A key principle of virtually every engineering licensure law, standard of professional conduct and code of ethics is that engineers should always and only practice in the technical areas in which they are competent. The licensing process uses education, experience and examination to identify those who have presumably acquired sufficient competence.
However, there is a loophole of sorts: In most jurisdictions, once engineers become licensed, it is entirely up to them to define their own areas of competence within the profession as a whole.
Engineers typically feel like they have a pretty good handle on this. We think that we can readily discern which assignments are within our capabilities, and which ones we ought to pass along to someone else or simply decline.
The truth, though, is that people routinely overestimate their competence without ever realizing it. Research has shown that it requires competence in order to detect competence (or incompetence). Incompetent persons are often unaware of their incompetence unless and until a mistake of sufficient magnitude makes it apparent.
Psychologists call this the Dunning-Kruger effect, after the authors of a landmark 1999 paper. David Dunning and Justin Kruger showed that human beings “tend to hold overly favorable views of their abilities in many social and intellectual domains.” More importantly, those who are incompetent are actually unable to realize it. In fact, the only dependable way to help others recognize their incompetence is to increase their competence!
This clearly refutes the nearly universal assumption that licensed engineers are inherently qualified to judge their own competence to carry out any particular engineering task. It is not a matter of an ethical deficiency, where individuals willfully provide services that they know are beyond their range of expertise; rather, it is a cognitive deficiency, where those individuals are quite innocently ignorant of the inadequacy of their knowledge and/or skills.
Since self-assessments of competence are so subjective (and unreliable), what is needed is an objective way to establish competence. Fortunately, one such method already exists: the discipline-specific technical examination. Assuming the test has been developed properly, someone who passes it is likely to practice competently within the particular branch of engineering that it covers —but not necessarily in others.
With this in mind, consider the case of structural engineering. In 2006, the National Council of Examiners for Engineering and Surveying (NCEES) convened a task force to review its examinations in that field. This highly competent group determined that an eight-hour test—the standard for all other disciplines—is insufficient to establish adequate competence to practice structural engineering; instead, 16 hours is the appropriate length. The task force also advocated modernizing the examination format and specifications to be consistent with current (and increasingly complex) building codes. NCEES adopted these recommendations, and the new test that resulted is being introduced nationwide in April.
This suggests that at least some engineers who passed an eight-hour examination in another discipline may be exhibiting the Dunning-Kruger effect when they deem themselves competent to practice structural engineering. Certainly, many individuals whose education and experience were heavily concentrated in the design of structures do have the necessary competence. But how is a client, code official or ordinary citizen supposed to tell the difference? Separate licensure of structural engineers helps to solve this problem. Several states already have taken this step, others are exploring it, and all should seriously consider it.
More broadly, engineering licensure laws, standards of professional conduct and codes of ethics probably need to be revisited. Engineers should no longer be expected to evaluate their own competence and practice accordingly; instead, they should be required to remain strictly within the scope of the examination that they passed to become licensed.
In light of the Dunning-Kruger effect, such a constraint is necessary for engineers to fulfill their most fundamental responsibility: holding paramount the safety, health and welfare of the public.