An official website of the United States government
his work-in-progress paper expands on a collaboration between engineering education researchers and machine learning researchers to automate the analysis of written responses to conceptually challenging questions in statics and dynamics courses (Authors, 2022). Using the Concept Warehouse (Koretsky et al., 2014), written justifications of ConcepTests (CTs) were gathered from statics and dynamics courses in a diverse set of two- and four-year institutions. Written justifications for CTs have been used to support active learning pedagogies which makes them important to investigate how students put together their problem-solving narratives of understanding. However, despite the large benefit that analysis of student written responses may provide to instructors and researchers, manual review of responses is cumbersome, limits analysis, and can be prone to human bias. In efforts to improve the analysis of student written responses, machine learning has been used in various educational contexts to analyze short and long texts (Burstein et al., 2020; Burstein et al., 2021). Natural Language Processing (NLP) uses transformer-based machine learning models (Brown et al., 2020; Raffel et al., 2019) which can be used through fine-tuning or in-context learning methods. NLP can be used to train algorithms that can automate the coding of written responses. Only a few studies for educational applications have leveraged transformer-based machine learning models further prompting an investigation into its use in STEM education. However, work in NLP has been criticized for heightening the possibility to perpetuate and even amplify harmful stereotypes and implicit biases (Chang et al., 2019; Mayfield et al., 2019). In this study, we detail the aim to use NLP for linguistic justice. Using methods like text summary, topic modeling, and text classification, we identify key aspects of student narratives of understanding in written responses to mechanics and statics CTs. Through this process, we seek to use machine learning to identify different ways students talk about a problem and their understanding at any point in their narrative formation process. Thus, we hope to help reduce human bias in the classroom and through technology by giving instructors and researchers a diverse set of narratives that include insight into their students’ histories, identities, and understanding. These can then be used towards connecting technological knowledge to students’ everyday lives.
more »
« less
WIP: Using Machine Learning to Automate Coding of Student Explanations to Challenging Mechanics Concept Questions
This work-in-progress paper describes a collaborative effort between engineering education and machine learning researchers to automate analysis of written responses to conceptually challenging questions in mechanics. These qualitative questions are often used in large STEM classes to support active learning pedagogies; they require minimum calculations and focus on the application of underlying physical phenomena to various situations. Active learning pedagogies using this type of questions has been demonstrated to increase student achievement (Freeman et al., 2014; Hake, 1998) and engagement (Deslauriers, et al., 2011) of all students (Haak et al., 2011). To emphasize reasoning and sense-making, we use the Concept Warehouse (Koretsky et al., 2014), an audience response system where students provide written justifications to concept questions. Written justifications better prepare students for discussions with peers and in the whole class and can also improve students’ answer choices (Koretsky et al., 2016a, 2016b). In addition to their use as a tool to foster learning, written explanations can also provide valuable information to concurrently assess that learning (Koretsky and Magana, 2019). However, in practice, there has been limited deployment of written justifications with concept questions, in part, because they provide a daunting amount of information for instructors to process and for researchers to analyze. In this study, we describe the initial evaluation of large pre-trained generative sequence-to-sequence language models (Raffel et al., 2019; Brown et al., 2020) to automate the laborious coding process of student written responses. Adaptation of machine learning algorithms in this context is challenging since each question targets specific concepts which elicit their own unique reasoning processes. This exploratory project seeks to utilize responses collected through the Concept Warehouse to identify viable strategies for adapting machine learning to support instructors and researchers in identifying salient aspects of student thinking and understanding with these conceptually challenging questions.
more »
« less
- Award ID(s):
- 2135190
- PAR ID:
- 10352607
- Date Published:
- Journal Name:
- ASEE Annual Conference proceedings
- ISSN:
- 1524-4644
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)The shift to remote teaching with the COVID-19 pandemic has made delivery of concept- based active learning more challenging, especially in large-enrollment engineering classes. I report here a modification in the Concept Warehouse to support delivery of concept questions. The new feature allows instructors to make students’ reasoning visible to other students by showing selected written explanations to conceptually challenging multiple-choice questions. Data were collected for two large-enrollment engineering classes where examples are shown to illustrate how displaying written explanations can provide a resource for students to develop multi-variate reasoning skillsmore » « less
-
null (Ed.)The shift to remote teaching with the COVID-19 pandemic has made delivery of concept-based active learning more challenging, especially in large-enrollment engineering classes. I report here a modification in the Concept Warehouse to support delivery of concept questions. The new feature allows instructors to make students’ reasoning visible to other students by showing selected written explanations to conceptually challenging multiple-choice questions. Data were collected for two large-enrollment engineering classes where examples are shown to illustrate how displaying written explanations can provide a resource for students to develop multi-variate reasoning skills.more » « less
-
In this work-in-progress paper, we continue investigation into the propagation of the Concept Warehouse within mechanical engineering (Friedrichsen et al., 2017; Koretsky et al., 2019a). Even before the pandemic forced most instruction online, educational technology was a growing element in classroom culture (Koretsky & Magana, 2019b). However, adoption of technology tools for widespread use is often conceived from a turn-key lens, with professional development focused on procedural competencies and fidelity of implementation as the goal (Mills & Ragan, 2000; O’Donnell, 2008). Educators are given the tool with initial operating instructions, then left on their own to implement it in particular instructional contexts. There is little emphasis on the inevitable instructional decisions around incorporating the tool (Hodge, 2019) or on sustainable incorporation of technologies into existing instructional practice (Forkosh-Baruch et al., 2021). We consider the take-up of a technology tool as an emergent, rather than a prescribed process (Henderson et al., 2011). In this WIP paper, we examine how two instructors who we call Al and Joe reason through their adoption of a technology tool, focusing on interactions among instructors, tool, and students within and across contexts. The Concept Warehouse (CW) is a widely-available, web-based, open educational technology tool used to facilitate concept-based active learning in different contexts (Friedrichsen et al., 2017; Koretsky et al., 2014). Development of the CW is ongoing and collaboration-driven, where user-instructors from different institutions and disciplines can develop conceptual questions (called ConcepTests) and other learning and assessment tools that can be shared with other users. Currently there are around 3,500 ConcepTests, 1,500 faculty users, and 36,000 student users. About 700 ConcepTests have been developed for mechanics (statics and dynamics). The tool’s spectrum of affordances allows different entry points for instructor engagement, but also allows their use to grow and change as they become familiar with the tool and take up ideas from the contexts around them. Part of a larger study of propagation and use across five diverse institutions (Nolen & Koretsky, 2020), instructors were introduced to the tool, offered an introductory workshop and opportunity to participate in a community of practice (CoP), then interviewed early and later in their adoption. For this paper, we explore a bounded case study of the two instructors, Al and Joe, who took up the CW to teach Introductory Statics. Al and Joe were experienced instructors, committed to active learning, who presented examples from their ongoing adaptation of the tool for discussion in the community of practice. However, their decisions about how to integrate the tool fundamentally differed, including the aspects of the tool they took up and the ways they made sense of their use. In analyzing these two cases, we begin to uncover how these instructors navigated the dynamic nature of pedagogical decision making in and across contexts.more » « less
-
This work-in-progress paper presents a joint effort by engineering education and machine learning researchers to develop automated methods for analyzing student responses to challenging conceptual questions in mechanics. These open-ended questions, which emphasize understanding of physical principles rather than calculations, are widely used in large STEM classes to support active learning strategies that have been shown to improve student outcomes. Despite their benefits, written justifications are not commonly used, largely because evaluating them is time-consuming for both instructors and researchers. This study explores the potential of large pre-trained generative sequence-to-sequence language models to streamline the analysis and coding of these student responses.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
