Stop Inventorying Skills: What Course Evaluation Should Ask About Career Adaptability

The short version

Even the more sophisticated, employability-aware course evaluations tend to do the same thing: inventory skills. Did you develop data analysis? Communication? Teamwork? It is a useful question, but it quietly assumes the world graduates enter is stable enough for a fixed skill list to matter. It is not. What best predicts how graduates fare across a volatile career is not a static inventory but career adaptability, the psychosocial resources that let a person navigate transitions, shocks and reinvention. And unlike "employability" in the abstract, career adaptability is a defined, measurable construct with evidence behind it. Course evaluation that ignores it is measuring the cargo and not the compass.

What career adaptability is (and why it beats a skills list)

Career adaptability comes from Mark Savickas's career construction theory, operationalised in the Career Adapt-Abilities Scale (CAAS), which Savickas and Porfeli validated with measurement equivalence across 13 countries (CAAS validation). It has four dimensions, easy to remember as the four Cs:

• Concern, awareness of and preparation for one's vocational future.
• Control, a sense of responsibility and agency over one's career.
• Curiosity, the disposition to explore possible selves and work environments.
• Confidence, self-efficacy to overcome obstacles and solve problems.

The reason this matters more than a skills checklist is empirical, not rhetorical. A meta-analysis by Rudolph, Lavigne and Zacher found that career adaptability shows incremental predictive validity, above and beyond other individual differences, for a range of career, work and well-being outcomes (meta-analysis). Related studies find career adaptability positively linked to self-perceived employability, with an especially strong contribution for those actively seeking work. A skill you learned in 2024 may be automated by 2028; the capacity to keep learning and re-steering is what carries a graduate through that change.

Why this is a course-evaluation question, not just a careers-service one

The instinctive objection is that adaptability is the careers service's job, not the curriculum's. But that is exactly the compartmentalisation that leaves it unmeasured. Programmes shape adaptability whether they intend to or not, through how much autonomy students get, whether assessment rewards exploration or only compliance, whether ambiguity is scaffolded or punished. If a programme is building, or eroding, the four Cs, its evaluation should be able to see that.

Most evaluations cannot, because they ask about satisfaction and content delivery. We have argued that course evaluation should move beyond satisfaction to skill development; career adaptability is the next step beyond that, because it captures the durable psychosocial resources a skills list misses. It also sits more honestly within the limits we have described: a course cannot directly measure employability, but it can measure proximal mediators, and adaptability is one of the best-evidenced proximal mediators there is. Where a signalling-versus-human-capital debate stalls on what a degree really adds, adaptability is a candidate answer that is both theorised and measurable.

"But isn't this scope creep, and can students even self-report it?"

Two strong objections, both worth taking seriously.

"Universities are not career coaches." Some academics will resist the idea that a history or physics module should be judged on whether it made students more career-confident. The concern is legitimate: education is not reducible to employability, and adaptability that is taught cynically becomes another box to tick. The response is to be precise about the claim. Career adaptability is not vocational training; concern, control, curiosity and confidence are educational goods in their own right, they describe an autonomous, self-directing learner, which is what a good degree produces regardless of its subject. Measuring them is not surrendering to the market; it is checking whether the programme develops agency, a value the academy already holds.

"Self-report of adaptability is unreliable." A student who feels confident is not necessarily adaptable, and self-perception is subject to the Dunning-Kruger and response-shift problems well documented in the measurement literature. This is real, and it constrains the claim. Two disciplines keep it honest: measure at the programme level over time, not from a single module, because adaptability is built across years and co-curricular life, so attributing a CAAS shift to one course is a mistake; and triangulate self-report with behavioural signals (internship uptake, help-seeking, elective breadth) rather than treating a self-rating as ground truth. Career adaptability belongs in the evaluation mix as a well-validated addition, not a replacement for everything else, and certainly not as a lagging metric that only arrives once it is too late to act on.

Building the four Cs, and asking about them

If a programme is going to be evaluated on adaptability, it is fair to ask how a course builds it in the first place, and the answer maps onto the four dimensions closely enough to guide both teaching and the questions you ask. Concern grows when a curriculum makes students look forward: capstone projects framed around where a field is heading, assessments that ask students to locate themselves in a professional trajectory rather than only recall content. Control grows with genuine autonomy: choice of topics, negotiated assessment, the experience of a decision having consequences the student owns. Curiosity is fed by breadth and permission to explore: elective range, exposure to adjacent disciplines, problems with more than one defensible answer. Confidence is built the way self-efficacy is always built, through mastery experiences scaffolded so that difficulty is challenging rather than defeating.

That mapping tells you what to ask. Instead of "how satisfied were you with the course," an adaptability-aware evaluation asks dimension-specific questions and then probes them: whether the course changed how clearly a student sees their professional future (concern), whether they felt real ownership over their learning choices (control), whether it opened up possibilities they had not considered (curiosity), whether they leave more willing to tackle problems they do not yet know how to solve (confidence). Each of these is a poor fit for a bare number and a good fit for a short conversation, because the useful information is in the why, not the score.

Timing matters as much as wording. Adaptability is formative by nature: a confidence deficit spotted at the end of a programme is a post-mortem, whereas the same signal caught mid-cycle is something a tutor can still act on. That is the argument for collecting this feedback while the cohort is present, not only in the exit survey, and for treating it as one input into a longitudinal picture rather than a verdict on a single module.

Where Koji fits

The four Cs are exactly the kind of construct a Likert checkbox flattens and a conversation reveals. "Rate your confidence, 1 to 5" tells you almost nothing; why a student feels more, or less, able to steer their career after a programme tells you what to change.

Koji for Education is suited to this because it does not stop at the number. Its AI-moderated conversational interviews can probe each dimension where it lives, whether a course expanded a student's sense of what they might become (curiosity), whether it gave them a felt sense of control over their path, whether it built confidence to tackle the unfamiliar, and follow up on the answer instead of averaging it. Its six structured question types let you pair a validated scale item with an open-ended probe; its automatic thematic analysis surfaces adaptability themes students raise unprompted; and its programme- and institution-level reporting matches the level at which adaptability should actually be assessed. Because feedback can be collected mid-cycle, a programme can act on a curiosity or confidence gap while the cohort is still present, not diagnose it after they graduate. The same conversational engine powers the general Koji platform for careers and graduate-outcomes teams running longitudinal studies.

The takeaway

A skills inventory tells you what a programme loaded into its graduates on the day they left. Career adaptability tells you whether they can keep navigating once the map changes, and the evidence says that capacity predicts outcomes better than the inventory does. Measuring it well means resisting two temptations: treating a self-rating as proof, and treating adaptability as vocational box-ticking. Done with that discipline, asking whether a course built concern, control, curiosity and confidence is not scope creep. It is finally evaluating the thing a degree is supposed to leave behind.