Search

With accelerating advances in artificial intelligence, it is clear that introducing K-12 students to AI is essential for preparation to interact with and potentially develop AI technologies. To succeed as the workers, creators, and innovators of the future, we argue students should encounter core concepts of AI as early as elementary school. However, building a curriculum that introduces AI content to K-12 students presents significant challenges, such as connecting to prior knowledge, developing curricula that are meaningful for students, and creating content that teachers feel confident to teach. To lay the groundwork for elementary AI education, we investigated the everyday experiences and ideas of students in grades 4 and 5 (ages 9 to 11) about AI to inform possible entry points for learning. This yielded themes around student conceptions, examples, and ethics of AI. For each theme, we juxtapose the student ideas with the teachers' reflections on those ideas as frames of reference to consider in co-designing curricular approaches.

This design case chronicles the efforts of an interdisciplinary team of researchers as they collaborated with middle grades science teachers and students to build and refine an epidemic disease curriculum module. The initial five-day design was delivered in five science classrooms at three nearby schools where researcher classroom observations and teacher feedback drove iterative refinements of the module's materials. The final design of this module consisted of four instructional days of modeling and simulation activities that integrate computational thinking practices, computer science concepts, and life sciences content. The paper aims to illustrate the design motivations to address contextual constraints such as tight curricular schedules and varied levels of exposure to programming for both teachers and students. The instructional materials presented in this design case were the result of a three-year long research-practice partnership with science teachers at nearby middle schools.

The increasing interest in computer science (CS) and CS-integrated STEM teaching and learning has created a need for assessment instruments that can be used to evaluate the efficacy of innovative instructional approaches to K-12 CS education. However, there is a lack of validated assessment tools aligned to core CS concepts for younger students. This paper reports on the development and validation of a CS concept assessment for middle grades (ages 11-13) students. A total of 27 multiple-choice items were developed, guided by focal knowledge, skills and abilities associated with the concepts of variables, loops, conditionals, and algorithms. These items were administered to 457 middle grades students. The items were presented in form of block-based programming code and administered in a week-long computational modeling intervention. A combination of classical test theory and item response theory approaches were used to validate the assessment. Based on results, it was found that only 24 items are considered valid and reliable items to measure CS conceptual understanding. The results also suggested that the assessment can be used as a pre and post-test to investigate students' learning gains. This work fills an important gap by providing a key resource for researchers and practitioners interested in assessing middle grades student CS conceptual understanding.

This exploratory paper highlights how problem-based learning (PBL) provided the pedagogical framework used to design and interpret learning analytics from "Crystal Island: EcoJourneys," a collaborative game-based learning environment centred on supporting science inquiry. In "Crystal Island: EcoJourneys," students work in teams of four, investigate the problem individually and then utilize a brainstorming board, an in-game PBL whiteboard that structured the collaborative inquiry process. The paper addresses a central question: how can PBL support the interpretation of the observed patterns in individual actions and collaborative interactions in the collaborative game-based learning environment? Drawing on a mixed method approach, we first analyzed students' pre- and post-test results to determine if there were learning gains. We then used principal component analysis (PCA) to describe the patterns in game interaction data and clustered students based on the PCA. Based on the pre- and post-test results and PCA clusters, we used interaction analysis to understand how collaborative interactions unfolded across selected groups. Results showed that students learned the targeted content after engaging with the game-based learning environment. Clusters based on the PCA revealed four main ways of engaging in the game-based learning environment: students engaged in low to moderate self-directed actions with: (1) high and (2) moderate collaborative sense-making actions; (3) low self-directed with low collaborative sense-making actions; and (4) high self-directed actions with low collaborative sense-making actions. Qualitative interaction analysis revealed that a key difference among four groups in each cluster was the nature of verbal student discourse: students in the low to moderate self-directed and high collaborative sense-making cluster actively initiated discussions and integrated information they learned to the problem, whereas students in the other clusters required more support. These findings have implications for designing adaptive support that responds to students' interactions with in-game activities.

This paper reports findings from the efforts of a university-based research team as they worked with middle school educators within formal school structures to infuse computer science principles and computational thinking practices. Despite the need to integrate these skills within regular classroom practices to allow all students the opportunity to learn these essential 21st Century skills, prior practice has been to offer these learning experiences outside of mainstream curricula where only a subset of students has access. We have sought to leverage elements of the research-practice partnership framework to achieve our project objectives of integrating computer science and computational thinking within middle science classrooms. Utilizing a qualitative approach to inquiry, we present narratives from three case schools, report on themes across work sites, and share recommendations to guide other practitioners and researchers who are looking to engage in technology-related initiatives to impact the lives of middle grades students.

This study examined students' genetics learning in a game-based environment by exploring the connections between the expectancy-value theory of achievement motivation and flow theory. A total of 394 secondary school students were recruited and learned genetics concepts through interacting with a game-based learning environment. We measured their science self-efficacy, science outcome-expectancy beliefs, flow experience, feelings of frustration, and conceptual understanding before and after playing the game, as well as their game satisfaction. Mixed-model ANOVA, correlation tests, and path analysis were run to answer our research questions. Based on the results, we found that the game had a significant impact on students' conceptual understanding of genetics. We also found an acceptable statistical model of the integration between the two theories. Flow experience and in-game performance significantly impacted students' posttest scores. Moreover, science outcome-expectancy belief was found to be a significant predictor of students' flow experiences. In contrast, science self-efficacy and pretest scores were found to be the most significant factors influencing the feeling of frustration during the game. The results have practical implications with regard to the positive role that an adaptive game-based genetics learning environment might play in the science classroom. Findings also underscore the role the teacher should play in establishing productive outcome expectations for students prior to and during gameplay.

Collaborative inquiry learning affords educators a context within which to support understanding of scientific practices, disciplinary core ideas, and crosscutting concepts. One approach to supporting collaborative science inquiry is through problem-based learning (PBL). However, there are two key challenges in scaffolding collaborative inquiry learning in technology rich environments. First, it is unclear how we might understand the impact of scaffolds that address multiple functions (e.g., to support inquiry and argumentation). Second, scaffolds take different forms, further complicating how to coordinate the forms and functions of scaffolds to support effective collaborative inquiry. To address these issues, we identify two functions that needed to be scaffolded, the PBL inquiry cycle and accountable talk. We then designed predefined hard scaffolds and just-in-time soft scaffolds that target the regulation of collaborative inquiry processes and accountable talk. Drawing on a mixed method approach, we examine how middle school students from a rural school engaged with Crystal Island: EcoJourneys for two weeks (N=45). Findings indicate that hard scaffolds targeting the PBL inquiry process and soft scaffolds that targeted accountable talk fostered engagement in these processes. Although the one-to-one mapping between form and function generated positive results, additional soft scaffolds were also needed for effective engagement in collaborative inquiry and that these soft scaffolds were often contingent on hard scaffolds. Our findings have implications for how we might design the form of scaffolds across multiple functions in game-based learning environments.

A distinctive feature of game-based learning environments is their capacity for enabling stealth assessment. Stealth assessment analyzes a stream of fine-grained student interaction data from a game-based learning environment to dynamically draw inferences about students' competencies through evidence-centered design. In evidence-centered design, evidence models have been traditionally designed using statistical rules authored by domain experts that are encoded using Bayesian networks. This article presents DEEPSTEALTH, a deep learning-based stealth assessment framework, that yields significant reductions in the feature engineering labor that has previously been required to create stealth assessments. DEEPSTEALTH utilizes end-to-end trainable deep neural network-based evidence models. Using this framework, evidence models are devised using a set of predictive features captured from raw, low-level interaction data to infer evidence for competencies. We investigate two deep learning-based evidence models, long short-term memory networks (LSTMs) and n-gram encoded feedforward neural networks (FFNNs). We compare these models' predictive performance for inferring students' knowledge to linear-chain conditional random fields (CRFs) and naïve Bayes models. We perform feature set-level analyses of game trace logs and external pre-learning measures, and we examine the models' early prediction capacity. The framework is evaluated using data collected from 182 middle school students interacting with a game-based learning environment for middle grade computational thinking. Results indicate that LSTM-based stealth assessors outperform competitive baseline approaches with respect to predictive accuracy and early prediction capacity. We find that LSTMs, FFNNs, and CRFs all benefit from combined feature sets derived from both game trace logs and external pre-learning measures.

Science learning is inherently multimodal, with students utilizing both drawings and writings to explain observations of physical phenomena. As such assessments in science should accommodate the many ways students express their understanding, especially given evidence that understanding is distributed across both drawing and writing. In recent years advanced automated assessment techniques that evaluate expressive student artifacts have emerged. However, these techniques have largely operated individually, each considering only a single mode. We propose a framework for the multimodal automated assessment of students' writing and drawing to leverage the synergies inherent across modalities and create a more complete and accurate picture of a student's knowledge. We introduce a multimodal assessment framework as well as two computational techniques for automatically analyzing student writings and drawings: a convolutional neural network-based model for assessing student writing, and a topology-based model for assessing student drawing. Evaluations with elementary students' writings and drawings collected with a tablet-based digital science notebook demonstrate that (1) each of the framework's two modalities provide an independent and complementary measure of student science learning, and (2) the computational methods are capable of accurately assessing student work from both modalities and offer the potential for integration in technology-rich learning environments for real-time formative assessment.

Successful problem-based learning (PBL) often requires students to collectively regulate their learning processes as a group and engage in socially shared regulation of learning (SSRL). This paper focuses on how facilitators supported SSRL in the context of middle-school game-based PBL. Using conversation analysis, this study analyzed text-based chat messages of facilitators and students collected during gameplay. The analysis revealed direct modeling strategies such as performing regulative processes, promoting group awareness, and dealing with contingency as well as indirect strategies including prompting questions and acknowledgment of regulation, and the patterns of how facilitation faded to yield responsibilities to students to regulate their own learning. The findings will inform researchers and practitioners to design prompts and develop technological tools such as adaptive scaffolding to support SSRL in PBL or other collaborative inquiry processes. [ABSTRACT FROM AUTHOR]