Nursing and Health Sciences (2011), 13, 184–191

Research Article

Turnover intention among Italian nurses: The moderating

roles of supervisor support and organizational support nhs_596 184..191

Maura Galletta, msc, phd,1 Igor Portoghese, msc, phd,1 Maria Pietronilla Penna, msc,1
Adalgisa Battistelli, msc, phd3 and Luisa Saiani, rn, bnsc, mnsc2
1
Department of Psychology, University of Cagliari, Cagliari, 2Department of Public Health and Community Medicine,
University of Verona, Verona, Italy and 3Laboratory Epsylon – Dynamics of Human Abilities and Health Behaviors,
University of Paul Valéry Montpellier, Montpellier, France

Abstract The purpose of this study was to examine the variables that are related to person–environment fit in Italian
nurses, highlighting the role of supervisor support and organizational support in the relationship between
nurses’ perceptions of care adequacy, job satisfaction, and turnover intention. Therefore, 1240 nurses from
three hospitals completed a self-administered questionnaire. The results showed that supervisor support and
organizational support act differently as moderators of the care adequacy–job satisfaction–turnover intention
relationship. Finally, job satisfaction was a mediating variable between care adequacy and turnover intention.
These findings have important implications for hospitals because they help to promote effective work envi-
ronments and to reduce turnover intention.

Key words care adequacy, Italy, job satisfaction, nurses’ turnover intention, organizational support, retention of nurses,
supervisor support.

INTRODUCTION individuals’ satisfaction, organizational functionality, and the

recruitment and retention of health workers (Schneider et al.,
Nurses’ retention is one of the main concerns in the health-
2000). Exploring the relationship between nurses’ percep-
care system, especially in times of chronic nursing shortages.
tions of their working environment and their professional
In the last 10 years, many countries have invested heavily in
needs, little research has been done on the predictor role of
resolving this phenomenon and shifted towards evidence-
care and resource adequacy (care adequacy) and the poten-
based practice, which has resulted in the need to study the
tial intervening factors that are related to turnover intention.
relationships among staffing turnover, nurse satisfaction,
Particularly, in the context of inadequate care and resources,
quality, and adequacy of care (Melnyk & Fineout-Overholt,
there are risks of errors, absenteeism, and high levels of
2010). Wagner (2010) asserted that nursing turnover is a dan-
staff turnover (Stone et al., 2006). Aiken et al. (2008), in their
gerous outcome because it is linked to the loss of individual
research on magnet hospitals, showed that perceptions of the
and organizational performance, a significant reduction in
quality and adequacy of care are important variables among
the quality of care, an increase in the workload of the staff
the numerous factors that influence nurses’ satisfaction. Spe-
members who survived the turnover, and a loss in morale
cifically, they suggested that nurses’ job satisfaction is linked
and further turnover. Actually, in Italy, a worrisome nursing
to their ability to provide high-quality care. This research
shortage has been observed and the Italian Federation of
highlights how care adequacy is related to both job satisfac-
Nurses has estimated a current shortage of 40,000 nurses
tion and turnover intention and examines the important role
(Destrebecq et al., 2009). There has been growth in the
of social exchange constructs (e.g. perceived support) in the
research that seeks to identify the most important predictors
above-mentioned relationships.
of nurses’ retention (Beecroft et al., 2008). However, accord-
ing to Takase et al. (2005: 211–212), researchers “tend to
examine the effects of environmental characteristics and LITERATURE REVIEW
nurses’ needs separately and assume that each or both of the
factors individually contribute to nurses’ work behavior.” In Theoretical perspectives
this regard, the work environment’s characteristics affect In this study, the Person–Environment (P–E) Fit Theory
(Schneider et al., 2000) was considered as the theoretical
framework. The central tenet of much of this work is that a
Correspondence address: Maura Galletta, Department of Psychology, University of
Cagliari, Via Is Mirrionis 1, 09123 Cagliari, Italy. Email: maura.galletta@gmail.com good fit between the person and the situation will generate
Received 24 February 2011; accepted 4 April 2011. positive outcomes, whereas a mismatch will produce negative

© 2011 Blackwell Publishing Asia Pty Ltd. doi: 10.1111/j.1442-2018.2011.00596.x

Nursing turnover intention 185

outcomes. Schneider et al. (2000) proposed a person-based employed, mediated through job satisfaction. Nurses who
framework to explain the process by which individuals are perceived their managers as supportive (e.g. recognition,
attracted to, selected by, and either leave or remain in orga- encouragement, flexible work schedules, and positive feed-
nizations. The main characteristic of the P–E Fit Theory is its back) reported greater scores in both job satisfaction and the
focus on the interaction between employees and their job or intention to remain employed (Sourdif, 2004). Moreover, per-
organization, rather than on the individual effects of personal ceptions of supervisory support were related to perceptions
or organizational characteristics, to explain the employees’ of the quality of care in the unit (Raikkonen et al., 2007).
work behavior. According to Schneider (1983: 35), “people Based on these considerations, PSS is likely to act specifically
are attracted to organizations because of organizational on the relationship between care adequacy and job satisfac-
goals, organizations select people who appear to be able to tion, which in turn is related to turnover intention. Perceived
help the organization achieve its goals, and people who organizational support is activated by a social exchange
achieve their own goals there will tend to remain in the process and through a norm of reciprocity, which can produce
organization.” However, a misfit can motivate workers to try a sense of employee obligation to aid the organization. This
to change the organizational characteristics to meet their might generate attitudinal and behavioral outcomes that
job needs, adjust themselves to the organization, or quit reciprocate favors, which lead to lower turnover (Tekleab
the organization to search for another compatible organiza- et al., 2005). As job satisfaction is a relevant antecedent of
tion (Takase et al., 2005). Therefore, a misfit between the job turnover intention (Lambert et al., 2001), POS is likely to act
and environmental supplies can lead to employees’ turnover specifically on the relationship between job satisfaction and
intention (Takase et al., 2008). For example, a misfit can occur turnover intention. Thus, supervisory and organizational
when nurses perceive that, in their unit, the main (profes- support can act as facilitators in the process of “matching” the
sional and organizational) mission to offer adequate care to employee with the work environment. Even though the lit-
patients is not ensured. In fact, studies on staff nurses in erature has shown the importance of PSS and POS in the
magnet hospitals identified the perceived care adequacy as relationship between job satisfaction, quality of care, and
one of the principal elements of a healthy, satisfying, and turnover intention, how PSS and POS can act differently as
productive work environment, which increased nurses’ reten- moderating variables in the care adequacy–job satisfaction–
tion (Kramer & Schmalenberg, 2002). Furthermore, in defin- turnover intention relationship has not been analyzed.
ing the quality of the relationship between the individual and
the organization, a growing body of work recognizes the AIM AND HYPOTHESES
important role of support from both a supervisor and the
The aim of the study was to examine the relationship
organization in decisions to stay (Maertz et al., 2007).
between the principal variables that are related to turnover
intention. Specifically, the moderating role of PSS on the
Perceived supervisor and organizational support relationship between care adequacy and job satisfaction and
the moderating role of POS on the relationship between job
An important factor in defining the essence of social
satisfaction and turnover intention were examined. Based on
exchange in employment relationships is supervisory and
the literature, the following hypotheses were tested:
management support. In fact, employees develop general
• Hypothesis 1a: care adequacy shows a positive relation-
views regarding the degree to which supervisors and the
ship with job satisfaction
organization value their contributions and care about their
• Hypothesis 1b: PSS shows a positive relationship with
well-being (Eisenberger et al., 1986; Kottke & Sharafinski,
job satisfaction
1988; Rhoades & Eisenberger, 2002), which are called “per-
• Hypothesis 2: PSS moderates the positive relationship
ceived supervisor support” (PSS) and “perceived organiza-
between care adequacy and job satisfaction, such that the
tional support” (POS), respectively. As supervisors act on
relationship is stronger when PSS is higher
behalf of the organization because they evaluate the employ-
• Hypothesis 3a: Job satisfaction shows a negative rela-
ees’ contribution and manage the reward systems, the treat-
tionship with turnover intention
ment that employees receive from their supervisor can
• Hypothesis 3b: POS shows a negative relationship with
impact on their turnover intention (Eisenberger et al., 1986;
turnover intention
Tekleab et al., 2005). In the healthcare context, a nursing
• Hypothesis 4: POS moderates the negative relationship
supervisor is the person who directly works with the nursing
between job satisfaction and turnover intention, such that the
staff in the unit and therefore has an impact on both the
relationship is weaker when POS is lower
staff’s working life and quality of care (Failla & Stichler,
• Hypothesis 5: Job satisfaction mediates the relationship
2008). They play a key role in coping daily with major
between care adequacy and turnover intention
challenges, as well as in demanding efficiency (Abdelrazek
Figure 1 represents the hypothesized relationships.
et al., 2010), and they can affect the quality in different ways,
including the effect that they have on job satisfaction or staff
METHOD
well-being (Hannan et al., 2001). In fact, a lack of support
from the direct supervisor could prompt nurses’ dissatis-
Design, participants, and data collection
faction and burnout. According to Tourangeau and Cranley
(2006), supervisor support is an important organizational A cross-sectional survey was conducted. A convenience
factor that indirectly affects the intention to remain sample of registered nurses was recruited for this study. For

© 2011 Blackwell Publishing Asia Pty Ltd.

186 M. Galletta et al.

items included “My supervisor really cares about my well-

being.” To measure care adequacy, four items that were
adapted from Nursing Work Index-Revised (Aiken & Patri-
cian, 2000) were used. The sample items included “Enough
time and staff to get the work done.” The Questionnaire of
Organizational Satisfaction from Cortese (2001) was used to
assess job satisfaction. Six high-loading items (loadings from
0.70–0.76) that related to satisfaction with job aspects were
selected. The sample items included “The personal growth
Figure 1. Representation of the hypothesized relationships. H1a derived from my job” and “Quality of my job.” Turnover
and H1b indicate the direct relationships of care adequacy and PSS intention was measured by two items that were adapted from
on job satisfaction; H2 indicates the moderating effect of PSS on the Hom et al. (1984). The sample items included “I am going to
relationship between care adequacy and job satisfaction; H3a and seek a job in another hospital next year” and “I intend to
H3b indicate the direct relationships of job satisfaction and POS on leave my unit to work in another unit in the same hospital
turnover intention; H4 indicates the moderating effect of POS on the next year.”
relationship between job satisfaction and turnover intention; H5
indicates the mediating effect of job satisfaction on the relationship
between care adequacy and turnover intention. H, hypothesis; POS, Data analysis
perceived organizational support; PSS, perceived supervisor support.
All the analyses were carried out by using SPSS software
program version 17.0 (SPSS: An IBM Company, Chicago, IL,
USA). An exploratory factor analysis (EFA) for the con-
the recruitment of the sample, hospitals within the public structs’ measurement was used and the assessment of the
sector and that were in urban areas were used. After contact- internal consistency of the measures was conducted by Cron-
ing the hospitals, the nurses for whom the administration bach’s alpha (a). Relative to hypotheses 2 and 4, moderation
gave formal approval to participate in the study were con- analyses were carried out via hierarchical regression analy-
tacted and recruited. In total, this research involved 1938 ses. A moderator is a variable that alters the strength of the
registered nurses from three hospitals in the north of Italy. Of relationship between an independent variable and a depen-
these nurses, 1240 completed the questionnaire, yielding a dent variable. Specifically, the authors believed that PSS
response rate of 64%. The data were collected by using a moderates the relationship between care adequacy and job
paper questionnaire, which was distributed to the nurses satisfaction. In the same manner, the authors expected to find
through the nurse supervisor of each unit. The participants an interaction between job satisfaction and POS with respect
anonymously and voluntarily completed the questionnaire to their effect on turnover intention. Relative to hypothesis 5,
during working hours. The nurses were given 3 weeks to com- a mediation analysis was carried out by the joint significance
plete and return their questionnaire in locked boxes. test (JST) (MacKinnon et al., 2002). A mediator (or interven-
ing) variable transmits the effect of an independent variable
to a dependent variable. The mediator causes variation in the
Instrument
outcome variable and itself is caused by the predictor vari-
The questionnaire included one part concerning personal able. In this study, the authors supposed that job satisfaction
data (sex, age, and tenure) and another part containing the intervenes in the relationship between care adequacy and
measurement scale of the variables that were used in this turnover intention.
study. As not all the scales that were used had been validated
previously in Italian, the translation–back-translation proce-
Ethical considerations
dure (Brislin, 1980) was used for those scales. A panel that
was made up of two bilingual linguistic experts and two The study was approved by the ethical committee of the
Italian nurse academics translated the original structure of Italian public health system. The participants were informed
the questionnaire independently from English into Italian. that their anonymity was assured. Informed consent to par-
After reviewing both translations, minimal changes were ticipate was assumed when the nurses returned a completed
made to the wording. A pilot study, involving 89 Italian questionnaire.
nurses, then was conducted to ensure further validity of the
instrument.
RESULTS
All the items were rated on a five-point scale, ranging
from 1 (“strongly disagree”) to 5 (“strongly agree”). To
Descriptive characteristics
measure POS, four high-loading items (loadings from 0.54–
0.67) from the Survey of Perceived Organizational Support Table 1 shows the mean values, standard deviations (SDs),
(SPOS) (Eisenberger et al., 1986) were used. The sample Cronbach’s alpha coefficients, and correlations among the
items included “My department cares about my opinions.” variables that were analyzed in this study. They all were sig-
To assess PSS, the SPOS was adapted by replacing the word nificantly correlated. The magnitude and direction of these
“organization” with the term “supervisor”. Five high-loading correlations were consistent with the predictions. Most of the
items (loadings from 0.78–0.87) were selected. The sample sample was female (81.45%). The average age of the women

© 2011 Blackwell Publishing Asia Pty Ltd.

Nursing turnover intention 187

Table 1. Means, standard deviations, and correlations among the study variables (n = 1240)

Mean SD TurInt JobSat CareAd PSS POS

TurInt 1.830 1.221 0.850

JobSat 3.344 0.678 -0.233** 0.870
CareAd 2.373 0.772 -0.149** 0.358** 0.820
PSS 2.964 0.874 -0.187** 0.485** 0.345** 0.920
POS 2.773 0.771 -0.197** 0.546** 0.371** 0.715** 0.860

**P < 0.01. The Cronbach’s alpha values are shown in italics. CareAd, care adequacy; JobSat, job satisfaction; POS, perceived organizational
support; PSS, perceived supervisor support; TurInt, turnover intention.

Table 2. Hierarchical regression results for the moderating effect of perceived supervisor support (PSS) on the relationship between care
adequacy and job satisfaction

95% CI
Regression step b SE t-value Lower limit Upper limit

Step 1
Education 0.01 0.018 0.41 -0.03 0.040
Tenure -0.08** 0.003 -2.87 -0.01 -0.003
R2 = 0.004*
Step 2
Care adequacy 0.22*** 0.026 8.55 0.17 0.270
PSS 0.40*** 0.026 15.37 0.35 0.450
DR2 = 0.27***
Step 3
Care adequacy ¥ PSS 0.09*** 0.023 3.69 0.04 0.130
DR2 = 0.008***
F = 13.64
Total R2(adj) = 0.29

*P < 0.05, **P < 0.01, ***P < 0.001. Standardized b-values from the final step of the regression analysis are presented. CI, confidence interval.

was 36.95 years (SD = 7.91) and the average age of the men Hierarchical regression analyses
was 37.31 years (SD = 8.19) (range: 21–60 years). The mean
experience levels were 14.16 years (SD = 8.88) in the profes- In order to test the interaction effects, two different regres-
sion (range: 1–38 years) and 7.84 years (SD = 6.92) in the unit sion models were carried out. Multiplicative terms for the
(range: 1–38 years). Table 1 shows that the Italian nurses per- standardized independent variables were created, as sug-
ceived moderate levels of care adequacy, job satisfaction, gested by Cohen and Cohen (1983). Then, the standardized
POS, and PSS and had low scores of turnover intention, on independent variables were introduced into the equation in
average (mean = 1.83, SD = 1.22). three steps. Regression equations were computed by entering
two control variables (education and tenure) in the first step
of each regression analysis. Next, the independent variable
(job satisfaction or care adequacy) and the moderator (PSS
Exploratory factor analysis
or POS) were introduced in step 2. The interaction term was
The discriminant validity of the constructs was tested introduced in step 3 (see Tables 2 and 3 for the results).
via an EFA with the maximum likelihood extraction Lastly, the significant interaction effects were represented
method. The results showed five factors with Eigenvalues by figures. Independent lines of regression were generated
> 1, explaining ~ 58% of the variance of the indicators. from the regression equation to represent the relation-
The Keiser-Meyer-Olkin measure of sampling adequacy was ship between the independent variable and the dependent
0.92 and the Bartlett’s test of sphericity was significant variable, defining the high values and the low values of the
(c2 [d.f. = 210] = 13,597.35, P < 0.001), indicating adequate moderator variable at relatively one SD above and below
intercorrelations among the items in the questionnaire and the mean. A JST was carried out to test the mediation effect.
the suitability of the factor analysis (Hair et al., 2006). The To assess the significance, three steps were carried out by
reliability coefficients of the measures were good (0.82–0.92), calculating: (i) the a path (care adequacy on job satisfaction);
revealing a strong internal consistency. (ii) the b path (job satisfaction on turnover intention),

© 2011 Blackwell Publishing Asia Pty Ltd.

188 M. Galletta et al.

Table 3. Hierarchical regression results for the moderating effect of perceived organizational support (POS) on the relationship between job
satisfaction and turnover intention

95% CI
Regression step b SE t-value Lower limit Upper limit

Step 1
Education 0.10*** 0.020 3.50 0.03 0.110
Tenure -0.26*** 0.003 -9.06 -0.04 -0.020
R2 = 0.09***
Step 2
Job satisfaction -0.19*** 0.032 -5.89 -0.26 -0.130
POS -0.01 0.033 -0.30 -0.07 0.050
DR2 = 0.04***
Step 3
Job satisfaction ¥ POS -0.05† 0.023 -1.93 -0.09 0.001
DR2 = 0.003†
F = 3.73
Total R2(adj) = 0.13

***P < 0.001. †P < 0.10. Standardized b-values from the final step of the regression analysis are presented. CI, confidence interval.

controlling for care adequacy effects in the equation; and (iii)

the mediated effect (the product of the a- and b-values [ab]).
If the ab product is significant (P = 0.05), mediation is
present. Furthermore, 95% confidence intervals were calcu-
lated in order to provide a range of estimates for the actual
mediated effect value.
Consistent with hypotheses 1a and 1b, care adequacy
and PSS were positively related to job satisfaction (b = 0.22,
P < 0.001 and b = 0.40, P < 0.001, respectively). When the
interaction term for care adequacy and PSS was entered in
step 3 of the first regression analysis, the positive effect of
care adequacy on job satisfaction was found to be moderated
by PSS. The interaction coefficient was significant (b = 0.09,
P < 0.001), with a significant increment in the explained vari-
ance (DR2 = 0.008, P < 0.001). In order to interpret the form
of interaction, the equation at the high and low levels of PSS Figure 2. Two-way interaction effect of care adequacy and per-
ceived supervisor support (PSS) on job satisfaction.
was plotted according to the procedure that was proposed by
Aiken and West (1991). The results showed that the form of
the interaction was in the expected direction. An increase in
the perceived care adequacy was significantly associated with
levels of POS were plotted. Figure 3 illustrates that the nurses
high job satisfaction; this relationship was attenuated by low
with high levels of job satisfaction showed low levels of turn-
PSS (Fig. 2). The nurses were more satisfied in response to
over intention and this relationship was stronger when the
higher levels of care adequacy when they had high PSS. Thus,
level of POS was high. Therefore, this study showed that the
hypothesis 2, about the moderating effect of PSS, also was
satisfied employees with a high perception of organizational
supported.
support had a lower desire to leave their unit. Thus, hypoth-
Relative to the second regression analysis, the results
esis 4 was supported.
indicated that job satisfaction was negatively related to turn-
Lastly, hypothesis 5, regarding the mediating effect of job
over intention (b = -0.19, P < 0.001), but POS showed an
satisfaction, also was supported. The coefficients of paths a
insignificant relationship with turnover intention (b = -0.02,
and b were statistically significant, as well as the ab indirect
not significant). Thus, hypothesis 3a was supported, but not
effect (Table 4).
hypothesis 3b. When the interaction term for job satisfaction
and POS was entered in step 3, it was marginally significant
(b = -0.05, P = 0.054). In fact, the job satisfaction ¥ POS inter-
DISCUSSION
action term explained a significant incremental portion of the
variance (DR2 = 0.003, P < 0.10). To examine the nature of the Driven by P–E Fit Theory, this study provides new insights
interaction, the regression lines for the relationship between into the constructs and their respective contribution to turn-
job satisfaction and turnover intention at the high and low over intention, the latter being considered historically as a

© 2011 Blackwell Publishing Asia Pty Ltd.

Nursing turnover intention 189

P–E relationship and to strengthen the relationship between

the analyzed predictors and variables. Although it is not pos-
sible to make causal inferences between the variables, this
research contributes to the identification of the relationship
between the individual and organizational variables that are
significantly related to the turnover intention process. In fact,
this study supports the findings of Aiken et al. (2001) that
care adequacy is related to job satisfaction and turnover
intention. Specifically, it was found that the nurses who per-
ceived the adequate care of patients and adequate resources
to provide for quality care in their own unit felt satisfied with
their activities and this satisfaction, in turn, was negatively
related to the desire to leave the unit. Moreover, it was found
that the role of the nurses’ supervisor significantly interacts
with care adequacy in relation to job satisfaction. The nurses
Figure 3. Two-way interaction effect of job satisfaction and per- who perceived high care adequacy and who had high PSS
ceived organizational support (POS) on turnover intention. were more satisfied with their work. Yet, organizational
support moderates the relationship between job satisfaction
and turnover intention: the higher the nurses’ job satisfac-
tion, the lower is their turnover intention, but the strength of
direct antecedent of the actual turnover (Waters & Roach, this relationship decreases when POS is low.
2006). Previous studies have shown that a supportive envi-
ronment is related to job satisfaction among nurses that, in
Practical implications
turn, is related to turnover intention (Cai & Zhou, 2009).
Nevertheless, there is no sufficient study on the aspects that The results of this research show the importance of develop-
are specifically related to work environments as impor- ing organizational management practices that are necessary
tant factors relating to the turnover process, such as care for success and general organizational efficacy. In order to
adequacy. The authors believe that this variable is important control turnover intention, organizations should implement
to consider, specifically in the nursing context that is charac- strategies that facilitate the satisfaction of employees’ needs
terized by different working units and different kinds of by promoting both high-quality nurse–supervisor and nurse–
activities, depending on the department. In several studies on organization relationships, activating significant experiences
magnet hospitals, care adequacy was one important aspect of autonomy and responsibility for employees, and feedback.
that was related to working quality, job satisfaction and It was found that organizational support is a key element in
nursing retention (Aiken et al., 2001). Relative to this aspect, nurses’ retention, but the relationship with supervisors is
it was found that care adequacy is related directly to job instrumental in job satisfaction. Ribelin (2003: 18) stated that
satisfaction. Moreover, this research adds to previous studies “nurses do not leave hospitals, they leave managers.” This
in relation to the importance of the moderating role of PSS means that the relationship with supervisors might be an
in the relationship between care adequacy and job satisfac- important factor in nurses’ decision to leave, as much as the
tion and of POS in the relationship between job satisfaction relationship with the unit. Yet, the perceived care adequacy
and turnover intention. Then, this study shows the different can be improved by focusing on the clinical competence of
contribution of these variables in the process related to turn- the staff, building cohesive teams, and promoting the staff’s
over intention. The supervisor role is important in order to freedom to alter delivery systems (Schmalenberg & Kramer,
promote the improvement processes of the unit, to take into 2009). Finally, enhancing job satisfaction through perceived
account nurses’ work-related needs, and to support the staff support and care adequacy can have important implications
members to perform better in their own work. A high or for the individual, the organization’s well-being, and the
low level of support varies the strength of the relationship patient’s health.
between care adequacy in the nurses’ own unit and job sat-
isfaction. Furthermore, it was found in this study that PSS is
Limitations of the study
an important determinant of job satisfaction, agreeing with
the result of McGilton et al. (2007). In contrast, POS concerns This study had a few limitations that might be addressed
the individual’s belief about the ability of the organization to in future research. First, in order to collect the data, self-
care for nurses’ well-being and to take into account their report measures were used without the integration of further
values. Thus, the individual’s perception about what the unit objective measures (absenteeism and actual turnover). This
does for its nurses represents an essential element of the aspect could raise doubts about the validity of the obtained
social exchange relationship between the workers and the data (Goffin & Gellatly, 2001). Furthermore, a truncated and
unit. In fact, although it was found that POS is not related unvalidated version of a variety of measures from different
directly to turnover intention, it does interact with nurses’ questionnaires was used. Despite the internal consistency of
job satisfaction in relation to the desire to leave their unit. the measures being good, it could be problematic to assess
Therefore, PSS and POS can help to define the quality of the the effective validity and reliability of the whole instrument.

© 2011 Blackwell Publishing Asia Pty Ltd.

190 M. Galletta et al.

Table 4. Mediation effect of job satisfaction on the relationship between care adequacy and turnover intention: results for the joint significance
effect

95% CI
Care adequacy effect Mediator effect on Lower limit of Upper limit of the
Mediator on the mediator (a) turnover intention (b) Indirect effect (ab) the mediated effect mediated effect

Job satisfaction 0.36* (0.03) -0.23* (0.03) -0.08* (0.01) -0.11 -0.06
Model summary
F R2(adj) P-value
43.58 0.07 < 0.001

*P < 0.001. CI, confidence interval.

Second, for this study, a sample of nurses from the north of Gagné and Deci (2005) stated that the extent to which orga-
Italy was used. Therefore, it is impossible to compare the nizations support their employees by satisfying their psycho-
measures of this study with data from other regions of Italy, logical needs could influence the employees’ behavior. They
which reduces the external validity of the research. In order documented that the need for job autonomy is associated
to obtain greater support for the model, it might be necessary with high motivational autonomy as it assigns responsibility
to replicate the study with nurses from different geographic and stimulates personal growth and it can determine higher
areas of Italy (e.g. nurses in central and southern areas). job satisfaction and commitment to the unit and lower turn-
Another limitation is that a longitudinal design was not used. over. In conclusion, the results of this study add new knowl-
As this research was cross-sectional in nature, one should be edge on more effective strategies to improve nurses’ quality
aware that it is not possible to explain the causal influence of work life and to reduce the likelihood of dysfunctional
(Mathieu & Taylor, 2006). Turnover is a result of dynamic turnover by creating work environments that support nursing
processes that need longitudinal-type studies to investigate practice and the actual demands of nurses.
its evolution across time. Future studies should consider such
methods and test the long-term effects of perceived support, ACKNOWLEDGMENTS
care adequacy, and job satisfaction over outcomes such as
actual turnover and absenteeism. The authors thank all the nurses who participated in this
research.

Future directions and conclusions

REFERENCES
This research highlights how PSS and POS can moderate
Abdelrazek F, Skytt B, Aly M, El-Sabour MA, Ibrahim N, Engström
differently the strength of the relationship between care M. Leadership and management skills of first-line managers of
adequacy and job satisfaction and between job satisfaction elderly care and their work environment. J. Nurs. Manag. 2010; 18:
and turnover intention, respectively; all the important ele- 736–745.
ments are related to a high P-E fit. Moreover, this study Aiken LH, Patrician P. Measuring organizational traits of hospitals:
shows how the individual’s evaluation of care adequacy is the Revised Nursing Work Index. Nurs. Res. 2000; 49: 146–153.
associated with higher job satisfaction that, in turn, is related Aiken LH, Clarke S, Sloane D et al. Nurses’ reports on hospital care
to low turnover intention. It also was found that job satisfac- in five countries. Health Aff. 2001; 20: 43–53.
tion has a central role. Future guidelines can contribute Aiken LH, Clarke SP, Sloane DM, Lake ET, Cheney T. Effects of
further to understanding the phenomenon and overcoming hospital care environment on patient mortality and nurse out-
comes. J. Nurs. Adm. 2008; 38: 223–229.
the limitations highlighted above. First, Eisenberger et al.
Aiken LS, West SG. Multiple Regression: Testing and Interpreting
(1986) proposed that employees who believe that they are Interactions. Newbury Park, CA: Sage, 1991.
supported by the organization tend to reciprocate with posi- Beecroft PC, Dorey F, Wenten M. Turnover intention in new gradu-
tive work attitudes and behaviors. This is because the per- ate nurses: a multivariate analysis. J. Adv. Nurs. 2008; 62: 41–52.
ceived support is linked to a stronger commitment to the Brislin RW. Translation and content analysis of oral and written
organization and feeling of obligation to reciprocate the materials. In: Triandis HC, Berry JW (eds). Handbook of Cross-
organization’s support by caring about the organization’s cultural Psychology. Boston: Allyn & Bacon, 1980; 389–444.
well-being (Rhoades & Eisenberger, 2002). Thus, future Cai C, Zhou Z. Structural empowerment, job satisfaction, and turn-
studies should examine the role of organizational commit- over intention of Chinese clinical nurses. Nurs. Health Sci. 2009; 11:
ment as an intervening factor, which can have a direct impact 397–403.
Cohen J, Cohen P. Applied Multiple Regression/Correlation Analysis
on turnover. Commitment is a dynamic process by nature;
for the Behavioral Sciences (2nd edn). Hillsdale, NJ: Erlbaum, 1983.
therefore, using a longitudinal method, including long-term Cortese CG. [First standardization of Organizational Satisfaction
cause–effect influences, could be useful to understand better Questionnaire (OSQ).] Risorsa Uomo 2001; 8: 331–350 (in Italian).
the existing relationship with turnover behaviors. Yet, it Destrebecq A, Terzoni S, Colosso C, Neri L, Brambilla G. Intention
would be interesting to expand studies on the examination of to leave nursing in a major Milan hospital: current situation and
the antecedents of perceived support and turnover intention. future perspectives. Med. Lav. 2009; 100: 109–119.

© 2011 Blackwell Publishing Asia Pty Ltd.

Nursing turnover intention 191

Eisenberger R, Huntington R, Hutchison S, Sowa D. Perceived Raikkonen O, Perala ML, Kahanpaa A. Staffing adequacy, supervi-
organizational support. J. Appl. Psychol. 1986; 71: 500–507. sory support and quality of care in long-term care settings: staff
Failla KR, Stichler JF. Manager and staff perceptions of the manag- perceptions. J. Adv. Nurs. 2007; 60: 615–626.
er’s leadership style. J. Nurs. Adm. 2008; 38: 480–487. Rhoades L, Eisenberger R. Perceived organizational support:
Gagné M, Deci EL. Self-determination theory and work motivation. a review of the literature. J. Appl. Psychol. 2002; 87: 698–714.
J. Organ. Behav. 2005; 26: 331–362. Ribelin PJ. Retention reflects leadership style. Nurs. Manage. 2003;
Goffin RD, Gellatly IR. A multi-rater assessment of organizational 34: 18–20.
commitment: are self-report measures biased? J. Organ. Behav. Schmalenberg C, Kramer M. Perception of adequacy of staffing. Crit.
2001; 22: 437–451. Care Nurse 2009; 29: 1–9.
Hair JJF, Black WC, Babin BJ, Anderson RE, Tatham RL. Schneider B. An interactionist perspective on organizational effec-
Multivariate Data Analysis (6th edn). Upper Saddle River, NJ: tiveness. In: Cameron KS, Whetten DA (eds). Organizational
Pearson-Prentice Hall, 2006. Effectiveness: A Comparison of Multiple Models. Orlando, FL:
Hannan S, Norman IJ, Redfern SJ. Care work and quality of care Academic Press, 1983; 27–54.
for older people: a review of the research literature. Rev. Clin. Schneider B, Smith DB, Goldstein HW. Attraction–selection–
Gerontol. 2001; 11: 189–203. attrition: toward a person–environment psychology of orga-
Hom PW, Griffeth RW, Sellaro CL. The validity of Mobley’s (1977) nizations. In: Walsh BW, Craik KH, Price RH (eds).
model of employee turnover. Organ. Behav. Hum. Perform. 1984; Person–Environment Psychology: New Directions and Perspectives
34: 141–174. (2nd edn). Mahwah, NJ: Erlbaum, 2000; 61–86.
Kottke JL, Sharafinski CE. Measuring perceived supervisory and Sourdif J. Predictors of nurses’ intent to stay at work in a university
organizational support. Educ. Psychol. Meas. 1988; 48: 1075–1079. health center. Nurs. Health Sci. 2004; 6: 59–68.
Kramer M, Schmalenberg CE. Magnet hospital staff nurses describe Stone PW, Larson EL, Mooney-Kane C, Smolowitz J, Lin SX, Dick
clinical autonomy. Nurs. Outlook 2002; 51: 13–19. AW. Organizational climate and intensive care unit nurses’ inten-
Lambert EG, Hogan NL, Barton SM. The impact of job satisfaction tion to leave. Crit. Care Med. 2006; 34: 1907–1912.
on turnover intent: a test of a structural measurement model using Takase M, Maude P, Manias E. Nurses’ job dissatisfaction and turn-
a national sample of workers. Soc. Sci. J. 2001; 38: 233–250. over intention: methodological myths and an alternative approach.
McGilton KS, Hall LM, Wodchis WP, Petroz U. Supervisory support, Nurs. Health Sci. 2005; 7: 209–217.
job stress, and job satisfaction among long-term care nursing staff. Takase M, Yamashita N, Oba K. Nurses’ leaving intentions: anteced-
J. Nurs. Adm. 2007; 37: 366–372. ents and mediating factors. J. Adv. Nurs. 2008; 62: 295–306.
MacKinnon D, Lockwood C, Hoffman J, West S, Sheets V. A Tekleab AG, Takeuchi R, Taylor MS. Extending the chain of rela-
comparison of methods to test mediation and other intervening tionships among organizational justice, social exchange, and
variable effects. Psychol. Methods 2002; 7: 83–104. employee reactions: The role of contract violations. Acad. Manage.
Maertz CP, Griffeth RW, Campbell NS, Allen DG. The effects of J. 2005; 48: 146–157.
perceived organizational support and perceived supervisor support Tourangeau AE, Cranley LA. Nursing intention to remain
on employee turnover. J. Organ. Behav. 2007; 28: 1059–1075. employed: understanding and strengthening determinants. J. Adv.
Mathieu JE, Taylor SR. Clarifying conditions and decision points for Nurs. 2006; 55: 497–509.
mediational type inferences in organizational behavior. J. Organ. Wagner CM. Predicting nursing turnover with catastrophe theory.
Behav. 2006; 27: 1031–1056. J. Adv. Nurs. 2010; 66: 2071–2084.
Melnyk BM, Fineout-Overholt E. Evidence-based Practice in Waters LK, Roach D. Job satisfaction, behavioral intention, and
Nursing and Healthcare: A Guide to Best Practice (2nd edn). New absenteeism as predictors of turnover. Pers. Psychol. 2006; 32:
York: Lippincott Williams & Wilkins, 2010. 393–397.

© 2011 Blackwell Publishing Asia Pty Ltd.