What Is Spatial Intelligence or Spatial Ability?
According to David Lohman of the University of Iowa, spatial ability can best be defined as the ability to "generate, retain, retrieve, and transform well-structured visual images." An example of a great inventor who used his high level of spatial ability to innovate was Nikola Tesla, who provided the basis for alternating current (AC) power systems. Tesla is said (or fabled) to have been able to visualize an entire working engine in his mind and be able to test each part over time to see what would break first. Rather than a great feat of mental math, one could consider this a great feat of mental imagery.
Why Don't the SAT, ACT, and GRE Include a Spatial Ability Measure?
The Scholastic Assessment Test (SAT), the American College Test (ACT), and the Graduate Record Examination (GRE) traditionally have included mathematical and verbal measures. In some of my research, along with work by my colleagues Rose Mary Webb of Appalachian State University and David Lubinski and Camilla Benbow of Vanderbilt University, we essentially found that over a half century of research on spatial ability has uncovered the significance it plays-in addition to math and verbal ability-In educational and occupational settings where it is essential, such as engineering, physics, math, and computer science. Therefore this clearly begs the question: why is there no spatial measure included in these standardized tests? Do we miss identifying a group of individuals who might be spatially talented but less mathematically and verbally talented?
Do Males and Females Differ on Average in Spatial Ability?
Diane Halpern of Claremont McKenna College, in her excellent book Sex Differences in Cognitive Abilities, has documented the evidence that males and females show a robust sex difference in spatial ability, favoring males.
Do Males and Females Differ in Math, Science Reasoning, and Spatial Ability in the Extreme Right Tail?
In a recent study I conducted along with my colleagues Megan Cacchio, Martha Putallaz and Matthew C. Makel of the Duke University Talent Identification Program, we examined data from 1981 to 2010 on the SAT and ACT among over 1.6 million students who were 12 years of age. We found that the male-female ratio among students in the extreme right tail who scored 700 or higher on the SAT-M (the top 0.01% in ability, a group with an average IQ level of about 180) was about 13 to 1 in the early 1980's but that it rapidly decreased in the first decade and has been stable at roughly 4 to 1 for the past 20 years (see the figure above). In addition to data from the SAT, we also had an independent sample who took the ACT, which includes not only a mathematics measure (ACT-M) but also a science reasoning measure (ACT-S). For the last 20 years, both the mathematics and science reasoning measures have demonstrated a male-female ratio of about 3 to 1.
Why Might This Matter For Male-Female Representation In High Level Math and Science Careers?
As I mentioned in an earlier post, in some of my research, even within the top 1% of mathematical ability for students who took the SAT-Math at age 12, when comparing the top quartile to the bottom quartile, there were significant differences between these groups about twenty years later in the earning of math and science outcomes, including Ph.D.s, publications, patents, and even securing tenure at a top university. Therefore, because we still find a sex difference on the SAT-Math, ACT-Math, and ACT-Science, math and science reasoning are likely still a part of the equation of explanation for the underrepresentation of women in high level math and science careers. Keep in mind that it is the individual's ability, not their sex, which matters in predicting these long-term outcomes.
However, because neither the SAT or the ACT include a spatial ability measure, we were not able to uncover whether there is a male-female difference in spatial ability in the extreme right tail. Because average spatial ability differences between males and females are quite robust it would make sense that there would also be a male advantage on spatial ability among the highest scorers. This is because small average differences usually translate into large differences in the tails. However, this is something that has not yet been empirically demonstrated and requires future research.
Can We Increase Spatial Ability, Perhaps Through Training?
The fact that we find these differences is intriguing but what is perhaps a more important question is what can we do about it? Shouldn't our goal be to help all men and women who have the ability, interest, and are passionate about math and science be able to pursue such a high level career in this area? In particular, what could we do to increase the numbers of women in high level math and science careers?
Now David Miller, a graduate student at University of California Berkeley, and Diane Halpern of Claremont McKenna College have conducted a fascinating study that examines whether spatial ability might be able to be increased through training. The authors examined highly gifted STEM (science, technology, engineering, and mathematics) undergraduates who completed twelve hours of spatial training and compared them to undergraduates that did not complete training.
According to Mr. Miller, these were the critical findings of the study:
1. "Compared to students in the control group, students in the training group showed larger improvements in spatial skills despite extremely high spatial skills prior to training."
2. "We found large gender differences in spatial skills prior to training, as many other researchers have. However, these gender differences were narrowed after training."
3. "Students in the training group had one-third of a letter grade higher GPA in a challenging calculus-based physics course."
4. "None of these training improvements lasted over eight to ten months."
Mr. Miller told me that "These results demonstrate that even highly gifted STEM undergraduates can benefit from spatial instruction, although twelve hours of instruction could have limited longitudinal effects."
He went on to say that, "This study's lack of longitudinal effects does not imply that such longitudinal effects are impossible. It is likely that we are not teaching these skills in ways that promote long-term retention and transfer to STEM courses. Although many research studies have found large short-term improvements in spatial skills, surprisingly little research has tried to understand how to promote the longevity of these training effects and directly promote STEM student success."
When I asked Dr. Halpern how research on spatial training might influence the representation of women in high level STEM careers, she said "I think that it will help increase the number of women in STEM fields."
So What Can We Do To Encourage More Women To Pursue High Level STEM Careers?
In addition to spatial training, are there other things we can do to encourage women to pursue math and science?
In the figure shown earlier, the male-female math ratio decreased quite rapidly from the early 1980's to the early 1990's. My colleagues and I are currently investigating what are likely to be the sociocultural factors responsible for the narrowing of the math male-female ratio and perhaps this is something we can use to help encourage more women to enter high level math and science careers.
Of course, although the male-female ratio declined rapidly, it has been roughly stable (and a difference has been present) for the last 20 years, therefore another key question is: If sociocultural factors were the key reason for the rapid decline, why is there still a male-female difference that appears to be stable? I don't have the answer to that question, but certainly we cannot rule out potential biases and barriers. Perhaps the wise words of Sir Alexander Cairncross might provide some insight:
"A trend is a trend is a trend
But the question is: will it bend?
Will it alter its course through some unforeseen force
And come to a premature end?"
© 2011 by Jonathan Wai
You can follow me on Twitter, Facebook, or G+. For more of Finding the Next Einstein: Why Smart is Relative go here.