Catálogo de publicaciones - libros

The effect of playing chess on problem solving was explored using Rasch scaling and hierarchical linear modelling. It is suggested that this combination of Rasch scaling and multilevel analysis is a powerful tool for exploring such areas where the research design has proven difficult in the past.

Explanatory style was measured with the Children’s Attributional Style Questionnaire (CASQ) in 243 students from Grades 3 to 9 on two occasions separated by almost three years. The CASQ was analysed with the Rasch model, with separate analyses also being carried out for the Composite Positive (CP) and Composite Negative (CN) subscales. Each of the three scales met the requirements of the Rasch model, and although there was some slight evidence of gender bias, particularly in CN, no grade level differences were found.

This Australian study developed and used scales to measure the strength and coherence of students’, teachers’ and scientists’ views, beliefs and attitudes in relation to science, technology and society (STS). The scales assessed views on: (a) science, (b) society and (c) scientists. The consistency of the views of students was established using Rasch scaling. In addition, structured group interviews with teachers provided information for the consideration of the problems encountered by teachers and students in the introduction of STS courses. The strength and coherence of teachers’ views on STS were higher than the views of scientists, which were higher than those of students on all three scales. The range of STS views of scientists, as indicated by the standard deviation of the scores, was consistently greater than the range of teachers’ views. The interviews indicated that a large number of teachers viewed the curriculum shift towards STS positively. These were mainly the younger teachers, who were enthusiastic about teaching the issues of STS. Some of the teachers focused predominantly upon covering the content of courses in their classes rather than discussing STS issues. Unfortunately, it was found in this study that a significant number of teachers had a limited understanding of both the nature of science and STS issues. Therefore, this study highlighted the need for the development of appropriate inservice courses that would enable all science teachers to teach STS to students in a manner that would provide them with different ways of thinking about future options. It might not be possible to predict with certainty the skills and knowledge that students would need in the future. However, it is important to focus on helping students to develop the ability to take an active role in debates on the uses of science and technology in society, so that they can look forward to the future with optimism.

This paper explores the application of the Rasch model in developing and subsequently analysing data derived from a series of rating scales that measures the preparedness of participants to engage in workplace rehabilitation. Brief consideration is given to the relationship between effect and learning together with an overview of how the rating scales were developed in terms of their content and processes. Emphasis is then placed on how the principles of Rasch scaling can be applied to rating scale calibration and analysis. Data derived from the application of the Workplace Rehabilitation Scale are then examined for the evidence of differentiated item function (DIF).

User satisfaction is considered to be one of the most widely used measures of information and communication technology (ICT) implementation success. Therefore, it is interesting to examine the possibility of creating a general measure of user satisfaction to allow for diversity among users and diversity in the ICT-related tasks they perform. The end user computing satisfaction instrument (EUCSI) developed by Doll and Torkzadeh (1988) was revised and used as a general measure of user satisfaction. The sample was 881 government employees selected from 144 organizations across all regions of Bali, Indonesia. The data were analysed with Rasch Unidimensional Models for Measurement (RUMM) software. All the items fitted the model reasonably well with the exception of two items which had the chi-square probability < 0.05 and one item which had disordered threshold values. The overall power of the test-of-fit was excellent.

In this paper we discuss complexities of measurement that can arise in a multidimensional situation. All of the complexities that can occur in a unidimensional situation, such as polytomous response formats, item dependence effects, and the modeling of rater effects such as harshness and variability, can occur, with a correspondingly greater degree of complexity, in the multidimensional case also. However, we will eschew these, and concentrate on issues that arise due to the inherent multidimensionality of the situation. First, we discuss the motivations for multidimensional measurement models, and illustrate them in the context of a state-wide science assessment involving both multiple choice items and performance tasks. We then describe the multidimensional measurement model (NMRCML). This multidimensional model is then applied to the science assessment data set to illustrate two issues that arise in multidimensional measurement. The first issue is the question of whether one should (or perhaps, can) design items that relate to multiple dimensions. The second issue arises when there is more than one form of multidimensionality present in the item design: should one use just one of these dimensionalities, or some, or all? We conclude by discussing further issues yet to be addressed in the area of multidimensional measurement.

Although models for the unfolding response processes are single-peaked, their information functions are generally twin peaked though in rare exceptions may be single-peaked. This contrasts with the models for cumulative response process which are monotonic and for which the information function is always single-peaked. in addition, in the cumulative models, the information is a maximum when the person and item locations are identical, whereas for most unfolding models, the information is minimum at this point. The general unfolding model (Luo, 1998, 2000) for dichotomous responses, of which all proposed probabilistic unfolding models are special cases, makes explicit two item parameters, one the location of the item, the other the latitude of acceptance which defines the thresholds between which the positive response is more likely than the negative response. The current paper carries on further studies of this general model, particularly the information function of the general model. First, the information function of this general unfolding model is resolved into two components, one related to the latitude of acceptance, the other related only to the distance between the person and item locations. The component relative to the latitude of acceptance has a maximum value at the affective thresholds, but is moderated by the operational function. Second, the contrasts between the information functions for unfolding and cumulative models is reconciled by showing that the key points for maximising the information is where the probability of the positive and negative responses are equal, which is the threshold where the person and item locations are identical in the cumulative models and are the two thresholds which define the latitude of acceptance in the unfolding models. As a result of the explication of these relationships, it is shown that some single peaked response functions have no defined information when the person is at the location of the item.

This chapter traces the developments in Rasch measurement, and its corresponding refinement in both its application and programs to compute pertinent item and person parameters, The underlying principles of conjoint measurement are discussed, and its implications for education and research in social sciences are highlighted.

Four sets of analyses were conducted on the 1996 Course Experience Questionnaire data. Conventional item analysis, exploratory factor analysis and confirmatory factor analysis were used, Finally, the Rasch measurement model was applied to this data set. This study was undertaken in order to compare conventional analytic techniques with techniques that explicitly set out to implement genuine measurement of perceived course quality. Although conventional analytic techniques are informative, both confirmatory factor analysis and in particular the Rasch measurement model reveal much more about the data set, and about the construct being measured. Meaningful estimates of individual students’ perceptions of course quality are available through the use of the Rasch measurement model. The study indicates that the perceived course quality construct is measured by a subset of the items included in the CEQ and that seven of the items of the original instrument do not contribute to the measurement of that construct. The analyses of this data set indicate that a range of analytical approaches provide different levels of information about the construct. In practice, the analysis of data arising from the administration of instruments like the CEQ would be better undertaken using the Rasch measurement model.