Characterizing potential subtypes and influencing factors of self-directed learning competence among clinical nurses in China by latent profile analysis

Abstract

Background

Nurses demonstrate varying levels of self-directed learning competence, which is influenced by multiple individual and contextual factors. Identifying profiles with varying levels of self-directed learning is essential for providing targeted support and training. However, there is limited research exploring the potential profiles of self-directed learning competence among nurses.

Objectives

This study aimed to identify distinct latent profiles of self-directed learning competence among nurses and analyze the influencing factors.

Methods

A cross-sectional study design was used. Nurses were recruited using a convenience sampling from four tertiary hospitals in Guangzhou, China, between August 2024 and February 2025. The Self-Directed Learning Competence Scale for Nurses was used to assess the self-directed learning Competence of nurses. Latent profile analysis was performed to identify different potential profiles. Pearson’s chi-square test and multinomial logistic regression were used to explore the factors influencing self-directed learning competence.

Results

A total of 740 nurses participated. Three latent profiles of self-directed learning competence were identified: low (n = 356, 48.1%), medium (n = 291, 39.3%), and high (n = 93, 12.5%). The self-directed learning competence of nurses was influenced by various factors, including Junior college and below (OR = 0.555, p= 0.043); Monthly number of night shifts = 3~4 (OR = 2.859, p = 0.029); Learning atmosphere = neutral (OR = 0.342, p = 0.018) and good (OR = 0.412, p = 0.038); Learning willingness (OR = 1.425, p <0.001), Difficulty of title promotion = difficult (OR = 2.628, p = 0.029) and Job stress (OR = 0.981, p <0.001).

Conclusion

The study revealed diverse profiles of self-directed learning competence among nurses. Nursing managers should design differentiated strategies based on these profiles. Enhancing organizational support and stimulating learning motivation can help improve nurses’ self-directed learning competence, thereby promoting their professional development and improving the overall quality of clinical nursing care.

Background

In the context of increasingly complex healthcare systems and the growing diversity of patient care demands, nurses are not only required to fulfill traditional caregiving responsibilities but also to continuously update their knowledge and enhance professional competencies to adapt to evolving clinical environments (Mlambo et al., 2021; Purvis, 2025). With the advancement of evidence-based practice and rapid clinical innovation, nurses are no longer passive implementers of care but are expected to engage in continuous learning and professional development (Middleton et al., 2022). Among the core competencies required, self-directed learning competence (SDLC) has emerged as essential for sustaining professional growth and ensuring high-quality care delivery (Vasli & Asadiparvar-Masouleh, 2024).

SDLC refers to an individual's ability to set learning goals, select appropriate resources, employ effective strategies, monitor progress, and regulate learning behavior autonomously, without external supervision (Xiao & Li, 2008). This competence is crucial in improving clinical performance and fostering career development. Studies (Bryant et al., 2023) have shown that over 90% of nurses enhance their competencies through self-directed learning. Nurses with high SDLC demonstrate superior performance in clinical decision-making, patient communication, and interprofessional collaboration (Huang et al., 2023; Pang et al., 2022).

However, despite its importance, many nurses struggle to develop and sustain this ability in practice. Under high workloads, staff shortages, and fragmented shift schedules, learning is often perceived as an additional burden rather than an opportunity for growth (Richard & Kim, 2024). Nurses frequently lack clear learning goals, structured plans, or access to effective support systems, which results in fragmented and short-term learning behaviors (Koca et al., 2024). Low self-efficacy and unclear career development pathways further weaken learning initiative (Jakobsson et al., 2023). In many cases, nurses only engage in learning to meet external requirements, such as mandatory training or title promotion, rather than from intrinsic motivation (Zhang et al., 2024). These issues foster passive, reactive learning that undermines long-term professional development and limits nurses’ adaptability in complex clinical environments.

Notably, SDLC is not a fixed trait, but a dynamic, evolving process shaped by individual and environmental factors. Variables such as personality traits, learning motivation, professional identity, and organizational support can significantly influence learning engagement (Pan et al., 2024). These influences may shift over time with changes in experience, emotional state, or life stage, resulting in significant variation among individuals. Yet, most current studies overlooked this heterogeneity, relying on aggregate scores and assuming population-level homogeneity (Chakkaravarthy et al., 2020; Huang et al., 2023).

To address these gaps, Latent Profile Analysis (LPA) provides a more appropriate method. LPA is a person-centered method that captures unobservable latent variables inaccessible to traditional measurement approaches, and it can identify nurses with different SDLC. Compared with other methods, it places more emphasis on individual differences. It classifies individuals based on their responses to different variables to identify distinct profiles with unique characteristics (Li et al., 2018). To our knowledge, most LPA applications to self-directed learning in nursing literature have focused on undergraduate student samples; studies applied LPA to characterize SDLC among clinical nurses and linked these profiles simultaneously to personal and workplace factors remain limited (Miao et al., 2024; Zhou et al., 2023). Therefore, we used LPA to determine the heterogeneity of SDLC among nurses. We explored the impact of sociodemographic characteristics, personality traits, learning willingness, and job stress on SDLC. By clarifying the characteristics and influencing mechanisms of different learning ability groups, the results of this study are expected to inform targeted training strategies, support personalized professional development, and contribute to the enhancement of nursing service quality.

Methods

Study Design

This study used a cross-sectional study design.

Sample/Participants

Nurses were recruited using a convenience sampling method from four tertiary hospitals in Guangzhou, China, between August 2024 and February 2025. Inclusion criteria were identified as follows: holding a valid nursing license issued by the National Health Commission of China, having at least six months of post-licensure experience in a hospital setting, and providing written informed consent. Nurses who were on leave or scheduled for further training during the study period were excluded. For this study, the sample size was calculated using the following formula (Ni et al., 2010): $N={\left(\frac{U_{\alpha }\sigma }{\delta }\right)}^2$. At a 95% confidence level, the critical value was U_α =1.96. The absolute error (δ) was set at 2, and the standard deviation (σ) was 20, based on data from a previous study on the development of SDLC for nurses (Pang et al., 2022). According to the formula, the required sample size was 384. To account for an anticipated 20% rate of invalid responses, the final target sample size was increased to 480 participants.

Instruments

Several instruments were used in this study, including:

Sociodemographic characteristics. General demographic information, including gender, age, marital status, educational level, monthly income, years of experience, professional title, years of working, employment form, monthly number of night shifts, clinical preceptors, participation in scientific research activities, a family member requires intensive care, difficulty of title promotion, learning atmosphere, and job satisfaction.

Self-directed learning competence. The scale was developed by Xiao and Li (2008) and used to assess nursing staff SDLC. It comprises 34 items and is divided into four dimensions: self-motivation belief, task analysis, self-monitoring and regulation, and self-evaluation. Each item is scored on a 5-point Likert scale, ranging from “Strongly Disagree” to “Strongly Agree,” with scores from 1 to 5, resulting in total scores ranging from 34 to 170. Higher scores indicate stronger competency. In the present study, the Cronbach’s α coefficient for the scale was 0.974.

Job stress. The scale was compiled by Li and Liu (2000) and used to assess job stress for nurses. It consists of five dimensions: nursing profession and work, workload and time allocation, work environment and equipment, patient care, and management and interpersonal relationships, comprising a total of 35 items. Responses were measured using a 5-point Likert scale, where higher scores indicated greater stress. The Cronbach’s α coefficient for the scale in this study was 0.958.

Learning willingness. Self-directed learning willingness was assessed using two questions. The first asked, “On average, how many hours per month do you spend learning medical-related knowledge outside of working hours?” The second question asked, “How willing are you to learn medical-related knowledge outside of working hours?” Willingness was measured on a Visual Analogue Scale (VAS) from 0 (no willingness) to 10 (very high willingness).

Data Collection

The questionnaire was distributed via the online platform. Instructions provided participants with the study’s purpose, core concepts, and assurances of anonymity and confidentiality. Participation was voluntary, and informed consent was obtained by requiring participants to click “Confirm” before proceeding. All items were mandatory, and the questionnaire could only be submitted after full completion. Each IP address was limited to a single submission. Responses completed in less than 200 seconds, patterned responses (e.g., selecting the same Likert scale option for all items), or inconsistent answers (e.g., contradictory responses to logically related questions) were deemed invalid after manual review by the research team.

Data Analysis

LPA was conducted using Mplus 8.3, with 34 items of the SDLC scale as the observed variables. The selection of a better model fit was based on the changes in the information criteria. The information index was Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and sample-size adjusted BIC (aBIC) to assess the profiles' accuracy, with lower values preferred. Entropy was used to determine classification accuracy, with values closer to 1 indicating greater accuracy. The Lo-Mendell–Rubin adjusted likelihood ratio test (LMRT) and bootstrapped likelihood ratio test (BLRT) were utilized to compare models with different class numbers, where a p-value <0.05 suggests that the k-class model is superior to the k-1 class model. Data were analyzed using SPSS 27.0 software, and comparisons between groups were made using the chi-square test and ANOVA. Multinomial logistic regression was used to analyze the factors influencing the potential profile of SDLC. A two-tailed p-value of <0.05 was considered statistically significant.

Ethical Considerations

This study was conducted in accordance with the Declaration of Helsinki and was approved by the First Affiliated Hospital of Guangzhou Medical University (Approval number: ES-2024-K099-01 on 24 July 2024). Participation was voluntary, and no physical harm was caused to the participants. All questionnaire responses were kept confidential.

Results

A total of 775 questionnaires were collected. After excluding 35 invalid responses, 740 valid questionnaires remained, yielding a 95.4% response rate.

Potential Profile Analysis of SDLC

Latent profile models with one to five classes were sequentially fitted using the 34 items of the SDLC scale as exogenous variables. Model fit indices are presented in Table 1. The AIC, BIC, and adjusted BIC values continued to decrease slightly for the fourth and fifth models; the third model achieved the highest entropy (0.987), indicating superior classification accuracy, and produced conceptually distinct and clinically meaningful profiles. In contrast, the fourth and fifth models fragmented one or more clinically coherent groups into small classes (e.g., one class accounting for only about 10% of the sample), raising concerns about overfitting and unstable estimates. Moreover, the incremental improvement in fit indices from the third to fourth or fifth model was small relative to the added model complexity and reduced interpretability. Balancing statistical fit, parsimony, and substantive meaning, we selected the third model. A profile plot was generated based on the scores of the four dimensions across the three classes of SDLC (see Figure 1). Nurses in Class 1 scored markedly lower than those in the other two classes and were categorized as the low SDLC group (n = 356, 48.1%). Nurses in Class 2 had intermediate scores across all dimensions and were labeled the medium SDLC group (n = 291, 39.3%). Nurses in Class 3 showed significantly higher scores across all dimensions and were categorized as the high SDLC group (n = 93, 12.5%).

Table 1.

Indicators for fitting the potential profile model of nurses’ self-directed learning competence

Model	AIC	BIC	aBIC	Entropy	p-value		Category probability (%)
					LMRT	BLRT
1	62260.107	62573.359	62357.435	-	-	-	-
2	50611.255	51085.740	50758.678	0.978	<0.001	<0.001	55.6/44.3
3	46658.141	47293.859	46855.660	0.987	<0.001	<0.001	48.1/39.3/12.5
4	45185.470	45982.421	45433.084	0.968	<0.001	<0.001	32.9/33.2/21.6/12.1
5	44303.635	45261.819	44601.345	0.966	0.01	<0.001	10.4/20.0/24.1/33.3/12.0

Note: aBIC, adjusted Bayesian information criterion; AIC, Akaike information criterion; BIC, Bayesian information criterion; BLRT, bootstrapped likelihood ratio test; LM, Lo–Mendell–Rubin.

Figure 1.

Overview of potential profiles of self-directed learning competence

Univariate Analysis of Potential Categories of SDLC

The results of the univariate analysis showed that the three groups of nurses were statistically significant (p < 0.05) in terms of participation in scientific research activities, difficulty of title promotion, learning atmosphere, job satisfaction, willingness to learn, and job stress, as shown in Table 2.

Table 2.

Univariate analysis of potential categories of self-directed learning competence

Characteristics	Category	C1 [%] (n = 356)	C2 [%] (n = 291)	C3 [%] (n = 93)	χ2/F	p
Personal
Gender	Male	42 (48.3)	36 (41.4)	9 (10.3)	0.494a	0.781
	Female	314 (48.1)	255 (39.1)	84 (12.9)
Age (years)	≤35	281 (49.3)	219 (38.4)	70 (12.3)	1.408a	0.495
	>35	75 (44.1)	72 (42.4)	23 (13.5)
Marital status	Married	134 (47.3)	119 (42.0)	30 (10.6)	2.331a	0.312
	Unmarried/ divorced / widowed	222 (48.6)	172 (37.6)	63 (13.8)
Educational level	Junior college and below	197 (48.9)	163 (40.4)	43 (10.7)	2.930a	0.231
	Undergraduate or above	159 (47.2)	128 (38.0)	50 (14.8)
A family member requires intensive care	Yes	134 (48.2)	104 (37.4)	40 (14.4)	1.591a	0.451
	No	222 (48.1)	187 (40.5)	53 (11.5)
Professional
Monthly income (RMB)	1000~5000	38 (10.7)	42 (46.2)	11 (12.1)	2.984b	0.811
	5001~10000	176 (48.1)	143 (49.1)	47 (12.8)
	10000~15000	110 (49.8)	85 (38.5)	26 (11.8)
	>15000	32 (51.6)	21 (33.9)	9 (14.5)
Professional title	RN	71 (41.5)	80 (46.8)	20 (11.7)	8.264b	0.082
	Junior RN	144 (53.1)	98 (36.2)	29 (10.7)
	Middle RN or above	141 (47.3)	113 (37.9)	44 (14.8)
Years of working	<5	80 (41.2)	93 (47.9)	21 (10.8)	8.842b	0.183
	5~10	147 (51.6)	102 (35.8)	36 (12.6)
	11~20	101 (49.8)	73 (36.0)	29 (14.3)
	>20	28 (48.3)	23 (39.7)	7 (12.1)
Employment form	Establishment	137 (49.5)	103 (37.2)	37 (13.4)	0.983a	0.526
	Non-establishment	219 (47.3)	188 (40.6)	56 (12.1)
Monthly number of night shifts	0	44 (39.3)	50 (44.6)	18 (16.1)	12.809b	0.235
	1~2	40 (48.2)	32 (38.6)	11 (13.3)
	3~4	38 (42.7)	37 (41.6)	14 (15.7)
	5~6	43 (44.8)	41 (42.7)	12 (12.5)
	7~8	94 (51.6)	63 (34.6)	25 (13.7)
	≥9	97 (54.5)	68 (38.2)	13 (7.3)
Tutor	Yes	63 (47.4)	134 (39.0)	47 (13.7)	0.706a	0.703
	No	193 (48.7)	157 (39.6)	46 (11.6)
Participate in scientific research activities	Yes	134 (44.8)	115 (38.5)	50 (16.7)	8.117a	0.017
	No	222 (50.3)	176 (39.9)	43 (19.8)
Difficulty of title promotion	Very easy	0 (0.0)	0 (0.0)	1 (100)	19.252b	0.014
	easy	26 (38.8)	28 (41.8)	13 (19.4)
	Neutral	130 (46.1)	120 (42.6)	32 (11.3)
	difficult	125 (48.8)	93 (36.3)	38 (14.8)
	Very difficult	75 (56.0)	50 (37.3)	9 (6.7)
Learning atmosphere	Very poor	3 (42.9)	4 (57.1)	0 (0.0)	23.904b	0.002
	Poor	19 (54.3)	12 (34.3)	4 (11.4)
	Neutral	180 (52.0)	133 (38.4)	33 (9.5)
	Good	128 (47.2)	109 (40.2)	34 (12.5)
	Very good	26 (32.1)	33 (40.7)	22 (27.2)
Job satisfaction	Very dissatisfied	13 (44.8)	15 (51.7)	1 (1.1)	27.805b	<0.001
	Dissatisfied	53 (59.6)	29 (32.6)	7 (7.9)
	Neutral	198 (51.2)	141 (36.4)	48 (12.4)
	Satisfied	85 (41.5)	93 (45.4)	27 (13.2)
	Very satisfied	7 (23.3)	13 (43.3)	10 (33.3)
Willingness to learn		5.57 ± 2.53	6.59 ± 2.53	7.93 ± 2.60	35.778b	<0.001
Job stress		97.70 ± 21.28	89.08 ± 21.57	79.06 ± 30.11	38.482b	<0.001

Note:

^a= F;

^b= χ²

Multifactorial analysis of potential categories of SDLC

Taking the potential category of SDLC as the dependent variable. (“low SDLC group” as the reference) Taking the statistically significant factors in the Multinomial logistic regression analysis and the significance of variables in the clinical as the independent variables. The results of the study showed that educational level, monthly number of night shifts, learning atmosphere, difficulty of title promotion, learning willingness, and job stress were influential factors of the potential profile of SDLC. Difficulty of title promotion = difficult (OR = 2.628, p = 0.029), Monthly number of night shifts = 3~4 (OR = 2.859, p = 0.029) and Learning willingness (OR = 1.425, p < 0.001) were more likely to be the high SDLC groups; Job stress (OR = 0.981, p < 0.001), Junior college and below (OR = 0.555, p = 0.043), Learning atmosphere = neutral (OR = 0.342, p = 0.018) and good (OR = 0.412, p = 0.038) were more likely to be the low SDLC group (Table 3).

Table 3.

Multifactorial analysis of potential categories of self-directed learning competence

Variables	Class 2 VS Class 1				Class 3 VS Class 1
	β	OR	95%CI	p	β	OR	95%CI	p
Educational level (ref: Undergraduate or above)  Junior college and below	0.014	1.014	(0.703,1.461)	0.701	-0.588	0.555	(0.314,0.982)	0.043
Monthly number of night shifts (ref: ≥9)  3~4	0.367	1.444	(0.805,2.590)	0.218	0.966	2.859	(1.115,7.331)	0.029
Learning atmosphere (ref: Very good)  Neutral  Good	-0.072 -0.120	0.931 0.887	(0.475,1.824) (0.459,1.714)	0.834 0.721	-1.072 -0.886	0.342 0.412	(0.141,0.830) (0.178,0.953)	0.018 0.038
Difficulty of title promotion (ref: Very difficult)  Difficult	0.031	1.032	(0.631,1.688)	0.901	0.966	2.628	(1.106,6.243)	0.029
Learning willingness	0.117	1.124	(1.052,1.201)	<0.001	0.354	1.425	(1.260,1.612)	<0.001
Job stress	-0.019	0.981	(0.973,0.990)	<0.001	-0.032	0.969	(0.956,0.981)	<0.001

Discussion

Three Latent Profile Analysis of SDLC among Nurses

This study used LPA to identify three distinct classes of SDLC among nurses: low (48.1%), medium (39.3%), and high (12.5%). Notably, it is essential to note that the low, medium, and high SDLC categories here are relative to the sample and derived from latent class analysis. These findings highlight the heterogeneity of SDLC levels within the nursing population. As shown in the study’s results, nurses in Class 3 exhibited the strongest performance across all four dimensions, particularly in task analysis and self-monitoring/regulation. These nurses demonstrated a proactive learning attitude and effective self-management. While advanced training and leadership opportunities can further enhance their abilities, it is also important to retain them (Vázquez-Calatayud et al., 2021).

Furthermore, these nurses often serve as role models and knowledge carriers, and their departure could result in a loss of expertise and continuity of care. This highlights the need for hospital leaders to actively implement retention strategies. Beyond providing advanced training, managers should create clear career development pathways, establish recognition and reward mechanisms, and offer leadership and decision-making opportunities. By combining professional growth with retention policies, organizations can better sustain the motivation of high SDLC nurses and secure their long-term contributions to patient care and team learning. Nurses in Class 2 displayed moderate and stable performance across dimensions, but lacked strong motivation and effective strategy application. This “conservative maintenance” state was reflected in clinical practice by a tendency to passively accept assigned learning tasks without a strong drive for self-improvement. Interventions should focus on stimulating intrinsic motivation, providing role models, and offering phased training with feedback to support their development (Nishimoto et al., 2023). In contrast, nurses in Group 1 scored significantly lower across all dimensions, with the lowest performance observed in task analysis. This may reflect a lack of clear goals and learning strategies, resulting in avoidance or disengagement. They often struggle with execution and tend to exhibit passive learning behaviors or avoidance (Xuan et al., 2024). For these nurses, targeted interventions should include motivational enhancement, emotional support, clear learning pathways, and team-based encouragement to foster self-efficacy and improve learning behaviors.

Impact of Demographic Characteristics on the SDLC of Nurses

Educational level

Higher educational attainment was associated with stronger SDLC, consistent with prior research (Wang & Dong, 2024). One possible explanation was that nurses with higher levels of education typically receive more comprehensive professional training during their formal studies, equipping them with more effective learning strategies and critical thinking skills. As a result, they were better prepared to engage proactively with new knowledge and technologies. Similarly, Zhao et al. (2024) found that nurses with higher academic qualifications often had greater expectations for their professional development, stronger intrinsic motivation, and a greater inclination to enhance their professional competence and competitiveness through continuous learning. Therefore, nursing managers should consider promoting self-directed learning by encouraging continuing education and establishing academic advancement pathways to support nurses in obtaining higher degrees, thereby fostering the ongoing development of their core competencies.

Monthly number of night shifts

Night shifts were associated with lower SDLC. Relevant research has shown that night shift work could disrupt nurses’ circadian rhythms, reduce sleep quality, and lead to fatigue accumulation, which may result in cognitive decline and decreased motivation to learn, thereby impairing their ability to engage in self-directed learning (Li et al., 2024a). The heavy consumption of cognitive resources and psychological energy limited learning effectiveness and could hinder professional development (Pang et al., 2020). Therefore, nursing managers should: firstly, limit consecutive night shifts to ensure adequate recovery time; secondly, establish peer learning or mentorship groups to enhance learning efficiency; and thirdly, evaluate temporary staffing solutions or reallocate administrative tasks to free up nursing staff’s time for learning. Interestingly, in this study, only nurses working 3–4 night shifts per month showed a significant association with higher SDLC. This suggests that the relationship between night shift frequency and SDLC may be complex. Our current data cannot fully explain this pattern, and the finding should be interpreted with caution. Further research is needed to clarify whether there is an optimal range of night shift frequency that supports nurses’ learning competence. Nonetheless, our results emphasize that night shift arrangements are an essential factor to consider when promoting SDLC in clinical settings.

Difficulty of title promotion

Difficulty of title promotion was one of the key driving forces behind the career development of nurses, and its level directly affects nurses’ motivation for self-directed learning and their capacity for professional growth. As promotion opportunities increase, so does the urgency to update professional knowledge, leading to a marked rise in learning initiatives. Research has shown (Wang et al., 2022) that title promotion was closely linked to nurses’ career advancement, salary, and social recognition; the more attainable the promotion, the more likely nurses were to perceive tangible rewards from their learning efforts, thus creating a positive feedback loop. Conversely, when promotion pathways are limited, highly competitive, or opportunities are scarce, nurses may recognize the importance of learning but feel demotivated due to a lack of concrete incentives (Chen et al., 2021). Furthermore, the fairness and transparency of promotion policies were critical factors influencing nurses’ enthusiasm for learning (Wu, 2021). Therefore, managers should tailor development pathways to meet the needs of nurses at different professional levels, diversify career advancement options, and enhance learning support systems to reduce ineffective internal competition. Additionally, understanding nurses’ genuine feedback on promotion systems and learning resources is essential for fostering a fair and motivating environment that supports the continuous development of SDLC.

Impact of Learning Environment on the SDLC of Nurses

This study found that both learning atmosphere and learning willingness were associated with SDLC, and their effects are often interrelated and mutually reinforcing. A positive learning atmosphere—characterized by active knowledge sharing, supportive team culture, and accessible educational opportunities—can enhance nurses’ interest in learning and facilitate the frequent use of self-directed learning strategies (Fooladvand & Nadi, 2019). Research has also shown that frequent participation in departmental knowledge-sharing activities—such as case discussions, nursing rounds, and specialized lectures—could enhance both the frequency and quality of nurses’ use of self-directed learning strategies (Li et al., 2024b). Furthermore, a supportive and encouraging team environment enabled nurses to seek assistance when facing learning difficulties (Pu et al., 2025), thereby reducing frustration and boosting their learning confidence. Another research also demonstrated that continuing education and in-service training within departments not only improve the overall quality of nursing services but also provide a solid foundation for enhancing nurses’ SDLC (Zhao et al., 2025).

At the same time, a willingness to learn plays a pivotal role in shaping how nurses engage with their professional development. As a core component of learning motivation, higher learning willingness directly enhances an individual’s initiative, persistence, and reflective capacity during the learning process (Yeung et al., 2024). Moreover, strong learning motivation helped nurses maintain a higher level of intrinsic drive and self-efficacy when facing clinical challenges, further reinforcing self-directed learning behaviors (Xuan et al., 2024). Research has shown that a supportive environment amplifies internal motivation, leading to greater engagement and improved learning outcomes (Yeung et al., 2024). In contrast, a poor or indifferent learning environment may suppress motivation, even among those who initially have strong learning intentions. Therefore, nursing managers should adopt a dual approach: externally, by cultivating a department-wide culture that encourages continuous learning through regular educational activities, feedback systems, and collegial support (Sarıköse & Türkmen, 2020); and internally, by tailoring individualized learning incentives to sustain nurses’ motivation. Establishing this synergy between the learning atmosphere and willingness can help shift nurses from passive to proactive learners, ultimately enhancing their SDLC and improving the quality of clinical care.

Impact of Job Stress on the SDLC of Nurses

Higher job stress was associated with lower SDLC (P<0.05). The primary sources of job stress for nurses include heavy workloads, patient safety risk management, complex interpersonal dynamics, and pressure related to career advancement. These factors contributed to prolonged exposure to high-stress environments, increasing both physical and psychological burdens. High levels of job stress not only diminished nurses’ motivation to learn but also consumed the time and cognitive resources necessary for effective self-directed learning. Research has shown that chronic stress can lead to psychological conditions such as job burnout and emotional exhaustion, which in turn reduce nurses’ active engagement in learning and professional development (He & Li, 2024; Li et al., 2025).

Job stress has been identified as a significant negative predictor of SDLC, an essential factor in improving nursing quality and promoting career growth (Feng et al., 2024). Therefore, nursing managers should closely monitor nurses’ workloads and psychological well-being, optimize staffing schedules, allocate human resources appropriately, and implement effective stress-reduction and psychological support mechanisms. These measures could help alleviate stress and boost nurses’ enthusiasm for self-directed learning. Additionally, interventions such as mental health workshops and the formation of peer learning support groups could help foster a work environment conducive to self-directed learning, ultimately contributing to the overall improvement of clinical nursing quality.

Limitations

First, the study recruited a relatively small sample size consisting solely of nurses from Guangdong Province. This may reduce statistical power and limit the generalizability of the findings to broader populations or regions. Second, due to the cross-sectional research design, this study could not establish causal relationships between the variables. Future research could consider adopting a longitudinal approach to explore the dynamic relationships among variables more thoroughly. Third, some potentially important confounding variables, such as nurses’ own health status, psychological well-being, and broader family demands, were not collected and could not be controlled for in the regression models. Finally, all data in this study were collected through self-report questionnaires, which may introduce recall errors and social desirability bias, potentially leading participants to overestimate positive behaviors or underreport negative ones.

Conclusion

This study analyzed the potential profiles of nurses’ SDLC and their influencing factors in tertiary hospitals in Guangzhou, China, using potential profile analysis. Nurses’ SDLC in tertiary hospitals can be categorized into three groups: low SDLC (48.1%), medium SDLC (39.3%), and high SDLC (12.5%). The results of the multi-categorical logistic regression analysis showed that educational level, Difficulty of title promotion, Monthly number of night shifts, Learning atmosphere, Learning willingness, and Job Stress were associated with the potential profile of nurses’ SDLC in tertiary hospitals. Nursing managers should pay attention to the level of nurses’ SDLC, develop targeted interventions based on nurse categories and influencing factors, and enhance core competencies in tertiary care hospitals, thereby improving SDLC and nursing service quality.

Funding Statement

Funding None.

Declaration of Conflicting Interest

The authors declare no conflict of interest in this study.

Authors’ Contributions

The first author contributed to the research design, data analysis, article drafting, and manuscript writing. The second author contributed significantly during the revision stage, including data re-analysis, result interpretation, and critical revision of the manuscript. The third author contributed by reviewing literature, performing data analysis, and collecting data. The fourth and fifth authors contributed by reviewing literature and collecting data. The sixth and seventh authors contributed by supervising research development and critically reviewing and revising the manuscript. All authors were responsible for each step of the research and read and approved the final draft of the manuscript.

Authors’ Biographies

Zhang Zhisheng, MNS, is a master’s student at the First Affiliated Hospital of Guangzhou Medical University.

Cai Mingju, RN, is a practitioner at the First Affiliated Hospital of Guangzhou Medical University.

Zhang Ruichu, MNS, is a master’s student at the First Affiliated Hospital of Guangzhou Medical University.

Wang Fang, MNS, RN, is a practitioner at the Shunde Hospital, Guangzhou University of Chinese Medicine.

Liao Shaona, MNS, RN, is a practitioner at the First Affiliated Hospital of Guangzhou Medical University.

Nie Anliu, MNS, RN, is a practitioner at the First Affiliated Hospital of Guangzhou Medical University.

Prof. Su Xiangfen, RN, is a Professor in the Master of Nursing program at the First Affiliated Hospital of Guangzhou Medical University.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declaration of Use of AI in Scientific Writing

The authors utilized AI in the writing process to enhance readability and eliminate grammatical errors, and then reviewed all interpretations to verify their accuracy.