Monday, January 19, 2015

Astro-Diversity: Post #1 – The Pipeline to Astronomy Degrees

Dr. Lisa M. Frehill [1] is an IPA at NSF in Strategic Human Capital Planning working as an Organizational Evaluation and Assessment Researcher.  Her home institution is Energetics Technology Center in St. Charles, MD, where she has completed science, technology, engineering and mathematics (STEM) workforce analysis and assessment and evaluation in support of the Office of Naval Research, the DoD STEM Development Office and the American Association for the Advancement of Science.  A past NSF awardee, Dr. Frehill was the PI and Program Director of the ADVANCE: Institutional Transformation program when she was an associate professor of sociology at New Mexico State University. She is an expert on diversity in STEM and on program evaluation. A forthcoming volume (co-edited with Willie Pearson, Jr. and Connie L. McNeely) titled Advancing Women in Science: An International Perspective is due winter 2015 from Springer.  In her free time, Lisa enjoys hiking, yoga, visiting family and baking.

This is the first in a series of posts about diversity in astronomy. The idea for the series emerged from conversations with Dr. Joan Schmelz, who is serving as an NSF program officer in the Division of Astronomy on loan from the University of Memphis. Joan has been involved in issues for women in astronomy and is interested in being attentive to how to more generally increase the diversity of her field. 

This first post will provide a view of the pipeline into college and bachelor’s degree attainment in both astronomy and physics, which is an important “feeder field.” Future posts will look at U.S. astronomy degrees in greater detail.  This post relies on institutionally-reported data in the U.S. Department of Education’s Integrated Postsecondary Education Data System (IPEDS) were accessed via the National Science Foundation WebCASPAR database tool. 

What does the STEM pipeline into college look like from a diversity standpoint?  The answer to this is a “glass half full/half empty.”  On the one hand, we have seen a significant narrowing of the sex gap in high school preparation in mathematics and sciences. Indeed, high school boys recently caught up with high school girls to earn an average of 7.4 credits in mathematics and science (Nord et al., 2011).  Girls (14 percent) and boys (12 percent) are equally likely to have taken a “rigorous” high school curriculum consisting of at least four years of English and mathematics (including pre-calculus or higher), and three years each of social studies, science (including biology, chemistry and physics), and foreign language.  These are important increases since 1990, when just 4 percent of girls and 5 percent of boys had taken a rigorous high school curriculum.  Science, not mathematics, continues to be a more important issue for girls.  An additional 15 percent of girls would have completed a rigorous curriculum by taking just one more science class, as compared to an additional 9 percent of boys.

On the other hand, there are persistent and significant ethnic [2] inequalities at the high school level that pose challenges for students interested in college fields like astronomy and physics.  Figure 1 compares data taken from high school transcripts in 1990 and 2009 for representative samples of U.S. high school graduates. Although the preparation level for college of all four ethnic categories shown has increased substantially since 1990, there are persistent disparities. Whereas 29 percent of Asian American high school graduates are well-prepared for college study, white students are about half as likely (14 percent) and African American and Hispanic/Latino students are far less likely (6 percent and 8 percent) to have completed a rigorous high school curriculum that will provide them with a firm foundation for college work.  Science classes still present a challenge for African American and Hispanic/Latino high school students.  Half or more of students from these two groups who had not completed a “standard” curriculum [3] had not done so for want of a single science class.
Figure 1. Completion of Rigorous High School Curriculum, 1990 and 2009
How does the demography of overall bachelor’s degree attainment compare with the U.S.? As shown in Figure 2, white men, and Asian American men and women earn bachelor’s degrees at about the same rate as the representation of these categories in the U.S. 20-24 year old population.  White women are overrepresented, accounting for 39 percent of degrees compared to 28 percent of the 20-24 year old U.S. population in 2013.  Women from the three underrepresented ethnic categories (African American, American Indian/Alaska Native, and Hispanic/Latino; also referred to as “URM”) represent 18 percent of 20-24 year olds but accounted for just 15 percent of bachelor’s degrees awarded in 2013.  Men from these same ethnic categories, however, were substantially underrepresented among the 2013 bachelor’s degree class (9 percent) versus the 19 percent of 20-24 year olds.
Figure 2. Demography: Bachelor’s Degrees and the U.S. 20-24 Year Old Population 2013
Figure 3 shows that the pool of students that “feeds” graduate programs in astronomy from U.S. bachelor’s degree programs in astronomy and physics is far from representative of the U.S. population.  White men accounted for half of the 2013 U.S. astronomy graduates and two-thirds of physics graduates while they constitute only 29% of the 20-24 year olds.  Overall, degree recipients in astronomy were more diverse compared to physics degree recipients.  At 26%, white women were more highly represented in bachelor’s astronomy than physics (15%), but were far below the expected value of 39 percent (i.e., the representation of white women amongst all bachelor’s degree recipients).  URM men were at population parity in astronomy and physics when compared to bachelor’s degree representation in 2013, but the rate of bachelor’s degree attainment was about half of population parity (i.e., 19% of 20-24 year olds but just 9% of 2013 bachelor’s degree recipients were URM men).  URM women were substantially less likely to earn a bachelor’s degree in astronomy or physics; while URM women accounted for 15% of all bachelor’s degree awarded in 2013, only 3% of astronomy bachelor’s degrees and 2% of physics bachelor’s degrees were earned by URM women. 

Figure 3. Demography of Astronomy and Physics Bachelor’s Degree Recipients, 2013

What does this mean? The literature about diversity emphasizes that--in addition to the obvious equity issues--diverse individuals working together on complex tasks often develop novel solutions to research problems.  Furthermore, a lack of diversity can become a self-fulfilling prophecy when multiple points of view are not represented.  The explanation for underrepresentation may often focus on deficiencies in “them” as either lacking interest or preparation as opposed to how barriers to diverse participation are erected and, potentially inadvertently, maintained.  Understanding the data on diversity is merely a first step towards wrestling with the more difficult question about how to improve the situation for all.  Subsequent posts in this series will examine trends in graduate degree attainment in astronomy, discussions of the role of gender in shaping the field, how ethnicity impacts access to the field, and success stories of promising programs to increase the diversity of the field.
[1] Acknowledgements: This material is based upon work while the author was serving at the National Science Foundation.  She is grateful to Agata Gluszek and Joan Schmelz for their helpful comments.

[2] Note: the term “ethnicity” refers to a category of people identified on the basis of a common ancestral, cultural, social or national experience.  Due to small population sizes, depending upon the data source, information about relatively small ethnic categories of the U.S. population may be unavailable (e.g., American Indians and Alaska Natives, Native Hawaiians and Pacific Islanders).  In addition, in some instances, for visualization purposes, three categories are often combined as “underrepresented minorities” (URM), which includes American Indians and Alaska Natives, African Americans and Hispanic/Latinos.  The history of ethnic categories and the labels used to reference various groups have varied over the years; we recognize this by referencing categories as demographic aggregations as opposed to groups, which connote belongingness and identity.  Finally, there is increasing research about “multiracials,” individuals who select more than one racial and or ethnic category on standard questionnaires.  Data availability for such a category, however, is currently limited. As U.S. ethnicity continues to change, more data may be available in the future.

[3] A standard curriculum includes four units of English and three each of social studies, science, and mathematics; a “midlevel” curriculum includes these requirements plus a year of foreign language along with the stipulation that the mathematics include geometry and Algebra I or II and that the science include at least two of biology, chemistry, and physics.