ANR220_proof 2 February 2017 1/7

Asian Nursing Research xxx (2017) 1e7

55
Contents lists available at ScienceDirect 56
57
58
Asian Nursing Research 59
60
journal homepage: www.asian-nursingresearch.com 61
62
63
Research Article 64
65
1 Mediating Role of Psychological Capital in Relationship between 66
2 67
3 Occupational Stress and Turnover Intention among Nurses at Veterans 68
4
5 Administration Hospitals in Korea 69
70
6 71
7 Q9 Hee-Yun Yim, RN, MSc, 1 Hyun-Ju Seo, RN, MPH, PhD, 2 Yoonhyung Cho, PhD, 3 72
8 JinHee Kim, RN, MPH, PhD 2, * 73
9 1 74
Nursing Department, Gwangju Veterans Administration Hospitals, Gwangju, South Korea
10 2
Department of Nursing, College of Medicine, Chosun University, Gwangju, South Korea 75
11 3
Department of Business Administration, College of Business, Chosun University, Gwangju, South Korea 76
12 77
13 78
14 a r t i c l e i n f o s u m m a r y 79
15 80
16 Article history: Purpose: The aim of this study was to evaluate the mediating role of psychological capital (PCP) in the 81
17 Received 6 October 2015 relationship between occupational stress and turnover intention in nurses. 82
Received in revised form
18 Methods: Data were collected from a sample of 447 nurses working at four Veterans Administration 83
4 January 2017
19 Hospitals throughout South Korea from July 1 to July 31, 2014. We collected data from the nurses using
Accepted 11 January 2017 84
20 the following surveys: the Short Form Korean-Occupational Stress Scale, the Korean version of the
85
21 Turnover Intention Scale, and the Korean version of the Psychological Capital Questionnaire. Multiple
Keywords:
linear regression analysis was performed to examine the mediating role of PCP.
86
22 nursing management 87
occupational stress
Results: The level of occupational stress was 1.8 0.2, the level of turnover intention was 3.3 0.9, and
23 88
psychological the PCP level was 3.9 0.5. There were signicant correlations among the three variables (occupational
24 stress, turnover intention, and PCP). PCP played a partial mediating role (b 0.216, p .008) in the 89
Q1 turnover intention
25 relationship between occupational stress and turnover intention (p < .001) among nurses working at the 90
26 Veterans Administration Hospitals. 91
27 Conclusion: Based on the ndings of this study, we recommend that South Korean hospitals offer 92
28 occupational stress management programs that incorporate relevant programs in efforts to strengthen 93
29 the overall components of PCP among nurses to reduce turnover intentions. Further studies are required 94
30 to determine the most effective intervention programs for hospital settings.
95
31 Copyright 2017, Korean Society of Nursing Science. Published by Elsevier. This is an open access article
96
32 under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
97
33 98
34 99
35 100
36 Introduction and intraprofessional conicts, emotional demands of caring, lack 101
37 102
of reward, role ambiguity, and shift work [5]. Research has
38 Stress is dened as an individual perceiving external demands demonstrated that occupational stress levels among nurses in Ko- 103
39 104
as being beyond his or her ability to cope with these demands [1]. rea are particularly high when working durations exceed 3 years,
40 More specically, stress relates both to individuals' perceptions of when nurses work three different shifts, when salaries are high, 105
41 106
the demands being made on them and to their perception of their and when nurses work on the general ward [4]. Occupational stress
42 capability to meet those demands [2]. Among health professionals, experienced by nurses can result in undesirable personal or orga- 107
43 nurses experience the highest levels of occupational stress [3]. The nizational outcomes, including increased medical expenditures, 108
44 occupational stress levels of nurses were reportedly 2.33e2.86 on a burnout, increased accidents, and job dissatisfaction [6]. Among 109
45 4-point scale [4]. Major sources of stress for nurses are heavy these negative outcomes, previous research has revealed a link 110
46 workloads, leadership and management issues, interprofessional between occupational stress and nurse turnover intention [7]. 111
47 The turnover intentions of nurses were reportedly 3.28e3.51 on 112
48 a 5-point scale [8]. General characteristics of nurses such as age, 113
49 * Correspondence to: JinHee Kim, RN, MPH, PhD, Department of Nursing, College educational level, work status, marriage status, type of work, 114
50 of Medicine, Chosun University, 309 Pilmun-daero, Dong-gu, Gwangju, 501-759,
experience, salary, and so on are reportedly related to turnover 115
51 South Korea.
intentions in Korea [9]. Nurse turnover is considered one factor that 116
E-mail address: jinheeara@chosun.ac.kr
52 117
53 118
http://dx.doi.org/10.1016/j.anr.2017.01.002
54 p1976-1317 e2093-7482/Copyright 2017, Korean Society of Nursing Science. Published by Elsevier. This is an open access article under the CC BY-NC-ND license (http:// 119
creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Yim H-Y, et al., Mediating Role of Psychological Capital in Relationship between Occupational Stress and
Turnover Intention among Nurses at Veterans Administration Hospitals in Korea, Asian Nursing Research (2017), http://dx.doi.org/10.1016/
j.anr.2017.01.002
 ANR220_proof 2 February 2017 2/7

2 H.-Y. Yim et al. / Asian Nursing Research xxx (2017) 1e7

1 negatively affects the quality of nursing care and patient outcome. been introduced as a positive resource in aiding nurses to overcome 66
2 Nurse turnover makes it difcult for human resources to maintain exhaustion and in helping to increase organizational commitment 67
3 or build a quality staff and can often lead to high costs for hiring and and customer orientation [18,19]. Although the feasibility of PCP is 68
4 training new nurses. The results of these effects leave organizations being assessed in the eld of nursing workforce management, no 69
5 with tremendous losses. Nurses who quit increase the workload for research has focused on the relationships among occupational 70
6 other nurses, which can often lead to a greater number of health stress, PCP, and turnover intention. This study yielded basic data 71
7 Q2 problems, emotional exhaustion, and accidents (e.g., needlestick related to building a workforce management plan to decrease nurse 72
8 injuries) [7]. Prevention of nurse turnover is a major challenge for turnover by identifying whether PCP plays a mediating role in the 73
9 human resource management in hospitals. relationship between occupational stress and turnover intention. 74
10 Instead of addressing the inadequacies of employees, employers 75
11 take an interest in cultivating positivity for employees to improve Objective 76
12 their work environment. Focusing on employees' positive qualities 77
13 is not merely an attempt to recognize the importance of positivity; The objective of this study was to evaluate the mediating role of 78
14 it is also an employer's attempt to adapt a more constructive PCP in the relationship between occupational stress and turnover 79
15 approach and to deviate from the mainstream approaches focused intention in nurses. To investigate this relationship, we measured 80
16 primarily on negative aspects [10]. the levels of occupational stress, turnover intention, and PCP in a 81
17 The positive organizational behavior approach is the study and large sample of nurses. We believe our ndings provide basic in- 82
18 application of positive human resource strengths and psychological formation for developing more successful nursing workforce 83
19 capacities that can be measured, developed, and effectively management strategies that will ultimately reduce nurse turnover 84
20 managed to improve performance in today's workplace [11]. Posi- intention. Specically, our study had three purposes: (a) to measure 85
21 tive organizational behavior and the increasing attention to em- the levels of occupational stress, turnover intention, and PCP in 86
22 ployees' positive characteristics have promoted a growing interest nurses; (b) to explore the differences associated with each of these 87
23 in psychological capital (PCP) within the eld of human resource factors; (c) to investigate the relationships among occupational 88
24 management. stress, turnover intention, and PCP in nurses; and (d) to identify 89
25 PCP is dened as an individual's positive psychological state of whether PCP plays a mediating role in the relationship between 90
26 development [11]. It has four components: efcacy, optimism, occupational stress and turnover intention in nurses. 91
27 hope, and resilience [11]. Efcacy is based on Bandura's Social 92
28 Cognitive Theory and is dened as one's conviction (or condence) Methods 93
29 about his or her abilities to mobilize the motivation, cognitive re- 94
30 sources, or courses of action needed to successfully execute a spe- Design 95
31 cic task within a given context [12,13]. According to Seligman 96
32 [14], optimism is the way in which a person interprets and un- This was a cross-sectional descriptive study aimed at identifying 97
33 derstands recent or past experiences. Snyder et al [15] dened hope the mediating role of PCP in the relationship between occupational 98
34 as a positive motivational state based on an interactively derived stress and turnover intention in nurses. 99
35 sense of successful agency (goal-directed agency) and pathways 100
36 (planning to meet goals). Hope does not merely involve willpower Participants and data collection 101
37 or the drive to achieve set goals; it also involves waypower, the 102
38 capacity to maneuver and work around challenges that stand in the The participants in this research were registered nurses working 103
39 way of achieving goals [10]. The last component, resilience, can be at four Veterans Administration Hospitals. South Korea has ve 104
40 understood as an individual's capacity to adapt after he or she has Veterans Administration Hospitals, but we collected data only from 105
41 experienced something negative or positive [11]. the four that were willing to cooperate with data collection. Re- 106
42 Efcacy affects how individuals perceive and interpret events. searchers visited the nursing department at each hospital to 107
43 Most human stress is dominated by beliefs about dealing with ef- request participation, and to explain the purpose of this study and 108
44 cacy [12]. An individual with a higher level of optimism is less the contents of the questionnaires. A total of 553 instructional 109
45 likely to experience symptoms of stress in the workplace [10]. Hope documents, along with consent forms and prepaid envelopes, were 110
46 provides individuals with a positive resource for stressful work distributed to nurses with the help of the nursing department 111
47 situations. An individual's level of hope protects him/her against (three hospitals received 130 documents each; the other hospital 112
48 perceptions of uncontrollability, vulnerability, and unpredictability received 163 documents). Nurses were informed of the purpose of 113
49 [16]. Resilience is the most important positive resource when the study and completed a written consent form prior to partici- 114
50 navigating a turbulent and stressful workplace [10]. Resilient in- pation. Of the 553 distributed questionnaires, 457 were collected, 115
51 dividuals are better equipped to cope with stressors in a workplace for a return percentage of 82.6%. Of the three hospitals that received 116
52 environment that continues to change. This is because these in- 130 questionnaires, one returned 103 (return percentage, 79.2%), 117
53 dividuals are open to new experiences, respond exibly to changing one returned 103 (return percentage, 79.2%), and one returned 108 118
54 demands, and demonstrate emotional stability when faced with (return percentage, 83.1%). The hospital that received 163 ques- 119
55 adversity [17]. tionnaires returned 143 questionnaires (return percentage, 87.7%). 120
56 PCP is negatively related to occupational stress, and decreased Of the returned questionnaires, 10 were excluded from this analysis 121
57 PCP, along with occupational stress, is thought to increase turnover due to various reasons: 6 questionnaires were missing answers for 122
58 intention [7,10]. Research has revealed that an increase in occupa- ve or more variables; the other 4 questionnaires had fewer than 123
59 tional stress results in a decrease in the overall PCP of employees, ve missing answers but included two or more with multiple re- 124
60 triggering consideration of turnover among nurses [10]. In the eld sponses. We believe that excluding these questionnaires, which 125
61 of nursing workforce management, active management of occu- were lled out with little sincerity, had no effect on the study re- 126
62 pational stress is required to prevent nurse turnover. Additionally, sults. All questionnaires were collected in a sealed envelope by mail 127
63 serious consideration of the use of PCP, a positive resource for or by the education head nurse with support from the division of 128
64 combating employee turnover that has been receiving much in- nursing service. 129
65 terest in the eld of workforce management, is needed. PCP has The data were collected from July 1st, 2014 to July 31st, 2014. 130

Please cite this article in press as: Yim H-Y, et al., Mediating Role of Psychological Capital in Relationship between Occupational Stress and
Turnover Intention among Nurses at Veterans Administration Hospitals in Korea, Asian Nursing Research (2017), http://dx.doi.org/10.1016/
j.anr.2017.01.002
 ANR220_proof 2 February 2017 3/7

H.-Y. Yim et al. / Asian Nursing Research xxx (2017) 1e7 3

1 Questionnaires and measurements guarantee the anonymity of nurses' identity, a prepaid envelope 66
2 was provided with the questionnaire and only personal informa- 67
3 Previously published and validated measures were used in this tion absolutely necessary for this study was collected. Participants 68
4 study. Questionnaires consisted of demographic questions (e.g., were informed that all information provided in the study would 69
5 gender, age, marital status, educational level, occupational level, remain condential and that all ndings would be reported 70
6 length of work experience as a nurse) and three scales that anonymously. 71
7 measured occupational stress, turnover intention, and PCP. 72
8 Data analysis 73
9 Occupational stress 74
10 Data were analyzed using the SAS statistical software, version 75
11 Occupational stress was dened as the harmful physical and 9.2 (SAS Institute, Cary, NJ, USA). First, general characteristics of 76
12 emotional responses that occur when the requirements of the job participants were expressed in terms of frequencies and percent- 77
13 do not match the capabilities, resources, or needs of the worker ages. Means and standard deviations were calculated to determine 78
14 [20]. To adjust to the South Korean context, occupational stress was the nurses' level of occupational stress, turnover intention, and PCP. 79
15 measured using the Short Form Korean Occupational Stress Scale To analyze differences in the levels of occupational stress, turnover 80
16 developed by Chang et al [21]. The scale is composed of 24 items intention, and PCP based on the respondent's general characteris- 81
17 from the following seven subscales: job demand, insufcient job 's test were
tics, a t test or one-way analysis of variance and Scheffe 82
18 control, interpersonal conict, job insecurity, organizational sys- utilized. Pearson's correlation coefcient was used to examine the 83
19 tem, lack of reward, and occupational climate. Each item is rated by relationships among occupational stress, turnover intention, and 84
20 a 4-point Likert scale ranging from 1 (strongly disagree) to 4 PCP. 85
21 (strongly agree). For this scale, high scores correspond to high levels To examine the mediating role of PCP in the relationship be- 86
22 of occupational stress. tween occupational stress and turnover intention, we utilized the 87
23 In the Chang et al [21] study, the Cronbach a score for occupa- method of Baron and Kenny [23]. They suggested that a mediating 88
24 tional stress ranged from .51e.82. In this study, Cronbach a for role can be identied by using the following regression equations: 89
25 occupational stress was .76. (a) regressing the mediator on the independent variable, (b) 90
26 regressing the dependent variable on the independent variable, 91
27 Turnover intention and (c) regressing the dependent variable on both the independent 92
28 variable and on the mediator. The following conditions must hold 93
29 Turnover intention was described as the process by which true to establish mediation from the results: (a) the independent 94
30 members of the nursing staff transfer or relinquish responsibilities variable must affect the mediator in the rst equation, (b) the in- 95
31 within their organization [7]. The Korean version of the Turnover dependent variable must affect the dependent variable in the sec- 96
32 Intention Scale was used to assess turnover intention of nurses. The ond equation, and (c) the mediator must affect the dependent 97
33 Turnover Intention Scale consists of four items, with each item on a variable in the third equation. If these conditions all hold as pre- 98
34 Q3 4-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly dicted, then the effect of the independent variable on the depen- 99
35 agree). For this scale, a higher score corresponded to higher rates of dent variable must be less in the third equation than in the second. 100
36 turnover intention. The reported score was an average of the scores Perfect mediation holds if the independent variable has no effect 101
37 for all four items. when the mediator is controlled. Partial mediation holds if the 102
38 According to a study conducted in South Korea by Kim and Lee, independent variable has an effect when the mediator is controlled. 103
39 the Cronbach a for turnover intention was .87 [8]. In this study, the 104
40 Cronbach a for turnover intention was .87. Results 105
41 106
42 PCP Baseline characteristics of participants 107
43 108
44 PCP has been dened as an individual's positive psychological Of the participants in this study, 98.7% were female, with the 109
45 state [22]. The Korean version of the Pychological Capital Ques- highest proportion in their 20s (43.6%), followed by those in their 110
46 tionnaire was used to assess the PCP level of nurses. It consists of 24 30s (34.5%) and 40s (17.9%). Approximately half (51.7%) were mar- 111
47 items on the following four subscales: self-efcacy (6 items), hope ried. In terms of educational level, the greatest proportion had a 112
48 (6 items), resilience (6 items), and optimism (6 items). Each item bachelor's degree (46.1%) and a slightly lower percentage had a 3- 113
49 was measured through a 6-point Likert scale ranging from 1 year college degree (43.0%). With regard to the occupational level, 114
50 (strongly disagree) to 6 (strongly agree). general-duty nurse (85.2%) was the most common position, while 115
51 Luthans et al [22] stated that the four components of PCP (ef- head nurse (5.8%) was the least common position. Nurses with 116
52 cacy, optimism, hope, and resilience) should not be examined more than 5 years, but less than 10 years, of work experience 117
53 separately but rather collectively, and that the factors have syner- (28.2%) were the most common, with having more than 15 years of 118
54 gistic effects, which can yield stronger explanatory power. Luthans experience (26.2%) being the next common. Nurses with less than 1 119
55 et al reported average scores from all 24 items, with higher scores year of experience comprised 3.1% of the participants (Table 1). 120
56 signifying an increased level of positive PCP [22]. 121
57 According to a study conducted in South Korea by Ko et al [18], Occupational stress based on participants' demographics 122
58 the Cronbach a for PCP was .93. In this study, Cronbach a for PCP 123
59 was .93. For all nurses, the average level of occupational stress was 124
60 1.8 0.2. Results showed that there were statistically signicant 125
61 Ethical considerations differences in the level of occupational stress based on occupational 126
62 level and length of job experience as a nurse (p .022, p .015). 127
63 This study was approved by the Bohun Medical Center's Insti- The general-duty nurses (1.8 0.2) reported a higher level of 128
64 tutional Review Board (Human 2014-06-02). All participants pro- occupational stress compared to that reported by head nurses 129
65 vided informed consent prior to answering the questionnaires. To (1.7 0.2). Although there were statistically signicant differences 130

Please cite this article in press as: Yim H-Y, et al., Mediating Role of Psychological Capital in Relationship between Occupational Stress and
Turnover Intention among Nurses at Veterans Administration Hospitals in Korea, Asian Nursing Research (2017), http://dx.doi.org/10.1016/
j.anr.2017.01.002
 ANR220_proof 2 February 2017 4/7

4 H.-Y. Yim et al. / Asian Nursing Research xxx (2017) 1e7

1 Table 1 Participant Demographics and Levels of Occupational Stress, Turnover Intention, and Psychological Capital (N 447). 66
2 67
Variable N % Occupational stress Turnover intention Psychological capital
3 68
4 M SD F or t p M SD F or t p M SD F or t p 69
5 Total 447 100.0 1.8 0.2 3.3 0.9 3.9 0.5 70
6 Gender 71
7 Male 6 1.3 1.7 0.4 0.45 .673 3.6 1.0 0.83 .405 4.0 0.6 0.35 .725 72
Female 441 98.7 1.8 0.2 3.3 0.9 3.9 0.5
8 73
Age
9 20e29 (a) 195 43.6 1.8 0.2 1.59 .192 3.4 0.8 9.97 < .001 3.8 0.5 25.21 < .001 74
10 30e39 (b) 154 34.5 1.8 0.2 3.4 0.8 a, b > c, d 4.0 0.5 a < c, d 75
11 40e49 (c) 80 17.9 1.8 0.2 3.0 0.9 4.2 0.5 b < c, d 76
50e59 (d) 18 4.0 1.7 0.2 2.6 1.0 4.5 0.3
12 77
Marital status
13 Married 231 51.7 1.8 0.2 0.89 .376 3.3 0.9 1.49 .138 3.8 0.5 0.54 < .001
78
14 Unmarried 216 48.3 1.8 0.2 3.2 0.9 4.1 0.5 79
15 Educational level 80
16 3-yr college (a) 192 43.0 1.8 0.3 2.00 .136 3.4 0.8 3.65 .027 3.8 0.5 19.39 < .001 81
Bachelor's (b) 206 46.1 1.8 0.2 3.2 0.9 4.0 0.5 a<b<c
17 82
Master's or higher (c) 49 11.0 1.8 0.2 3.1 1.1 4.3 0.5
18 Occupational level 83
19 General-duty nurse (a) 381 85.2 1.8 0.2 3.84 .022 3.4 0.8 16.58 < .001 3.9 0.5 25.28 < .001 84
20 Supervising nurse (b) 40 8.9 1.8 0.2 a>c 2.8 0.8 a > b, c 4.3 0.5 a < b, c 85
Head nurse (c) 26 5.8 1.7 0.2 2.6 1.0 4.4 0.4
21 86
Length of job experience as a nurse (yrs)
22 < 1 (a) 14 3.1 1.6 0.2 2.86 .015 2.4 0.5 9.00 < .001 3.8 0.5 15.75 < .001
87
23  1, but < 3 (b) 54 12.1 1.8 0.2 3.2 0.8 a < c, d, e 3.8 0.5 a, b, c, d, e < f 88
24  3, but < 5(c) 69 15.4 1.8 0.3 3.5 0.9 c, d > f 3.8 0.5 89
25  5, but < 10 (d) 126 28.2 1.8 0.2 3.5 0.8 3.8 0.5 90
 10, but < 15 (e) 67 15.0 1.8 0.2 3.4 0.8 4.0 0.5
26 91
 15 (f) 117 26.2 1.8 0.2 3.0 0.9 4.3 0.5
27 92
28 93
29 94
in the level of occupational stress among groups based on their followed by nurses who had earned a bachelor's degree (4.0 0.5)
30 95
length of work experience, this study did not examine these dif- or those who had earned a 3-year college degree (3.8 0.5). Nurses
31 96
ferences in any further detail (Table 1). at higher occupational levels had higher levels of PCP compared to
32 97
nurses at lower levels did. This observation is supported by the high
33 98
Turnover intention based on participants' demographics levels of PCP reported by supervising nurses (4.3 0.5) and head
34 99
nurses (4.4 0.4) compared with that of general-duty nurses
35 100
The average level of turnover intention for nurses was 3.3 0.9. (3.9 0.5). In addition, nurses with more than 15 years of experi-
36 101
This study found statistically signicant differences in the level of ence (4.3 0.5) had higher levels of PCP than did those with less
37 102
turnover intention among nurses based on their age, educational than 15 years of experience (Table 1).
38 103
39 level, occupational level, and length of work experience as a nurse 104
(p < .001, p .027, p < .001, p < .001, respectively). Nurses in their Correlations
40 105
41 20s and 30s had a higher level of turnover intention (3.4 0.8) 106
compared to nurses in their 40s (3.0 0.9) and 50s (2.6 1.0). Occupational stress and turnover intention had a statistically
42 107
Among all educational levels, there were statistically signicant signicant, positive correlation (r .52, p < .001). Occupational
43 108
differences in the level of turnover intention; however, this study stress and PCP had a statistically signicant, negative correlation
44 109
did not examine the contributing factors. The general-duty nurses (r .47, p < .001). PCP also had a statistically signicant, negative
45 110
(3.4 0.8) had higher levels of turnover intention compared to that correlation with turnover intention (r .39, p < .001) (Table 2).
46 111
47 of nurses at all other occupational levels. In terms of the length of 112
work experience as a nurse, those with 1e15 years of experience Mediating role of PCP in relationship between occupational stress
48 113
had a higher level of turnover intention than did those with less and turnover intention
49 114
50 than 1 year of work experience (2.4 0.5). Furthermore, those with 115
3e10 years of experience had a higher level of turnover intention To examine the mediating role of PCP in the relationship be-
51 116
compared to nurses with more than 15 years of work experience tween occupational stress and turnover intention, three regression
52 117
(3.0 0.9) (Table 1). analyses were conducted (Table 3) [23]. First, we observed a sta- Q4
53 118
tistically signicant negative impact of occupational stress on PCP
54 119
(b 0.984, p < .001). Second, we observed that occupational
55 PCP based on participants' demographics 120
56 121
57 The average PCP level was 3.9 0.5. Nurses' level of PCP was Table 2 Correlations among Occupational Stress, Turnover Intention, and Psychological 122
58 signicantly different across ages, marital status, educational levels, Capital (N 447). 123
59 occupational levels, and length of work experience as a nurse 124
Variable OS TI PCP
60 (p < .001). Nurses in their 40s (4.2 0.5) and 50s (4.5 0.3) had a 125
61 higher level of PCP than did those in their 20s (3.8 0.5) and 30s r (p) r (p) r (p) 126
62 (4.0 0.5). Nurses who were unmarried (4.1 0.5) had a higher OS 1 127
63 level of PCP than did those who were married (3.8 0.5). In terms TI .52 (< .001) 1 128
64 of the impact of educational level on PCP, nurses who had earned PCP .47 (< .001) .39 (< .001) 1 129
65 more than one master's degree (4.3 0.5) had the highest PCP, Note. OS occupational stress; PCP psychological capital; TI turnover intention. 130

Please cite this article in press as: Yim H-Y, et al., Mediating Role of Psychological Capital in Relationship between Occupational Stress and
Turnover Intention among Nurses at Veterans Administration Hospitals in Korea, Asian Nursing Research (2017), http://dx.doi.org/10.1016/
j.anr.2017.01.002
 ANR220_proof 2 February 2017 5/7

H.-Y. Yim et al. / Asian Nursing Research xxx (2017) 1e7 5

1 Table 3 Mediating Role of Psychological Capital in Relationship between Occupational For these reasons, management of occupational stress should be an 66
2 Stress and Turnover Intention (N 447). important aspect of the management of nurses. 67
3 Stepsa b (p) t R2 Adjusted R2 F (p) The nurses in this study reported levels of turnover intention 68
4 similar to those reported in a previous study conducted in Korea [8]. 69
1. OS / PCP 0.984 (< .001) 11.35 0.357 0.350 48.96 (< .001)
5 However, the level of turnover intention among nurses in this study 70
2. OS / TI 1.860 (< .001) 12.59 0.317 0.310 40.92 (< .001)
6 3. OS, PCP / TI 0.328 0.319 35.78 (< .001) was higher than those reported by nurses within the United States 71
7 PCP / TI 0.216 (.008) 2.69 and Europe, and among workers in various sectors in the United 72
8 OS / TI 1.647 (< .001) 9.87 States [10,26,27]. This suggests that the level of turnover intention 73
9 Note. OS occupational stress; PCP psychological capital; TI turnover intention. among Korean nurses is more likely to be higher than that in other 74
a
10 Values were adjusted for age, marital status, educational level, and occupational developed countries. Turnover intention is a strong predictor of 75
11 level. actual turnover in nurses [26]. The chronic shortage of nurses in 76
12 South Korea because of high rates of nurse turnover is an important 77
13 issue that must be addressed by the government. To tackle this 78
stress had a statistically signicant positive impact on turnover
14 problem, the Korean government has been helping in a number of 79
intention (b 1.860, p < .001). Lastly, a model that considered both
15 ways, including expanding the number of nursing schools and 80
occupational stress, as an independent variable, and PCP, as the
16 nding work for unemployed registered nurses. Despite such ef- 81
mediator, revealed that PCP had a statistically signicant negative
17 forts, the Korean government has failed to increase the number of 82
impact on turnover intention (b 0.216, p .008) The model also
18 practicing nurses. High staff turnover threatens the ability to secure 83
indicated that the b value of occupational stress equaled 1.647,
19 quality human resources and leads to huge costs for recruitment 84
which was lower than the b value from the second analysis (1.860).
20 and hiring. In addition, there is a decrease in the effectiveness and 85
Furthermore, this model highlighted the statistically signicant
21 productivity of the organization [7]. Therefore, this is not only an 86
impact of occupational stress on turnover intention (p < .001).
22 issue that needs to be addressed by the government, but hospitals 87
Therefore, our results suggest that PCP plays a partial mediating
23 also need to play an active role in devising plans and strategies to 88
role in the relationship between occupational stress and turnover
24 prevent the loss of high-quality personnel. 89
intention.
25 The results of this research revealed that turnover intention 90
26 levels among nurses in their 20s and 30s were higher than those of 91
27 Discussion nurses in their 40s and 50s, and also differed according to educa- 92
28 tional levels. Additionally, turnover intention was higher among 93
29 The chronic shortage of nurses in South Korea is a domestic and general-duty nurses than among supervising nurses and head 94
30 international issue. This research was conducted to demonstrate nurses, and those with less than 1 year of job experience as nurses 95
31 the importance of PCP by investigating its mediating role in the had higher levels of turnover intention compared with those with 96
32 relationship between occupational stress and turnover intention more than 1 year of job experience. A review of turnover intention 97
33 and to provide fundamental information for the development of among Korean nurses identied the following factors affecting 98
34 the nursing workforce management plan, which is essential in turnover intention level including age, educational level, work 99
35 preventing nurse turnover. status, marriage status, type of work, level of experience, and salary 100
36 The levels of occupational stress reported by nurses in this study [9]. These ndings are similar to those of the current study, but the 101
37 were lower than those reported previously [4]. The reason for this review did not clarify which values in terms of age, educational 102
38 difference might be that nurses enrolled in this study were from level, and so on were associated with the highest levels of turnover 103
39 Veterans Administration Hospitals, which tend to have more pa- intention because the individual studies included in this issue were 104
40 tients with chronic than with acute diseases. This affects the inconsistent, and the review could not directly compare the results. 105
41 workload of nurses, as nurses at hospitals that have more patients Therefore, it is difcult for us to directly compare our ndings with 106
42 with chronic diseases have lower workloads than those at hospitals that review. Research on turnover intention has been conducted for 107
43 having more patients with acute diseases. a long time within the Division of Nursing. In the future, it will be 108
44 The results also revealed that general-duty nurses had higher necessary to identify the characteristics of high-risk groups using 109
45 levels of occupational stress than head nurses did, and that occu- systematic reviews and/or meta-analyses of previous research 110
46 pational stress differed by length of job experience. Because head ndings. 111
47 nurses earn higher wages than general-duty nurses do, and wages Nurses in this study reported levels of PCP similar to those re- 112
48 increase with length of job experience, these ndings appear to ported in previous studies [18]. However, the levels were lower 113
49 support previous research indicating that high annual wages (> than those reported in research that targeted physicians, as well as 114
50 $25,000) are related to high occupational stress levels [4]. Together, workers in various sectors, in the United States [10,28]. Collectively, 115
51 these results suggest that hospitals should provide occupational Korean nurses appear to have lower PCP levels than do individuals 116
52 stress management programs for nurses, including head nurses, in other occupations. 117
53 with more job experience. PCP, which consists of efcacy, optimism, hope, and resilience, is 118
54 Study participants reported levels of occupational stress similar positively associated with employees' exhibiting desirable attitudes 119
55 to those reported by workers in Korean conglomerates [24]. Overall, (satisfaction, commitment, well-being), behaviors (citizenship 120
56 15.2%e18.4% of Korean workers are exposed to high levels of behavior), and performance. It is negatively associated with un- 121
57 occupational stress that can cause health problems, and the level of desirable attitudes (cynicism for change, stress, anxiety, turnover 122
58 occupational stress reported by nurses is sufcient to affect their intention) and behaviors. Efcacy may derive from mastery expe- 123
59 physical health [25]. rience, modeling, social persuasion, and physiological/psychologi- 124
60 Certain levels of stress are known to promote personal growth cal arousal [12]. Optimism is amenable to development through a 125
61 and increase productivity [10]. Alternatively, stress can have three-step process, including leniency toward the past, apprecia- 126
62 negative effects on both individual nurses and the organization tion for the present, and opportunity seeing for the future [14]. 127
63 (e.g., higher turnover, burnouts, increases in accidents). It is likely Hope can be built and supported via training interventions that 128
64 these errors occur when individuals are exposed to levels of stress focuses on designing goals, creating pathways, and overcoming 129
65 that cannot be managed, or accumulate into a chronic condition [6]. obstacles [16]. Resilience may be built using positive emotions and 130

Please cite this article in press as: Yim H-Y, et al., Mediating Role of Psychological Capital in Relationship between Occupational Stress and
Turnover Intention among Nurses at Veterans Administration Hospitals in Korea, Asian Nursing Research (2017), http://dx.doi.org/10.1016/
j.anr.2017.01.002
 ANR220_proof 2 February 2017 6/7

6 H.-Y. Yim et al. / Asian Nursing Research xxx (2017) 1e7

1 altering the relative levels of perceived risk and personal assets limitations. First, this study asked nurses employed in four of the 66
2 [10]. Avey et al [10] proposed that combining the positive resources ve Veterans Administration Hospitals in Korea to participate, and 67
3 of efcacy, optimism, hope, and resilience could help with the so was considered a convenience sample. For this reason, it may be 68
4 development of practical guidelines for managing and better un- difcult to make generalizations about all clinical nurses from our 69
5 derstanding the stress epidemic plaguing the workplace. To in- ndings. Second, this was a cross-sectional study, meaning that it 70
6 crease desirable attitudes, behaviors, and performance among was unable to account for alternative explanations of the order of 71
7 employees, an intervention program that strengthens all compo- variables. To understand the interactions between variables fully, a 72
8 nents of PCP must be developed and implemented. longitudinal study that accounts for the chronological context to 73
9 The analysis of the relationships among occupational stress, determine the cause-and-effect relationships among occupational 74
10 turnover intention, and PCP revealed a moderate negative rela- stress, PCP, and turnover intention is needed. Lastly, relationships 75
11 tionship between PCP and occupational stress, and a low negative between variables may be overstated due to the use of a self- 76
12 relationship between PCP and turnover intention. These ndings reported questionnaire, which can lead to social desirability bias. 77
13 are consistent with those of prior research, and are related to PCP's These limitations must be considered when interpreting the results 78
14 role as a partial mediator in the relationship between occupational of this study. 79
15 stress and turnover intention [10]. A previous study demonstrated 80
16 that organizational commitment partially mediated the relation- Conclusion 81
17 ship between PCP and customer orientation, and also yielded the 82
18 same ndings regarding relationships [18]. To develop nurse workforce management strategies that reduce 83
19 Analysis of the mediating effects of PCP in the relationship be- nurse turnover intention, this study measured levels of PCP, occu- 84
20 tween nurse occupational stress and turnover intentions veried pational stress, and turnover intention among nurses, and identi- 85
21 that PCP had a partial mediating effect. Thus, increased nurse ed the mediating role of PCP in the relationship between 86
22 occupational stress levels result in indirectly and directly lowered occupational stress and turnover intention. Our results indicated 87
23 levels of PCP and increases in turnover intention. Therefore, one that nurses working in Korea experience high levels of occupational 88
24 important result of this study is the suggestion that, because PCP stress and turnover intention and low levels of PCP compared with 89
25 plays a partial mediating role between occupational stress and those in other occupations. Statistically signicant associations 90
26 turnover intention, efforts must be made to increase PCP levels, were observed among all three variables, and PCP was shown to 91
27 specically, occupational stress management programs to lower have a partial mediating role in the relationship between occupa- 92
28 turnover intention. tional stress and turnover intention. Based on the results of this 93
29 Previous research has demonstrated that individual-level pro- study, we recommend that South Korean hospitals operate occu- 94
30 grams for occupational stress management using cognitive- pational stress management programs as the main program and 95
31 behavioral intervention and relaxation techniques are more effec- incorporate accompanying programs that strengthen the 96
32 tive than are organizational-level programs [29]. Furthermore, a overall components of PCP among nurses to reduce turnover 97
33 team approach, including workers, health administrators, and intention. Further studies are required to determine the most 98
34 managers is more effective than individual-approach programs. effective intervention programs for implementation in hospital 99
35 Luthans et al [30] developed instructions for strengthening the settings. 100
36 components of PCP using a Web-based intervention program. The 101
37 intervention program includes two parts. The rst develops and Conict of interest 102
38 enhances resilience and efcacy by dening the concepts of resil- 103
39 ience and efcacy and then presenting ways to implement them in No conict of interest has been declared by the authors. 104
40 the workplace. Afterward, subjects consider personal work-related 105
41 situations in their own organization and carry out self-reection 106
Acknowledgment Q5
42 exercises. The second part is intended to improve hope and opti- 107
43 mism by helping workers to identify realistic, challenging goals and 108
This research was supported by research fund from Chosun
44 to break this goal into smaller goals to increase the capacity for 109
University, 2014.
45 hope. As goals appear more feasible, expectations of success in the 110
46 applicable areas increase, ultimately inuencing levels of optimism 111
References
47 and hope. 112
48 Based on the ndings of these previous studies, occupational 1. Folkman S. Stress: appraisal and coping. Encyclopedia of behavioral medicine. 113
49 stress management programs should incorporate personal-level Springer; 2013. Q6 114
50 team approaches using cognitive-behavioral intervention and/or 2. Clancy J, McVicar AJ. Physiology and anatomy: a homeostatic approach. 2nd ed. 115
London: CRC Press; 2002. p. 611e33. Q7
51 relaxation techniques as the main strategy for reducing nurse 3. Sullivan T, Kerr M, Ibrahim S. Job stress in healthcare workers: highlights from
116
52 turnover intention [29,30]. This kind of program should also the National Population Health Survey. Hosp Q. 1998;2(4):34e40. 117
53 incorporate the PCP component-strengthening program suggested 4. Kwon KJ, Lee S-H. Occupational stress and coping styles as factors affecting the 118
burnout of clinical nurses. J Korean Acad Nurs Adm. 2012;18(4):383e93.
54 by Luthans et al [30]. Finally, this merged program should be http://dx.doi.org/10.11111/jkana.2012.18.4.383. Korean.
119
55 assessed to verify its effectiveness. Currently, research on PCP 5. Ruotsalainen J, Serra C, Marine A, Verbeek J. Systematic review of interventions 120
56 within the division of nursing is in the initial stages. More research for reducing occupational stress in health care workers. Scand J Work Environ 121
Health. 2008:169e78.
57 is needed to investigate the causal relationship and to explore the 122
6. Goetzel RZ, Carls GS, Wang S, Kelly E, Mauceri E, Columbus D, et al. The rela-
58 application and effects of PCP in lowering nurse turnover intention tionship between modiable health risk factors and medical expenditures, 123
59 so as to produce concrete evidence for incorporating PCP into absenteeism, short-term disability, and presenteeism among employees at 124
60 Novartis. J Occup Environ Med. 2009;51(4):487e99. 125
nursing workforce management plans.
http://dx.doi.org/10.1097/JOM.0b013e31819eb902
61 The results of this study demonstrated that the PCP has medi- 7. Hayes LJ, O'Brien-Pallas L, Dufeld C, Shamian J, Buchan J, Hughes F, et al. Nurse 126
62 ating effects in the relationship between nurse occupational stress turnover: a literature reviewdan update. Int J Nurs Stud. 2012;49(7):887e905. 127
63 and turnover intention. The ndings are signicant with regard to http://dx.doi.org/10.1016/j.ijnurstu.2011.10.001 128
8. Kim SH, Lee MA. Effects of emotional labor and communication competence on
64 the importance of PCP in workforce management strategies to turnover intention in nurses. J Korean Acad Nurs Adm. 2014;20(3):332e41. 129
65 lower nurse turnover intention. However, the study has several http://dx.doi.org/10.11111/jkana.2014.20.3.332. Korean. 130

Please cite this article in press as: Yim H-Y, et al., Mediating Role of Psychological Capital in Relationship between Occupational Stress and
Turnover Intention among Nurses at Veterans Administration Hospitals in Korea, Asian Nursing Research (2017), http://dx.doi.org/10.1016/
j.anr.2017.01.002
 ANR220_proof 2 February 2017 7/7

H.-Y. Yim et al. / Asian Nursing Research xxx (2017) 1e7 7

1 9. Kim JK, Kim MJ. A review of research on hospital nurses' turnover intention. 21. Chang SJ, Koh SB, Kang D, Kim SA, Kang MG, Lee CG, et al. Developing an 28
J Korean Acad Nurs Adm. 2011;17(4):538e50. occupational stress scale for Korean employees. Korean J Occup Environ Med.
2 29
http://dx.doi.org/10.11111/jkana.2011.17.4.538. Korean. 2005;17(4):297e317. Korean.
3 10. Avey JB, Luthans F, Jensen SM. Psychological capital: a positive resource for 22. Luthans F, Avolio BJ, Avey JB, Norman SM. Positive psychological capital: 30
4 combating employee stress and turnover. Hum Resour Manag. 2009;48(5): measurement and relationship with performance and satisfaction. Pers Psy- 31
5 677e93. http://dx.doi.org/10.1002/hrm.20294 chol. 2007;60(3):541e72. http://dx.doi.org/10.1111/j.1744-6570.2007.00083.x 32
11. Luthans F, Youssef CM, Avolio BJ. Psychological capital: developing the human 23. Baron RM, Kenny DA. The moderatoremediator variable distinction in social
6 competitive edge. New York: Oxford University Press; 2006. psychological research: conceptual, strategic, and statistical considerations. 33
7 12. Schwarzer R. Self-efcacy: thought control of action. New York: Routledge; 2014. J Pers Soc Psychol. 1986;51(6):1173. 34
8 13. Stajkovic AD, Luthans F. Social cognitive theory and self-efcacy: implications for 24. Park H, Jung H-S. Health behaviors by job stress level in a large-sized company 35
Q8 motivation theory and practice. Motivation and Work Behavior. 2003;126:140. with male and female workers. J Korean Acad Nurs. 2010;40(6):852e62.
9 14. Seligman ME. Learned optimism: how to change your mind and your life. New http://dx.doi.org/10.4040/jkan.2010.40.6.852. Korean.
36
10 York: Vintage Books; 2011. 25. Choi ES, Ha Y. Work-related stress and risk factors among Korean employees. 37
11 15. Snyder CR, Irving LM, Anderson JR. Hope and health: measuring the will and J Korean Acad Nurs. 2009;39(4):549e61. 38
the ways. In: Snyder CR, Forsyth DR, editors. Handbook of social and clinical http://dx.doi.org/10.4040/jkan.2009.39.4.549. Korean.
12 psychology. New York: Pergamon; 1991. p. 285e305. 26. Han GH, Jekel M. The mediating role of job satisfaction between leader-
39
13 16. Snyder C, Lopez SJ, Shorey HS, Rand KL, Feldman DB. Hope theory, measure- emember exchange and turnover intentions. J Nurs Manag. 2011;19(1):41e9. 40
14 ments, and applications to school psychology. Sch Psychol Q. 2003;18(2):122. http://dx.doi.org/10.1111/j.1365-2834.2010.01184.x 41
http://dx.doi.org/10.1521/scpq.18.2.122.21854 27. Van der Heijden B, Kmmerling A, Van Dam K, Van der Schoot E, Estryn-
15 har M, Hasselhorn H. The impact of social support upon intention to leave
42
17. Tugade MM, Fredrickson BL. Resilient individuals use positive emotions to Be
16 bounce back from negative emotional experiences. J Pers Soc Psychol. among female nurses in Europe: secondary analysis of data from the NEXT 43
17 2004;86(2):320. http://dx.doi.org/10.1037/0022-3514.86.2.320 survey. Int J Nurs Stud. 2010;47(4):434e45. 44
18. Ko JO, Park SK, Lee MH. Factors affecting burnout among clinical nurses ac- http://dx.doi.org/10.1016/j.ijnurstu.2009.10.004
18 45
cording to positive psychological capital. J Korean Acad Nurs Adm. 2013;19(2): 28. Liu L, Chang Y, Fu J, Wang J, Wang L. The mediating role of psychological capital
19 304e14. http://dx.doi.org/10.11111/jkana.2013.19.2.304. Korean. on the association between occupational stress and depressive symptoms 46
20 19. Kim IS, Seo RB, Kim BN, Min AR. The effects of positive psychological capital, among Chinese physicians: a cross-sectional study. BMC Public Health. 47
21 organizational commitment, customer orientation in clinical nurses. J Korean 2012;12(1):219. http://dx.doi.org/10.1186/1471-2458-12-219 48
Acad Nurs Adm. 2015;21(1):10e9. 29. Kim JH. A meta-analysis of effects of job stress management interventions
22 http://dx.doi.org/10.11111/jkana.2015.21.1.10. Korean. (SMIs). J Korean Acad Nurs. 2007;37(4):529e39. Korean. 49
23 20. National Institute for Occupational Safety and Health. Stress at work [Internet]. 30. Luthans F, Avey JB, Patera JL. Experimental analysis of a web-based training 50
24 Atlanta (GA): Author; 2014 [cited 2017 Jan 2]. Available from: http://www.cdc. intervention to develop positive psychological capital. Acad Manag Learn Educ. 51
gov/niosh/docs/99-101/ 2008;7(2):209e21. http://dx.doi.org/10.5465/AMLE.2008.32712618
25 52
26 53
27 54

Please cite this article in press as: Yim H-Y, et al., Mediating Role of Psychological Capital in Relationship between Occupational Stress and
Turnover Intention among Nurses at Veterans Administration Hospitals in Korea, Asian Nursing Research (2017), http://dx.doi.org/10.1016/
j.anr.2017.01.002