Home » Featured

2020 Bachelor’s Degree Recipients in Statistics Show Resilience

1 October 2021 3,277 views One Comment
Steve Pierson, ASA Director of Science Policy

This spring’s respondents to the ASA survey of 2020 bachelor’s graduates reported a challenging, stressful, and prolonged job search. Nevertheless, the results show an unemployment rate (as of March 2021) roughly the same as the corresponding rate for the 2016 and 2017 cohort respondents, median salaries approximately $10,000 larger, and graduates still being hired by a wide variety of companies in many sectors. In addition, those in the work force reported high job satisfaction and that their undergraduate training prepared them well.

Respondents also responded generously to open-ended questions asking about their job searches and what undergraduate experiences likely helped, what they would have done differently, and what advice they have for current students. Those who took full-time jobs spoke highly of the value of internships, hackathons, other extra-curricular statistics experiences, and career fairs. They also wished they’d had more classes in computer science, data science, and mathematics.

The ASA previously surveyed the 2016 and 2017 bachelor’s graduates (and the 2018 master’s graduates). The surveys provide snapshots of the job market experienced by graduates during the dramatic growth in statistics degrees earned annually over the past decade—nearly a six-fold increase for bachelor’s degrees—and allow insights into how student interests, the job market, and curriculum may be changing. This summary will highlight apparent changes, with the effects of the COVID-19 pandemic in the results described in the sidebar at the end of this article.

The ASA also surveyed the 2020 master’s graduated in statistics and biostatistics, which will be reported later this fall in Amstat News.

The results here are intended to be helpful to the following three main audiences:

  1. Current statistics students as they complete their remaining studies and plan for post-graduation
  2. Potential statistics students as they decide on a major(s)
  3. Faculty and administrators as they advise students, design curricula, and allocate resources

To be especially helpful to the third audience, department-specific reports are provided to departments with sufficient numbers of responding graduates. This year, 24 departments received such reports.

Survey Respondents Overview

This year, 425 graduates from 57 universities participated in the survey, which was distributed in the spring and asked about employment or graduate student status as of March 1, 2021. Answering the entire questionnaire were 321 students. These numbers are larger than for the surveys of 2016 and 2017, when approximately 270–295 graduates from roughly 50 universities participated. The number of 2020 bachelor’s graduates in statistics and biostatistics was 4,490, as reported by the National Center for Education Statistics (NCES) for the 2019–2020 academic year.

Of the 319 individuals who responded to the gender identity question, 131 were female, 182 male, two other, and four prefer not to say. The 41 percent who are female is close to the 43 percent in the NCES data.

For citizenship data—where NCES data shows 33 percent of 2020 bachelor’s degrees going to nonresidents—only 15 percent (49) in the ASA survey reported being non-US citizens (23 from China, seven from South Korea, and five from India) and 85 percent (269) reported being US citizens. These numbers follow closely those for the combined numbers from the survey of the 2016 and 2017 graduates. One exception is that Canada had the third-highest number for those years.

The universities with the most bachelor’s degree recipients participating in the survey were Carnegie Mellon University (43), University of California, Santa Barbara (43), University of California, Davis (31), Brigham Young University (28), University of Minnesota (18), and Purdue University (17).

Information About Undergraduate Studies

A sizable portion of the participating graduates had busy academic schedules, with both courses and non-classroom engagements including statistics. Roughly a third of respondents—117—said they graduated with a double major, with the most common companion majors being economics and mathematics (Table 1). Data science is new as a second major. Fifty-six percent (184 graduates) indicated they minored in a field, the most common being computer science (38), mathematics (37), economics (19), and business administration (7). In past years, mathematics and business administration were the most common minors.

To probe the outside-the-classroom learning experiences as undergraduates, the survey asked about internships, research, and capstone projects and how the graduates regarded such experiences for their education and career. One hundred graduates said they completed a thesis or capstone project, 146 an off-campus internship or industrial co-op, 113 on-campus research, 58 DataFest or a hackathon, 17 an NSF Research Experience for Undergraduates, 27 other summer research on another campus, 72 conference presentation or attendance, 92 a graduate course, 40 on-campus internship, and 18 consulting or freelancing. Sixty-one said they served as a teaching assistant.

For these experiences, graduates were asked, “How do you feel your above experience(s) affected your current situation?” The open-ended responses were overwhelmingly positive. The following are a few of the nearly 200 responses:

  • “My capstone project gave me some good material to talk about during job interviews.”
  • “The above experiences allowed me to be competitive in my graduate school applications, and I would not be enrolled in my current department without those experiences.”
  • “Working with baseball data helped confirm I was interested in a career in it.”
  • “Hackathons were by far the best experience, which contributed to me getting a job.”
Figure 1. Word cloud for free responses to what students would have done differently.

Figure 1. Word cloud for free responses to what students would have done differently.

    Despite the generally engaging learning schedules inside and outside the classroom, respondents were unrestrained when asked what they would do differently. Common themes from the 181 responses were taking more computer science, programming/coding, or database courses and, to a lesser extent, mathematics and data science courses. Some of these themes are captured in the word cloud in Figure 1. Responses included the following:

    • “I would secure myself internships in the summer for more experience.”
    • “I would have looked to engage in research opportunities.”
    • “Learn a lot more software programming related to statistical analysis.”
    • “I would have tried to double major.”
    • “Start developing relationships with professors and advisers earlier.”
    • “Attend more nonacademic activities.”

    Many of the respondents also commented they wouldn’t have done anything differently.

    The advice provided for current statistics students tracked with the responses above from graduates about what they would have done differently. Programming and coding skills were most frequently recommended, followed by gaining experience through internships and related activities. More mathematics, double majoring, and learning more statistical programming (e.g., R, SQL) were also recommended, as reflected in the following select responses:

    • “There are a lot more things that a statistics major could do than you could think of, so be open-minded and take any opportunities you could while in college.”
    • “Get an internship!!!”
    • “Get a double major in either CS/finance/liberal arts.”
    • “Do data science projects on your own using data from internet resources like Kaggle to learn new skills and learn how to independently solve issues faced when doing data analysis.”
    • “Do side projects and make a well-documented website/GitHub where you provide examples of your work. Be proactive in the job search and networking. Being a strong communicator can set you apart.”

    Students were generally positive about how well their undergraduate program prepared them. Eighty-six percent agreed or strongly agreed that their program prepared them to effectively analyze and interpret data critically using statistical models, and 79 percent agreed or strongly agreed that their program prepared them to effectively communicate both orally and in written form results of statistical analyses to a variety of audiences. Eighty-five percent agreed their program prepared them to effectively analyze and interpret data critically using computational methods, which is an improvement compared to the 2016 and 2017 figures of 77 percent; there were only minor differences for the other two items in this category.

    Regarding their job search, students generally took advantage of help from their statistics departments and faculty or the on-campus career counseling center. Sixty percent of those who responded said their statistics department or the faculty provided career guidance, and 52 percent said they used their on-campus career counseling center. Of those who used a campus career counseling center, 59 percent found it useful. Many of the open responses reflected help for improving résumés. The 98 comments are available in the supplemental material, with the following being a small selection:

    • “They helped me revise my résumé and cover letter about 10 times and helped with mock interviews.”
    • “The career center on campus was helpful in helping me construct a usable résumé and providing opportunities like career fairs for me to get a sense of the job market situation prior to graduation.”
    • “Helped when I got my offer to decide whether I want to accept and how I should negotiate.”

    After their undergraduate studies, twice as many respondents received jobs as continued their studies, roughly the same as for the classes of 2016 and 2017. See Table 2, which also provides information about those who are unemployed, left the US, have internships, or are full-time volunteers. The 2020 number of unemployed and seeking a job is up a percentage point, and the number of full-time volunteers down a percentage point.

    One hundred nineteen listed themselves in full-time degree programs, and another 53 said they were planning to begin a degree program in the future. The next two sections provide more information about those employed and those enrolled as students.

    Of the 328 who characterized their undergraduate field (Table 3), 62 percent listed statistics, applied statistics, or biostatistics (down 15 percentage points from the 2016 and 2017 surveys of graduates.) Those listing data science were up 17 percentage points to 19 percent for the class of 2020. In the “other” category for the 2020 graduates, six of the 10 respondents specified statistics and machine learning.

    Employed

    The respondents who obtained jobs seem to be well paid and generally satisfied with their jobs. They also work for a diverse and large group of employers with generally unique job titles. The jobs fit into a variety of sectors, shown in Table 4. Nearly three-quarters are in a company or business, as they were in the previous survey for bachelor’s graduates. The survey of 2020 graduates asked for more sector specificity, revealing 23 percent of the 185 respondents working in the technology sector, 17 percent in insurance, and 12 percent in finance or banking. Of the 42 (23 percent) in the category “Other Private Sector,” there was a wide assortment provided, with the most common being some form of consulting.

      The median salary for the 121 employed full-time and providing salaries is $69.9k, the 25th and 75th percentiles being $59.0k and $80.0k, which is approximately $10k higher than the corresponding figures from the 2016 and 2017 respondents. For an approximate comparison, see the corresponding graph at the American Institute of Physics website showing middle 50 percent salary ranges for 15 fields for the class of 2018, along with other useful information.

      The median salary was the highest for those employed in the technology sector, as shown in Table 5. The median salary by state had a wide variation, as shown in Table 6 for states with an n of five or greater. The corresponding information from the survey of 2016 and 2017 graduates is included in the table for reference and, to the extent possible given nonrepresentative samples and other factors, comparison.

      There was less variation in median salary by undergraduate field, as shown in Table 7. The breakdown of salary by gender shows strong parity for men and women at the 1st, 2nd, and 3rd quartiles (Table 8). For those listing themselves as employed and reporting hourly salary, the median hourly rate was $19.50, with the middle 50 percent ranging from $15.75 to $25.08.

      When describing job search experiences and what helped them locate and secure the position, responses varied, with many citing internships, career fairs, career centers, and job search sites (see corresponding word cloud in Figure 2). Following are select responses:

      • “Secured full-time offer from internship.”
      • “[My university’s] career fair was very helpful. All contact with employers I had was through there.”
      • “I applied to lots of different companies directly on their websites. I also significantly prepared for the interview process by reviewing relevant technical concepts and by also reading more about the company.”
      • “I was quick to respond to recruiters and, since I had no prior, relevant work experience, I put my school projects at the top of the résumé.”
      • “I primarily found job applications on LinkedIn and then started interview processes. I learned SQL on my own through online classes, and that helped with several of my technical interview challenges.”
      Figure 2. Word cloud for free responses to describe job search experiences, including what helped you locate and secure the position.

      Figure 2. Word cloud for free responses to describe job search experiences, including what helped locate and secure the position.

      Students were also asked what types of experiences, training, or other qualifications they thought would have helped secure a position. Internships came through the strongest. One graduate responded, “More statistics internship, more projects, more coding classes.”

      Job satisfaction in a number of categories is quite high. Eighty-two percent of respondents were very or somewhat satisfied with the position they held at the time of filling out the survey. Seventy-seven percent were satisfied with their salary and benefits, 91 percent with their job security, 77 percent with the opportunity for advancement, 67 percent with the intellectual challenge, and 82 percent with their level of responsibility. These levels are roughly the same as those for the survey of 2016 and 2017 graduates with the exception of intellectual challenge, which dropped seven percentage points.

      As noted above, the diversity of job titles and number of companies employing statisticians is impressive. There were 112 unique job titles among 170 job titles reported. The most common job titles were data analyst (10), actuarial analyst (10), data scientist (10), software engineer (9), and analyst (8). Except for software engineer, these were also the most common job titles for the previous cohort. Listed were 151 unique employers (of a total 166). Axtria and Epic were the only companies employing more than two of the respondents.

      The 163 graduates who categorized themselves as employed and provided the state were employed in 29 states and the District of Columbia.


      The questionnaire also asked those employed about the frequency of the skills they use in their work. Consistent with the survey of the 2016 and 2017 graduates, the most-used technical skills—as shown in Table 9—are technical problem-solving and data analysis. Managing and querying databases, performing quality control, and programming are in the next tier of most-used technical skills. Experimental and survey design are the least used technical skills. For interpersonal, communication, and management skills (Table 10), teamwork is again the most often used skill by far, followed by working with clients, project management, and writing.

      The survey also asked which tools/languages new bachelor’s graduates use and with what frequency. The survey revealed at least weekly use by those providing answers: SQL (47 percent used weekly or daily), Python (33 percent), R (22 percent), Tableau (19 percent), and SAS (nine percent). Twenty-nine respondents reported weekly or daily use of other tools. See Table 11.

      Postgraduate Study

      Of the 166 respondents, 103 listed themselves as being in a graduate program—35 in statistics, 17 in biostatistics, 21 in data science, and 30 others in 15+ other programs. Computer science, environmental/geophysical/industrial engineering, and business administration were the most common with three each. Twelve others were in fields not included in the drop-down list for that question. Eighteen of the graduates were enrolled in doctoral programs and 70 in master’s, though 24 of the students intend to earn a doctorate degree.

      For the 72 graduates who reported a teaching or research assistant or fellowship as their primary support, the median annual stipend was $20.0k (25th percentile: $16.0; 75th: $25.0k).

      The ASA will conduct a survey of bachelor’s graduates again in two or three years and welcomes input on the frequency of the survey and all aspects of the survey, from questions and administration to reporting and dissemination. Reports on previous ASA surveys of bachelor’s and master’s graduates are available.

      Survey Administration

      The survey was conducted by the Statistical Research Center of the American Institute of Physics (AIP). The ASA provided department names and contacts for the departments granting statistics or biostatistics degrees according to our records. For both the bachelor’s and master’s graduates, AIP reached out to 232 departments, receiving the names and contact information for 2,001 bachelor’s graduates, who received up to four invitations to participate in the survey. Departments that did not provide contact information for graduates were asked to distribute survey invitations to their alumni.

      Class of 2020 Graduates into Pandemic Job Market

      Class of 2020 graduates faced an uncertain job market with the economy reeling from the effects of the COVID-19 pandemic. While the survey did not have pandemic-specific questions, heightened job search challenges were attributed to it. In the free response questions, many bachelor’s students commented on “hiring freezes from the COVID pandemic,” “companies dropp[ing] positions due to the COVID-19 pandemic,” fewer job postings, and canceled interviews. Several respondents mentioned the stress and difficulties caused by the pandemic.

      The effects of the pandemic on the job market may also have come through in the responses to the questions for when a bachelor’s graduate started the position held on March 1, 2021. For the 2016 and 2017 graduates, 60 percent started their positions in May through August. The corresponding figure was 50 percent for the 2020 graduates. Twenty-five percent of the 2020 graduates reporting employment started their positions from September to December, compared to 19 percent for the 2016 and 2017 graduates.

      The unemployment rate for the 2020 graduates as of March 1, 2021, was a percentage point higher than that for the previous cohort (Table 2), which could be due to the pandemic.

      1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)
      Loading...

      One Comment »

      • Eric said:

        Where is Table 5?