United States

Recently, many efforts to improve undergraduate STEM instruction encourage team formation. While team-based approaches are promising for enacting high quality and sustainable outcomes, they are also “risky.” Our research group is currently developing a survey instrument for teams. This research builds on our prior work to develop an instructional change teams model. The model includes five team processes – strategic leadership, egalitarian power dynamics, team member commitment, effective communication, clear decision-making processes, and three emergent states – shared vision, psychological safety, team cohesion. Through the survey development, we aim to explore the interrelationships between all the major pieces in our model and provide formative feedback about team member collaboration. In this poster, I will describe the key parts of our emerging survey instrument and share preliminary findings from the psychometric evaluation of the pilot data. Finally, I will describe opportunities for teams to engage with our work.

AAPT SM 22

A survey for assessing instructional change teams in undergraduate STEM

The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. 

AAAI 2022

poster

AAPT meetings are a great way to reflect on your teaching, recommit to best practices and re-energize your career with new research, products, and techniques. Listen to internationally known plenary speakers, attend sessions, workshops, and special activities. This meeting is an excellent opportunity to share your expertise and learn from and network with leaders in the field.



We report on a study comparing the impact on student attitudes about science and scientific reasoning skills of two instructional interventions applied in a single semester general education physics course. The first intervention had students watch videos on physics topics and complete worksheets testing their comprehension of the concepts and the scientific reasoning shown. The second intervention had students manipulate PhET simulations and complete worksheets based on the simulations. The results of these instructional interventions are presented for two control, three video treatment, and four PhET simulation semesters. The CLASS and Lawson’s CTSR were used for pre- and posttests to assess changes in students’ attitudes about science and students’ scientific reasoning skills. Pre-post shifts in student attitudes were 3% toward more expert-like attitudes with the video interventions and 6.3% with the PhET simulations, and improvements of scientific reasoning skills of 4.2% with the videos and 7% with the PhET simulations.

Student attitude changes and curricular benefits from two instructional interventions

In introductory physics, we often encourage students to see themselves as part of the physics community. To do so, we might highlight inclusive aspects of our classroom culture and the broader physics community. In this poster, we offer a lesson plan that we designed to critique certain ways of knowing in physics. In this lesson plan, we critique gendered science practices, norms, and assumptions as existing in defined boxes. The lesson is structured to support students in noticing, naming, and rejecting these boxes as they have experienced them in life and in their STEM classes. We then offer an alternative view in which STEM classes and culture do not conform to gendered practices and expectations, but instead open avenues for broader inclusion. We aim with this lesson and reflection on science to support students in developing agency to see their ways of being as an integral part of our community.

Rejecting the Gendered Boxes of Science

Current ethics training shows little improvement in ethical decision making. Recent studies have shown that values play a much bigger role in science than what is normally accepted. We believe that by having discussions about the values found in decisions in science, we can improve scientists’ ethical decision making. We are measuring better ethical decision making by seeing how scientists’ awareness of ethical values and level of moral reasoning changed. We formed a fellowship of fifteen science faculty where they discussed the values embedded in various scientific norms to bring awareness to the values inherent in research practices. In order to see how the fellowship impacted the scientists' ethical reasoning, we conducted pre/post interviews where we asked “what would you do” in different ethical vignettes. We determined that scientists’ identified more values and we are still looking at their ethical reasoning.

The Effect of Value-Focused Discussions on Scientists' Ethical Decision Making

To boost student autonomy, traditional lectures have been replaced by a more active format integrating different student programs at the University of Cologne: Learning by Teaching (LbyT). The LbyT concept was researched and developed by using Action Research with the *teacher as reflective practitioner* model (1). The university faculty were simultaneously in the role of researchers as well as teachers exploring their own practices. Initial focus was on their role as teachers and the scope and sequence of student activities. In further development of the courses, the agency of the roles of mentors and tutors unfolded. We analyzed and present the evolution these developments by considering the individual roles. (2) (3) Beyond, we present experiences for tutoring concepts in the typical first semester courses in experimental physics for more than 10 years, where physics students can gain initial experience teaching small learning groups starting their studies.

The unfolding roles of mentors and tutors in Learning-by-Teaching model

The Departmental Action Leadership Institute (DALI) is an initiative that works with physics programs to implement institutional change within their departments. Each participating program sends two faculty members who attend ongoing DALI meetings for one year to learn and reflect on effective change practices. During that year, these change leaders facilitate local teams based on the Department Action Team (DAT) model. We present outcomes associated with individuals’ professional growth, based on survey data, comparing current states of change leaders’ abilities and their retrospective accounts prior to the DALI. These outcomes span areas such as change leaders’ self-efficacy in facilitating DATs, partnering with students, and joint data sense-making. These data also highlight aspects of DALI participants attribute their growth to, as well as what could be improved. We hope this talk serves as inspiration and lessons learned for others partnering with faculty to develop capacity for change in physics programs.

Growing as a change agent: Slowing down and facilitating teams

The practice of partnering with students on departmental projects and initiatives has grown within STEM and Physics programs. These collaborations between faculty and students span multiple contexts, from curriculum and classroom partnerships to institutional change efforts. While these partnerships produce more student-centered results and are able to engage a wider set of departmental stakeholders, they place students in complex roles in which they must navigate complicated systems. This presentation will attempt to summarize the literature student partnership initiatives and situate two ongoing research projects within this literature. The first concerns students who are members of Departmental Action Leadership Institute (DALI) affiliated Departmental Action Teams (DAT), working on institutional change within their programs. The second study is focused on undergraduates who take up the role of Network Fellows within the Access Network, an inter-institutional coalition that works towards a vision of a diverse, equitable, inclusive, and accessible STEM community.

Students' roles in faculty-student partnerships

Likert-style surveys are a widely used research instrument to assess respondents' preferences, beliefs, or experiences. This poster showcases a new methodology in which network analysis can be used to evaluate the interconnectedness of items in Likert-style surveys. The unique features of this approach are explored through an application of the methodology to the Aspects of Student Experience Scale (ASES) data set and a comparison of the results to principal component analysis. I show how a meaningful network based on student survey item response similarity is successfully created. I also demonstrate how modular analysis can be used in order to identify larger themes built from the connections between particular survey items. Our proposed methodology is widely applicable and provides a new way to investigate phenomena assessed by Likert-style surveys.

Network analysis of Likert-style surveys

Significant attention in the PER community has been paid to student cognition and reasoning processes in undergraduate quantum mechanics. Until recently, however, these same topics have remained largely unexplored in the context of emerging interdisciplinary quantum information science (QIS) courses. We conducted exploratory think-aloud interviews with 24 students in an upper-division quantum computing course at a large R1 university crosslisted in physics and computer science, as well as 5 graduate students in a similar graduate-level QIS course offered in physics. We classify and analyze students' responses to a pair of questions regarding the fundamental differences between classical and quantum computers, interpreting students' responses as a possible window into the underlying ontologies of quantum mechanics students develop through such coursework.

Investigating student interpretations of differences between classical and quantum computers

The application of various machine learning algorithms and techniques to the prediction of student performance in an introductory mechanics class is discussed (N = 2061 students). The dataset used for prediction includes institutional variables (GPA, credit hours) and in-class variables gathered from the first two weeks of class (homework and clicker scores). These variables are used to predict which students are at risk of failing the course (receiving a D or F). The effectiveness and implementations of random forest, logistic regression, and voting ensemble classification models are discussed. Various model optimization methods are explored including decision threshold tuning, model regularization, synthetic upsampling, and variable selection techniques. The best performing models achieved a balanced accuracy of around 80%.

Machine Learning Techniques for Classifying Physics Performance

This work is part of a multi-institution collaboration developing a fair, valid, and reliable research-based conceptual fluids assessment, the Fluids Conceptual Evaluation (FCE). The FCE will utilize two-tier multiple-choice items covering both fluids statics and fluids dynamics. Rasch analysis will be used to create different equivalent versions of the FCE, allowing instructors to give just the statics questions, just the dynamics questions, or a combination of both. In this poster we will describe the development of student-based distractors for the FCE. IPLA instructors interested in serving as a pilot test site should contact Project Lead Dawn Meredith.

Premium content

Next from AAPT SM 22

Student attitude changes and curricular benefits from two instructional interventions

Similar lecture

Team Correlated Equilibria in Zero-Sum Extensive-Form Games via Tree Decompositions