- 201 684 7712
Amruth Kumar
Ramapo College of New Jersey, Computer Science, Faculty Member
- Professor of Computer Science with research interests in Artificial Intelligence, Intelligent Tutoring Systems, Compu... moreProfessor of Computer Science with research interests in Artificial Intelligence, Intelligent Tutoring Systems, Computer Science education research and Computer Science education. Author of problets (problets.org) and epplets (epplets.org) - software tutoring suites for introductory programming in C /Java/Cedit
Research Interests:
Research Interests: Engineering, Computer Science, Signal Processing, Complexity, Theoretical Computer Science, and 13 moreApplication of Artificial Intelligence in Water Sciences Engineering, Probability, Failure Analysis, Representation, Ontologies, Fault diagnosis, Data Structures, Classes, Scalability, Model based Reasoning, Contents, Representation Politics, and Space Complexity
Research Interests:
Research Interests:
Research Interests:
Research Category/Full paper) – We analyzed the data collected in a Computer Science I course to quantify the relationship between programming projects, code-tracing assignments and course grade when online tests and closed lab... more
Research Category/Full paper) – We analyzed the data collected in a Computer Science I course to quantify the relationship between programming projects, code-tracing assignments and course grade when online tests and closed lab instruction were used in the course. We found that completion of programming projects was positively and moderately correlated with course grade; each completed project contributed nearly one sign grade to the course grade; the grade of students who had completed at least a given number of projects was four sign grades better than of those who had not and the difference was statistically significant; the mean course grade ranged from F for those who had completed 1 or fewer projects to A- for those who had completed 9 or more projects; and completion of later projects was indicative of higher grade in the course. Similarly, completion of code-tracing assignments was positively, but weakly correlated with course grade; the grade of students who had completed at least a given number of assignments was one letter grade better than of those who had not, and the difference was statistically significant; and the mean course grade ranged from C- for those who had completed 6 or fewer assignments to B+ for those who had completed 11 or more assignments. Concurrence among the course objectives, classroom instruction, assessment techniques, programming projects and assignments may be a pre-requisite for obtaining the results of this study. (Abstract)
Research Interests:
This Research to Practice Full Paper presents the experiences and lessons learned from five programs that provide financial awards and a holistic student support structure to low-income, academically talented students in Science,... more
This Research to Practice Full Paper presents the experiences and lessons learned from five programs that provide financial awards and a holistic student support structure to low-income, academically talented students in Science, Technology, Engineering, and Mathematics (STEM). This report synthesizes the experiences of a diverse set of institutions, both public and private, that vary in size and geographic location. We have experience supporting students from a range of disciplines with an emphasis on students studying Computer Science. The goals of this work are to (1) outline the decisions that must be considered when designing a financial award program; (2) describe the interventions we have implemented and underline the institutional contexts that have led to their success; (3) describe the unique challenges posed by the COVID pandemic; and (4) highlight key elements necessary for successful program implementation. We specifically discuss the challenges we have encountered when implementing existing best practices. We report observations and results, some of which buttress those reported in the literature. Our work is intended to serve as a guide for educators who wish to implement programs to support students from financially disadvantaged and/or historically marginalized groups. By sharing our experiences and pain points, we hope to make it easier for them to design and implement effective programs adapted to their institutional needs and contexts.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
(Research Category/Full paper) — A recent study conducted at the University of Oklahoma, a large research university attempted to use the grades on initial Computer Science courses to predict the success of Computer Science majors. We... more
(Research Category/Full paper) — A recent study conducted at the University of Oklahoma, a large research university attempted to use the grades on initial Computer Science courses to predict the success of Computer Science majors. We attempted to reproduce this study in a mid-sized liberal arts institution. We analyzed 15 years of data of students (majors as well as non-Computer Science majors) who had taken introductory Computer Science courses. We found that the better the grade on Computer Science I, the introductory course in the major, the better the cumulative GPA of the student upon graduation, and this applied to Computer Science majors as well as non-majors. All the students who had successfully graduated with a Computer Science degree had earned at least a C grade in the first three required courses: Computer Science I, Computer Science II and Data Structures. When we considered grades on six of the required courses in the Computer Science sequence, we found that students generally earned the same or lower grade on each subsequent course. Therefore, the performance of Computer Science majors on the first three courses in the required course sequence can reasonably be used as predictors of their success in the major. Finally, we found that Math SAT score was a good predictor of student success in Computer Science I as well as obtaining an undergraduate degree regardless of the major. Our study generalizes the results of the previous study and strengthens the results by finding that they are statistically significant. (Abstract)
Research Interests:
(Research Category/Full paper) — A recent study conducted at the University of Oklahoma, a large research university attempted to use the grades on initial Computer Science courses to predict the success of Computer Science majors. We... more
(Research Category/Full paper) — A recent study conducted at the University of Oklahoma, a large research university attempted to use the grades on initial Computer Science courses to predict the success of Computer Science majors. We attempted to reproduce this study in a mid-sized liberal arts institution. We analyzed 15 years of data of students (majors as well as non-Computer Science majors) who had taken introductory Computer Science courses. We found that the better the grade on Computer Science I, the introductory course in the major, the better the cumulative GPA of the student upon graduation, and this applied to Computer Science majors as well as non-majors. All the students who had successfully graduated with a Computer Science degree had earned at least a C grade in the first three required courses: Computer Science I, Computer Science II and Data Structures. When we considered grades on six of the required courses in the Computer Science sequence, we found that students generally earned the same or lower grade on each subsequent course. Therefore, the performance of Computer Science majors on the first three courses in the required course sequence can reasonably be used as predictors of their success in the major. Finally, we found that Math SAT score was a good predictor of student success in Computer Science I as well as obtaining an undergraduate degree regardless of the major. Our study generalizes the results of the previous study and strengthens the results by finding that they are statistically significant. (Abstract)
Research Interests:
Research Interests:
We propose principles behind component-ontological representation of function. In this representation, the function of a component is expressed in terms of its ports. The representation has many advantages:(i) The function of a component... more
We propose principles behind component-ontological representation of function. In this representation, the function of a component is expressed in terms of its ports. The representation has many advantages:(i) The function of a component can be represented in isolation of its environment. Therefore, libraries of function models can be built. These models are re-usable.(ii) The function model of a complex device can be built by composing the function models of its components. This helps preserve the fidelity of representation. ...
Research Interests:
This paper identifies design guidelines for the application of evolutionary techniques to the task of generating practice problems for learners in an Intelligent Tutoring System. To this end, we designed experiments that progressively... more
This paper identifies design guidelines for the application of evolutionary techniques to the task of generating practice problems for learners in an Intelligent Tutoring System. To this end, we designed experiments that progressively incorporated an increasing number of the characteristics we expect to find in our target application. These features included noisy evaluations, overspecialization, and the need to mitigate user fatigue resulting from interactive evaluations of practice problems. As we did so, we evaluated the potential of recent breakthroughs in coevolutionary learning theory and identified the tradeoff specific to educational applications.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
We have developed n web-based tutor for teaching and testing counfer-controlled loop concepts in C++. The tutor is designed to promote problem-based learning. It repeatedly generates problems, grades user's answers and... more
We have developed n web-based tutor for teaching and testing counfer-controlled loop concepts in C++. The tutor is designed to promote problem-based learning. It repeatedly generates problems, grades user's answers and providesfeedback about the correct answers. This paper describes the design ofthe tutor. outlines a test that we used to evaluate its eflectiveness. and presents the results of he test. The test confirmed our hypothesis I hat using the tutor helps improve student learning. The improvement is stofistically significant. This tutor can be usedforpraciice or testing in Computer Science I.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests: Engineering, Electronic Engineering, Computer Science, Information Technology, Education, and 15 moreComputer Engineering, Higher Education, Learning and Teaching, Curriculum, Literature Review, Computer Programming, Programming, CS, Computing Education, Novice Programming, Introductory Programming, Mainstream, E Learning, Meta Analysis and Systematic Review, and Enabling
We conducted a study to evaluate the effect of using software tutors on the self-efficacy of students - in particular, whether the type of activity covered by the software tutor correlated with any improvement in self-efficacy after using... more
We conducted a study to evaluate the effect of using software tutors on the self-efficacy of students - in particular, whether the type of activity covered by the software tutor correlated with any improvement in self-efficacy after using the tutor along the levels of Bloom's taxonomy; and whether any differential effects could be observed in the improvement of self-efficacy among the sexes and racial groups. The study was conducted over four semesters using two different software tutors. The collected data was analyzed using paired sample t-test and 2 × 2 ANOVA for three different sets of students - those who used the tutor, those who needed to use the tutor and those who learned one or more concepts by using the tutor. We found that if a significant difference was found among taxonomic levels of self-efficacy after using a software tutor, the improvement was statistically significantly greater on the taxonomic level directly relevant to the topic/activity of the tutor than any other level on Bloom's taxonomy except comprehension. Such a difference was found at least among those who actually learned one or more concepts using the tutor, if not everyone who used the tutor. In most cases, the improvement in self-efficacy resulting from the use of the software tutor was indistinguishable across sexes and racial groups.
Research Interests:
Research Interests: Computer Science and Writ
We wanted to cover mobile computing in our curriculum without incurring the costs of adding a new course to the curriculum or hiring a new instructor to the faculty roster. We did so through collateral learning, by incorporating mobile... more
We wanted to cover mobile computing in our curriculum without incurring the costs of adding a new course to the curriculum or hiring a new instructor to the faculty roster. We did so through collateral learning, by incorporating mobile computing into the projects of two existing upper-level Computer Science courses: Organization of Programming Languages and Artificial Intelligence. We recount our experience using collateral learning of mobile computing over the last four years: motivation, logistics, course-specific details, reception by students, results of course evaluations, challenges faced and solutions devised. Our experience affirms that collateral learning is an excellent option for incorporating emerging topics such as drone programming, cybersecurity and parallel computing into the curriculum at resource-strapped Computer Science departments.
Research Interests:
Research Interests:
Research Interests:
AI-ED community has hewed to rigorous evaluation of software tutors and their features. Most of these evaluations were done in-ovo or in-vivo. Can the results of these evaluations be replicated in in-natura evaluations? In our experience,... more
AI-ED community has hewed to rigorous evaluation of software tutors and their features. Most of these evaluations were done in-ovo or in-vivo. Can the results of these evaluations be replicated in in-natura evaluations? In our experience, the evidence for such replication has been mixed. We propose that the features of tutors that are found to be effective in-ovo/in-vivo might need motivational supports to also be effective in-natura. We speculate that some features may not transfer to in-natura use even with supports. Recognition of these issues might bridge the gap between AI-ED community and educational community at large.
Research Interests:
We present a preliminary investigation into applying dimension extraction methods from coevolutionary algorithm theory to the analysis of student-problem performance in a computer programming instruction context. Specifically, we explore... more
We present a preliminary investigation into applying dimension extraction methods from coevolutionary algorithm theory to the analysis of student-problem performance in a computer programming instruction context. Specifically, we explore using the dimension extraction coevolutionary algorithm (DECA) from coevolution and co-optimization theory, which identifies structural relationships amongst learners and tests by constructing a geometry encoding how learner performance can be distinguished in fundamentally different ways. While DECA was developed for software learners and tests, its foundational ideas can in principle be applied to data generated by human students taking real tests. Here we apply DECA's dimension-extraction algorithm to student-problem data from four semesters of an introduction to programming course where students used an online software tutor to solve a number of predesigned problems. Dimension extraction reveals structures (dimensions) that partially align w...
Research Interests:
Do students retain the programming concepts they have learned using software tutors over the long term? In order to answer this question, we analyzed the data collected by a software tutor on selection statements. We used the data of the... more
Do students retain the programming concepts they have learned using software tutors over the long term? In order to answer this question, we analyzed the data collected by a software tutor on selection statements. We used the data of the students who used the tutor more than once to see whether they had retained for the second session what they had learned during the first session. We found that students retained over 71% of selection concepts that they had learned during the first session. The more problems students solved during the first session, the greater the percentage of retention. Even when students already knew a concept and did not benefit from using the tutor, a small percentage of concepts were forgotten from the first session to the next, corresponding to transience of learning. Transience of learning varied with concepts. We list confounding factors of the study.
Research Interests:
Parsons puzzles are popular for programming education. Identifying the strategies used by students solving Parsons puzzles is of interest because they can be used to determine to what extent students use the strategies typically... more
Parsons puzzles are popular for programming education. Identifying the strategies used by students solving Parsons puzzles is of interest because they can be used to determine to what extent students use the strategies typically associated with programming expertise, and to provide feedback and monitor the progress of students in a tutor. We propose solution sequence as an approximation of the student’s strategy for solving Parsons puzzles. It is scalable in terms of both the size of the puzzle and the number of students solving the puzzle. We propose BNF grammar to represent desirable puzzle-solving strategies associated with expert programmers. This representation is extensible and agnostic to the puzzle-solving strategies themselves. Finally, we propose a best match parser that matches a student’s solution sequence against the BNF grammar of a desirable strategy and quantifies the degree to which the student’s solution conforms to the desirable strategy. As a proof of concept, we...
Research Interests:
In response to student feedback on a tutor on Parsons puzzles on the programming concept of sequence, we incorporated three features meant to improve the motivation of the student solving the puzzles. We compared the performance of... more
In response to student feedback on a tutor on Parsons puzzles on the programming concept of sequence, we incorporated three features meant to improve the motivation of the student solving the puzzles. We compared the performance of students before and after introducing these features. We found that introduction of motivational supports did not affect pre-post improvement, and therefore, the amount of learning. Students who were provided motivational supports spent more time per puzzle than those who were not.
Research Interests:
This Birds-of-a-Feather session is for anyone interested in the NSF Scholarships in STEM (S-STEM) program, including current and former Principal Investigators (PIs) and those planning to apply. The S-STEM program funds scholarships and... more
This Birds-of-a-Feather session is for anyone interested in the NSF Scholarships in STEM (S-STEM) program, including current and former Principal Investigators (PIs) and those planning to apply. The S-STEM program funds scholarships and activities to support low-income, academically talented students in STEM. Any institution of higher education may apply, and the program supports a variety of projects. Designing and implementing a successful S-STEM project is challenging. The goal of this session is to catalyze a community of practice for S-STEM PIs. It will provide an opportunity to discuss lessons learned and best practices for proposal writing, project implementation, and providing student support. Specific topics to be discussed include the following: (1) Understanding the solicitation requirements and common proposal mistakes; (2) Scholar recruitment and data-driven approaches for selection; (3) Cohort building including activities for students from different majors or class years and integration of new students into existing cohorts; and (4) Remediation strategies including proactive interventions and peer support. Session leaders will introduce each topic; participants will then join a breakout group discussion of one topic. Lastly, participants will be invited to join a Slack workspace dedicated to S-STEM best practices and lessons.