The adequate nurse-patient ratio is essential for maintaining quality nursing care and maximizing patient outcomes. However, the global nursing shortage has become a significant concern. The nursing workforce in South Korea exhibits a structural imbalance balance between supply and retention. As of the year 2022, the number of licensed nurses working in clinical settings in South Korea is significantly smaller than that in the Organization for Economic Co-operation and Development (OECD) countries (4.9 vs. 8.4 per 1,000 population), while the number of new nursing graduates is greater than in OECD countries (44.9 vs. 33.5 per 100,000 population) [1]. These figures suggest that, despite the sufficient output of new graduates, a persistent shortage of clinical nurses remains in practice.
High turover rates have been identified as a primary factor contributing to this discrepancy. The turnover rate of Korean nurses is approximately three times higher than that of nurses in the United States as of 2016 [2]. As a result, average tenure of Korean nurses in 2019 was only 7.8 years [3]. Low wages and heavy workloads have frequently been identified as major factors contributing to short tenure [4]. These results strongly indicate nursing graduates in South Korea are likely to leave in their workplace within a relatively short period. High turnover rates among nurses have a considerable negative impact on patient outcomes, including lower activities of daily living, worsening existing pressure ulcers or developing new ones, prolonged hospitalization, and increased medical errors [5, 6], ultimately contributing to rising healthcare costs [5]. Accordingly, nurse turnover can be regarded as a factor that affects not only individual healthcare institutions but also the overall efficiency and quality of the national healthcare system.
It was reported that the turnover rate of newly graduated nurses (NGNs) is higher than that those with more than five years of working experience [7]. NGNs are often experience a substantial level of job-related stress during their short transition period from students to competent nurses, which lasts for a couple of years after graduation [8]. They are required to adapt to their new roles as professional healthcare personnel, learn clinical knowledge and build relationships with various types of colleagues. Dissatisfaction with job experiences and difficulty in adapting to their new roles may contribute to high turnover rates. Lee [8] reported that 25% of nurses left their job within one year after graduation.
Turnover intention is a strong predictor of actual turnover among nurses and approximately 20% of nurses expressing turnover intention ended up leaving their positions [9]. Turnover intention is considered a complex concept that involves economic, psychological, and organizational outcomes, all of which depend on the interaction of various factors [10]. Factors related to NGNs’ turnover intention include work schedules, the desirability of the hospital, opportunities for professional development, becoming a part of a team and availability of practical support [11]. A recent study reported that the turnover intention of NGNs in South Korea was 12.8% [7]. In contrast, a meta-analysis of 12 studies including 8,593 participants revealed a wide variation in NGNs’ turnover intention, ranging from 6% to 61% across eight countries [12]. This meta-analysis identified both individual-level factors (e.g., age, work experience, education level, proactive personality, and undergraduate clinical education) and job-related factors (e.g., heavy workload, job stress, work type, and benefits) as significant determinants of NGNs’ turnover intention.
Among NGNs, the initial job experience is often described as a period of transition or adaptation. Studies on NGNs’ turnover have reported that transitional shock, which is characterized by the gap between learned theory and practical application, overwhelming workloads, lack of social support, and conflicts with coworkers, significantly contributes to turnover intention [13, 14]. These findings suggest that a wide range of factors contribute to the turnover intention of NGNs. Therefore, accurately understanding this phenomenon is crucial for maximizing organizational efficacy and enhancing the quality of nursing care, which is directly linked to positive patient outcomes.
Most of previous studies regarding turnover intention of NGNs have limitations due to small sample sizes and geographical restrictions. Furthermore, since many studies have verified only a limited number of variables selected by researchers, it is difficult to comprehensively identify the influencing factors as a whole [15]. Accurately predicting turnover intention using large datasets that incorporate multifaceted factors is crucial for developing more effective strategies to increase retention rates of NGNs.
Analyzing large datasets using the machine learning methods can more accurately predict influencing factors by considering the interactions and nonlinear relationships among various independent variables [16]. In particular, decision tree models, which are a type of machine learning algorithm, provide a valuable analytical framework for predicting related factors by identifying meaningful patterns and rules. These models have demonstrated effectiveness in predicting relevant factors to support decision-making in healthcare settings, and often outperformed traditional methods in specific contexts [17, 18]. Specifically, the Chi-square Automatic Interaction Detection (CHAID) decision trees employ the chi-square test of association to assess the goodness of fit between observed and expected outcomes, thereby determining splitting rules and node formation. The CHAID decision tree has a unique tree structure that can generate non-binary trees, which means some splits may have more than two branches, making it particularly suitable for analyzing larger datasets [19].
The conceptual framework of this study (Fig. 1) is grounded in Mobley’s turnover theory, which explains the decision-making process of dissatisfied employees considering leaving an organization [20], and in social exchange theory, a fundamental paradigm for understanding organizational behaviors [21]. Within this framework, turnover intention is defined as the voluntary thoughts and behaviors of organizational members regarding leaving the organization [22].
Fig. 1
Conceptual framework of this study
Based on previous literature, we identified four domains of factors relevant to NGNs: personal, college-related, job-related, and physical and mental health. Personal factors include age, gender, education level, and marital status, with younger age, male gender, higher education, and limited work experience consistently associated with greater turnover intention [23, 24].
College-related factors are defined as variables associated with a college or university environment that impact NGNs’ turnover intention. Literature emphasizes the importance of college education in preparing nursing students not only for clinical nursing after graduation but also for minimizing transitional shock in the workplace, therefore satisfaction for nursing education lead to low intention to leave [25]. More recently, a cross-national study of 10 countries demonstrated that NGNs’ educational satisfaction was significantly related to job satisfaction [26]. Job-related factors encompass employment status, income, incentive payments, workplace benefits, and job satisfaction. Salary is consistently reported as a significant factor influencing both turnover intention and actual turnover among nurses [27, 28]. Job satisfaction plays a crucial role in the decision-making process regarding leaving an organization within Mobley’s turnover model [29]. It has been identified as a strong predictor of both turnover intention and actual turnover in multiple studies [30, 31]. Physical and mental health factors include physical limitations and emotions that nurses frequently experienced during the past month with evidence that higher job stress, sleep disturbances, poor subjective health, and burnout significantly increase NGNs’ turnover intention [7, 32, 33]. Taken together, these factors illustrate how structural conditions, educational experiences, and individual health and demographic variables interact to shape NGNs’ turnover intention. Table 1 summarizes the definitions of these critical variables and the empirical evidence regarding their mechanisms of influence, which together form the basis for the conceptual framework of this study.
Table 1 Critical variables of the conceptual framework
The purpose of this study is to identify key predictors of turnover intention among NGNs within the first year of employment in the current workplace and examine how these interact with work values, personal factors, college-related factors and physical-mental health using a decision tree model based on the data from the 2016–2020 Graduate Occupational Mobility Survey (GOMS). Kim and Lee [34] also analyzed the 2010 GOMS data to examine turnover intention among NGNs focusing mainly on personal and job-related factors; however, the present study uses more recent datasets, incorporates college-related, physical and mental health-related variables, and applies the CHAID technique to capture interaction effects.