Assessment in the context of foreign language learning can be difficult and time-consuming for instructors. Distinctive from other domains, language learning often requires teachers to assess each student's ability to speak the language, making this process even more time-consuming in large classrooms which are particularly common in post-secondary settings; considering that language instructors often assess students through assignments requiring recorded audio, a lack of tools to support such teachers makes providing individual feedback even more challenging. In this work, we seek to explore the development of tools to automatically assess audio responses within a college-level Chinese language-learning course. We build a model designed to grade student audio assignments with the purpose of incorporating such a model into tools focused on helping both teachers and students in real classrooms. Building upon our prior work which explored features extracted from audio, the goal of this work is to explore additional features derived from tone and speech recognition models to help assess students on two outcomes commonly observed in language learning classes: fluency and accuracy of speech. In addition to the exploration of features, this work explores the application of Siamese deep learning models for this assessment task. We find that models utilizing tonal features exhibit higher predictive performance of student fluency while text-based features derived from speech recognition models exhibit higher predictive performance of student accuracy of speech.

It has been shown in multiple studies that expert-created on-demand assistance, such as hint messages, improves student learning in online learning environments. However, there are also evident that certain types of assistance may be detrimental to student learning. In addition, creating and maintaining on-demand assistance are hard and time-consuming. In 2017-2018 academic year, 132,738 distinct problems were assigned inside ASSISTments, but only 38,194 of those problems had on-demand assistance. In order to take on-demand assistance to scale, we needed a system that is able to gather new on-demand assistance and allows us to test and measure its effectiveness. Thus, we designed and deployed TeacherASSIST inside ASSISTments. TeacherASSIST allowed teachers to create on-demand assistance for any problems as they assigned those problems to their students. TeacherASSIST then redistributed on-demand assistance by one teacher to students outside of their classrooms. We found that teachers inside ASSISTments had created 40,292 new instances of assistance for 25,957 different problems in three years. There were 14 teachers who created more than 1,000 instances of on-demand assistance. We also conducted two large-scale randomized controlled experiments to investigate how on-demand assistance created by one teacher affected students outside of their classes. Students who received on-demand assistance for one problem resulted in significant statistical improvement on the next problem performance. The students' improvement in this experiment confirmed our hypothesis that crowd-sourced on-demand assistance was sufficient in quality to improve student learning, allowing us to take on-demand assistance to scale. [This paper was published in: "L@S '20, August 12-14, 2020, Virtual Event, USA," ACM, 2020, pp. 115-124.]

The use of computer-based systems in classrooms has provided teachers with new opportunities in delivering content to students, supplementing instruction, and assessing student knowledge and comprehension. Among the largest benefits of these systems is their ability to provide students with feedback on their work and also report student performance and progress to their teacher. While computer-based systems can automatically assess student answers to a range of question types, a limitation faced by many systems is in regard to open-ended problems. Many systems are either unable to provide support for open-ended problems, relying on the teacher to grade them manually, or avoid such question types entirely. Due to recent advancements in natural language processing methods, the automation of essay grading has made notable strides. However, much of this research has pertained to domains outside of mathematics, where the use of open-ended problems can be used by teachers to assess students' understanding of mathematical concepts beyond what is possible on other types of problems. This research explores the viability and challenges of developing automated graders of open-ended student responses in mathematics. We further explore how the scale of available data impacts model performance. Focusing on content delivered through the ASSISTments online learning platform, we present a set of analyses pertaining to the development and evaluation of models to predict teacher-assigned grades for student open responses.

We examine the use of computer-based learning in the classroom and the effect of immediate feedback on student performance. Since it is well known in educational research that it is possible to observe a Matthew Effect in which the rich get richer, we wanted to see if feedback was useful for low prior knowledge students, as defined by students whose pretest score was at or below the median. In this counterbalanced randomized controlled trial, 243 tenth and eleventh grade mathematics students were exposed to one of two conditions, as we measured their learning from: 1) immediate feedback (where the computer told them correctness and they could also ask for hints) or 2) practice only (where they received feedback only after taking a posttest). Results suggest that immediate feedback from computer-based learning tasks benefit both high and low prior knowledge students, with low prior knowledge students exhibiting greater gains. The implications of these findings support further investigation into the use of computer-based learning tasks that provide immediate feedback.

Persistence is a crucial trait for learners. However, a common issue in mastery learning is that persistence is not always productive, a construct termed wheel-spinning. In this paper, we extend on prior work to develop wheel-spinning detectors in the ASSISTments learning system that distinguish between non-persistence, productive persistence and wheel-spinning. To understand how quickly we can detect each state, we use data from different numbers of practice opportunities and compare model performance across student-problem set pairs. We identify that a model constructed using data from the first nine practice opportunities outperforms models using less practice data. However, it is possible to differentiate students who will eventually wheel-spin from learners who will persist productively using data from only the first three opportunities. Wheel-spinning can be differentiated from non-persistence from the first five opportunities, and non-persistence can be differentiated from productive persistence from the first seven opportunities. These results show that early differentiation between wheel-spinning and productive persistence is feasible. These detectors relied upon hint requests, the correctness of prior opportunities, and the amount of practice and time on the skill. Identifying predictive features offer insights into the impact of in-system behaviors on wheel-spinning and guide the system design.

A prominent issue faced by the education research community is that of student attrition. While large research efforts have been devoted to studying course-level attrition, widely referred to as dropout, less research has been focused on finer-grained assignment-level attrition commonly observed in K-12 classrooms. This later instantiation of attrition, referred to in this paper as stopout, is characterized by students failing to complete their assigned work, but the cause of such behavior are not often known. This becomes a large problem for educators and developers of learning platforms as students who give up on assignments early are missing opportunities to learn and practice the material which may affect future performance on related topics; similarly, it is difficult for researchers to develop, and subsequently difficult for computer-based systems to deploy interventions aimed at promoting productive persistence once a student has ceased interaction with the software. This difficulty highlights the importance to understand and identify early signs of stopout behavior in order to provide aid to students pre-emptively to promote productive persistence in their learning. While many cases of student stopout may be attributable to gaps in student knowledge and indicative of struggle, student attributes such as grit and persistence may be further affected by other factors. This work focuses on identifying different forms of stopout behavior in the context of middle school math by observing student behaviors at the sub-problem level. We find that students exhibit disproportionate stopout on the first problem of their assignments in comparison to stopout on subsequent problems, identifying a behavior that we call refusal, and use the emerging patterns of student activity to better understand the potential causes underlying stopout behavior early in an assignment.

The increased usage of computer-based learning platforms and online tools in classrooms presents new opportunities to not only study the underlying constructs involved in the learning process, but also use this information to identify and aid struggling students. Many learning platforms, particularly those driving or supplementing instruction, are only able to provide aid to students who interact with the system. With this in mind, student persistence emerges as a prominent learning construct contributing to students success when learning new material. Conversely, high persistence is not always productive for students, where additional practice does not help the student move toward a state of mastery of the material. In this paper, we apply a transfer learning methodology using deep learning and traditional modeling techniques to study high and low representations of unproductive persistence. We focus on two prominent problems in the fields of educational data mining and learner analytics representing low persistence, characterized as student stopout, and unproductive high persistence, operationalized through student wheel spinning, in an effort to better understand the relationship between these measures of unproductive persistence (i.e., stopout and wheel spinning) and develop early detectors of these behaviors. We find that models developed to detect each within and across-assignment stopout and wheel spinning are able to learn sets of features that generalize to predict the other. We further observe how these models perform at each learning opportunity within student assignments to identify when interventions may be deployed to best aid students who are likely to exhibit unproductive persistence.

Considerable attention has been given to methods for knowledge estimation, a category of methods for automatic assessment of a student's degree of skill mastery or knowledge at a specific time. Knowledge estimation is frequently used to make decisions about when a student has reached mastery and is ready to advance to new material, but there has been little work to forecast how far a student is from mastery or predict how much more practice the student will need before he or she will reach mastery. This article presents a method for predicting the point at which a student will reach skill mastery within an adaptive learning system, based on current approaches to estimating student knowledge. We apply this technique to two popular methods of modeling student learning - Bayesian knowledge tracing and performance factors analysis - and compare prediction correctness. Potential applications and future steps for improving the method are discussed.

Handwriting practice is the most time-consuming activity for learners of Chinese as a foreign language (CFL). CFL instructors report allocating at least one third of their course time to handwriting practice although it prevents students from engaging in meaningful communication, especially in the earliest stages of learning. Given the amount of time students spend in a college course is relatively fixed, the preregistered study presented herein examines the best use of students' time when primary goals are word acquisition and communication. This work replicates a pilot study examining CFL word recognition in an online learning environment (ASSISTments) and the effects of supplemental handwriting practice. We examined word acquisition and recognition while manipulating condition (no-handwriting practice and with-handwriting practice), and posttest test point (1 [immediate], 2 [1 day delay], and 3 [1 week delay]). Two-way repeated measures analyses of variance revealed significant main effects for both condition and posttest test point in online and on-paper measures of word recognition and handwriting. Potential implications for CFL instruction and directions for future work are discussed.

Identifying the mathematical skills or knowledge components needed to solve a math problem is a laborious task. In our preliminary work, we had two expert teachers identified knowledge components of a state-wide math test and they only agreed only on 35% of the items. Previous research showed that machine learning could be used to correctly tag math problems with knowledge components at about 90% accuracy over more than 100 different skills with five-fold cross-validation. In this work, we first attempted to replicate that result with a similar dataset and were able to achieve a similar cross-validation classification accuracy. We applied the learned model to our test set, which contains problems in the same set of knowledge component definitions, but are from different sources. To our surprise, the classification accuracy dropped drastically from near-perfect to near-chance. We identified two major issues that cause of the original model to overfit to the training set. After addressing the issues, we were able to significantly improve the test accuracy. However, the classification accuracy is still far from being usable in a real-world application.