So-called “coding boot camps” are often pitched as an alternative to four-year degree programs. Yet new research suggests this is more often not the case and that such boot camp programs are increasingly acting as an auxiliary to college degrees.
Coding schools have seen tremendous growth over the past three years (up by 64 percent according to the 2016 analysis by Course Report). The schools offer brief (10-14 week), intense curricula and promise considerable salaries for their graduates (ranging from $50,000 to over $100,000), while often advertising that no prior coding experience is required for admission.
Certainly, a 10-14 week program represents a fraction of the coursework associated with a four-year computer science undergraduate degree. However, some criticism of the sector seems to operate upon the inherent assumption that going to college or going to a boot camp is an “either/or” scenario for students interested in learning software development.
This is hardly the case.
For the past year, we have been examining the learning opportunities, admission criteria and learner profiles of various coding boot camps nationally. Funded through a pair of National Science Foundation Core Research and Development Grants, our research suggests that while coding boot camps are hardly learning environments equivalent to an alternative to a four-year degree, neither are they the egalitarian learning environments open to all comers that many purport to be in their advertising.
What do we mean? From August through December 2016, we conducted a series of nine national focus groups and six interviews with computer science faculty, coding boot camp instructors and administrators, as well as hiring managers from a range of software development companies. Participants were from over 20 different states, ranging from cities as large as Chicago and New York to smaller towns in North Carolina, California and throughout the Midwest. Perhaps contrary to expectations, coding boot camps are hardly exclusive to the U.S. coasts.
The discussions showed that the country’s top coding boot camps are not so much an alternative to a college degree as a supplement to one. Camps are filled with highly-motivated students committed to making themselves more marketable and economically successful than what their college degrees provided.
That surprised us. And in some cases, the code camp developers we talked to were equally surprised. According to the director of growth and operations at a Colorado boot camp, “(G)oing into this, we had assumptions that a lot of people were going to be doing this boot camp instead of college, and that's actually not the case. When we were looking at the data, we saw that most of our students actually had a college degree, so I would say around 80 percent.” A director of curriculum and instruction at a Chicago-based boot camp was quick to add that it in addition to bachelor degrees, many students had master’s degrees, including one with a master’s in computer science from Yale University. The student was taking the boot camp course to brush up his coding skills.
Based on our early research, it appears that the top coding boot camp programs are as competitive (if not even more competitive) than national undergraduate and post-graduate programs in computer science.
Rich Kochman of the Washington D.C.-based Adaptive Consulting has been monitoring the growth of boot camps for the past three years and sees their future as highly tied to ever-more competitive admission criteria. “Frankly,” he said, “a limited amount of the population has the requisite mathematical and critical thinking skills to be trained as a successful coder in a time period of 12 weeks. This is a tight time frame, and ultimately each boot camp is judged by its success in delivering on their implicit promise to place students in jobs upon program completion; thus, they can't afford to let in too many ‘low potentials’ through the door."
The solution? Advertise widely and vet thoroughly. Camps we spoke to reported having online applications that took up to 12 hours to complete, evaluating applicants with multiple coding puzzles as part of the pre-assessment packet and conducting candidate interviews for up to three hours.
Within our focus groups, one theme of discussion that ran through both camp and university groups was the requisite software development skills each environment offered their students. This question of necessary skills came up 383 times among the code camp participants, while among computer science faculty members it came up 323 times -- commensurate numbers given that we spoke with two more boot camp administrators than computer science faculty members. Here, both code camp participants and college faculty members explained that they offered their students a range of cognitive skills (algorithms and data structures) as well as interpersonal skills through internships and team-based projects. However, while colleges provided those types of opportunities in capstone courses, the majority of code camps expected their students to engage in team-based, cooperative learning environments by the second week of their 10- to 12-week programs.
As any good teacher knows, effective group work is neither easy to teach nor naturally occurring, so code camps’ expectations of their student admits are significant, particularly given the short timeline.
To the extent that coding camps expect such collaborative aplomb from their students almost immediately offers some insight as to why participating code camps were four times more likely than colleges and universities to focus on student recruitment and admission (165 mentions versus 35 mentions in the focus group transcripts and, remarkably, nearly seven times more likely to focus on the learning profiles of their students (136 versus 20). Of course, some of this wide discrepancy stems from the fact that undergraduate computer science programs largely rely on external admission departments who recruit from the same pool of high school students. These entering college students, while diverse across certain measures (race, ethnicity, socioeconomic background), are largely homogenous in terms of age and educational background. Code camps have no such homogeneity in their applicant pool. Based on our focus groups, code camp applicants range in age from 19 to 74 and in education from high school credential to master’s and doctoral degrees.
What is the leading common denominator among this diverse group of code-camp applicants? It may be a college degree. The camps we spoke with self-reported between 66 percent (on the low end) to 80 percent (on the high end) of their students having college degrees. Not surprisingly, some of the top colleges and universities in the country recently figured this out. As an August 2016 Inside Higher Ed article noted, Northeastern University in Boston has developed its own boot camp, while other institutions such as Northwestern University, Bellevue College and the wider University of Texas system all have partnered with external coding boot camps to offer their own version of intense, practically-minded coding coursework. Top providers such as General Assembly, Revature and Trilogy are actively recruiting college and universities they can partner with, working directly with their students to add on software development certifications and thus producing graduates who earn more upon graduation. Just this past June, the boot camp Trilogy, a six-month long program, reported raising $30 million in venture capital financing to continue to grow its list of 25 university partners.
What is clear from our research on industry expectations for new hires is that college degrees very much matter. All of the 14 hiring managers and directors we interviewed this past fall through the same series of national focus groups reiterated the importance of teamwork, clear communication and timeliness -- so called “soft” skills. Where do these skills come from? Hiring managers ultimately had different answers here, but they were collectively uniform on the necessity of requiring a college degree for entry-level software development positions to ensure such soft skills are already in place. While the debate around the merit of a four-year college degree continues, in the field of software engineering, a bachelor degree still matters -- a lot.
Working with the analytics software company Burning Glass, we found that the top skills companies were looking for from entry level software developers in 2016 were programming language-specific. (Java, Microsoft C#, C++, SQL and JMS were the top in advertised positions nationally.) However, simply having proficiency in one such programming language was not sufficient. Over the past three months, software developer job postings overwhelmingly required a bachelor’s degree (98 percent). And 23 percent of software developer positions required a graduate or professional degree, while only 3 percent listed an associate degree as a minimum educational requirement. Yet of the 38,000 postings tagged “software developer” nationally over the past three months, only half (49 percent) likewise posted a minimal educational requirement, which may suggest that companies may be more flexible with degree requirements. Nonetheless, the current data is clear: If software development companies are explicitly weighing educational attainment as part of the hiring process, even among junior level developer positions, associate and high school degrees simply do not cut it.
With all the discussion of alternative educational environments, it is easy to casually dismiss the importance of a college degree. A recent New York Times article states, “At the nation’s coding boot camps, the pitch has nothing to do with a liberal arts education. This thriving niche of the for-profit education business tightly links courses to the job market.” And yet this is not the case. While coding boot camps may not be pitching a liberal arts education, they are certainly pitching their services to liberal arts graduates.