Search

There has been great progress towards Reinforcement Learning (RL) approaches that can achieve expert performance across a wide range of domains. However, researchers have not yet applied these models to learn expert models for educationally relevant tasks, such as those taught within tutoring systems and educational games. In this paper we explore the use of Proximal Policy Optimization (PPO) [25] for learning expert models for tutoring system tasks. We explore two alternative state and action space representations for this RL approach in the context of two intelligent tutors (a fraction arithmetic tutor and a multicolumn addition tutor). We compare the performance of these models to a computational model of learning built using the Apprentice Learner architecture. To evaluate these models, we look at whether they achieve mastery and how many training opportunities they take to do so. Our analysis shows that at least one PPO model is able to successfully achieve mastery within both tutors, suggesting that RL models might be successfully applied to learn expert models for educationally relevant tasks. We find that the Apprentice model also achieves mastery, but requires substantially less training (thousands of times less examples) than PPO. Finally, we find that there is an interaction between the PPO representation and task (one representation is better for one tutor and the other representation is better for the other tutor), suggesting that the design of the state and action representations for RL is important for success. Our work showcases the promise of RL for expert model discovery in educationally relevant tasks and highlights limitations and challenges that need further research to overcome. [For the full proceedings, see ED615472.]

Instructional Systems Design is the practice of creating of instructional experiences that make the acquisition of knowledge and skill more efficient, effective, and appealing [18]. Specifically in designing courses, an hour of training material can require between 30 to 500 hours of effort in sourcing and organizing reference data for use in just the preparation of course material. In this paper, we present the first system of its kind that helps reduce the effort associated with sourcing reference material and course creation. We present algorithms for document chunking and automatic generation of learning objectives from content, creating descriptive content metadata to improve content-discoverability. Unlike existing methods, the learning objectives generated by our system incorporate pedagogically motivated Bloom's verbs. We demonstrate the usefulness of our methods using real world data from the banking industry and through a live deployment at a large pharmaceutical company. [For the full proceedings, see ED593090.]

The influence of three sets of instructions on teacher/course evaluation by student raters was investigated. Students did not differ in their teacher/course evaluation ratings when the instructions specified the evaluation results would be used: (1) only by the instructor; (2) by the administration; or (3) by students for course selection purposes. Secondarily, the effects of rank and class level were also examined. It was found that the evaluation of graduate courses was significantly higher than that of undergraduate courses. With respect to academic rank, graduate teaching assistants received higher ratings than did either assistant or full professors. (Author/LBH)

A logically sound and methodologically easy technique for improving the construct validity of instruments and developed for relatively new concepts is described. It is based on the premise that if an instrument is valid, so should be each item within it and, therefore, the latter should sustain in field setting the behavioral implications derived from the theory underlying the construct. This idea was examined for 87 items of seven factorially "pure" scales, using a nearly random sample of about 7,000 subjects, and testing 10 such hypotheses for each item. The results are described. (Author/BB)

Effects of the physical and social environment on the feeding of severely and multiply handicapped students are discussed. The role of such physical environmental factors as sound, visual stimuli, light, movement, touch, clothing and furniture is considered. Among the social influences operating are the student's physical abilities and disabilities, current health, likes and dislikes, and past feeding activities. Suggestions are made for altering the environment to enhance the development of feeding skills. (CL)

This study explored the potent sources of stress for a sample of teachers, investigated the differences in stress predictors for men and women, and examined the consequences of stress for men and women. Twenty-five public school teachers participated; sixteen of these were women and nine were men. The study was conducted through interviews with the teachers. Stressors were categorized under three broad categories: job-related factors, interpersonal factors, and general organizational factors. The interviews sought to determine the perceived role demands that an individual is unable to satisfy and the extent of symptoms of strain. Results indicated that quantitative role overload was the stress most frequently experienced by the subjects, followed by role ambiguity and role conflict. Stressors generally tended to be more potent for women than for men. Job variety, job responsibility, and interpersonal characteristics were found to be salient stressors for women. No stressor appeared to be salient for men. Organizational stressors were minimally potent among both men and women. While women were more likely to report the experience of role stress than were men, stress resulted in the experience of psychological strain for both men and women. Stress had predictable effect on men. For women, however, role stress was inversely related to physiological and behavioral strain symptoms. (JD)

This working group is a follow-up initiative that will allow the mathematics education community to engage with adaptations of lesson study designed to address equity in the mathematics classroom. This year we will examine four adaptations (Teaching Through Problem-solving, Focal Students and Equity Commentator from the California Action Network for Mathematics Excellence and Equity (CANMEE), Rights of the Learner, and Continuous Improvement Lesson Study (CILS)), identifying impacts on equitable teaching practices (i.e, practices that involve all students in cognitively demanding mathematics). Participants will view a video-recorded research lesson that grounds the discussion of the adaptations. The group aims to publish an edited volume on equity-focused lesson study, with a broader goal of supporting participants to develop lesson study that integrates equity across different contexts (e.g. K-12 teachers, preservice teachers, graduate students, and mathematics teacher educators). [ABSTRACT FROM AUTHOR]

This is a report on the working group aimed at providing a structure for Mathematics Teacher Educators (MTEs) to engage in Continuous Improvement Lesson Study (CILS), a process of MTE professional development that involves working collaboratively to design a lesson, improve preservice teacher (PST) learning, and MTE practice. [ABSTRACT FROM AUTHOR]

In soil, mites (Acari) are the most abundant and diverse arthropods. They play an important ecological role in maintaining soil health. Among acarines, mesostigmatids are the most abundant group occurring in soil. The present study was undertaken in order to explore the diversity and seasonal occurrence of mesostigmatid mites in three tea gardens maintaining different agro-practices (viz., conventional, organic and biorational) in Assam India. The survey was done over 2 years (August 2016–July 2018). Mesostigmatid mites were extracted by using the heat desiccation method in a Tullgren funnel. The study revealed the occurrence of 41 species of mesostigmatid mites belonging to 17 families and 31 genera in the three tea gardens. Populations of mesostigmatid mites exhibited a fluctuating trend being higher in monsoon (June- September), slightly declining in post monsoon period (October- November), moderate during winter (December- February), and noticeably reduced in summer (March-May). Shannon- diversity index (H′) was found to be highest in the organic tea garden (3.01), followed by biorational tea garden (2.76), and conventional tea garden (2.41). [ABSTRACT FROM AUTHOR]