Implementing Restricted Maximum Likelihood Estimation in Structural Equation Models

Structural equation modeling (SEM) is now a generic modeling framework for many multivariate techniques applied in the social and behavioral sciences. Many statistical models can be considered either as special cases of SEM or as part of the latent variable modeling framework. One popular extension is the use of SEM to conduct linear mixed-effects modeling (LMM) such as cross-sectional multilevel modeling and latent growth modeling. It is well known that LMM can be formulated as structural equation models. However, one main difference between the implementations in SEM and LMM is that maximum likelihood (ML) estimation is usually used in SEM, whereas restricted (or residual) maximum likelihood (REML) estimation is the default method in most LMM packages. This article shows how REML estimation can be implemented in SEM. Two empirical examples on latent growth model and meta-analysis are used to illustrate the procedures implemented in OpenMx. Issues related to implementing REML in SEM are discussed.     

Context matters: increasing understanding with interactive Clicker Case studies

Although interactive technology is presumed to increase student understanding in large classes, no previous research studies have empirically explored the effects of Clicker Cases on students?? performance. A Clicker Case is a story (e.g., a problem someone is facing) that uses clickers (student response systems) to engage students in understanding the meaning of the science contained within the story. Using an experimental randomized Solomon design across 11 institutions, we found that Clicker Cases increased student understanding more than PowerPoint lectures in large introductory biology classrooms, although there was variation across institutions and topics. By examining student performance in conjunction with faculty experience, we found that strong Clicker Cases created dissonance, captured attention and involved students in interpreting data or making decisions. This study provides a model for collaborative research across multiple institutions and demonstrates the need for using multiple institutions and topics in research on education.     

How can action learning contribute to social capital?

This paper explores the contribution that action learning can make to the formation of social capital via experiences of action learning projects in NHS Pathology Services in the UK. The paper describes the development of action learning practice in recent years, reviews the notion of social capital and considers how action learning might contribute to its formation. A case study of action learning in the development of a locally unified pathology service is used to illustrate the processes by which actions and learnings may be transferred and extended from sets to contribute to organisational learning in wider systems and networks.     

Resources, capabilities, risk capital and the creation of university spin-out companies

The commercialization of university research has become an increasingly important issue, given concerns regarding licensing and the university's desire to maximize the returns to intellectual property (IP). In this paper, we assess the impact of university resources and routines/capabilities on the creation of spin-out companies. We find that both the number of spin-out companies created and the number of spin-out companies created with equity investment are significantly positively associated with expenditure on intellectual property protection, the business development capabilities of technology transfer offices and the royalty regime of the university. These results highlight the importance not just of resource stocks, but also of developing appropriate capabilities of technology transfer officers in spinning-out companies. The results suggest that universities and policy-makers need to devote attention to the training and recruitment of technology officers with the broad base of commercial skills.     

Evidence-based practice: the use and abuse of research

The authors of this position paper argue that there is currently very little evidence-based practice in relation to learners with severe learning disabilities (SLD) and profound and multiple learning disabilities (PMLD), and that which there is, has often been badly used and/or abused. More specifically, we argue that relevant educational research undertaken so far has a strong tendency towards: (i) conflating the need for common strategies to be universally used in teaching, with the ‘need’ for a common curriculum; (ii) quoting research that applies to children with certain types of SEND as though it applies to all children with SEND; (iii) assuming there is a homogeneity of learning disability among people with the same condition (for example Down's syndrome, autism) and (iv) encouraging assumptions that any academic progress, no matter how small the gain, is axiomatically superior in value for all pupils. The authors conclude that there is need for a new look at ‘evidence-based practice’ for these populations.     

A framework for monitoring instructional environments in a virtual world

Virtual worlds are gaining momentum as a platform for delivering simulation‐based educational experiences to students. However, a key aspect of virtual world‐based education that has received little attention is recording and analyzing students' in‐world actions. This capability is essential for assessing what students have learned through their simulation experience, and engaging the students in post‐simulation reflective learning. In this research, we present a framework for recording and analyzing students' actions in a virtual world. This framework is based upon pedagogical theories of exploratory and experiential learning, and is defined in a virtual‐world agnostic manner. The framework consists of two parts: (1) the Avatar Capabilities Model, which defines the educationally relevant actions that a student can take within a virtual world and (2) the Simulation Capture and Analysis toolkit that records and analyzes these actions, from an educational perspective. These analyses provide instructors with systematically collected evidence of the students' actions during their virtual world experience. This alleviates the need for instructors to directly observe students, thereby allowing for the scaling‐up of virtual worlds use in education. We have demonstrated the usefulness of the tool via a pilot study, with two students, in an emergency medical education context.     

Determining the Overall Impact of Interruptions During Online Testing

With an increase in the number of online tests, interruptions during testing due to unexpected technical issues seem unavoidable. For example, interruptions occurred during several recent state tests. When interruptions occur, it is important to determine the extent of their impact on the examinees’ scores. There is a lack of research on this topic due to the novelty of the problem. This article is an attempt to fill that void. Several methods, primarily based on propensity score matching, linear regression, and item response theory, were suggested to determine the overall impact of the interruptions on the examinees’ scores. A realistic simulation study shows that the suggested methods have satisfactory Type I error rate and power. Then the methods were applied to data from the Indiana Statewide Testing for Educational Progress‐Plus (ISTEP+) test that experienced interruptions in 2013. The results indicate that the interruptions did not have a significant overall impact on the student scores for the ISTEP+ test.     

Physical education for health and wellbeing: a discourse analysis of Scottish physical education curricular documentation

This paper provides an analysis of the discourses associated with physical education in Scotland's Curriculum for Excellence. We implement a poststructural perspective in order to identify the discourses that underpin the physical education sections of the Curriculum for Excellence ‘health and well-being’ documentation. Discourses related to physical activity and health are particularly prominent, along with a related concern with motor skill development. Our focus lies with the ways in which these discourses are likely to be taken up and deployed within Scottish educational establishments. The paper thus explores the ways in which these discourses might ‘work’ to produce specific effects on practitioners and pupils. This involves speculating about how practitioners and students might engage in specific practices relative to these discourses. We conclude that the discourses identified lend themselves to interpretation and negotiation in multiple ways in the context of Scottish physical education, with specific consequences for the experiences and subjectivities of practitioners and children.     

The Learning Transfer Big Picture

Do not forget learning transfer while crafting your learning and development initiatives. A research‐driven model allows your organization to adapt quickly and make real‐time adjustments to your learning strategy.