In a previous Forbes post, we considered the disconcerting reality that while we find ourselves in the midst of an unemployment crisis—one characterized by highly educated candidates who are unable to find skilled work—employers from myriad industries are nevertheless reporting that they’re unable to find the talent profiles they’re really looking for. And evidence indicates that this talent gap primarily refers not to an absence of technical skills, but to an absence of “soft skills,” or what we’ll call 21^st century skills. These primarily refer to interpersonal and general analytic abilities like: teamwork, empathy, leadership, negotiation, adaptability, and problem solving.

This is highly useful information to students and educators, but employers can also learn a lot from this research about how to hire successful candidates. The problem is that 21^st century skills are very difficult to assess with any kind of rigor, especially before one can evaluate a candidate on the job. Can a candidate think innovatively? Collaborate with other team members? Assimilate feedback and coaching? Will the candidate get along with her team members and other colleagues? Will she bring personality strengths to the table that the current team might be lacking? Will the candidate be adaptable to new environments and successfully integrate with teams? It is very difficult to reduce these questions to discrete qualifications and quantifiable metrics in the same way we can assess recognized degrees and numerical grades.

Certainly some approaches exist. For example, businesses have used “type”-based personality tests for decades in attempts to measure the 21^st century skills of prospective candidates, assuming that certain personality types would correlate with high performance. One example is the Jung Typology Profiler for Workplace, which purports to measure qualities such as “Power” (leadership index), “Assurance,” “Visionary,” “Rationality,” and so forth.

The reality is that personality tests such as these have serious methodological flaws and lack the statistical reliability to predict performance among prospective employees. In fact, the makers of the Myers-Briggs Type Indicator (MBTI), a closely-related profiler that also has its origins in Jungian typology, clearly state in their ethical guidelines that “It is unethical, and in many cases illegal, to require job applicants to take the Indicator if the results will be used to screen out applicants.”

It’s clear that we need 21^st century methods to assess 21^st century skills. Unfortunately, that seemingly simple idea proves to be much trickier in practice than it is in theory.

Tools for talent development do not work for pre-employment screening

Part of the problem is that many companies are using the wrong tools for the job.   There is a fundamental difference between tools intended to develop existing teams and tools used for pre-employment selection.

While typical personality tests are poor tools for pre-employment screening, there is evidence that newer tests can help managers better develop and deploy the talent they have already hired. For example, Gallup’s StrengthsFinder 2.0 is a tool that helps individuals understand and describe their own talents, and is commonly used by managers to understand and capitalize on the strengths of those they hire. More importantly, it is methodologically sound, and its reliability and validity are backed up by clear evidence.

For example, Facebook uses StrengthsFinder in a clever way to deploy talent efficiently. Regardless of the job openings they have available, Facebook simply hires the smartest people it can find, then uses StrengthsFinder results to understand their talents and create a job tailored to the candidate.

One might naturally assume that the same type of test that helps identify and develop strengths in an existing team could also be used to assess suitable candidates for entry into that team. In the words of Gallup, “Absolutely not… A development-oriented assessment such as StrengthsFinder is markedly different from selection tools because its purpose is not to assess whether an individual is suited for a particular job or role. Instead, it aims to provide talent insights for developing strengths within roles.”

Pre-employment selection tools can predict employee performance on the job

Many pre-employment selection tools succeed at predicting performance because they have a completely different design than development tools like personality tests. Instead of seeking general traits and preferences, selection tools are tailored to a particular job in a particular organization, and are statistically calibrated to provide reliable predictive results (i.e., candidates who score highly on these tests also tend to perform well after they’re hired). In addition to the StrengthsFinder development tool, Gallup also offers these pre-employment selection tools, which include analytic services to ensure the validity and predictive value of the measures for candidate screening.

Pairin, Inc. is another organization that seeks to combine the personality test approach with specialized testing (for specific jobs, values, culture, etc.) as part of a pre-employment selection system. Using the Job Pairin System, employers can assess the presence of around 100 coachable/changeable behaviors such as emotional intelligence, leadership, attraction of followers, and even character.

A new spin on the behavioral interview

While services from Gallup and Pairin provide strong, evidence-based methods, the debate on using metrics to assess 21^st century skills will certainly continue. For good or bad, it is unlikely that the traditional way to measure 21^st century skills – the behavioral interview – will be unseated anytime soon. (Behavioral interviews are those that include questions like “Tell me about a time when you worked effectively under pressure.”)

Certainly, behavioral interviewing has problems of its own – for example, canned and otherwise disingenuous responses are all too common. While most companies still use a behavioral interviewing approach, those with top hiring practices tend to put their own clever spin on the questions to weed out rote responses and thus generate better insights from candidates. Questions like this tend to yield a wider spread between canned responses and those that show more nuance and self-awareness. Google also uses behavioral interviews, but structures them in a way that allows them to perform analytics and prove that certain responses predict employee performance.

Beyond the interview

Whatever method companies use to assess 21^st century skills in prospective employees, it’s important that they reflect on the key principles behind the assessment approaches. Findings from research on 21^st century skills provide an extremely valuable lens through which companies can view all interactions with candidates. Consider: What can you teach a new hire on the job, and what can you not teach? With information now abundantly available to us, almost anyone can learn basic Photoshop skills, for example, via online seminars. But what about abstract but indispensable skills like connectedness and empathy – can they be taught on the job?

When you interviewed the person who is now your highest performer, how did you know she would outperform the rest? Did you spot her innate ability to relate to other people, her ability to intuit the needs of different kinds of people? What were the indicators that she possessed those abilities?  We’d like to hear your thoughts on these questions and your experiences with hiring for 21^st century skills.

This post was written by Kate Jenkins, editor at Ashoka and editor-in-chief of The Intentional Quarterly, and James Ray, a Strategy Consultant for Nonprofit, Human Services, and International Development clients at Accenture It was originally published in Forbes and was legally licensed through the NewsCred publisher network  This post does not necessarily represent the views of Accenture. Learn more about SmartRecruiters, your workspace to find and hire great people.