Publications
Collaborative learning is a key skill for student success, but simultaneous monitoring of multiple small groups is untenable for teachers. This study investigates whether automatic audiobased monitoring of interactions can predict collaboration quality. Data consist of hand-labeled 30-second segments from audio recordings of students as they collaborated on solving math problems. Two types of features were explored: speech activity features, which were computed at the group level; and prosodic features (pitch, energy, durational, and voice quality patterns), which were computed at the speaker level. For both feature types, normalized and unnormalized versions were investigated; the latter facilitate real-time processing applications. Results using boosting classifiers, evaluated by F-measure and accuracy, reveal that (1) both speech activity and prosody features predict quality far beyond chance using majority-class approach; (2) speech activity features are the better predictors overall, but class performance using prosody shows potential synergies; and (3) it may not be necessary to session-normalize features by speaker. These novel results have impact for educational settings, where the approach could support teachers in the monitoring of group dynamics, diagnosis of issues, and development of pedagogical intervention plans. © 2016, International Speech Communications Association. All rights reserved.
Studies of skill development often describe a process of cumulative advantage, in which small differences in initial skill compound over time, leading to increasing skill gaps between those with an initial advantage and those without. We offer evidence of a similar phenomenon accounting for differential patterns of research skill development in graduate students over an academic year and explore differences in socialization that accompany diverging developmental trajectories. As predicted, quantitative analysis indicated large effect sizes for skill gains after controlling for initial performance levels. Qualitative analyses indicated that students with initial advantages were more likely to report greater demands of independence by their advisors and see more extensive value in research tasks comparable to those assigned their less skilled peers.
Although research has consistently shown that higher levels of educational attainment are associated with better mental health at midlife, we know little about specific aspects of education from adolescence and the transition to adulthood that help to produce educational mental health disparities. This study examines the extent to which academic marginalization in high school, as defined by non-academic math course-taking, lowered educational expectations, and repeated course failures, are related to depressive symptoms in midlife. Using data from the 2014 follow-up survey of the High School and Beyond study, I find that high school students who took a non-academic math curriculum had significantly greater depressive symptoms at midlife, net of selection factors and subsequent degree attainment. Furthermore, failing three or more courses in high school or lowering expectations for high school graduation were also independently associated with depressive symptoms at mid-life, although the models suggest that the effect of course failures work through the subsequent likelihood of completing high school. These results highlight the importance of developing a more nuanced understanding of the association between education and mental health, and demonstrate that academic outcomes as early as in adolescence have implications for levels of depressive symptoms much later in the life course.
Intelligent tutoring systems are known for providing customized learning opportunities for thousands of users. One feature of many systems is differentiating the amount of practice users receive. To do this, some systems rely on a threshold of consecutive correct responses. For instance, Khan Academy used to use ten correct in a row and now uses five correct in a row as the mastery threshold. The present research uses a series of randomized control trials, conducted in an online learning platform (eg., ASSISTments.org), to explore the effects of different thresholds of consecutive correct responses on learning. Results indicate that despite spending significantly more time practicing there is no significant difference on learning between two, three, four, or five consecutive correct responses. This suggests that systems, and MOOCS, can employ the simple rule of two or three consecutive correct responses when determining the amount of practice provided to users.
Prerequisite skill structures have been closely studied in past years leading to many data-intensive methods aimed at refining such structures. While many of these proposed methods have yielded success, defining and refining hierarchies of skill relationships are often difficult tasks. The relationship between skills in a graph could either be causal, therefore, a prerequisite relationship (skill A must be learned before skill B). The relationship may be non-causal, in which case the ordering of skills does not matter and may indicate that both skills are prerequisites of another skill. In this study, we propose a simple, effective method of determining the strength of pre-to-post-requisite skill relationships. We then compare our results with a teacher-level survey about the strength of the relationships of the observed skills and find that the survey results largely confirm our findings in the data-driven approach.
Cumulatively, participation in optional science learning experiences in school, after school, at home, and in the community may have a large impact on student interest in and knowledge of science. Therefore, interventions can have large long-term effects if they change student choice preferences for such optional science learning experiences. To be able to track K-12 students' intentions to participate in optional science learning experiences, we developed a new measure of science choice preferences in early adolescence. The present study with 284 5th and 894 6th graders from diverse school contexts (i.e., from the Bay Area and the Pittsburgh area) illustrates the value of applying Item Response Theory analyses to develop a measurement instrument. These analyses established the overall reliability of the instrument and each item in the scale, as well as the generalizability of the scale and individual items across subgroups by gender, by ethnicity, and by achievement levels in science. Further, preferences to participate in science were shown to be separate from preferences to participate in mathematics, engineering, or medicine. Finally, the science choice preferences measure is validated through replicated positive correlations with levels of science interest, self-efficacy, and learning achievement. (c) 2015 Wiley Periodicals, Inc. J Res Sci Teach 52: 686-709, 2015.
this study, we examined the influence of achievement goals and scaffolding on self-regulated learning (SRL) and achievement within MetaTutor, a multi-agent intelligent tutoring system. Eighty-three (N = 83) undergraduate students were randomly assigned to either a control or prompt and feedback condition and engaged in a 1-h learning session with MetaTutor to learn about the human circulatory system. Process and product data were collected from all participants prior to, during, and following the session. MANCOVA analyses revealed that students in the prompt and feedback condition deployed more SRL strategies and spent more time viewing relevant science material compared to students in the control condition. Results also revealed a significant interaction between achievement goals and condition on achievement outcomes, such that learners adopting a dominant performance-approach demonstrated higher achievement in the prompt and feedback condition. Findings are discussed in relation to the role of motivation in self-regulated learning within computer-based learning environments. Implications for the design of pedagogical agents are also discussed. (C) 2015 Elsevier Ltd. All rights reserved.
The Open Quantum Materials Database (OQMD) is a high-throughput database currently consisting of nearly 300,000 density functional theory (DFT) total energy calculations of compounds from the Inorganic Crystal Structure Database (ICSD) and decorations of commonly occurring crystal structures. To maximise the impact of these data, the entire database is being made available, without restrictions, at www.oqmd.org/download. In this paper, we outline the structure and contents of the database, and then use it to evaluate the accuracy of the calculations therein by comparing DFT predictions with experimental measurements for the stability of all elemental ground-state structures and 1,670 experimental formation energies of compounds. This represents the largest comparison between DFT and experimental formation energies to date. The apparent mean absolute error between experimental measurements and our calculations is 0.096 eV/atom. In order to estimate how much error to attribute to the DFT calculations, we also examine deviation between different experimental measurements themselves where multiple sources are available, and find a surprisingly large mean absolute error of 0.082 eV/atom. Hence, we suggest that a significant fraction of the error between DFT and experimental formation energies may be attributed to experimental uncertainties. Finally, we evaluate the stability of compounds in the OQMD (including compounds obtained from the ICSD as well as hypothetical structures), which allows us to predict the existence of similar to 3,200 new compounds that have not been experimentally characterised and uncover trends in material discovery, based on historical data available within the ICSD.
This article provides an overview of learning progressions (LP) and assesses the potential of this line of research to improve geography education. It presents the merits and limitations of three of the most common approaches used to conduct LP research and draws on one approach to propose a first draft of a LP on map reading and interpretation. It also highlights findings from LP research that may be especially significant for future work in geography education. The article concludes with a description of a new research project, GeoProgressions, to build capacity for LP research in geography.
In evaluating research investments, it is important to establish whether the expertise gained by researchers in conducting their projects propagates into the broader economy. For eight universities, it was possible to combine data from the UMETRICS project, which provided administrative records on graduate students supported by funded research, with data from the U.S. Census Bureau. The analysis covers 2010-2012 earnings and placement outcomes of people receiving doctorates in 2009-2011. Almost 40% of supported doctorate recipients, both federally and nonfederally funded, entered industry and, when they did, they disproportionately got jobs at large and high-wage establishments in high-tech and professional service industries. Although Ph.D. recipients spread nationally, there was also geographic clustering in employment near the universities that trained and employed the researchers. We also show large differences across fields in placement outcomes.


