A greater understanding of cognitive abilities in career fields, and the preceding college majors, is an ongoing effort. Investment theory outlines how people make implicit or explicit choices about investment of time, efforts, and interests, and the related outcomes of achievement in specific areas. When a person has abilities to meet the demands of an activity, they feel competent and are more likely to continue that activity, thus supporting a competence need as outlined in self-determination theory. Likewise, the abilities-demands aspect of person-environment fit for career engagement and satisfaction has been shown in previous research (Greguras & Diefendorff, 2009).
This study compares cognitive ability patterns among popular college majors. People in different college majors have profiles of abilities that might show strengths, or tilt, in numeric versus verbal abilities (Coyle, 2018). Coyle found that non-g residuals of verbal and numeric were related to choosing a major or occupation in humanities or STEM fields. The current study also includes the important role of a spatial ability factor (Wai, Lubinski, & Benbow, 2009).
Specifically, this study considers group factors as they relate to a general factor of intelligence, and group factor non-g residuals that are higher/lower than, the person’s level of g. We wanted to know how the general and group factor non-g residuals for a variety of college majors would map onto a STEM vs humanities hypothesis. The specific college majors that were included in this study were: Computer Science, Engineering, Science, Art, Communication, Social Science, and Business.
The sample for this study was young adults and adults who were a mean age of 23.6 years old at the time of the initial measures of the study, and a mean age of 29 years old at the time of the follow-up survey that was either five or ten years after their initial testing. Cognitive abilities were measured using the Johnson O’Connor Research Foundation standard battery of tests. We selected those respondents who indicated that their college major had been a good fit for their abilities, in order to include the concept of person-environment fit. We had complete information from 288 respondents.
First, mean levels of g were found for each college major. Then, we looked at the non-g residuals for the verbal factor: there were higher-relative residuals for the expected majors of arts, communications, and social sciences. For the numeric factor, interestingly the STEM related fields remained around zero, but the business majors had a higher-relative non-g residual for numeric. For the spatial factor, two majors stood out from the others as having higher-relative non-g residuals--as expected they were computer science and engineering majors.
In conclusion, most of the predicted patterns were supported, but spatial ability had a unique contribution to some patterns. These patterns benefited from also considering the general factor of intelligence in addition to group factor non-g residuals. The general factor shared collinearilty with the initial numeric group factor, thus did not result in influence from a numeric factor non-g residual for some cases when it might have been expected.
