Research

Welcome to the Learning Analytics & Psychology in Education Lab (LAPeL)!
Our aim is to help ensure that digital technologies can be used to promote equitable outcomes for underserved populations.
Learning Analytics

We use big data generated in digital learning environments to help design, build, and evaluate technologies that support learning in various educational contexts.

Educational Data Science

Data science techniques, such as natural language processing (NLP), enable us to study various phenomena that potentially impact learning and/or learners at scale.

Visualizing Learning Data

Data driven educational insights are often communicated through visualizations. Our lab studies how visualizations can impact learning and motivation.

Refutation Texts

Keeping to our roots in psychology, we also design studies focused on refuting various misconceptions that may impact on policy, teaching, and learning.

Our Mission

Our Team

Learn more about the members of the Learning Analytics and Psychology Lab
Stephen J. Aguilar
Stephen J. Aguilar
Assistant Professor of Education
Founder of LAPeL. Stephen’s research focuses on the design and efficacy learning analytics technologies, and how they can be used in ways that promote digital equity and educational achievement among ethnic and racial minority students.
Clare Baek
Clare Baek
Ph.D. Student
Urban Education Policy
(Educational Psychology)
Clare has worked as a mentor teacher and department chair at a STEM academy, and as an AP Computer Science reader for Educational Testing Service. At USC, she is excited to research issues related to motivation in STEM, digital technologies and students with disabilities, and bridging educational equity gaps for underserved communities.
Anamely Salgado
Anamely Salgado
Ph.D. Student
Urban Education Policy
(Educational Psychology)
Aidan Gray
Aidan Gray
Research Assistant
Aidan is a computer science undergrad interested in artificial intelligence and machine learning. He hopes to continue his time in academia as a data science researcher.

Publications

  • Sexual harassment in academe is underreported

    Abstract:

    What factors predict the underreporting of sexual harassment in academe? We used logistic regression and sentiment analysis to examine 2,343 reports of sexual harassment involving members of university communities. Results indicate students were 1.6 times likely to not report their experiences when compared to faculty. Respondents in the life and physical sciences were 1.7 times more likely to not report their experiences when compared to respondents in other disciplines. Men represented 90% of the reported perpetrators of sexual harassment. Analysis of respondents’ written accounts show variation of overall sentiment based on discipline, student type, and the type of institution attended, particularly with regard to mental health. Our results suggest that institutional and departmental barriers driven by power asymmetries play a large role in the underreporting sexual harassment among students—especially those in STEM disciplines.

    Cite:

    Aguilar, S.J., Baek, C. (2020). Sexual Harassment in Academe is Underreported, Especially by Students in the Physical Sciences. PLoS One. DOI: 10.1371/journal.pone.0230312

    PDF:

    Sexual harassment in academe is underreported

  • Motivated Information Seeking and Graph Comprehension Among College Students

    Abstract:

    Learning Analytics Dashboards (LADs) are predicated on the notion that access to more academic information can help students regulate their academic behaviors, but what is the association be-tween information seeking preferences and help-seeking practices among college students? If given access to more information, what might college students do with it?

    We investigated these questions in a series of two studies. Study1 validates a measure of information-seeking preferences—the Motivated Information-Seeking Questionnaire (MISQ)—-using a college student sample drawn from across the country (n = 551). Ina second study, we used the MISQ to measure college students’ (n=210) performance-avoid (i.e., avoiding seeming incompetent in relation to one’s peers) and performance-approach (i.e., wishing to outperform one’s peers) information seeking preferences, their help-seeking behaviors, and their ability to comprehend line graphs and bar graphs—two common graphs types for LADs.

    Results point to a negative relationship between graph comprehension and help-seeking strategies, such as attending office hours, emailing one’s professor for help, or visiting a study center—even after controlling for academic performance and demographic characteristics. This suggests that students more capable of reading graphs might not seek help when needed. Further results suggest a positive relationship between performance-approach information-seeking preferences, and how often students compare themselves to their peers.

    This study contributes to our understanding of the motivational implications of academic data visualizations in academic settings, and increases our knowledge of the way students interpret visualizations. It uncovers tensions between what students want to see, versus what it might be more motivationally appropriate for them to see. Importantly, the MISQ and graph comprehension mea-sure can be used in future studies to better understand the role of students’ information seeking tendencies with regard to their interpretation of various kinds of feedback present in LADs.

    Cite:

    Aguilar, S.J., Baek, C. (2019). Motivated Information Seeking and Graph Comprehension Among College Students. Proceedings of the 9th International Conference on Learning Analytics & Knowledge (LAK’19). ACM, New York, NY. Tempe, AZ.

    PDF:

    Motivated Information Seeking and Graph Comprehension Among College Students

  • Refutation Texts: A New Approach to Changing Public Misconceptions about Education Policy

    Abstract:

    Individuals often have misconceptions about education policy issues. Prior research has shown that refutation texts can address misconceptions in other areas (e.g., climate change, GMOs); this study is the first to explore whether participants’ views on controversial education policies—the Common Core State Standards and charter schools—are similarly malleable through refutation text interventions. Results of two experiments show that refutation texts reduced participants’ misconceptions and increased their correct conceptions about both policy issues. These impacts persisted for at least a week in both cases. Our findings hold promise for policymakers, implementers, and researchers seeking to buttress support for policies through direct, evidence-based refutation texts.

    Cite:

    Aguilar, S.J., Polikoff, M., Sinatra, G. (2019). Refutation Texts: A New Approach to Changing Public Misconceptions about Education Policy. Educational Researcher. DOI: 10.3102/0013189X19849416

    PDF:

    Refutation Texts: A New Approach to Changing Public Misconceptions about Education Policy

  • Hispanic Students’ Sense of Control in Relation to Post-Secondary Enrollment Outcomes

    Abstract:

    U.S. Hispanics are the fastest growing minority population pursuing post-secondary education, yet their bachelor degree attainment lags behind other ethnic–racial groups. Previous work supports the theory that having a high locus of control (LOC) can enable persistence in challenging post-secondary settings. We examine LOC as a potential mitigate against low college enrollment, and hypothesize that Hispanic students’ capability to enroll in post-secondary institutions (e.g., community college, 4-year colleges), in the face of personal, academic, and financial challenges, is likely predicated on their belief that they control their academic futures. We modelled college enrollment using a path-model using a generalized structural equation modeling (GSEM) approach. Our findings indicate that LOC decreases the likelihood of Hispanic students’ post-secondary enrollment. This work advances the current state of knowledge on how we understand Hispanic students’ transition to college, and informs the development of potential interventions supporting the academic success of this growing and significant community.

    Cite:

    Aguilar, S.J., Kim, A.Y. (2019). Hispanic Students’ Sense of Control in Relation to Post-Secondary Enrollment Outcomes. Social Sciences, 8, 67. DOI: 10.3390/socsci8020067

    PDF:

    Hispanic Students’ Sense of Control in Relation to Post-Secondary Enrollment Outcomes

  • When public opinion on policy is driven by misconceptions, refute them

    Abstract:

    The Common Core Standards, or a close variant, are the standards of record in approximately 40 states. Once popular, the standards have seen their support decline on both the right and the left. Previous research suggests that Americans hold a number of misconceptions about the standards, and that these misconceptions are strongly related to their support or opposition.

    We see misconceptions about policies as a need worth addressing. If the public does not understand a policy (or even worse, misunderstands it), they may accept or reject it based on misinformation. To begin to tackle this issue, we test an approach called a “refutation text” meant to correct people’s misconceptions about an issue. While refutation texts have been widely used to correct misconceptions about controversial science issues (e.g., global warming, GMOs), to our knowledge they have never been tested to correct misconceptions about policy.

    We use a sample of respondents from Amazon’s Mechanical Turk and test the impact of a refutation text on respondents’ misconceptions about the standards. We also test whether the refutation text reduces partisan views about the standards. Finally, we follow up with participants one week later to see whether the effects persist.

    Our data confirm that substantial misconceptions about the standards continue to exist. In fact, very few respondents held correct conceptions about five aspects of Common Core standards. Our data also confirm that views toward the standards are tepid—very close to the middle of the scale on a 5-point oppose-to-support scale.

    However, our results suggest that the simple refutation text we created substantially reduces people’s misconceptions about Common Core and increases their correct conceptions. Even a week later, there are large differences between those who read the refutation text and those who read a control text in their conceptions and misconceptions about the standards. Furthermore, the refutation text reduced to zero the partisan effect on support for the standards. Finally, the text improved people’s attitudes toward the standards.

    Cite:

    Aguilar, S. J., Polikoff, M. S., & Sinatra, G. M. (2018). When public opinion on policy is driven by misconceptions, refute them. Brookings Evidence Speaks Reports, 2(36), 1-11

    PDF:

    When public opinion on policy is driven by misconceptions, refute them

  • Structured Generation of Technical Reading Lists

    Abstract:

    Learners need to find suitable documents to read and prioritize them in an appropriate order. We present a method of automatically generating reading lists, selecting documents based on their pedagogical value to the learner and ordering them using the structure of concepts in the domain. Resulting reading lists related to computational linguistics were evaluated by advanced learners and judged to be near the quality of those generated by domain experts. We provide an open-source implementation of our method to enable future work on reading list generation.

    Cite:

    Gordon, J., Aguilar, S., Sheng, E., & Burns, G. (2017). Structured generation of technical reading lists. In Proceedings of the 12th Workshop on Innovative Use of NLP for Building Educational Applications (pp. 261-270).

    PDF:

    Structured Generation of Technical Reading Lists

  • Visualizations & College Students’ Academic Motivation

    Abstract:

    This qualitative study focuses on capturing students’ understanding two visualizations often utilized by learning analytics-based educational technologies: bar graphs, and line graphs. It is framed by Achievement Goal Theory—a prominent theory of students’ academic motivation—and utilizes interviews (n = 60) to investigate how students at risk of college failure interpret visualizations of their potential academic achievement. Findings suggest that visualizations only containing information about students themselves (i.e., self-focused affordances) evoked statements centered on mastering material. Visualizations containing information about students and a class average (i.e., comparative information), on the other hand, evoked responses that disheartened students and/or made them feel accountable to do better. Findings from this study suggest the following guidelines for designing visualizations for learning analytics-based educational technologies: (1) Never assume that more information is better; (2) anticipate and mitigate against potential misinterpretations—or harmful alternative interpretations—of visualizations; and (3) always suggest a way for students to improve. These guidelines help mitigate against potential unintended consequences to motivation introduced by visualizations used in learning analytics-based educational technologies.

    Cite:

    Aguilar, S.J. (2017). Examining the Relationship Between Comparative vs. Self‐Focused Academic Data Visualizations in At‐Risk College Students’ Academic Motivation. Journal of Research on Technology in Education. DOI: 10.1080/15391523.2017.1401498

    PDF:

    Visualizations & College Students’ Academic Motivation

  • Learning Analytics & Social Justice in Education

    Abstract:

    We are still designing educational experiences for the average student, and have room to improve. Learning analytics provides a way forward. This commentary describes how learning analytics-based applications are well positioned to meaningfully personalize the learning experience in diverse ways. In so doing, learning analytics has the potential to contribute to more equitable and socially just educational outcomes for students who might otherwise be seen through the lens of the average student. Utilizing big data, good design, and the input of the stakeholders, learning analytics techniques aim to develop applications for the sole purpose of reducing the classroom size to 1. Over time, these digital in- novations will enable us to do away with a model of education that teaches toward the non-existent average student, replacing it with one that is more socially just——one that addresses the individual needs of every student.

    Cite:

    Aguilar, S.J. (2017). Learning Analytics: At the Nexus of Big Data, Digital Innovation, and Social Justice in Education. TechTrends. DOI: 10.1007/s11528‐017‐0226‐9

    PDF:

    Learning Analytics & Social Justice in Education
Go Back

Read more of Dr. Aguilar's work

Google Scholar

Get in touch