Internship Participation in the United States by Student and School Characteristics,1994 to 2017

Abstract

College internships have become seemingly ubiquitous, yet evidence of when and for whom participation has changed over time remains limited. This visualization uses data from the College Senior Survey, an annual nationwide exit survey of baccalaureate graduates, to document internship participation by student and school characteristics between 1994 and 2017. Recent graduates are more than twice as likely to participate as those from the mid-1990s; however, the results indicate significant and sustained inequalities for lower income students, first-generation students, public school students, and students from less selective schools. These ongoing participation gaps for students with less individual and institutional privilege underscore the need to consider internship access as a form of educational and labor market disadvantage.

In 1994, the New York Times published an article that posed a provocative question: “Who would want a job that may offer no pay, will last only a few weeks and more than likely mean a lot of drudge work?” The answer was college students: those who “clamor” for the opportunity to gain contacts and job experience as part of an internship. Yet in the early 1990s, the author estimated that only 9 percent of graduates hired by companies had internship experience (Goss 1994). Internships, by available accounts, were relatively anomalous 30 years ago. ¹

They have become seemingly ubiquitous since then, but knowledge of when internship participation has changed over time and for whom remains limited, despite evidence that internships both facilitate employment and reproduce inequality (Armstrong and Hamilton 2013; Arum and Roksa 2014; Frenette 2013; Jack 2019; Rivera 2016). Historically, longitudinal or repeat cross-sectional internship-specific information has not been collected using publicly available data sources such as population-based surveys, education cohort studies, or federal administrative databases. ² Yet this information is crucial for establishing baseline information about the role of internships for job seekers as well as within educational institutions, work organizations, and the labor market.

These analyses use data from the 1994–2017 College Senior Survey (CSS) (HERI 2021a; HERI 2021b), an annual nationwide exit survey of baccalaureate graduates, to document time trends in internship participation by student (racial/ethnic background, income, and first-generation status) and school (selectivity, institutional control) characteristics. Overall participation and participation by sex are presented in the Online Supplement.

Results

Figure 1 presents five panels, arranged by (1) student income quartile, (2) student first-generation status, (3) student race/ethnicity, (4) school institutional control, and (5) school selectivity. See the Online Supplement for additional information about Methods.

Figure 1.

Percentage of graduating seniors participating in at least one collegiate internship, by year and by student and school characteristics. Data are from the 1994–2017 College Senior Survey, collected annually as a nationwide exit survey among graduating seniors (n = 332,832). The left side of each panel presents the percentage of students participating in an internship by year, displayed in three-year moving averages. The right side of each panel presents the percentage-point change between the first and last moving averages over time. Additional information about methods, along with unaveraged annual percentages, hypothesis tests, overall participation, and differences by sex, is available in the Online Supplement.

Overall internship participation, on the basis of the CSS, has increased dramatically in the past three decades, from 29.9 percent in 1994 to 70.5 percent in 2017. This equates to a 41 percentage-point increase, or 136 percent. Under the assumption that the internship participation rate in the CSS matches the internship participation rate nationwide, the baccalaureate class of 2017 completed at least 1,379,000 college internships before graduating. ³

Student Characteristics

These trends are not equal by student sociodemographic characteristics. Students in the top income quartile are significantly more likely than those in lower income quartiles to participate in internships in all observed years after 1998 and have the highest percentage-point increase in participation over time. With few years of exception, the same is true for continuing-generation students compared with first-generation students. Differences by student racial/ethnic background are more complicated. Although non-Hispanic Asian and White students have the largest percentage-point change in participation, in some years Black students have the highest participation, and in other years there are no significant differences among groups. Finally, female-identified students tend to be more likely to intern than male-identified students, although the latter experience a larger percentage-point increase over time.

School Characteristics

These trends are also not equal by students’ school characteristics. Although students from less selective schools were more likely to intern than students from more selective schools in 1994 and 1995, these trends reversed in 1997. Not only did students from the most selective schools experience the greatest percentage-point change over time, but in the 2000s and 2010s, they were often at least 10 percentage-points more likely than those from less selective schools to intern. In nearly every year, students from private versus public schools are also significantly more likely to participate (despite the larger percentage-point increase over time among public school students).

Conclusions

This visualization documents the prevalence of internship participation among baccalaureate graduates in the United States over time and by student and school characteristics. Recent graduates are more than twice as likely to participate as those in the mid-1990s. However, lower income and first-generation students, and those from public and less selective schools, remain significantly less likely to participate than their more privileged peers.

Supplemental Material

sj-docx-1-srd-10.1177_23780231221121058 – Supplemental material for Internship Participation in the United States by Student and School Characteristics, 1994 to 2017

Supplemental material, sj-docx-1-srd-10.1177_23780231221121058 for Internship Participation in the United States by Student and School Characteristics, 1994 to 2017 by Carrie L. Shandra in Socius

Supplemental Material

sj-pptx-3-srd-10.1177_23780231221121058 – Supplemental material for Internship Participation in the United States by Student and School Characteristics, 1994 to 2017

Supplemental material, sj-pptx-3-srd-10.1177_23780231221121058 for Internship Participation in the United States by Student and School Characteristics, 1994 to 2017 by Carrie L. Shandra in Socius

Supplemental Material

sj-xlsx-2-srd-10.1177_23780231221121058 – Supplemental material for Internship Participation in the United States by Student and School Characteristics, 1994 to 2017

Supplemental material, sj-xlsx-2-srd-10.1177_23780231221121058 for Internship Participation in the United States by Student and School Characteristics, 1994 to 2017 by Carrie L. Shandra in Socius