Earlier this month, a survey was sent out to Olin students asking about choice of major and involvement in Olin clubs and organizations. For this issue, we focused on the distribution of students’ current majors, as well as what students’ intended majors were prior to starting Olin.
A total of 209 people responded to the survey. 55% percent of respondents were female, 44% were male, and 1% identified as other. There was a relatively equal response rate from each class year.
Figure 1 shows a distribution of students’ majors. ME appears to be the most popular major at Olin, followed by ECE. Of the E:Concentration majors, E:C and E:Bio are the most popular concentrations. Excluding the 7% of respondents who are currently undecided, 54% of respondents identify with the “traditional” majors (ME or ECE), 34% are doing a well-defined concentration major (E:Bio, E:C, E:Design, E:Sys, E:MatSci, E:Robo), and 12% have chosen a self-designed major.
Figure 2 shows the distribution of what respondents intended to study before entering Olin. Compared with the current distribution of majors, presented above, we see some interesting trends – particularly with the two “traditional” majors. There is a significant increase in the number of people who intended to be an ECE compared with the number who currently are pursuing this major – from 16% to 23%. Conversely, 35% percent of respondents intended to be mechanical engineers, compared with the 28% who currently identify as ME.
Although the number of students undecided dropped from 18% before entering Olin to 7% currently, the number of people identifying as ME decreased by seven percentage points. This suggests a trend of students switching out of ME into ECE and other majors. The p-value of this change in mechanical engineering majors is 0.02, showing that this result is statistically significant.
Why are people leaving mechanical engineering for other majors? Perhaps people had perceptions of mechanical engieering before entering Olin that did not align with their experience at Olin. Some may have been inspired to become an engineer by doing work on mechanical projects in high school, but found a passion for a different field after being exposed to other types of work. It is also possible that the reason could have to do with the Olin mechanical engineering curriculum, or with the availability of jobs in other fields after Olin. This survey did not ask respondents why they switched from their intended majors, so all of these potential reasons are just hypotheses.
We then tested for differences in major choice between female and male students. (The 1% who identified as “other” in gender were not statistically significant and therefore not considered). Because more females students responded to the survey than male students, results were normalized to match Olin’s current gender distribution. Figure 3 shows the percentage distribution of each major by gender. Between the two “traditional” majors, no significant gender trends emerge. Both ME and ECE are roughyly 50% male and 50% female. However, trends emerge with some of the E:concentration majors. E:Bio and E:Design are significantly female heavy, while E:Robo is significantly male heavy. E:C and E:Sys also present some male bias.
There are two possible explanations for this data: either some aspects of the Olin curriculum/culture are creating this bias, or the bias exists before students come to Olin. To determine which hypothesis is correct, we examine the gender distributions of students’ intended majors before coming to Olin.
From Figure 4 it appears that the distribution of students’ intended majors is roughly the same as the distribution of students’ current majors. Significantly more females than males wanted to major in bioengineering, and more males than females wanted to be robotics majors. Some majors became less gendered – both ECE and E:C are currently less gendered than students’ intended majors made them out to be. It is possible, however, that Olin is creating the gender bias in E:Design. This is evident because a biased gender distribution emerged, even though only one respondent intended to be a design engineer. Due to the small data set, it is impossible to tell whether these results are significant.
If you want access to the data set or have suggestions for further analysis, email email@example.com.