By Lina Sorg
Fostering diverse participation from women, people of color, and other underrepresented groups within the fields of science, technology, engineering, and mathematics (STEM) is an ongoing challenge (see Figure 1). Despite financial resources and funded recruitment efforts, most engineering programs continue to suffer from a lack of representation. While previous studies have examined student preparedness or methods of instruction, some researchers believe that a lack of accessibility in engineering criteria might be responsible for low retention. During a minisymposium at the 2021 SIAM Conference on Computational Science and Engineering, which took place virtually this week, Sharon Tettegah of the University of California, Santa Barbara presented a set of mathematical models that examine the various characteristics of engineering curricula. She utilized surveys, focus groups, and web crawling techniques to establish a typology of content and better understand students’ preferences and expectations for education. “We want to enhance diversity across the board, within our group, and also outside our group,” Tettegah said.
Figure 1. The value of enhancing diversity.
Tettegah conceptualized this project when she was a proposal officer at the National Science Foundation. She noticed that proposal submissions lacked diversity and subsequently realized that a comprehensive examination of engineering curricula had never occurred. Wondering about the essential characteristics of engineering teams, Tettegah decided to examine the issue with a creative mindset. She began by administering a Torrance Test of Creative Thinking (TTCT) to undergraduate students. The TTCT utilizes both a verbal and figural component to identify and evaluate one’s creative potential, and Tettegah wished to determine creativity’s connection to factors like academic engagement, self-efficacy, interest in STEM, and curricular representation. Engineering curricula are typically comprised of word problems, equations, two-dimensional and three-dimensional diagrams, animations, and interactive simulations. Students who took the test selected their preferences from these categories.
Figure 2. Word clouds representing the most common words and phrases in engineering curricula.
After collecting TTCT data, Tettegah’s group related the data to curricula propensities and displayed it in various ways. This process was quite laborious and involved examining the course catalogs of 35 universities; manually extracting their information; and generating word clouds, graphs, and tables to categorize the most common words and themes (see Figure 2). “Differential equations” was the most frequently cited phrase, which is unsurprising since it is a routine topic in engineering programs (see Figure 3).
Figure 3. “Differential equations” was the most frequently cited phrase in engineering curricula.
To supplement this curricula-based insight, the team developed a web crawler to automatically discover, fetch, and process syllabi that are publicly available on the internet. Tettegah has thus far collected over 900 syllabi from seven different universities and more than 10 engineering departments. Her group has also developed a text mining tool to identify and extract topics from the gathered syllabi. “What does engineering education look like?” Tettegah asked. “If things are happening differently somewhere else, do they have a higher retention rate of underrepresented students? No one has ever thought to aggregate this data.”
Next, Tettegah plans to catalog the syllabi on CUE4CHNG’s Science Gateway community. She encouraged attendees to complete surveys for both instructors and students or contribute course material at cue4chng.org. All survey information is valuable in determining student preferences; certain students expressed a propensity for videos because they can explore them on their own time, others like interactive content and simulations, and some prefer traditional equation-based lectures. “We really need to look at this to find out what is happening outside of what we already study in terms of users themselves and how they’re interacting with their preferences for the curricula,” Tettegah said. Doing so will bring institutions one step closer to broadening participation in undergraduate engineering programs.
|
Lina Sorg is the managing editor of SIAM News.
|