One aspect of higher order social cognition is empathy, a psychological construct comprising a cognitive (recognizing emotions) and an affective (responding to emotions) component. The complex nature of empathy complicates the accurate measurement of these components. The most widely used measure of empathy is the Interpersonal Reactivity Index (IRI). However, the factor structure of the IRI as it is predominantly used in the psychological literature differs from Davis’s original four-factor model in that it arbitrarily combines the subscales to form two factors: cognitive and affective empathy. This two-factor model of the IRI, although popular, has yet to be examined for psychometric support. In the current study, we examine, for the first time, the validity of this alternative model. A confirmatory factor analysis showed poor model fit for this two-factor structure. Additional analyses offered support for the original four-factor model, as well as a hierarchical model for the scale. In line with previous findings, females scored higher on the IRI than males. Our findings indicate that the IRI, as it is currently used in the literature, does not accurately measure cognitive and affective empathy and highlight the advantages of using the original four-factor structure of the scale for empathy assessments.
Emerging developments in artificial intelligence present significant opportunities to enhance equity and access to science and mathematics assessment content for students with disabilities. Artificial intelligence (AI) technologies may have the potential to support test developers in creating more inclusive assessments that better measure what students know and can do. But they must also consider the potential accessibility challenges or introduction of construct-irrelevant variance posed by these technologies. The purpose of this article is to provide a conceptual overview of the issues to be considered when creating and implementing large-scale science and mathematics assessments for students with disabilities. We discuss how AI has been utilized in large-scale assessments to date and describe the opportunities and potential pitfalls in the stages of the process: assessment design, development, administration, scoring, reporting, and data use. This article concludes with proposed priorities for research that will advance the responsible practice of AI in large-scale assessment that is inclusive, fair, and valid for students with disabilities. This article contributes to the growing body of information on AI applications for assessment by identifying the roles that AI can play in science and mathematics assessment practices and demonstrating how AI can inform approaches to equitable science, technology, engineering, and mathematics (STEM) learning.
Diagnostic assessments measure the knowledge, skills, and understandings of students at a smaller and more actionable grain size than traditional scale-score assessments. Results of diagnostic assessments are reported as a mastery profile, indicating which knowledge, skills, and understandings the student has mastered and which ones may need more instruction. These mastery decisions are based on probabilities of mastery derived from diagnostic classification models (DCMs).This report outlines a Bayesian framework for the estimation and evaluation of DCMs. Findings illustrate the utility of the Bayesian framework for estimating and evaluating DCMs in applied settings. Specifically, the findings demonstrate how a variety of DCMs can be defined within the same conceptual framework. Additionally, using this framework, the evaluation of model fit is more straightforward and easier to interpret with intuitive graphics. Throughout, recommendations are made for specific implementation decisions for the estimation process and the assessment of model fit.
1 November 2005 in SAE International Journal of Advances and Current Practices in Mobility
The future of vehicle safety will benefit greatly from precrash detection - the ability of a motor vehicle to predict the occurrence of an accident before it occurs. There are many different sensor technologies currently available for pre-crash detection. However no single sensor technology has demonstrated enough information gathering capability within the cost constraints of vehicle manufacturers to be used as a stand alone device. A proposed solution consists of combining information from multiple sensors in an intelligent computer algorithm to determine accurate precrash information. In this paper, a list of sensors currently available on motor vehicles and those that show promise for future development is presented. These sensors are then evaluated based on cost, information gathering capability and other factors. Cost sensitivity is lower in large commercial vehicles than sensitivity is lower in large commercial vehicles than in personal vehicles due to their higher initial cost and longer life span making them a good candidate for early adoption of such a system. This work forms the basis for ongoing research in developing an integrated object detection and avoidance precrash sensing system.
6 November 2023 in Journal of Open Source Software
In educational and psychological research, we are often interested in discrete latent states of individuals responding to an assessment (e.g., proficiency or non-proficiency on educational standards, the presence or absence of a psychological disorder). Diagnostic classification models (DCMs; also called cognitive diagnostic models [CDMs]) are a type of psychometric model that facilitates these inferences (Rupp et al., 2010; von Davier & Lee, 2019). DCMs are multi-dimensional, meaning that we can classify respondents on multiple latent attributes within a profile of skills. A Q-matrix is then used to define which items on the assessment measure each attribute. Using the pre-defined latent profiles and the Q-matrix, DCMs then estimate the probability that respondents are in profile, or have the corresponding pattern of proficiency, or presence, of the attributes. This means that DCMs are able to provide fine-grained feedback on specific skills that may need additional instruction in an educational context, or particular symptoms that may be contributing to a diagnosis in a psychological context. Finally, because DCMs are classifying respondents rather than placing them along a performance continuum, these models are able to achieve more reliable results with shorter test lengths (Templin & Bradshaw, 2013), reducing the burden on respondents. Given these benefits, the goal of measr is to make DCMs more accessible to applied researchers and practitioners by providing a simple interface for estimating and evaluating DCMs.
26 September 2019 in Applied Measurement in Education
As the use of diagnostic assessment systems transitions from research applications to large-scale assessments for accountability purposes, reliability methods that provide evidence at each level of reporting are needed. The purpose of this paper is to summarize one simulation-based method for estimating and reporting reliability for an operational, large-scale, diagnostic assessment system. This assessment system reports the results and associated reliability evidence at the individual skill level for each academic content standard and broader content strands. The system also summarizes results for the overall subject using achievement levels, which are often included in state accountability metrics. Results are summarized as measures of association between true and estimated mastery status for each level of reporting.
About Me
Assistant Director of Psychometrics at Accessible Teaching, Learning, and Assessment Systems (ATLAS).
Assistant Director of Psychometrics at Accessible Teaching, Learning, and Assessment Systems (ATLAS).