Nursing students rated their practicum skills highly, but the map shifted under the numbers
A cross-sectional study of 699 third-year nursing students in northern Taiwan found a sharp mismatch between how students valued fundamental practicum competencies and how satisfied they felt with their own performance. The overall ratings were high across domains, yet the analytic picture was less tidy than the top-line scores suggest. In particular, the nursing-process domain sat above most other competencies in absolute terms but still trailed the sample mean on both importance and self-rated performance satisfaction.
The study used an importance–performance analysis framework to separate perceived priority from perceived execution. That approach can be useful in nursing education because it does not assume that what students care about most is also where they feel most secure. Here, the sample grand means were 4.679 for importance and 4.503 for performance satisfaction, and those means served as the crosshairs for the quadrant analysis. Bootstrap resampling added another layer of scrutiny, and the resulting classifications were not uniformly stable.
Communication and biomedical-science domains moved around enough across bootstrap samples to raise caution about over-reading any single quadrant placement. In practical terms, the map itself was somewhat fragile.
A large cohort, but still one department and one moment in time
The cohort came from a single Department of Nursing, where all 730 eligible third-year students were invited during October to November 2023. Of those, 28 did not participate, 702 returned consent forms and questionnaires, and 3 incomplete questionnaires were excluded, leaving the final analytic sample of 699. The design was cross-sectional, so it captured perceptions after completion of the summer fundamental nursing practicum rather than changes over time.
Participants were drawn from 16 classes and had rotated through 29 medical-surgical wards or practicum units at 22 accredited teaching-hospital campuses. Supervision came from 29 school-employed clinical instructors, with one instructor overseeing seven students per rotation. Even with that structure, participant-level identifiers for class, ward, instructor, hospital, or rotation were not collected, which meant cluster effects could not be modeled. The analysis therefore had to stop short of the kind of multilevel inference that would be needed to separate individual sentiment from unit-level influence.
High internal consistency may reflect redundancy as much as reliability
The Core Competence in Fundamental Nursing Practicum Scale performed with extremely high internal consistency: alpha and omega were .987 and .987 for importance, and .985 and .986 for performance satisfaction. Those values sound reassuring at first glance, but they also suggest the item set may contain overlapping content. When internal consistency becomes that high, the scale may be measuring the same idea in several slightly different ways rather than spanning a broad competence domain with crisp separation.
That possibility matters in education research because a scale can be statistically coherent and still be too repetitive to add much interpretive precision. If items are highly redundant, total scores may look stable while the instrument contributes less nuance than intended. The study’s own framing treated the findings as perceptions rather than objective competence, and that distinction is clinically important when the goal is to understand readiness for practice.
The nursing-process domain carried the clearest gap
Among the competency domains, nursing process stood out with a paired importance–performance gap of 0.229, and the 95% confidence interval ran from 0.179 to 0.280. This was not a trivial numerical sliver; it marked a consistent separation between what students believed mattered and how satisfied they felt with their ability to deliver it. Because the scale used the sample grand means as crosshairs, the domain also fell below both thresholds despite its high absolute ratings.
A domain can be viewed as important in the abstract and still feel underpowered in real practice. For nursing students, that gap often traces back to the cognitive load of moving from classroom algorithms to bedside decision-making: assessment, planning, intervention, and evaluation must happen under time pressure, around real patients, and under supervision. The study does not prove where the weakness originates, but the pattern fits a familiar educational bottleneck in which process knowledge is understood yet not fully internalized as routine bedside action.
Learning stress rose in the lowest achievement group
Exploratory analyses using the adapted learning-stress questionnaire found higher reported stress scores among students in the lowest academic-achievement category. The relationship was examined with analysis of variance and independent-samples t tests, and the investigators applied Benjamini–Hochberg false-discovery-rate correction. Adjusted linear models used heteroskedasticity-consistent HC3 standard errors.
The findings point in the same direction often seen in educational settings: students who struggle academically tend to report more stress, whether because the workload feels heavier, confidence is lower, or both. Still, the data remain cross-sectional, and no causal direction can be assigned. Lower grades may contribute to stress, stress may impair academic performance, or both may arise from a third factor such as prior preparation, sleep disruption, or coping style.
Counseling history was descriptive, not a test of counseling benefit
The study also compared stress scores by psychological-counseling history. Those comparisons were descriptive only and were not designed to assess whether counseling was effective. Without a randomized design or a longitudinal structure, counseling history is vulnerable to confounding by indication; students who seek counseling often differ from those who do not in ways that matter long before the questionnaire is completed.
A separate psychometric issue emerged as well. The prespecified two-factor structure for the adapted stress questionnaire did not hold up in confirmatory factor analysis, whether estimated by maximum likelihood or ordinal diagonally weighted least squares. Within-cluster dependence could not be modeled because the relevant identifiers were not collected. The result is a stress measure that may be informative as a broad descriptive tool, but not one that can be treated as a settled construct in this dataset.
Written comments added texture, not a new conclusion
All 699 analyzed participants answered both open-ended questions, and after non-substantive entries were removed, 217 non-mutually-exclusive segment-to-theme assignments remained for thematic sorting. The open-ended material was useful as a supplement, but it did not overturn the central quantitative message. Students perceived the practicum as important, rated their own performance highly, and still left identifiable room for educational refinement in selected domains.
What the study cannot show is whether those perceptions match observed bedside competence. It also cannot say whether changing the practicum structure would improve stress, skill acquisition, or future clinical performance. The data are best read as a detailed snapshot of student perception, with enough variability and psychometric uncertainty to caution against overconfident conclusions from a single cohort at a single institution.