The emergence of smart technology has given rise to classroom response systems that offer instructors quick access to students’ understanding. Recently, this technology has been used to explore student understanding of introductory-level geological concepts using click-on-diagram (COD) response-style questions. The current study extends this line of research and investigates student conceptions of groundwater and surface water flow. Participants were predominantly undergraduate non-science majors enrolled in an introductory geology course at one Midwest and one Canadian university. Course instructors presented a set of six diagram-based questions where students were asked to make a prediction. Students were shown plan view river channel diagrams (straight and curved channels) and a cross-sectional diagram of the subsurface and landscape with a house and septic tank (source of pollution). They answered by clicking on the diagram in response to a prompt. When shown the river channel diagrams, students were asked to predict the location of the greatest erosion and fastest water speed. When shown the cross-sectional landscape diagram, students were asked where the pollution from the septic tank would end up after either 1 month or several months. Results of the study showed robust misconceptions related to water velocity and erosion, especially for curved river channels. Approximately 40% of students misidentified the inside of a river bend as having the highest level of erosion. Misconceptions were found related to pollution transport as well, with approximately one-third of students correctly selecting a location within a plume which accounted for both gravity and topography (i.e. local groundwater flow). This study demonstrates that a substantial portion of incoming undergraduate students do not reason about surface and groundwater flow using accurate physical principles. Future work will further investigate the sources of errors or alternative reasoning behind student’s answers to the questions in this study.

Given the importance of fresh water, we investigated undergraduate students’ understanding of water flow and its consequences. We probed introductory geology students’ pre-instruction knowledge using a classroom management system at two large research-intensive universities. Open-ended clicker questions, where students click directly on diagrams using their smart device (e.g., cell phone, tablet) to respond, probed students’ predictions about: (1) groundwater movement and (2) velocity and erosion in a river channel. Approximately one-third of students correctly identified groundwater flow as having lateral and vertical components; however, the same number of students identified only vertical components to flow despite the diagram depicting enough topographic gradient for lateral flow. For rivers depicted as having a straight channel, students correctly identified zones of high velocity. However, for curved river channels, students incorrectly identified the inside of the bend as the location of greatest erosion and highest velocity. Systematic errors suggest that students have mental models of water flow that are not consistent with fluid dynamics. The use of students’ open-ended clicks to reveal common errors provided an efficient tool to identify conceptual challenges associated with the complex spatial and temporal processes that govern water movement in the Earth system.

This article reports the results of a randomized control trial of a semester-long intervention designed to promote ninth-grade science students’ use of text-based investigation to create explanatory models of biological phenomena. The main research question was whether the student participants in the intervention outperformed the students in the control classes, as assessed by several measures of comprehension and application of information to modeling biological phenomena not covered in the instruction. A second research question examined the impact on the instructional practices of the teachers who implemented the intervention. Multilevel modeling of outcome measures, controlling for preexisting differences at individual and school levels, indicated significant effects on the intervention students and teachers relative to the controls. Implications for classroom instruction and teacher professional development are discussed.

We now have almost no filters on information that we can access, and this requires a much more vigilant, knowledgeable reader. Learning false information from the web can have dire consequences for personal, social, and personal decision making. Given how our memory works and our biases in selecting and interpreting information, now more than ever we must control our own cognitive and affective processing. As examples: Simply repeating information can increase confidence in its perceived truth; initial incorrect information remains available and can continue to have an effect despite learning the corrected information; and we are more likely to accept information that is consistent with our beliefs. Information evaluation requires readers (a) to set and monitor their goals of accuracy, coherence, and completeness; (b) to employ strategies to achieve these goals; and (c) to value this time- and effort-consuming systematic evaluation. Several recommendations support a reasoned approach to fake news and manipulation.

This article describes several approaches to assessing student understanding using written explanations that students generate as part of a multiple-document inquiry activity on a scientific topic (global warming). The current work attempts to capture the causal structure of student explanations as a way to detect the quality of the students’ mental models and understanding of the topic by combining approaches from Cognitive Science and Artificial Intelligence, and applying them to Education. First, several attributes of the explanations are explored by hand coding and leveraging existing technologies (LSA and Coh-Metrix). Then, we describe an approach for inferring the quality of the explanations using a novel, two-phase machine-learning approach for detecting causal relations and the causal chains that are present within student essays. The results demonstrate the benefits of using a machine-learning approach for detecting content, but also highlight the promise of hybrid methods that combine ML, LSA and Coh-Metrix approaches for detecting student understanding. Opportunities to use automated approaches as part of Intelligent Tutoring Systems that provide feedback toward improving student explanations and understanding are discussed.

Das Verständnis und die Evaluation von Argumenten: Zur Rolle allgemeiner kognitiver FähigkeitenZusammenfassung. Personen, denen die notwendigen Fähigkeiten für die korrekte Bewertung von Argumenten fehlen, ziehen ein Leben lang daraus Nachteile. Forschungsbefunde zeigen, dass einfache Tutorien für viele Studierende (ca. 30 %) keinen Nutzen bringen. Wir berichten Daten zur Frage, wie sich durch allgemeine Fähigkeiten (z. B. Vokabelwissen, Leseverständnis, analytisches Schlussfolgern) das Erlernen der Argumentevaluation vorhersagen lässt. In Studie 1 konnte gezeigt werden, dass – obwohl alle drei kognitiven Fähigkeiten einige Aspekte von Argumentverständnis und Evaluationsfähigkeit vorhersagen – das Vokabelwissen sowohl die Argumentationsfähigkeit und nicht durch das Wissen um die Bedeutung einer spezifischen Aussage, die in der Aufgabe gestellt wurde, erklärt. Die Ergebnisse zeigen auf, dass der gezielte Einsatz von Vokabelwissen und allgemeiner lexikalischer Qualität diesen Studierenden, die nicht von einem einfachen Tutorium profitieren, helfen könnte.

Educational standards put a renewed focus on strengthening students' abilities to construct scientific explanations and engage in scientific arguments. Evaluating student explanatory writing is extremely time-intensive, so we are developing techniques to automatically analyze the causal structure in student essays so that effective feedback may be provided. These techniques rely on a significant training corpus of annotated essays. Because one of our long-term goals is to make it easier to establish this approach in new subject domains, we are keenly interested in the question of how much training data is enough to support this. This paper describes our analysis of that question, and looks at one mechanism for reducing that data requirement which uses student scores on a related multiple choice test.

We examined students' understanding of the causes of a scientific phenomenon from a multiple-document-inquiry unit. Students read several documents that each described causal factors that could be integrated to address the given writing task of explaining the causes of change in average global temperature. We manipulated whether the document set included a document that took a position on climate change policies and whether a reading/writing prompt focused only on understanding the causes (“explain how and why recent temperature changes are occurring”) or also included a solution-related addendum (“and what we can do about it”). The results suggest that including a policy-related document can lead to poorer learning outcomes for the causes of climate change, with evidence that students focused on policy in lieu of, rather than in relation to, a causal understanding of the issue.

In the US in particular, there is an increasing emphasis on the importance of science in education. To better understand a scientific topic, students need to compile information from multiple sources and determine the principal causal factors involved. We describe an approach for automatically inferring the quality and completeness of causal reasoning in essays on two separate scientific topics using a novel, two-phase machine learning approach for detecting causal relations. For each core essay concept, we initially trained a window-based tagging model to predict which individual words belonged to that concept. Using the predictions from this first set of models, we then trained a second stacked model on all the predicted word tags present in a sentence to predict inferences between essay concepts. The results indicate we could use such a system to provide explicit feedback to students to improve reasoning and essay writing skills.

The purpose of this study was to examine online processes used during reading that may contribute to comprehension. One hundred twenty-four fourth-grade students with proficient decoding but poor reading comprehension skills responded to narrative and informational texts using a thinkaloud procedure. Analysis of think-aloud transcripts revealed that readers made significantly more textbased connections while reading narratives and more knowledge-based connections while reading informational texts. Findings have implications for further research on reading comprehension processes. Novice Literary Interpretations: Prompting and Processing Matter Candice Burkett, Susan R. Goldman Abstract. Research suggests literary novices are inept at interpreting literary works (Graves & Frederiksen, 1991; Zeitz, 1994). The current study investigated novices’ literary interpretations for a short story. Results indicated more interpretations when prompted than during initial reading despite evidence of elaborative processing, attention to literary devices, and adequate story comprehension. Furthermore, elaborative processing was positively related to interpretations. These results implicate differences between experts and novices in what is entailed in “reading” a literary work.

With an increasing focus on science and technology in education comes an awareness that students must be able to understand and integrate scientific explanations from multiple sources. As part of a larger project aimed at deepening our understanding of student processes for integrating multiple sources of information, we are developing machine learning and natural language processing techniques for evaluating students’ argumentative essays. In previous work, we have focused on identifying conceptual elements of the essays. In this paper, we present a method for inferring the causal structure of student essays. We used a standard parser to derive grammatical dependencies of the essay and converted them to logic statements. Then a simple inference mechanism was used to identify concepts linked to syntactic connectors by these dependencies. The results suggest that we will soon be able to provide explicit feedback that enables teachers and students to improve comprehension.