Monday, November 24, 2014

Spatial Skills, STEM, and the Gender Gap*

Most engineering faculty have highly developed 3-D spatial skills and may not understand that others can struggle with a topic they find so easy. Furthermore, they may not believe that spatial skills can be improved through practice, falsely believing that this particular skill is one that a person is either “born with” or not. They don’t understand that they probably developed these skills over many years.                             
 —Sheryl Sorby
 
One of the most persistent gender gaps in cognitive skills is found in the area of spatial skills, specifically on measures of mental rotation, where researchers consistently find that men outscore women by a medium to large margin (Linn & Petersen, 1985; Voyer et al., 1995). While no definitive evidence proves that strong spatial abilities are required for achievement in STEM careers (Ceci et al., 2009), many people, including science and engineering professors, view them as important for success in fields like engineering and classes like organic chemistry. The National Academy of Sciences states that “spatial thinking is at the heart of many great discoveries in science, that it underpins many of the activities of the modern workforce, and that it pervades the everyday activities of modern life” (National Research Council, Committee on Support for Thinking Spatially, 2006, p.1).
 
Sheryl Sorby, a professor of mechanical engineering and engineering mechanics at Michigan Technological University, has studied the role of spatial-skills training in the retention of female students in engineering since the early 1990s. She finds that individuals can dramatically improve their 3-D spatial-visualization skills within a short time with training, and female engineering students with poorly developed spatial skills who receive spatial visualization training are more likely to stay in engineering than are their peers who do not receive training.
 
Sorby became interested in the topic of spatial skills through her personal difficulty with spatial tasks as an engineering student. In an interview with AAUW, Sorby described her experience:
I was blessed with the ability to do academic work. When I got to college, I was getting A’s in all of my classes, getting 97 on chemistry exams where the average was in the 50s, and then my second quarter, I took this engineering graphics course, and it was the first time in my entire life that I couldn’t do something in an academic setting. I was really frustrated, and I worked harder on that class than I did on my calculus and my chemistry classes combined.
 
A few years later, when Sorby was working on a doctorate in engineering, she found herself teaching the same course that she had struggled with: “While I was teaching this class, it seemed anecdotally to me that a lot of young women had the same issues with this class that I had had. They just struggled, they didn’t know what they were doing, they were frustrated, and I had a number of them tell me: ‘I’m leaving engineering because I can’t do this. I really shouldn’t be here.’ ”
 
After she earned a doctorate in engineering mechanics in the early 1990s, Sorby connected with Beverly Baartmans, a math educator at Michigan Tech, who introduced her to research on gender differences in spatial cognition, and Sorby began to understand her own and her  students’ challenges with spatial visualization in a new way. As a result, Sorby and Baartmans formulated the following research question: If spatial skills are critical to success in engineering graphics, and graphics is one of the first engineering courses that students take, and women’s spatial skills lag behind those of their male counterparts, will women become discouraged in this introductory course at a disproportionate rate and drop out of engineering as a result?
 
To answer this question, Sorby and Baartmans, with funding from the National Science Foundation, developed a course in spatial visualization for first-year engineering students who had poorly developed spatial skills. The researchers’ intention was to increase the retention of women in engineering through this course, which focused on teaching basic spatial-visualization skills, including isometric and orthographic sketching, rotation and reflection of objects, and cross sections of solids.
 
Spatial skills gender gap
In one of their first studies in 1993, Sorby and Baartmans administered the Purdue Spatial Visualization Test: Rotations (PSVT:R) (Guay, 1977) along with a background questionnaire to 535 first-year Michigan Tech engineering students during orientation. Sorby’s analysis of the results of the test and the background questionnaire showed that previous experience in design-related courses such as drafting, mechanical drawing, and art, as well as play as children with construction toys such as Legos, Lincoln Logs, and Erector Sets, predicted good performance on the PSVT:R. Another factor that predicted success was being a man. Women were more than three times as likely as their male peers to fail the test, with 39 percent of the women failing the test compared with 12 percent of the men (Sorby & Baartmans, 2000).
 
Sorby then selected a random sample of 24 students (11 women and 13 men) who failed the PSVT:R test to participate in the pilot offering of the spatial-visualization course. During a 10-week period, these students took a three-credit course that included two hours of lecture and a two-hour computer lab each week. Lectures covered topics such as cross sections of solids, sketching multiview drawings of simple objects, and paper folding to illustrate 2-D to 3-D transformations. In the lab, students used solid-modeling computer-aided design (CAD) software to illustrate the principles presented during the lectures. At the end of the course, students took the PSVT:R again. The results were remarkable. Students’ test scores improved from an average score of 52 percent on the PSVT:R before taking the class to 82 percent after taking it. This is approximately 10 times the improvement that would be expected of someone taking the PSVT:R a second time with no training (ibid.) and three to four times the improvement that Sorby had seen among her students as a result of taking an engineering graphics or computer-design course.
 
Sorby is quick to point out that her course does not help people become perfect at spatial visualization; rather, the training brings students’ scores up to the average score for all engineering students. This finding is particularly relevant for women in STEM fields because, although no gender differences appeared in average pre- or post-test scores among the students taking the course, as explained above, a much larger percentage of women failed the test initially.
 
Sorby and her colleagues continued to offer this course through 1999 to engineering freshmen who failed the PSVT:R. Each year, students’ scores on the PSVT:R increased by 20 to 32 percentage points on average after taking the course. In 2000 Sorby condensed the training into a one-credit course that met once each week for 14 weeks for a two-hour lab session. She found similar results: students’ PSVT:R scores increased 26 percentage points on average after the training among the 186 students who took the course between 2000 and 2002 (Sorby, 2009).
 
Lego Pink: marketing to girls
In 2004 and 2005 Sorby conducted a study with non-engineering first-year students at Michigan Tech and pilot studies with high school and middle school students and in each case found that students’ spatial scores improved with training. Other universities, such as Virginia Tech and Purdue, are now offering the spatial-visualization course, and the National Science Foundation has funded the Women in Engineering ProActive Network (WEPAN) to make the course available to students at 30 additional universities by 2014. Sorby, along with Baartmans and Anne Wysocki, published a multimedia software-workbook package, Introduction to 3D Spatial Visualization, in 2003, which contains content similar to the course and is available to the general public to guide anyone interested in improving her or his 3-D spatial visualization skills.
 
Sorby has produced striking findings on spatial skills and retention of female engineering students. She found that among the women who initially failed the PSVT:R and took the spatial-visualization course between 1993 and 1998, 77 percent (69 out of 90) were still enrolled in or had graduated from the school of engineering. In comparison only 48 percent (77 out of 161) of the women who initially failed the PSVT:R and did not take Sorby’s course were still enrolled or had graduated from the school of engineering. Much of Sorby’s analysis is based on nonrandom samples of students since, after the first year, students opted to take the course rather than being randomly assigned. Therefore, the women who remained in engineering after taking the course may have been more motivated to succeed in engineering to begin with, and the higher retention rate could be a result of their motivation rather than the course. Nonetheless, Sorby’s findings were consistent and compelling enough to convince the departmental chairs and the dean at Michigan Tech to require the spatial-skills course for all students who fail the PSVT:R during orientation, starting in fall 2009. Sorby will soon be able to isolate the impact of the course itself on retention since all students who fail the test are now required to take the course, and the students are no longer self-selected.
 
Sorby believes that well-developed spatial skills can help retain women in engineering and help attract more girls to STEM. She sees well-developed spatial skills as important for creating confidence in one’s ability to succeed in math and science courses and ultimately in a STEM career, because spatial skills are needed to interpret diagrams and drawings in math and science textbooks as early as elementary school. In a pilot study Sorby found that middle school girls who took a spatial-visualization course took more advanced-level math and science courses in high school than did girls who did not take the course. Sorby recommends that this training happen by middle school or earlier to make a difference in girls’ choices.
 
Sorby’s research shows that with training, women and men achieve consistent and large gains in tests of spatial skills. First-year engineering students, undergraduate students outside engineering, high school students, and middle school students have all shown improvement with training. Sorby’s work demonstrates that spatial skills can indeed be developed through practice.
 
Recommendations

Parents, AAUW volunteers, and teachers, especially engineering educators, can help young people, especially girls, develop their spatial skills in the following ways:
 
• Explain to young people that spatial skills are not innate but developed.
• Encourage children and students to play with construction toys, take things apart and put them back together again, play games that involve fitting objects into different places, draw, and work with their hands.
• Use handheld models when possible (rather than computer models) to help students visualize what they see on paper in front of them.

*This is Chapter 5 of the 2010 report entitled, Why So Few? Women in Science, Technology, Engineering, and Mathematics, by the American Association of University Women (AAUW).
 
References
 
Ceci, S. J., Williams, W. M., & Barnett, S. M. (2009). Women’s underrepresentation in science:
Sociocultural and biological considerations. Psychological Bulletin, 135(2), 218–61.
 
Guay, R. (1977). Purdue Spatial Visualization Test: Rotations. West Lafayette, IN: Purdue
Research Foundation.
 
Linn, M. C., & Petersen, A. C. (1985). Emergence and characterization of sex differences in spatial ability: A meta-analysis. Child Development, 56(6), 1479–98.
 
Sorby, S. A. (2009). Educational research in developing 3-D spatial skills for engineering students. International Journal of Science Education, 31(3), 459–80.
 
Sorby, S. A., & Baartmans, B. J. (2000). The development and assessment of a course for enhancing the 3-D spatial visualization skills of first year engineering students. Journal of Engineering Education, 89(3), 301–07.
 
Voyer, D., Voyer, S., & Bryden, M. P. (1995). Magnitude of sex differences in spatial abilities: A meta-analysis and consideration of critical variables. Psychological Bulletin, 117(2), 250–70.