Predictors of nursing students’ satisfaction with the clinical learning environment in Kosovo: A multivariate analysis using the Clinical Learning Environment, Supervision and Nurse Teacher framework

Abstract

Background

A supportive clinical learning environment is essential in nursing education; however, evidence regarding predictors of student satisfaction in Southeast European higher education contexts remains limited.

Objective

To identify demographic and item-level CLES+T predictors of nursing students’ overall satisfaction with the clinical learning environment in Kosovo.

Methods

A cross-sectional study was conducted between April and May 2025 among 300 undergraduate nursing students. Data were collected using the Clinical Learning Environment, Supervision and Nurse Teacher (CLES+T) scale, while overall student satisfaction was assessed using a separate single-item measure. Data were analyzed using descriptive statistics, non-parametric tests, and multiple linear regression analyses. The primary demographic model included sex, year of study, and economic status, while a second model included all 34 CLES+T items simultaneously.

Results

The demographic model was statistically significant, F(4, 295) = 35.894, p < 0.001, explaining 32.7% of the variance in satisfaction (R² = 0.327; adjusted R² = 0.318). Compared with first-year students, second- and third-year students reported significantly lower satisfaction, while sex and Economic Status were not significant predictors. The CLES+T item-level model was statistically significant, F(34, 265) = 18.319, p < 0.001, explaining 70.2% of the variance (R² = 0.702; adjusted R² = 0.663). The strongest positive predictors were positive joint meetings with the mentor and nurse teacher (β = 0.415), feeling comfortable going to the ward at the start of the shift (β = 0.335), and sufficient meaningful learning situations (β = 0.249). Because several item-level predictors showed elevated multicollinearity, individual coefficients, particularly negative coefficients, should be interpreted cautiously.

Conclusion

Nursing students’ satisfaction was associated with both academic year and specific relational, supervisory, and pedagogical characteristics of the clinical learning environment. Supportive mentorship, meaningful learning opportunities, and effective integration of theory and practice remain important components of positive clinical learning experiences.

Background

The training of nurses responsible for general care shall comprise at least three years of study or 4,600 hours of theoretical and clinical training. The duration of the clinical training shall include at least one-half of the minimum duration of the overall program (European Higher Education Area, 1999; European Parliament & Council of the European Union, 2005). Nursing education focuses particularly on the clinical aspect of training, where trainee nurses learn as part of a team, alongside nursing professionals, supervisors, and other healthcare team members, in direct contact with healthy or sick individuals in hospital theatres, ICUs, wards, and other care settings (European Parliament et al., 2016).

In community and hospital contexts alike, nursing students are expected to organize, deliver, and evaluate comprehensive nursing care based on the knowledge and skills they have acquired (Chan, 2002). Clinical learning environments (CLEs) represent approximately 50% of the educational experience for nursing students and are designated spaces where students engage in actual or simulated patient care. These environments enable students to integrate theoretical knowledge into practice, develop competence as entry-level nurses, and cultivate essential skills, attitudes, and decision-making abilities (Arkan et al., 2018; D’Souza et al., 2013; Flott & Linden, 2016; Şahan & Güven, 2020).

For students to feel motivated and satisfied during their clinical placements, managers of healthcare facilities and faculties must promote supportive learning and supervisory environments where supervisors, tutors, and staff recognize their responsibility in student education and foster optimal, positive learning experiences (Rodríguez-García et al., 2021). It has been demonstrated that the student–supervisor relationship significantly influences students’ satisfaction with the clinical learning environment (Pitkänen et al., 2018; Warne et al., 2010) and is therefore a crucial determinant of learning in clinical settings (Lee & Chiang, 2021). Furthermore, students evaluate their clinical learning environment and mentoring more positively when they have a named mentor, discuss learning goals with the mentor, and receive appropriate support during clinical practice (González-García et al., 2021; Saukkoriipi et al., 2020).

Motivation, physiological needs, and the overall learning environment are among the key factors that affect nursing students’ success in clinical practice (Hutchinson, 2003; Pournamdar et al., 2015). Additionally, students become acquainted with the realities of the nursing profession and its responsibilities within these environments (Clarke et al., 2003). Nevertheless, the clinical learning environment has been identified as a key determinant influencing students’ decisions to either continue or abandon the nursing profession (Joolaee et al., 2015). The complexity of these environments and the stressful events that occur within them make them difficult to manage (Papp et al., 2003).

A constructive clinical learning environment—characterized by structured student integration, an educational culture in which staff actively participate in student supervision, and collaboration between preceptors and nursing educators—is essential for student success (Demirören et al., 2008). Educators play a pivotal role in bridging theoretical knowledge with practical skills, and an enhanced educational role contributes to improved learning outcomes (Alharbi & Alshammari, 2024).

Guidance, support, stimulation, and the facilitation of learning across diverse practice settings form an integral part of the nurse educator’s clinical teaching role (WHO, 2016). Through this process, undergraduate nursing students have the opportunity to practice nursing care, acquire necessary competencies, internalize professional values, and develop interpersonal skills (Oermann et al., 2017). Nursing students may experience fear and anxiety within clinical settings, highlighting the importance of supportive supervision and structured mentorship approaches that promote emotional security and facilitate effective learning (Joolaee et al., 2015).

Despite the extensive literature on clinical learning environments, there remains a lack of empirical evidence specifically examining predictors of nursing students’ satisfaction using the Clinical Learning Environment, Supervision and Nurse Teacher (CLES+T) framework, particularly in Southeast European contexts such as Kosovo. In this study, developing European higher education contexts refer to higher education systems in transitional Southeast European countries, such as Kosovo, where nursing education is undergoing continuous structural and curricular reforms, and where empirical evidence on clinical learning environments remains limited compared with more established Western European systems. While previous studies have explored general perceptions of clinical learning environments, fewer have applied multivariate analytical approaches to identify independent predictors of satisfaction across both educational and contextual domains (Flott & Linden, 2016; Warne et al., 2010). This highlights a critical gap in understanding how different dimensions of the clinical learning environment interact to influence student satisfaction.

The CLES+T framework provides a multidimensional conceptual model that captures key aspects of the clinical learning environment, including the pedagogical atmosphere, supervisory relationships, leadership style, nursing premises, and the role of the nurse teacher (Saarikoski et al., 2008). These dimensions were used to inform the selection of variables included in the regression analysis, allowing for a theoretically grounded examination of how different components of the clinical environment contribute to overall student satisfaction.

In addition to the CLES+T framework, this study is conceptually informed by broader clinical learning environment theory, which views student satisfaction as shaped by the interaction of individual characteristics, interpersonal relationships, pedagogical processes, and organizational features of the clinical setting (Flott & Linden, 2016). Within this perspective, demographic factors may influence students’ expectations and adaptation to clinical practice, while supervisory support, pedagogical atmosphere, and nurse teacher involvement are expected to directly affect students’ perceptions of satisfaction in clinical learning environments.

Consistent with this conceptual perspective, demographic variables such as age, sex, year of study, clinical practice site, and socioeconomic status were included based on previous evidence suggesting that individual characteristics and contextual conditions may influence students’ perceptions, expectations, and adaptation to clinical learning environments (Chan, 2002; Papastavrou et al., 2016). For example, students at different stages of their education may experience varying levels of confidence and support needs, while contextual and socioeconomic factors may shape access to learning opportunities and perceived satisfaction within clinical settings.

This study aimed to identify predictors of nursing students’ satisfaction with the clinical learning environment in Kosovo. Specifically, the study examined associations between demographic variables (age, sex, year of study, clinical practice site, and socioeconomic status) and overall student satisfaction, and evaluated the item-level contribution of all 34 CLES+T items. To further clarify the analytical focus, this research examined which demographic factors were associated with nursing students’ satisfaction, which individual CLES+T items significantly predicted student satisfaction, and to what extent the demographic and CLES+T item-level models explained variation in satisfaction outcomes.

Methods

Study Design

A cross-sectional survey design using a standardized self-report measure was conducted from April 1 to May 31, 2025, among undergraduate nursing students. This study was conducted and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (von Elm et al., 2007). The research took place in Kosovo at a higher education institution offering undergraduate nursing education, within a clinical training system that integrates theoretical instruction with supervised clinical practice across primary, secondary, and tertiary healthcare settings.

Sample/Participants

The study included 300 undergraduate nursing students across all three academic years, representing a full response rate from the available cohort. A census sampling approach was used, comprising all available undergraduate nursing students within the institution. Eligible participants were undergraduate nursing students enrolled in the first, second, or third academic year who had completed at least one clinical placement and were present during the scheduled data collection sessions. The survey was administered in a controlled classroom setting during regular academic sessions on a voluntary basis, with informed consent obtained from all participants.

Participants were recruited directly during regularly scheduled class sessions, where all eligible students present were invited to participate immediately after being briefed on the study’s purpose. Data were collected during required attendance periods, ensuring access to all eligible students and eliminating the need for follow-up reminders. To minimize any potential perception of coercion associated with data collection during compulsory attendance, students were informed that attendance at the academic session did not imply an obligation to participate and that survey completion was entirely voluntary and free from academic consequences. Questionnaires were completed independently and without the presence of teaching faculty, thereby reducing the possibility that students might perceive participation as linked to academic evaluation or faculty relationships.

The sample size was substantial and balanced across study years. Regarding age, the majority of participants (n = 234; 78.0%) were aged 20–21 years, followed by students aged 18–19 years (n = 60; 20.0%), and a small proportion aged 22–23 years (n = 6; 2.0%). With respect to sex, the sample was predominantly female (n = 254; 84.7%), while male participants accounted for a smaller proportion (n = 46; 15.3%). Students were evenly distributed across academic years, with 100 students (33.3%) enrolled in each of the first, second, and third years of study, reflecting the actual cohort structure of the nursing program.

Clinical practice placements varied by care level: the majority of students were assigned to secondary care settings (n = 198; 66.0%), followed by tertiary care (n = 101; 33.7%), and a minimal number in primary care (n = 1; 0.3%). The limited representation of students in primary care settings is acknowledged as a limitation that may affect the comparability of clinical placement groups. Socioeconomic status was self-reported as follows: 4 students (1.3%) reported low economic status, 207 (69.0%) average, 59 (19.7%) above average, 21 (7.0%) high, and 9 (3.0%) very high. Only one student (0.3%) reported having a disability, while 299 (99.7%) reported no disability. Regarding overall satisfaction with the clinical learning environment, 105 students (35.0%) reported being very satisfied, 85 (28.3%) satisfied, 76 (25.3%) moderately satisfied, 26 (8.7%) dissatisfied, and 8 (2.7%) very dissatisfied.

Data Collection

The survey required approximately 10 to 15 minutes to complete and was self-administered in designated classroom sessions scheduled outside formal examinations or assessment activities. Questionnaires were completed in quiet environments using paper-based forms under standardized conditions, without the presence of teaching faculty, to minimize social desirability bias and ensure independent responses.

Instrument

Data were collected using the Clinical Learning Environment, Supervision and Nurse Teacher Scale (CLES+T), originally developed by Saarikoski et al. (2008). The CLES+T is a validated and widely applied instrument designed to assess nursing students’ perceptions of their clinical learning experiences. It provides a comprehensive, multidimensional evaluation through 34 items distributed across five core dimensions: the pedagogical atmosphere on the ward, the leadership style of the ward manager, the premises of nursing on the ward, the supervisory relationship, and the role of the nurse teacher.

Responses were recorded using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). In this study, student satisfaction was measured as a separate single-item variable reflecting overall satisfaction with the clinical learning environment, which served as the dependent variable in the regression analysis. Another section of the questionnaire addressed socio-demographic characteristics, including age, sex, year of study, clinical practice location, economic status, and disability status.

The instrument was administered in its original format and was deemed appropriate for the study context. Permission to use the instrument was obtained from the original developers.

The internal consistency of the CLES+T scale was assessed using Cronbach’s alpha, demonstrating good to excellent reliability across all subscales. Specifically, Cronbach’s alpha was 0.870 for the pedagogical atmosphere (9 items), 0.927 for the supervisory relationship (8 items), 0.895 for the role of the nurse teacher (9 items), 0.786 for the leadership style of the ward manager (4 items), and 0.776 for the premises of nursing on the ward (4 items), indicating satisfactory internal consistency.

Data Analysis

All statistical analyses were conducted using IBM SPSS Statistics (version 25). Descriptive statistics were used to summarize demographic characteristics and CLES+T questionnaire responses. Frequencies and percentages were calculated for categorical variables, while means, standard deviations, medians, and minimum/maximum values were reported for ordinal and continuous variables.

The Shapiro–Wilk test was applied to assess the normality of data distribution. Because most variables did not meet the assumption of normality, non-parametric tests were employed for bivariate analyses. Associations between demographic variables (age group, sex, year of study, type of clinical placement, economic status, and disability status) and satisfaction with clinical learning were examined using Kendall’s Tau correlation coefficient, which is particularly appropriate for ordinal variables with tied ranks and provides a robust estimate in datasets with many tied observations. To compare satisfaction levels across independent subgroups (e.g., sex), the Mann–Whitney U test was utilized.

To identify demographic predictors of student satisfaction, multiple linear regression analysis was performed. Year of study was treated as a categorical variable and entered using dummy coding, with first-year students serving as the reference category (Year2: second year = 1, otherwise = 0; Year3: third year = 1, otherwise = 0). Sex, economic status, and year of study were included in the primary demographic regression model based on the theoretical framework and preliminary analyses.

Clinical practice site was initially considered for the same model; however, cross-tabulation revealed substantial overlap with the year of study (all third-year students were assigned to tertiary healthcare, whereas all first-year students and 99 of 100 second-year students were assigned to secondary or primary healthcare). When Year3 and clinical practice site were entered simultaneously, severe multicollinearity was observed (VIF = 70.47 and 69.61, respectively; tolerance = 0.014 for both). Therefore, clinical practice site was not retained simultaneously with the year of study in the primary regression model to avoid unstable and uninterpretable estimates. A sensitivity analysis was subsequently conducted using sex, economic status, and clinical practice site as predictors (excluding year of study) to assess the association between clinical placement site and student satisfaction. Collinearity diagnostics, including tolerance and variance inflation factor (VIF), were examined for all models. For analyses involving the clinical practice site, the single participant assigned to primary healthcare (n = 1; 0.3%) was combined with the secondary healthcare category, resulting in a binary variable coded as 0 = secondary/primary healthcare and 1 = tertiary healthcare, thereby preventing the estimation of a regression coefficient from an inadequately represented category.

To examine the item-level contribution of the clinical learning environment, a second multiple linear regression model was conducted in which all 34 CLES+T items were entered simultaneously using the Enter method. Because the items conceptually overlap, tolerance and VIF were assessed for all item-level predictors to evaluate multicollinearity and coefficient stability.

Sex was coded as binary (0 = female, 1 = male) to clarify directionality in correlation and regression analyses. Predictor selection was theory-driven and informed by preliminary correlations. Regression assumptions—linearity, independence, and residual normality—were checked. Factors associated with variance in students’ clinical learning satisfaction were identified using p < 0.05 for statistical significance.

Ethical Considerations

The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and was approved by the Ethics Committee of Rezonanca College (approval number: AD-6022/24). All participants were fully informed about the purpose and objectives of the study and voluntarily agreed to participate by providing written informed consent.

Results

A total of 300 nursing students participated. Most were female (84.7%), aged 20–21 (78.0%), and placed in secondary care settings (66.0%). Students were evenly distributed across academic years. Overall, 63.3% were satisfied or very satisfied with clinical learning, while 25.3% were moderately satisfied and the remainder were dissatisfied or very dissatisfied (Table 1).

Table 1.

Demographic characteristics of the participants (N = 300)

General characteristics	Category	n	%
Age	18–19	60	20.0
	20–21	234	78.0
	22–23	6	2.0
Sex	Female	254	84.7
	Male	46	15.3
Academic year	1st year	100	33.3
	2nd year	100	33.3
	3rd year	100	33.3
Clinical practice site	Primary healthcare	1	0.3
	Secondary healthcare	198	66.0
	Tertiary healthcare	101	33.7
Economic status	Low	4	1.3
	Average	207	69.0
	Above average	59	19.7
	High	21	7.0
	Very high	9	3.0
Are you a person with a disability	Yes	1	0.3
	No	299	99.7
Satisfaction with clinical learning	Very dissatisfied	8	2.7
	Dissatisfied	26	8.7
	Moderately satisfied	76	25.3
	Satisfied	85	28.3
	Very satisfied	105	35.0

The distribution of students’ responses to the CLES+T scale items is presented in Table 2. Overall, responses indicate generally positive perceptions across all five dimensions of the clinical learning environment. In the pedagogical atmosphere dimension, the majority of students reported agreement or strong agreement with statements related to staff approachability, participation in discussions, and the overall learning environment. Similarly, the supervisory relationship dimension showed high levels of positive responses, particularly regarding supervisors’ attitudes, feedback, and mutual respect. The role of the nurse teacher received the highest levels of agreement, especially in relation to integrating theory into practice and supporting students’ learning needs. Positive perceptions were also observed in the leadership style of the ward manager and the premises of nursing on the ward, although some variability was noted across specific items.

Table 2.

Nursing students’ responses to the CLES+T scale

CLES+T Dimension/Item	Strongly disagree n (%)	Disagree n (%)	Neutral n (%)	Agree n (%)	Strongly agree n (%)
Pedagogical atmosphere
The staff members were easy to approach	4 (1.3)	14 (4.7)	97 (32.3)	122 (40.7)	63 (21.0)
I felt comfortable going to the ward at the start of my shift	10 (3.3)	17 (5.7)	99 (33.0)	122 (40.7)	52 (17.3)
During staff meetings (e.g., before shifts) I felt comfortable taking part in the discussions	5 (1.7)	18 (6.0)	73 (24.3)	109 (36.3)	95 (31.7)
There was a positive atmosphere on the ward	4 (1.3)	21 (7.0)	103 (34.3)	103 (34.3)	69 (23.0)
The staff members were generally interested in student supervision	7 (2.3)	38 (12.7)	124 (41.3)	82 (27.3)	49 (16.3)
The staff learned to know the students by their personal names	42 (14.0)	42 (14.0)	76 (25.3)	72 (24.0)	68 (22.7)
There were sufficient meaningful learning situations on the ward	7 (2.3)	23 (7.7)	74 (24.7)	110 (36.7)	86 (28.7)
The learning situations were multidimensional in terms of content	2 (0.7)	53 (17.7)	98 (32.7)	106 (35.3)	41 (13.7)
The ward can be regarded as a good learning environment	7 (2.3)	12 (4.0)	81 (27.0)	94 (31.3)	106 (35.3)
Supervisory relationship
My supervisor showed a positive attitude towards supervision	9 (3.0)	4 (1.3)	83 (27.7)	102 (34.0)	102 (34.0)
I had appropriate individual supervision	14 (4.7)	41 (13.7)	81 (27.0)	112 (37.3)	52 (17.3)
I continuously received feedback from my supervisor	12 (4.0)	33 (11.0)	82 (27.3)	104 (34.7)	69 (23.0)
Overall, I was satisfied with the supervision I received	4 (1.3)	20 (6.7)	111 (37.0)	85 (28.3)	80 (26.7)
The supervision was based on an equal relationship and promoted my learning	6 (2.0)	41 (13.7)	90 (30.0)	88 (29.3)	75 (25.0)
There was a mutual interaction in the supervisory relationship	10 (3.3)	18 (6.0)	99 (33.0)	104 (34.7)	69 (23.0)
The supervisory relationship was characterized by mutual respect and approval	2 (0.7)	8 (2.7)	74 (24.7)	118 (39.3)	98 (32.7)
The supervisory relationship was characterized by a sense of trust	4 (1.3)	20 (6.7)	96 (32.0)	106 (35.3)	74 (24.7)
Role of the nurse teacher
The nurse teacher was capable of integrating theoretical knowledge into daily nursing practice	2 (0.7)	2 (0.7)	103 (34.3)	102 (34.0)	91 (30.3)
The nurse teacher was able to achieve the learning objectives in the clinical unit	6 (2.0)	2 (0.7)	93 (31.0)	125 (41.7)	74 (24.7)
The nurse teacher helped me reduce the gap between theory and practice	2 (0.7)	12 (4.0)	65 (21.7)	115 (38.3)	106 (35.3)
The nurse teacher was seen as a member of the nursing team	6 (2.0)	11 (3.7)	54 (18.0)	92 (30.7)	137 (45.7)
The nurse teacher was able to bring his/her expertise to the clinical team	6 (2.0)	10 (3.3)	81 (27.0)	102 (34.0)	101 (33.7)
The nurse teacher and the clinical team worked together to support my learning	11 (3.7)	11 (3.7)	97 (32.3)	89 (29.7)	92 (30.7)
The joint meetings between me, my mentor, and the nurse teacher were a positive experience	2 (0.7)	21 (7.0)	91 (30.3)	102 (34.0)	84 (28.0)
The atmosphere during the meetings was pleasant	6 (2.0)	9 (3.0)	101 (33.7)	99 (33.0)	85 (28.3)
The focus of the meetings was on my learning needs	10 (3.3)	10 (3.3)	94 (31.3)	107 (35.7)	79 (26.3)
Leadership style of the ward manager
The ward manager regarded the staff on her/his ward as a key resource	23 (7.7)	23 (7.7)	69 (23.0)	111 (37.0)	74 (24.7)
The ward manager was a member of the team	41 (13.7)	23 (7.7)	72 (24.0)	100 (33.3)	64 (21.3)
Feedback from the ward manager could easily be considered as a learning situation	16 (5.3)	53 (17.7)	57 (19.0)	85 (28.3)	89 (29.7)
Individual efforts of staff members were appreciated	11 (3.7)	32 (10.7)	80 (26.7)	106 (35.3)	71 (23.7)
Premises of nursing on the ward
The ward’s nursing philosophy was clearly defined	7 (2.3)	46 (15.3)	88 (29.3)	88 (29.3)	71 (23.7)
Patients received individual nursing care	21 (7.0)	17 (5.7)	104 (34.7)	88 (29.3)	70 (23.3)
There were no problems in the flow of information related to patient care	6 (2.0)	19 (6.3)	70 (23.3)	145 (48.3)	60 (20.0)
The nursing documentation (e.g., care plans, daily recording of nursing procedures, etc.) was clear	9 (3.0)	5 (1.7)	63 (21.0)	104 (34.7)	119 (39.7)

Note: Responses were measured using a five-point Likert scale ranging from 1 = strongly disagree to 5 = strongly agree.

As presented in Table 3, the mean scores of the CLES+T dimensions ranged from 3.58 to 3.91, indicating overall positive perceptions of the clinical learning environment. The highest mean was observed for the role of the nurse teacher (M = 3.91, SD = 0.72), reflecting strong academic support and effective integration of theory into clinical practice. The lowest mean appeared for the leadership style of the ward manager (M = 3.58, SD = 0.92), which, although lower, still represented a satisfactory level of managerial engagement within the clinical setting.

Table 3.

Descriptive statistics of the CLES+T dimensions

CLES+T Domain	Mean	SD	Min	Max	Q1	Q3
Pedagogical atmosphere	3.65	0.71	1.00	5.00	3.20	4.20
Supervisory relationship	3.74	0.81	1.00	5.00	3.20	4.40
Role of the nurse teacher	3.91	0.72	1.00	5.00	3.40	4.60
Leadership style of the ward manager	3.58	0.92	1.00	5.00	3.00	4.20
Premises of nursing on the ward	3.77	0.79	1.00	5.00	3.20	4.40

Correlation analysis using Kendall’s Tau revealed statistically significant negative associations between student satisfaction and year of study (r = -0.404, p < 0.001), clinical practice site (r = -0.188, p < 0.001), and sex (r = -0.169, p = 0.002). Based on the coding scheme used for sex (0 = female, 1 = male), the negative correlation indicates that female students tended to report higher satisfaction scores than male students. Similarly, lower academic years and certain clinical placement categories were associated with higher reported satisfaction levels. These findings reflect statistical associations only and should not be interpreted as causal relationships. No significant correlations were observed with age group, economic status, or disability status (Table 4).

Table 4.

Correlation between student satisfaction and demographic variables (N = 300)

Variable	r	p
Age group	−0.075	0.152
Sex	−0.169**	0.002
Year of study	−0.404**	<0.001
Clinical practice setting	−0.188**	<0.001
Economic status	0.046	0.371
Disability status (Yes/No)	−0.024	0.650

Note: r = Kendall’s Tau correlation coefficient;

^*p < 0.05,

^**p < 0.001.

Following re-specification of the year of study as a categorical variable using dummy coding, with first-year students as the reference category, the primary demographic regression model included Sex, Year2, Year3, and Economic Status. The regression model was statistically significant, F(4, 295) = 35.894, p <0 .001, and explained 32.7% of the variance in student satisfaction (R² = 0.327; adjusted R² = 0.318). Year2 was a significant negative predictor of satisfaction (B = -1.414, β = -0.617, p < 0.001), indicating lower satisfaction among second-year students compared with first-year students. Similarly, Year3 was a significant negative predictor (B = -1.206, β = -0.527, p < 0.001), indicating lower satisfaction among third-year students compared with first-year students. Sex (B = 0.037, β = 0.012, p = 0.807) and Economic Status (B = 0.098, β = 0.070, p = 0.153) were not statistically significant predictors.

Collinearity diagnostics indicated no problematic multicollinearity in the revised primary model, with tolerance values ranging from 0.679 to 0.965 and VIF values ranging from 1.036 to 1.472. Clinical practice site was not entered simultaneously with the year of study because of near-perfect confounding between these variables in the sample (Table 5).

Table 5.

Multiple linear regression predicting student satisfaction from demographic variables (N = 300)

Predictor	B	SE	β	t	p	95 % CI for B	Tolerance	VIF
Constant	4.340	0.277	—	15.647	<0.001	3.794–4.885	—	—
Sex	0.037	0.151	0.012	0.245	0.807	−0.260–0.334	0.897	1.115
Year2	−1.414	0.133	−0.617	−10.658	<0.001	−1.675–-1.153	0.679	1.472
Year3	−1.206	0.131	−0.527	−9.224	<0.001	−1.464–-0.949	0.699	1.431
Economic status	0.098	0.068	0.070	1.433	0.153	−0.037–0.233	0.965	1.036

Note: Year1 was the reference category for Year of Study. Clinical Practice Site was not entered simultaneously with Year of Study because of near-perfect confounding between Year3 and clinical placement site.

As a sensitivity analysis, a separate regression model including Sex, Economic Status, and Clinical Practice Site, without the year of study, was conducted. This model was statistically significant, F(3, 296) = 7.318, p < 0.001, and explained 6.9% of the variance in student satisfaction (R² = 0.069; adjusted R² = 0.060). Clinical Practice Site was significantly associated with student satisfaction (B = -0.475, β = -0.208, p < 0.001), and Sex was also statistically significant (B = -0.445, β = -0.148, p = 0.009), whereas Economic Status was not statistically significant (B = 0.095, β = 0.067, p = 0.238). Collinearity was not problematic in this sensitivity model, with VIF values ranging from 1.013 to 1.032. For this sensitivity analysis, primary healthcare was combined with secondary healthcare because only one participant was assigned to the primary healthcare category.

Furthermore, the multiple linear regression model including all 34 CLES+T items was statistically significant, F(34, 265) = 18.319, p < 0.001, explaining 70.2% of the variance in overall student satisfaction (R² = 0.702; adjusted R² = 0.663). Compared with the primary demographic model (R² = 0.327; adjusted R² = 0.318), the CLES+T item-level model explained a substantially larger proportion of the variance in overall student satisfaction (R² = 0.702; adjusted R² = 0.663).

Several item-level predictors were statistically significant. The strongest positive predictor was positive joint meetings with the mentor and nurse teacher (V32; β = 0.415, p < 0.001), followed by feeling comfortable going to the ward at the start of the shift (I3; β = 0.335, p < 0.001), sufficient meaningful learning situations (I7; β = 0.249, p < 0.001), individual nursing care (III15; β = 0.217, p = 0.001), and a supervisory relationship characterized by mutual respect and approval (IV24; β = 0.209, p = 0.002).

Significant negative coefficients were observed for staff learning students’ personal names (I6; β = -0.217, p = 0.001), nurse teacher and clinical team support of learning (V31; β = -0.196, p = 0.006), multidimensional learning situations (I8; β = -0.160, p = 0.010), and meetings focused on students’ learning needs (V34; β = -0.133, p = 0.023). Collinearity diagnostics showed tolerance values ranging from 0.126 to 0.470 and VIF values ranging from 2.129 to 7.941. Because several item-level predictors showed elevated collinearity, individual regression coefficients, particularly negative coefficients, should be interpreted cautiously as they may partly reflect shared variance among the CLES+T items (Table 6).

Table 6.

Multiple linear regression predicting overall student satisfaction from CLES+T item-level predictors (N = 300)

Predictor	B	SE	β	t	p	95 % CI for B	Tolerance	VIF
Constant	0.150	0.278	—	0.539	0.590	−0.397 to 0.697	—	—
I1 - Staff members were easy to approach	0.201	0.084	0.164	2.394	0.017	0.036 to 0.366	0.239	4.177
I2 - Comfortable taking part in staff discussions	0.010	0.077	0.008	0.124	0.902	−0.142 to 0.161	0.247	4.044
I3 - Comfortable going to the ward at the start of the shift	0.373	0.077	0.335	4.835	<0.001	0.221 to 0.525	0.234	4.265
I4 - Positive atmosphere on the ward	−0.141	0.076	−0.123	−1.846	0.066	−0.291 to 0.009	0.255	3.924
I5 - Staff interested in student supervision	−0.089	0.068	−0.081	−1.296	0.196	−0.223 to 0.046	0.290	3.444
I6 - Staff learned to know students by their personal names	−0.176	0.053	−0.217	−3.349	0.001	−0.279 to -0.072	0.269	3.721
I7 - Sufficient meaningful learning situations on the ward	0.266	0.071	0.249	3.728	<0.001	0.126 to 0.407	0.253	3.955
I8 - Learning situations were multidimensional in content	−0.181	0.070	−0.160	−2.577	0.010	−0.319 to -0.043	0.291	3.433
I9 - Ward regarded as a good learning environment	0.139	0.069	0.128	2.016	0.045	0.003 to 0.275	0.279	3.580
II10 - Ward manager regarded staff as a key resource	0.001	0.055	0.001	0.010	0.992	−0.108 to 0.110	0.320	3.128
II11 - Ward manager was a member of the team	−0.040	0.053	−0.047	−0.747	0.456	−0.144 to 0.065	0.286	3.502
II12 - Feedback from ward manager regarded as a learning situation	−0.022	0.064	−0.025	−0.339	0.735	−0.149 to 0.105	0.210	4.771
II13 - Individual efforts of staff members were appreciated	−0.088	0.066	−0.087	−1.337	0.182	−0.218 to 0.042	0.266	3.761
III14 - Ward’s nursing philosophy was clearly defined	0.168	0.069	0.168	2.443	0.015	0.033 to 0.304	0.237	4.216
III15 - Patients received individual nursing care	0.210	0.063	0.217	3.316	0.001	0.085 to 0.335	0.263	3.805
III16 - No problems in the flow of information related to patient care	−0.079	0.058	−0.066	−1.356	0.176	−0.194 to 0.036	0.470	2.129
III17 - Nursing documentation was clear	0.053	0.069	0.048	0.775	0.439	−0.082 to 0.189	0.295	3.393
IV18 - Supervisor showed a positive attitude towards supervision	−0.060	0.072	−0.054	−0.831	0.407	−0.201 to 0.082	0.271	3.691
IV19 - Appropriate individual supervision	0.136	0.070	0.135	1.953	0.052	−0.001 to 0.274	0.235	4.264
IV20 - Continuously received feedback from supervisor	0.144	0.066	0.143	2.191	0.029	0.015 to 0.273	0.263	3.799
IV21 - Overall satisfaction with the supervision received	−0.178	0.105	−0.161	−1.700	0.090	−0.385 to 0.028	0.126	7.941
IV22 - Supervision based on an equal relationship and promoted learning	−0.050	0.089	−0.049	−0.556	0.579	−0.225 to 0.126	0.146	6.828
IV23 - Mutual interaction in the supervisory relationship	−0.099	0.078	−0.091	−1.262	0.208	−0.253 to 0.055	0.215	4.658
IV24 - Supervisory relationship characterized by mutual respect and approval	0.263	0.082	0.209	3.189	0.002	0.101 to 0.425	0.261	3.828
IV25 - Supervisory relationship characterized by a sense of trust	0.005	0.078	0.004	0.066	0.948	−0.148 to 0.158	0.243	4.111
V26 - Nurse teacher integrated theoretical knowledge into daily nursing practice	−0.011	0.093	−0.008	−0.114	0.909	−0.193 to 0.172	0.210	4.773
V27 - Nurse teacher achieved learning objectives in the clinical unit	−0.118	0.086	−0.094	−1.376	0.170	−0.286 to 0.051	0.240	4.158
V28 - Nurse teacher helped reduce the theory–practice gap	0.205	0.083	0.169	2.456	0.015	0.041 to 0.369	0.239	4.188
V29 - Nurse teacher was seen as a member of the nursing team	−0.158	0.084	−0.142	−1.888	0.060	−0.322 to 0.007	0.199	5.016
V30 - Nurse teacher brought expertise to the clinical team	−0.018	0.067	−0.016	−0.273	0.785	−0.150 to 0.113	0.321	3.120
V31 - Nurse teacher and clinical team worked together to support learning	−0.205	0.074	−0.196	−2.755	0.006	−0.352 to -0.059	0.222	4.506
V32 - Joint meetings with mentor and nurse teacher were a positive experience	0.475	0.093	0.415	5.114	<0.001	0.292 to 0.657	0.171	5.838
V33 - Atmosphere during the meetings was pleasant	0.122	0.089	0.107	1.379	0.169	−0.052 to 0.296	0.188	5.314
V34 - Focus of the meetings was on learning needs	−0.146	0.064	−0.133	−2.289	0.023	−0.272 to -0.020	0.333	2.999

Note: All 34 CLES+T items were entered simultaneously using the Enter method. Model fit: R =0 .838, R² = 0.702, adjusted R² = 0.663, F(34, 265) = 18.319, p < 0.001. VIF values are reported to assess multicollinearity among the CLES+T item-level predictors. Reference categories were not applicable because all predictors in this model were individual CLES+T items treated as numerical variables.

Discussion

The way students perceive their clinical placements has a substantial impact on their learning outcomes. To establish quality-based nursing practices, it is essential to provide nursing students with an effective and supportive clinical learning environment (Arkan et al., 2018; Bodys-Cupak, 2021; D’Souza et al., 2013). The clinical learning environment should promote learning, facilitate the application of theoretical knowledge, and support students in becoming competent and confident healthcare professionals (Flott & Linden, 2016; Guejdad et al., 2022).

The present study identified several statistically significant item-level predictors of nursing students’ satisfaction with the clinical learning environment, many of which reflected relational, supervisory, and pedagogical aspects of the CLES+T framework. Overall, the findings suggest that students’ satisfaction is closely associated with the quality of interpersonal interactions, mentorship, learning opportunities, and support received within clinical settings.

Among the strongest item-level predictors, positive joint meetings with the mentor and nurse teacher (β = 0.415, p < 0.001) and feeling comfortable going to the ward at the start of the shift (β = 0.335, p < 0.001) demonstrated the importance of supportive interpersonal relationships and emotional security within the clinical learning environment. Previous research has similarly shown that supportive supervisory relationships and meaningful mentor–student interactions are strongly associated with students’ satisfaction in clinical learning environments (Cant et al., 2021; Papastavrou et al., 2016). These findings reinforce the importance of consistent guidance, constructive feedback, and supportive supervision during clinical education.

Furthermore, sufficient meaningful learning situations (β = 0.249, p < 0.001), a supervisory relationship characterized by mutual respect and approval (β = 0.209, p = 0.002), and staff approachability (β = 0.164, p = 0.017) were positively associated with overall student satisfaction. The perception of the ward as a good learning environment was also a significant positive predictor (β = 0.128, p = 0.045). These findings reinforce the importance of a supportive clinical climate characterized by accessibility, mutual respect, and meaningful learning opportunities. Evidence indicates that clinical environments characterized by pedagogical support, professional accessibility, and constructive interpersonal dynamics are highly valued by nursing students (Warne et al., 2010).

In addition, individual nursing care (β = 0.217, p = 0.001), a clearly defined nursing philosophy (β = 0.168, p = 0.015), and continuous feedback from the supervisor (β = 0.143, p = 0.029) were significant positive predictors of student satisfaction. These findings highlight the importance of clear professional values, patient-centered nursing practice, and constructive supervisory feedback in creating a coherent and supportive clinical learning environment. Similar findings have been reported in previous studies examining clinical learning quality and student satisfaction (Mikkonen et al., 2017).

Moreover, the contribution of the nurse teacher in reducing the theory–practice gap (β = 0.169, p = 0.015) reinforces the important role of structured clinical guidance in connecting theoretical preparation with clinical practice. Previous research has demonstrated that effective clinical mentoring and appropriate integration of theoretical and practical learning are associated with students’ confidence, motivation, and positive perceptions of the clinical learning environment (Cant et al., 2021; Papastavrou et al., 2016).

Collectively, these findings support the relational and pedagogical foundations of the CLES+T framework. Previous studies using the CLES+T instrument have similarly emphasized the importance of supervisory relationships, continuity of mentorship, and nurse teacher involvement in shaping students’ perceptions of clinical learning quality (Magnani et al., 2014; Papastavrou et al., 2016; Saarikoski et al., 2008).

Several CLES+T items demonstrated statistically significant negative regression coefficients, including staff learning students’ personal names (β = -0.217, p = 0.001), nurse teacher and clinical team support of learning (β = -0.196, p = 0.006), multidimensional learning situations (β = -0.160, p = 0.010), and meetings focused on students’ learning needs (β = -0.133, p = 0.023). These findings should be interpreted with particular caution. Collinearity diagnostics for the 34-item model showed tolerance values ranging from 0.126 to 0.470 and VIF values ranging from 2.129 to 7.941, indicating substantial overlap among several CLES+T predictors. Therefore, the direction and magnitude of individual coefficients, particularly the negative coefficients, may partly reflect shared variance, suppression effects, or instability arising from the simultaneous inclusion of conceptually related items rather than inherently negative effects of these educational practices.

Accordingly, these negative coefficients should not be interpreted as evidence that the corresponding clinical or educational behaviors reduce student satisfaction. Rather, they illustrate the complexity of disentangling the independent contribution of closely related components of the clinical learning environment. This interpretation is particularly relevant for multidimensional constructs such as the CLES+T, in which several items assess conceptually overlapping aspects of supervision, pedagogical support, interpersonal relationships, and clinical learning.

Overall, the findings underscore the importance of structured supervision, respectful communication, meaningful learning opportunities, and supportive pedagogical environments in shaping students’ satisfaction with clinical learning. Mentorship quality, constructive feedback, clearly defined nursing values, and effective integration of theory and practice remain important components of positive clinical learning experiences. Educational institutions and clinical partners should therefore continue to strengthen mentorship structures, promote continuity in supervision, and foster an atmosphere of openness and mutual respect.

The item-level CLES+T regression model demonstrated substantial explanatory power, accounting for 70.2% of the variance in overall student satisfaction (R² = 0.702; adjusted R² = 0.663). However, given the observed multicollinearity among several CLES+T items, the strong overall model fit should be distinguished from the interpretation of individual regression coefficients. Consequently, individual item-level estimates should be interpreted cautiously and in the context of the conceptual overlap among the CLES+T predictors.

Limitations of the Study

This study should be interpreted with consideration of several limitations. First, the research was conducted within a single nursing education institution, which may limit the generalizability of the findings. Differences in clinical supervision structures and teaching cultures across other universities may influence students’ perceptions. Another limitation relates to the use of self-reported data, which depend on students’ subjective interpretations and may be affected by recall or social desirability bias. Although the CLES+T instrument is internationally validated, data were collected in an academic setting during scheduled sessions, which may have encouraged participants to provide more socially acceptable or favorable responses. In addition, cultural and organizational factors specific to the local context may have influenced how participants interpreted and rated certain items. In addition, because all study variables were measured using the same self-reported instrument and collected from the same respondents at one time point, the findings may be subject to common-method bias, which could have inflated observed associations among variables.

An additional limitation concerns the measurement of overall student satisfaction as a single-item outcome. Because satisfaction was assessed using a single item, internal consistency reliability, such as Cronbach’s alpha, could not be evaluated. Single-item measures may be more susceptible to random measurement error and may provide a less reliable and less comprehensive assessment of the construct than validated multi-item measures. Future studies should therefore consider using a validated multi-item measure of student satisfaction to improve measurement reliability and provide a more comprehensive assessment of the construct.

A further consideration involves the study design. Being cross-sectional, it captures students’ perceptions at a single point in time and therefore cannot establish causal relationships between predictors and satisfaction. Longitudinal or mixed-methods studies could provide deeper insight into how satisfaction with clinical learning evolves and how mentoring practices contribute to long-term professional development.

An additional limitation concerns the statistical complexity of the regression models. Given the relatively large number of predictors included, particularly in the CLES+T regression model, in relation to the sample size, there is a possibility of model overfitting and reduced stability of coefficient estimates. Although the sample size met general analytical requirements, future studies with larger and more diverse samples are recommended to improve model robustness and statistical precision. Finally, this study focused solely on students’ perspectives. Including the viewpoints of clinical mentors, nurse teachers, and supervisors in future research would provide a more comprehensive understanding of how relational and pedagogical factors interact to shape the clinical learning environment. A further limitation relates to the highly uneven distribution of clinical placement sites, particularly the minimal number of students assigned to primary healthcare settings (n = 1). This imbalance limits the generalizability of findings related to clinical placement comparisons and reduces the reliability of conclusions drawn for primary care settings.

Conclusion

This study identified academic year and several item-level characteristics of the clinical learning environment as significant correlates of nursing students’ overall satisfaction. In the demographic model, second- and third-year students reported lower satisfaction than first-year students, whereas sex and economic status were not statistically significant predictors. At the item level, positive joint meetings with the mentor and nurse teacher, feeling comfortable at the start of the clinical shift, meaningful learning situations, individual nursing care, and respectful supervisory relationships were among the strongest positive predictors. The CLES+T item-level model explained a substantial proportion of the variance in overall satisfaction. However, because several CLES+T items showed elevated multicollinearity, the direction and magnitude of individual coefficients, particularly negative coefficients, should be interpreted cautiously and should not be considered evidence that the corresponding educational practices reduce student satisfaction. Overall, the findings emphasize the importance of supportive mentorship, constructive feedback, meaningful learning opportunities, respectful supervision, and effective integration of theory and practice. Strengthening these aspects of the clinical learning environment may contribute to more positive learning experiences for nursing students.

Funding Statement

Funding This study did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The research was conducted independently as part of the authors’ academic and professional responsibilities.

Declaration of Conflicting Interest

The authors declare that there are no conflicts of interest related to the conduct, authorship, or publication of this study.

Author Contribution

Conception and design (MO, EP, HG); data analysis and interpretation (MO, EP, HG); manuscript drafting (MO, HG); critical revision of the manuscript (EP); final approval of the manuscript (MO, EP, HG).

Author Biography

Musa Oruqi, PhD, MD, is a Professor of Surgery in the Department of Nursing at Alma Mater Europaea – Rezonanca College, Prishtina, Kosovo. He is actively involved in nursing education and academic training, with extensive experience in surgical practice, teaching, and academic supervision. His professional interests include surgical care, clinical education, and interdisciplinary collaboration between surgery and nursing practice.

Elizana Petrela, PhD, is a Professor in the Department of Public Health, Faculty of Medicine, at the University of Medicine, Tirana, Albania. She has extensive experience in public health research and biostatistics, with a substantial record of peer-reviewed publications and scientific contributions in health sciences. Her professional interests include health promotion, epidemiology, and statistical methodologies applied to clinical and public health research.

Hasan Gashi, MSc, RN, is a Lecturer in the Department of Nursing at Alma Mater Europaea – Rezonanca College, Prishtina, Kosovo, and a PhD candidate at the University of Novo Mesto, Faculty of Health Sciences, Slovenia. He is engaged in undergraduate nursing education and has professional experience in clinical nursing practice and academic instruction. His areas of interest include nursing education, clinical skills development, and quality improvement in healthcare services.

Data Availability

The data supporting the findings of this study were collected using a structured questionnaire. Due to ethical and privacy considerations, the data are not publicly available. However, the data can be made available from the corresponding author upon reasonable request.

Declaration of Use of AI in Scientific Writing

The authors utilized Gemini during the writing process to enhance readability and eliminate grammatical errors. However, the authors took full responsibility for the content.