STEM in America: Data, Approaches, and Questions (Part I)

Our economy needs STEM skills.

Not only does it need STEM skills, but it rewards people who have them – significantly more than those who don’t.

As educators, we have two major responsibilities on the STEM front:

  • Create STEM exposure opportunities for students who might lack access, and
  • Make those opportunities as rich and engaging as possible.

Determining how to do those things successfully and on a large scale (and not just on a case-by case basis) requires looking at some large-scale data – some of which we have, and some of which we don’t.

This multi-part series will look at some of the data we have, some of the problems those data suggest, some solutions currently in use and development, and where to look to answer our next questions.

Part I: The State of STEM

For reference, in 2012 the US resident population of 18-24 years old was 63% white, 5% Asian and Pacific Islander, 12% black, 17% Hispanic or Latino, less than 1% (0.7%) Native American/Alaskan Native, and 2% mixed-race (non-Hispanic). [Note: to my knowledge, all race and ethnicity data is self-reported, and Hispanic/Latino may refer to any race. Some citizens identify as more than one race or ethnicity, and percentages may not add up to 100 in all data sets.] Data used for the charts in this section comes from the NSF, and is freely available here. (Thanks, NSF!)

The 4-year college and university population in 2012 looked like this:Pie chart showing US 4-year college population by race, ethnicity, sex, and citizenshipWe would hope to see that the makeup of the university population is similar to the makeup of the national population of college age, and in fact it is, though the categories don’t line up quite perfectly. Nevertheless, women outnumber men by a greater margin as college students than as residents in most groups.

Now let’s look at the numbers of those same students at US four-year colleges who intend to pursue, and are later awarded, bachelor’s degrees in STEM fields.Percentage of 2012 freshmen intending to pursue STEM majorsIn every single racial or ethnic group, women have higher college attendance than men (sometimes by too small a margin to appear in the percentages), while men significantly outnumber women in intent to major in STEM fields.

Looking at these data, it isn’t clear that any particular demographic group needs STEM encouragement more than another. We can see that all women are lagging, but white women more so than other women; and that Asian men lead all other groups. It doesn’t appear that minorities are disadvantaged with regards to white students, as one might assume; in fact, white, black, and Hispanic interest in STEM degrees is similar, while Asian interest is considerably greater and American Indian/Alaska native interest is only somewhat less.

This holds true for degree recipients. The percentage of graduating students for the same year shows a very similar pattern; most groups graduated at a slightly lower rate than those of major intent.Percent of 2012 bachelors graduating with STEM degreesWhen we break STEM down, however, we see greater disparities. STEM is an umbrella term covering a wide variety of fields of study in science, technology, engineering and math. But math degrees frequently produce teachers, while “science” covers nuclear physics, geology, and biology – everyone from rocket scientists to petrochemists to family doctors to zoologists. Engineers are also an internally-varied group. Unsurprisingly, then, demographics can look quite different from one STEM field to another.

Here’s a graph of data for those same 2012 graduates, but this time looking specifically at engineering degrees.2012 engineering degrees by race, ethnicity, sex, and citizenshipNot only has the disparity between male and female students grown, but some greater racial and ethnic disparities have appeared, as well. Black men are the least likely men to be awarded an engineering degree, though they are still more likely to receive that degree than all women except temporary resident women, who have the same likelihood. American Indian and Alaska Native men are also among the least likely men to receive engineering degrees, while temporary resident men are by far the most likely, followed by Asian men. White and Hispanic men are roughly equal, both close to the average for the total population. On the other hand, there is much less disparity among women, who are largely equal in their very low likelihood of receiving engineering degrees. Just about 1% of women bachelors of almost all groups are receiving a degree in engineering. Only Asian women and temporary resident women buck that trend, at 3% and 4%, respectively.

We see similar effects with psychology degrees, with the gender imbalance reversed:2012 psychology degrees by race, ethnicity, sex, and citizenship

Racial and ethnic imbalances are not as pronounced for psychology degrees as for engineering degrees, but they do follow a similar pattern: whereas just 1% of almost all groups of women received engineering degrees, all groups of men in psychology are within one percentage point of the male average (3%). There is more variation among groups of women in psychology, with Hispanic women leading all groups at 10% and temporary resident women trailing at 5%. This 5-point spread, however, is dwarfed by the 15-point spread among men in engineering (11 among permanent residents and citizens only).

In Part II, we’ll consider some potential conclusions to be drawn from these data that might inform efforts to even out these disparities. We’ll look at some employment data and ask questions to further direct our research, such as:

  • What are the benefits of an engineering degree vs other STEM degrees?
  • Can we identify factors that affect the diversity of STEM fields?

Stay tuned.

One thought on “STEM in America: Data, Approaches, and Questions (Part I)

Leave a Reply

Your email address will not be published. Required fields are marked *