STEM Learning for Children in Informal Learning Environments

Presented by Tikyna Dandridge

Informal learning environments are settings that provide valuable learning experiences outside of traditional classroom settings. According to the National Research Council, informal learning environments can be categorized into three primary settings: everyday experiences, designed settings, and program settings. These environments play an important role in children’s learning, as they provide spaces for children to engage in self-directed learning. In an independent study course titled Informal STEM Learning Experiences for Children, a summative literature review is conducted on informal learning and qualitative data collection and analysis is completed for two larger research projects. The first project is analyzing a how families engage in computational thinking during an engineering design task at a science museum. The second is a report of experiences from observational data collection of a summer engineering program for minorities. This presentation will focus on some key lessons and findings from the course.

“What I did this summer”  

Presented by Jessica Rush

Her discussion will focus on how learning intensely about case study research helped her with other parts of her research this summer including research observations and managing three classes.  Also, her independent study helped her define her passions and focus on potential dissertation themes.

Application of a new clustering method to MOOC pre-surveys

Presented by Taylor Williams

MOOCs (Massive Open Online Courses) frequently ask their learners to complete a pre-survey before they begin a course.  These surveys typically include a variety of questions, including, for example, questions about each learner’s motivations, goals, educational history, and demographics.  These responses can be treated as a set of high dimensional data.  In this project, Williams used a recently published clustering technique (i.e., RP1D) which addresses a major problem encountered when trying to cluster high dimensional data—data sparsity.  In the context of high dimensional data, data sparsity occurs when that data includes dozens or hundreds of dimensions, such that even a large number of individual cases (e.g., a data set of thousands of MOOC learners) will not obviously clump into groups. That is, the data appear too spread out within the high dimensional space to cluster and typical clustering techniques fail.  Using the RP1D method, he identified many promising cluster pairs within the sparse high dimensional pre-survey data from the FutureLearn MOOC platform.  In this presentation, Williams will present the RP1D method, how he implemented it in R, how he applied it to the FutureLearn pre-survey data, and the resulting preliminary findings.

Biographies

Tikyna Dandridge is a first-year doctoral student in the School of Engineering Education. She received her B.S. in mechanical engineering technology from Alabama A&M University and her M.S. in mechanical engineering from Purdue University. Passionate about engineering education experiences for children, she joined the INSPIRE Institute as a graduate fellow and researches how children engage in computational thinking in informal learning environments. Other research interests include diversity in engineering education, engineering identity, and culturally relevant design. In her spare time, she tutors students in underserved communities, conduct design workshops, and mentors undergraduates pursuing degrees in engineering. Her interests include cooking, swimming, traveling and hiking.

Originally from Fayetteville, Georgia, Jessica Rush earned her undergraduate degree from Penn State with a focus in Supply Chain and Information Systems and a minor in international business. She attended Purdue University, receiving an MBA with specialization in Sustainability and Operations. Before business school, Jessica spent a summer in Haiti, delivering shoes to those in need and creating a more efficient supply chain for urban water projects.  Jessica has worked for many successful companies including Schneider Electric, Unilever, and Georgia Pacific.  Currently, Jessica is completing her Ph.D. in Engineering Education at Purdue University to focus on effective methods of corporate outreach in STEM to minorities communities.  In her free time, she writes children’s books, teaches yoga to children, travels, hikes, and plays with her dogs Sasha and Mr. Coconut.

Taylor Williams is an Engineering Education Ph.D. student advised by Dr. Kerrie Douglas. His current research interests include how to improve and best use deep learning to improve student learning. He hopes to utilize machine learning to assist those who teach by lessening their burden of important and necessary but difficult or tedious assessment activities. The goal is to provide improved and personalized learning opportunities to students along with actionable and insightful information to those teaching engineering.