). The battery of tests was administered again following training to provide data relating to changes in these measures which should be described as indirect since no specific training was directed towards them. For both groups of subjects it was found that verbal self-instruction training resulted in significantly greater improvement on several of the measures. These changes were identifie...

The effects of reducing correlation of external anchors on test equating methods for the equivalent groups and non-equivalent groups designs☆

Abstract Six equating methods were compared: a one-parameter Item Response Theory (IRT) method; two equipercentile methods (direct and by frequency estimation); and three linear methods (Tucker, Levine Equally Reliable and Levine Unequally Reliable) in a situation in which different forms were administered to different groups, thus necessitating the use of an anchor test. The groups were simulated as either equivalent groups or groups of variable ability representing the two types of class groupings that can exist in schools (i.e. parallel or streamed classes). The correlation between the ability measured by an external anchor and the tests to be equated was systematically manipulated. A discrepancy index summarised the discrepancy of each equating method from an IRT criterion, an equipercentile criterion, and from each other. Large discrepancies were interpreted with the aid of graphs and discussed in terms of examinee indifference to the alternative transformations. The direct equipercentile and Levine Unequally Reliable methods were the only methods that consistently increased their level of the discrepancy from criterion following reduction in correlation for the two equatings examined in the equivalent groups design. For the non-equivalent groups design, a reduction in correlation resulted in a systematic effect in favour of those taking an easier form (usually the less able) for all equating methods. What was observed, however, was that for small reductions in correlation, the discrepancy of some of the equating methods from the IRT criterion was reduced. The implications of these findings are discussed and recommendations made for further work.