Why we should train workers like we train machine learning algorithms
The evolution of workforce opportunity in the United States depends on the future of education and our commitment to far-reaching, equitable federal reform. Unfortunately, policy conversations at the federal and state levels about transforming education systems to meet future workforce demands have focused disproportionately on a skills agenda, largely ignoring behavioral competencies that often complement and enhance the value of technical skills.
This misguided approach equates 21st-century workforce development with skills acquisition, which only serves to reinforce a two-tiered workforce: those who are best positioned to acquire and monetize their skills will be granted mobility and long-term security while all others continue to be stranded on the bottom rung of the socioeconomic ladder.
An agenda dedicated exclusively to skills acquisition cannot deliver the dynamic workforce employers require for several reasons. Technology is progressing rapidly. Today’s widely used tools may be obsolete six months from now, yet it is impossible to predict which specific technical skills will be required for future jobs. Training students solely on technical skills does not guarantee they will possess the behavioral competencies necessary to perform in real-world workplaces. Investing scarce public resources into training for skills with an uncertain shelf life and fluctuating value from employer to employer is poor public policy.
Which skills do employers really want?
Developing intelligent policies to combat workforce inequality requires acknowledging that employer demand for “skills” actually refers to a constellation of content knowledge, technical abilities, and applied intelligence.
Per the National Association of Colleges and Employers’ 2018 Job Outlook survey, eight out of 10 employers reported that applicants’ problem-solving and teamwork abilities influenced hiring decisions; only six out of 10 employers reported the same for technical skills. An earlier Economist Intelligence Unit survey revealed similar findings: problem-solving and the ability to work as part of a team were the top workforce skills CEOs considered when making hiring decisions.
This set of priorities has emerged in many other employer attitudinal surveys, a trend that prompts several questions: Are problem-solving and teamwork actually “skills,” or are they behavioral competencies? If problem-solving and teamwork are discrete skills, then how do we train for them? If they are competencies, do these survey findings suggest that employers are conflating skills with behaviors?
Problem-solving and the ability to work in team settings are more accurately defined as cognitive skills or, better yet, behavioral competencies related to yet distinct from mechanical, task-oriented skills. These competencies describe functional behavioral patterns. Effective problem-solving involves integrating factual knowledge with experiential contexts to create a solution. We marvel at the successes of machine learning algorithms, but problem recognition and the freedom to learn from failure are what make these algorithms so effective in solving domain-specific problems. From the success of machines, we can infer (with increasing confidence) that the realization of optimal solutions requires exposure to diverse experiences. Humans need the same exposure that we give machines.
Employer demands for problem-solving and teamwork abilities encompass a desire for dynamic workers who can combine content knowledge and experiences in real-world contexts. Despite a bevy of behavioral research indicating “little transfer of training from one type of problem to another or even across different versions of the same problem,” we cling to the idea that mechanically training for problem-solving and collaborative teamwork is possible, measurable, generalizable, and transferable to unpredictable real-life environments.
Training for technical skills without behavioral competency and context adaptations ensures that only individuals whose résumés reflect an appropriate mix of specific skills and behavioral competencies will have access to employment opportunities. This pattern erects labor market barriers and compounds disparities. Weinberger (2014 “The Increasing Complementarity between Cognitive and Social Skills,” The Review of Economics and Statistics) confirmed the existence of a wage premium for two cohorts (1972 and 1992) of young white men whose high school tenure included participation in sports and other leadership activities, which she argues helped cultivate their social skills prior to entering the labor market. Even after controlling for a host of socioeconomic factors, including college attendance, math scores, and family background, her results remained robust: “the labor market increasingly favors workers with strong endowments of both cognitive and social skills.”
The skills acquisition discourse must recognize the distinction between technical (mechanical) skills and behavioral competencies, the latter of which are context-dependent. A skills–competency typology that articulates the importance of context will ensure that students from all backgrounds have the freedom to achieve economic, career (horizontal and vertical), and spatial mobility. In this reformulated agenda, acquisition of technical skills will remain a priority, but the addition of behavioral competencies will underscore the need for experiential learning and shift the focus from learning outcomes to learning inputs.
Additionally, education reforms must include employer feedback, which is essential to the social valuation of skills and competencies. Constructive criticism will allow us to design effective lifelong learning models that blend technical skills proxied by content knowledge with behavioral competencies as measured by experiential learning. The result of improved educational outcomes is jobs!
Empirical studies have begun to quantify the link between the social competencies employers have identified as valuable above and beyond technical skills. Deming and Kahn (2017) found that “a [one] standard deviation increase in demand for social skill is associated with a 5 percent wage increase.” Deming (2015) corroborated this finding and added evidence of occupational sorting to the skills–competency debate. Using a combination of O*NET occupational and National Longitudinal Survey of Youth data, he noted that in addition to a wage differential benefiting socially adept individuals, workers generally sort themselves into occupations based on their cognitive and social skills. According to Deming’s research, workers who self-select into occupations requiring high-level social skills “earn about 3.9 percent higher wages” than those in positions requiring fewer social skills.
Training for behaviors that matter
The need for individuals with the right mix of skills and behavioral competencies is not solely a workforce issue; it is also permeating higher education, as colleges and universities seek ideal students who are creative, dynamic, and possess high-order analytical skills. College applicant evaluations have hence expanded beyond standardized test scores to a more holistic set of parameters. Unfortunately, this shift has complicated the transition from high school to higher education for disadvantaged students, who often lack the exposure to applied experiences necessary to develop these behavioral competencies.
Here, we can again learn from machines. Exposure to experiential learning underpins the success of artificial intelligence and machine learning algorithms. An algorithm’s performance improves incrementally from high-frequency exposure to a variety of data and context adaptations from which the computer tries, successfully and unsuccessfully, to identify the appropriate contextual response to a situation or problem. Computer science experts call this process “reinforcement learning.” Students need similar high-frequency exposure to reinforcement learning, at the K-12 and tertiary levels, to become dynamic, problem-solving, collaborative applicants who meet universities’ and employers’ expectations. This metamorphosis does not transpire through magic; it requires education reform.
Experiential learning at its core involves reframing the delivery of factual content into an applied model that allows “an individual to gain practical experience relevant to their academic training.” Such models provide ideal settings for cooperation and likely develop teamwork abilities more effectively than a monochromatic training program. In a must-read critique of our current approach to educational training, Hoffman (2017) noted that “all young people, not just those from low-income families or with weak academic preparation, need more information about careers, more structure in their school-to-post-secondary pathways, and much more experience of the workplace than ever before.”
Educational siloes divorced from real-world contexts will not help students develop the employment competencies and dispositions that everyone—irrespective of socioeconomic standing—needs to be successful in our increasingly tech-centric world. Behavioral competencies carry economic consequences related to wage disparities as well as the two-tiered workforce into which future workers will self-select based on their technical and social skills.
Makada Henry-Nickie is the David M. Rubenstein Fellow for governance studies, race, prosperity, and inclusion initiative at Brookings. This article originally appeared here.
TimeControl 8 Raises the Bar in Automated Time Management