Is the Deep Learning Approach Effective in Building Nursing Students' Professional Identity?

1. INTRODUCTION

Population aging, rising multi-morbidity rates, and the growing demand for high-quality, patient-centered services make the health system overwhelming worldwide. Nurses are indispensable in delivering safe and effective care. However, nurse shortages and high turnover rates continue to aggravate across the world [1]. In China, problems such as nurse shortage, nurse retention, and regional imbalance are more serious than in other countries [2]. One study from China indicates that more than 69.4% of nurses intend to leave their jobs [3]. Many new nursing graduates see their job more as a way to earn a salary, not as a long-term professional identity [4]. Given this clear identity crisis, this study focuses on nursing students, not nurses.

The development of nursing professional identity is an ongoing process, and education plays an important role in it. During the knowledge acquisition and training, students begin to build their view of nursing. A positive view of nursing helps students understand nursing and see how nurses differ from other health professions [5]. Nursing students develop their professional identity gradually through several key elements, such as what they learn in the classroom, the clinical practice skills, and the interactions they have with other health workers [6]. Some educators argue that nursing educators should focus on professional identity from the early stage of training [7]. Simultaneously, there is a growing emphasis within nursing education on developing interdisciplinary competencies. In this context, experts believe that students need strong, deep learning abilities to handle future challenges [8].

Deep learning ability is an important concept in educational psychology. It means learning for true understanding, not simply memorizing. This concept captured our research interest because it involves genuine interest, thoughtful reflection, and critical analysis [9], which are important in integrating knowledge into a coherent conceptual network [10]. For nursing students, deep learning occurs when they critically examine theories, relate them to clinical practice, and apply them in real situations [11]. Critical thinking and knowledge integration can help students practice safely in the clinical stage and develop a meaningful sense of nursing professional identity [12]. Each attempt supports them to build a flexible network of knowledge that can be used in various settings, and finally transit from beginners to prepared professionals [13] through collaboration and communication with other healthcare professionals during the process of deep learning and practical application. Students can enhance their understanding of their roles, thereby strengthening their professional identity [14]. However, the process by which deep learning ability shapes professional identity is still not fully understood [15].

Professional self-efficacy is a meaningful point rooted in Bandura’s Social Cognitive Theory. It describes a person’s confidence in finishing tasks, overcoming challenges, and coping with work pressure [16]. Numerous researches point out that nurses with higher self-efficacy tend to communicate effectively [17] and deliver higher quality care [18]. They are also better at building positive relationships with patients [19]. These nurses even experience less burnout and show stronger profession commitment [20]. All these facts support the correlation between professional self-efficacy and professional identity. In nursing education, self-efficacy helps students bridge their theoretical knowledge and clinical experience. This integration gradually deepens individuals’ professional identity from freshmen to nurse profession, and enhances their understanding of nursing [21]. While multiple studies confirm a positive relationship between professional self-efficacy and professional identity, the underlying process still needs further clarification [22].

The research team supposes that deep learning plays a key role in professional self-efficacy and professional identity. Simply put, students with stronger beliefs in their own abilities are more likely to engage in deep learning, leading to a better understanding of the nursing profession. This process, in turn, helps them develop a stronger and more stable professional identity. To test this assumption, this research was designed to examine whether deep learning acts as a mediator between self-efficacy and professional identity. The results could provide valuable insights for both future research and nursing education practice.

2. METHODS

3. RESULTS

3.3. Mediation Analysis of DLACS as a Mediator Between PSENS and PISNS

The mediation analysis, controlling for gender, grade year, only child status, family address, cadre status, first choice of major, and clinical experience, indicated that PSENS significantly predicted PISNS both directly and indirectly through DLACS. Specifically, PSENS was positively associated with DLACS (a = .56; 95% boot CI = .49- .63) and DLACS was positively associated with PISNS (b = .14; 95% boot CI = .00- .26). The direct effect of PSENS on PISNS remained significant (c' = .56; 95% boot CI = .48- .65). The indirect effect was significant (ab = .08; 95% boot CI = .00- .15), indicating partial mediation with DLACS accounting for 12.0% of the total effect. Significant covariate effects revealed that non-first-choice students reported lower PISNS (β = -2.39; p < .050). Senior students’ PISNS were lower than those of lower-grade students (β = -.99; p < .050). Students with cadre status had lower DLACS (β = -3.17; p < .050). The models demonstrated excellent fit: PISNS model (R² = .76; F = 85.65; p < .001). These results are presented in Table 4 and Fig. (1).

Table 4.

Effect Type	Path	Estimates	BootSE	95% BootCI	Effect Size	Model Fit Indicators
Direct effect	PSENS → PISNS (c')	.56	.05	.48 → .65	88.0%	F = 85.65, p < .001
Indirect effect	PSENS → DLACS → PISNS	.08	.04	.00 → .15	12.0%	R2 = .76
Total effect	PSENS → PISNS (c)	.64	.02	.59 → .69	100.0%	MSE = 35.46
Covariate effects
Cadre status	→ DLACS	-3.17	1.20	-5.54 → -.76	---	P < .050
First choice (No)	→ PISNS	-2.39	.75	-3.88 → -.92	---	P < .050
Grade	→ PISNS	-.99	.47	-1.93 → -.09	---	P < .050

Note: All analyses controlled for general information. Bootstrap sample = 5000.

PISNS: Professional Identity Scale for Nursing Students;

PSENS: Professional Self-Efficacy Scale for Nursing Students;

DLACS: Deep Learning Ability Scale for College Students.

4. DISCUSSION

The study found that nursing students showed moderate levels of professional self-efficacy, deep learning ability, and professional identity, which aligns with prior studies [27]. Some research reported a temporary increase in professional identity during the pandemic [28]. But most studies suggest that professional identity has stayed at a moderate level for a long time [29]. This shows how difficult it is to build a strong professional identity among nursing students, which highlights the need to take action earlier. Helping nursing students develop a strong sense of identity may encourage them to stay in the nursing profession after graduation. Nurse retention is not only a task for hospitals and health systems. Schools and Universities also play an important role in the early stage of cultivation.

Our results confirmed a significant positive correlation among the three concepts. The deep learning ability partially mediates the relationship between professional self-efficacy and professional identity. This mediating effect remained significant even when other factors were controlled. This may be interpreted as students with higher self-efficacy are more likely to adopt deep learning approaches. Through their deep learning, they construct a powerful network of complex knowledge and become more skilled at solving clinical problems. The positive emotional feedback can further strengthen their confidence and clarify their identity as nurses [30].

These findings have direct implications for nursing education. Curricula should move beyond knowledge transmission to actively cultivate deep learning abilities [31]. For example, teachers can link theory with practice and encourage students to reflect on clinical experiences. After guiding students with enough reflective practices, instructors provide different problem-solving practices in real or simulated settings. This will help strengthen their deep learning skills. Furthermore, evaluating teaching based on students’ deep learning outcomes could also improve nursing programs [32].

When considering a deep learning teaching strategy, educators should be more careful. Although online learning has become widespread in recent years, fostering deep learning in digital formats can be challenging. Course design, meaningful interactions, and learning assessment are all needed to be considered comprehensively [33]. Research indicates that clear guidelines and sufficient time for peer discussion are important in online environments [34]. Digital tools may also be helpful when used appropriately [35].

In contrast, face-to-face classes benefit more from a supportive and collaborative atmosphere. Such an environment promotes the kind of reflection that leads to deeper understanding. Increasing educational practice evidence shows that a blended approach may be the most effective way. Because it can address students' variable needs [36].

According to the results, nursing students should take part in deep learning activities more often. It can help them reflect on what they learn and apply knowledge and skills in different situations. On the other hand, Educators should integrate deep learning theories and relevant metrics into course design. Clear guidance, active learning tasks, new teaching methods, and proper use of technology and AI tools can all support this goal [37].

An unexpected finding was that students who served as student leaders showed lower deep learning ability, possibly because of role overload [38]. These students may spend much time on management tasks, which reduces the time and mental energy needed for deep learning. In addition, senior students and those who did not choose nursing as their first choice demonstrated significantly lower professional identity. For senior students, this may be related to the “reality shock” and emotional depletion and pressures experienced during clinical practice. When real work conditions do not match their idealized expectations, they can not handle these conflicts effectively [39]. For students who entered nursing without a strong interest, conflict between personal goals and external pressure may weaken their sense of professional identity [40]. More research is needed to understand these issues.

5. IMPLICATIONS

This study confirms that deep learning ability plays a partial mediating role in the development of professional identity of nursing students, filling the gap in this research area. But for further understanding of deep learning in the nursing field, following directions is worth exploring in the future. Future researchers could try to develop a tool specific to assessing nursing students’ deep learning ability. Nursing educators are responsible to construct a clear framework of deep learning in nursing education practice. Subsequently, examine how courses based on deep learning theory affect nursing competence. It may also be useful to evaluate teaching effectiveness from the perspective of deep learning outcomes. Researchers can also build an assessing framework for the deep learning ability of nursing educators in the future.

6. LIMITATIONS

We have to acknowledge that there are several limitations of this study. The sample came from a single city in China, which limits how widely the results can be applied. The tool we used to assess deep learning ability was not specifically for nursing students, which means it may affect accuracy. The sample size of this study is small, which may limit the stability of subgroup analyses. Some potential influences on professional identity might be neglected.

CONCLUSION

Deep learning ability is an important process to enhance professional self-efficacy and develop professional identity. The mediating role of deep learning ability examined in this study provides a direction for nursing educational exploration. In future nursing education, deep learning ability may be considered as a key competence and evaluation indicator. Future research should continue to refine its meaning, measurement, and development in the changing context of nursing education.

REFERENCES