ORGANIZATIONAL COMMITMENT AND INTENTION TO LEAVE AMONG NURSES : THE MEDIATING ROLE OF MORAL OBLIGATION

In today’s fast-paced economic competition, committ ed and loyal employees are important for the profi t-oriented organizations to gain and sustain their competitiveness. Even for non-profi t-oriented organizations like hospitals and universities, these committ ed and loyal employees will lead to bett er quality service and improved organizational performance. Thus, it is vital for the organizations to sustain employees’ commitment because once the employees’ commitments decrease, it would be diffi cult for organizations to retain their employees. This study investigated the relationship between organizational commitment and intention to leave among nurses in Malaysian public hospitals, and determined the role of moral obligation as a mediator on the relationship between organizational commitment and intention to leave. The study focused on permanent nurses working in public hospitals in Peninsular Malaysia. The results supported that organizational commitment was signifi cantly and negatively related to intention to leave. The macro results also showed that moral obligation was a mediator in the relationships between organizational commitment and the intention to leave. The results were crucial to be looked into so that management and employers could have ample understanding and guidelines if they were to draft retention strategies. Even though many studies had been conducted on the intention to leave, most of them were conducted in developed countries and this study is believed to enhance the literature gap since it has an emphasis in the Malaysian context.


Introduction
There have been abundant studies on intention to leave.However, at a time when nursing shortage has become a global issue (Buchan & Calman, 2004) and a serious problem in the vast majority of postindustrialized countries (Gallett a, Portoghese & Batt istelli, 2011), the issue seems to be relevant and still needs special att ention.In many developing countries, the situation is worsening as many trained nurses hop over to the private sector or leave their own countries to serve in developed countries due to bett er salaries and benefi ts.The issue of nursing shortage in most parts of the world, having high turnover rate and losing a number of nurses to the private sector and foreign countries also suggests that, instead of looking at the workers' turnover itself, it is bett er for the organization to detect the feeling of intent to leave among their employees so that retention strategies could be drafted and implemented.The turnover rate itself could thus possibly be minimized.
The shortage of nurses has become a critical issue all over the world (Elsevier Foundation, 2010;Khaliq, Broyle & Mwachofi , 2009;Brush, 2008;Yearwood, 2007).The situation is worsening, especially in poorresource countries (Elsevier Foundation, 2010;Pillay, 2007;Yearwood, 2007) where by many trained nurses hop over to the private sector or leave their own countries to serve in foreign countries due to bett er off ers and benefi ts.To mitigate the problem of employees' turnover in the nursing fi eld, some researchers suggested that it was bett er for employers to examine nurses' intention to leave instead of studying turnover itself.
As revealed by literature, there are many reasons that contributed to workers' intention to leave.However, this paper would only focus on one of the most prominent variables -organizational commitment and its relationship with intention to leave (Kumar & Koh, 2011;Tatcher, Stepina & Boyle, 2003;Yamada, 2002).This study would also investigate the role of moral obligation as a mediating variable in the relationship between organizational commitment and intention to leave.Even though there have been many researches that relate organizational commitment and intention to leave, this paper off ers a new perspective for it discusses the relationship between organizational commitment and intention to leave specifi cally among nurses in public hospitals in the Malaysian context.

Overview
In studying turnover and shortage in certain professions, recent decades have witnessed an increasing interest in understanding why employees leave (Van der Heijden, Van Dam & Hasselhorn, 2007).However, knowing the reasons for leaving is too late to stop the actual leaving and detecting the presence of intention to leave feeling is said to be bett er in supporting the eff ort to prevent or minimize employees' leaving.In fact, voluntary turnover is said to be best predicted by the employees' intention to leave (Price, 2001).Thus, further understanding and investigation of the existence of the intention to leave feeling is crucial.In other words, it is much more important to scrutinize the factors that make them intend to leave their organization.
It is important to know the signals to turnover by studying the employees' intention to leave so that necessary preparations and preventive actions can be taken.Consequently, this study proceeds with the aim to examine the relationship between organizational commitment and the intention to leave among Malaysian public hospital nurses and at the same time, this study will also investigate the role played by moral obligation as a mediating variable.

Intention to Leave
Turnover and intention to leave are two diff erent concepts.Intention to leave involves an individual's perception towards leaving while turnover involves the act of an individual actually leaving the organization or profession (Hinshaw & Atwood, 1984).An employee is said to have the intention to leave when he or she has serious consideration to leave his or her current job (Castle, 2007).
Many studies on intention to leave have been conducted to examine the intention to leave and the factors relating to intention to leave in various fi elds.Many researchers have att empted to answer the questions of what really determines the employees' intention to leave by investigating possible antecedents of the employees' intention to leave.However, there is no standard reason why employees leave an organization or profession (Ongori, 2007) and there has been litt le consistency in the fi ndings to the question of what really determines the employees' intention to leave.

Organizational Commitment and Intention to Leave
Organizational commitment is said to be an important variable in the discussion of intention to leave since it is a popular belief that the more committ ed the employee is, the more likely he or she will stay loyal to the organization.In other words, it is less likely that he or she will leave for another job or organization.The statement is supported by the study of Slatt ery and Selvarajan (2005).In their study, they found that organizational commitment is negatively associated with turnover intention.
Results of some other studies also showed that more committ ed employees would be less likely to leave an organization (Nogueras, 2006;Simmons, 2005;Larkey et al., 1995).This is also supported by some local researchers (Ishak Ismail, 2006;Zulkifl i Kassim, 1997;Mohd Fikri , 1997) who found that the higher the degree of commitment among employees, the less likely they would express the intention to leave.In fact, organizational commitment is found to be critical (Khatri et al., 2001) and has a negative correlation with turnover intention (Rahman, Naqvi & Ramay, 2008).
The literature review shows that studies on organizational commitment have increased and are continuously gaining importance among researchers all over the world.This is because more and more organizations realize that human resources are their most valuable assets and having loyal and committ ed employees can ensure that they stay competitive and survive.In fact, some researchers (Colakoglu, Culha & Atay, 2010;Meyer & Allen, 2004) claimed that employees who have higher emotional att achment which is one of the measures for aff ective commitment to their organization, show bett er performance and contribute more meaningfully than those with lower emotional att achment.This kind of employees will also enhance the organizational competitiveness and guarantee the survival of the organizations (Ongori, 2007).In short, it is a valuable bonus for the company to have employees with a high level of organizational commitment.These highly-committ ed employees would feel that they owe so much to the organization and in return, they would give full support, eff ort and cooperation to the organization and would be less likely to leave.

Moral Obligation
There are limited references or past research that really discussed or focused on moral obligation as one of the variables in the study of intention to leave.Prestholdt et al. (1987), Sharkey (1994) and Lane, Mathews and Prestholdt (1990) have contributed to the understanding of the role that moral obligation plays in examining the intention to leave.However, the relevance of moral obligation in predicting intention depends very much on the type of behaviour, profession and the sample being studied (Sharkey, 1994).In nursing, moral obligation and work responsibilities are assumed to be the two dimensions that complement each other (Cronqvist et al, 2004).Yearwood (2007) also seemed to agree that moral obligation could provide a basis to tackle the problem of internal as well as external nurse migration.

Problem Statement
Proper staffi ng and a stable nursing workforce will ensure the quality of nursing care received by patients and the quality of the healthcare system.Thus, having a suffi cient number of nurses is important.However, many articles and journals have discussed the nursing shortage, which has become a global issue.The existing and predicted national and global shortage of nurses would surely bring negative impact on society's health-care needs.The shortage at the national and international levels is exacerbated when nurses, especially from the public sector, are lured by off ers from private hospitals or even foreign hospitals (Ministry of Health, 2008;Missourri State Nursing Board, 2008).

Research Objectives
The objectives of this study are: 1.
To investigate the relationship between organizational commitment and intention to leave among nurses in Malaysian public hospitals.2.
To determine the role of moral obligation as a mediator on the relationship between organizational commitment and intention to leave.

Methodology
This quantitative research involves the use of a questionnaire in collecting the primary data.Most of the questions used in the survey were based on established existing instruments with some alterations and modifi cations to the original questions made to suit the study due to cultural factors and practice.The questions used in this study applied the multi-item approach where a few questions were asked under one particular variable.The questions were grouped under four major sections.Section A measured respondents' level of organizational commitment, section B measured moral obligation, section C measured the respondents' intention to leave and Section D required the respondents to provide their demographic data.
The survey used was a self-administered survey and 700 sets of these self-administered questionnaires were sent to selected major hospitals in Peninsular Malaysia.Prior to that, the ethical approval for the study and its method was fi rst obtained.Since this study involved nurses working at public hospitals as respondents, permission and approval were obtained from the Malaysian Ministry of Health and a few other important related institutions such as the Malaysian Research Ethical Committ ee (MREC), the National Institute of Health (NIH) and the respective hospitals.

Population, Sample and Sample Size
Nurses who were working permanently in public hospitals were the population sample of this study.They were chosen as the sample since this group of workers was said to be more likely to opt for off ers from private health institutions (News Straits Times, 2007) or even accept off ers to serve overseas (Ministry of Health, 2004, Missouri State Nursing Board, 2008).
This study made use of the systematic proportionate stratifi ed random sampling, which involves dividing the population into homogeneous subgroups and then taking a simple random sample in each subgroup.The use of this stratifi ed random sampling was decided upon in order to reduce the potential bias as well as to ensure that the sample would refl ect the general population.Since in Malaysia nurses are categorized into fi ve main groups -matrons, sisters, staff nurses, community nurses and assistant nurses, using the proportionate stratifi ed random sampling was preferred so that the sample would be able to represent not only the overall population, but also the subgroups of the population.
As of March 2008, there were about 75,000 active nurses in Malaysia (Ministry of Health, 2008).According to Sekaran (2003) if the size of the population is 75,000, the sample size (S), should be 382.However, 700 sets of questionnaires were delivered to the selected hospitals, considering that the response rate would not be 100 per cent.

Measurements
The existing instruments with established reliability and validity were used to measure the variables involved in this study.To measure organizational commitment of the respondents, the revised scale of TCM (Three-Component Model Employee Commitment Survey) developed by Meyer, Allen and Smith (1993) was used.These scales measure the three types of organizational commitment of the respondents: aff ective commitment, continuance commitment and normative commitment.There were six statements for each scale (18 questions altogether) and employees indicated the strength of their agreement by selecting a number ranging from 1 (strongly disagree) to 7 (strongly agree).Aff ective commitment was measured by Questions 1 to 6.Under this aff ective commitment scale, three (3) questions were positively worded and three (3) questions were negatively worded.Continuance commitment was measured by Questions 7 to 12 and under this continuance commitment scale, all questions were positively worded.Normative commitment was measured by Questions 13 to 18.Under this normative commitment scale, fi ve (5) questions were positively worded and only one (1) was negatively worded.
To indicate whether an employee was committ ed or not towards their organization, the mean was used as the benchmark.Scores that are lower than the mean indicate low or no commitment and scores that are higher than the mean indicate good or higher commitment.
To measure moral obligation, two questions were adopted from Prestholdt et al. (1987) with two additional questions added.Two questions were added to this instrument because the original instrument only contained two items and its Cronbach's Alpha was only 0.45 which was below the minimum acceptable value.Thus, att empts were made to increase the number of items and at the same time increase the value of Cronbach's Alpha of the instrument.Those two additional items went through the proper process of item development and both of them fulfi lled both the construct and the content validity.The addition of the two questions increased Cronbach's Alpha to 0.51.Nunally (1967) and Smith and Glass (1987) claimed that 0.50 is still an acceptable value for a reliability test.All the questions used were rated on a seven-point scale (Likert Scale) ranging from 1 (strongly disagree) to 7 (strongly agree).Scores that are lower than the mean indicate low or no moral obligation feeling among the respondents and scores that are higher than the mean indicate good or higher feeling of moral obligation.
The Anticipated Turnover Scale (ATS) which was developed by Hinshaw and Atwood (1984) was used to measure the respondents' intention to leave.This self-report Anticipated Turnover Scales (ATS) instrument contained twelve items relating to the respondents' anticipated length of time to leave and their certainty of leaving their jobs.However, the word 'position' in the questions was changed to 'organization' since this study intended to measure nurses' intention to leave the organization, not their profession.
The internal consistency reliability was 0.84 for the scale.Furthermore, this ATS instrument has been used repeatedly and successfully (Miller, 2007) in the study of intention to leave.All the questions used were rated on a seven-point scale (Likert Scale) ranging from 1 (strongly disagree) to 7 (strongly agree).The score is the simple sum of all the items in the scale divided by the number of items in the total scale.To indicate whether an employee is experiencing the feeling of leaving his/her organization, the mean was used to determine the existence of the intention to leave of the employee.Respondents with scores above the mean would be considered as having the intention to leave while those whose scores fall below the mean would not be considered as having the intention to leave.

Data Analysis
For data analysis, the responses were keyed into SPSS and analyzed using descriptive statistics such as frequency distribution, mean and standard deviation to show trends in the data.Correlation analysis and hypothesis testing were then conducted by applying SPSS Pearson Correlation and SPSS macro which was provided by Preacher and Hayes (2004).

Validity and Reliability
Principal Components Analysis (PCA) factor loadings show whether items are actually measuring what they should measure.Items with larger factor loadings provide bett er measures of the underlying property than do items with smaller factor loadings.As suggested by Hair (1998), loadings are considered signifi cant if they are +0.50 or greater (Manning & Munro, 2004).The questions used in this study were considered valid since all the questions fell under their respective dimensions.As for reliability, if the variable has an internal consistency reliability (Cronbach's Alpha) of at least 0.60 (Nunally, 1978) or 0.50 (Smith & Glass, 1987) the measurement is suffi cient and reliable for research purposes.The Cronbach's Alpha for all dimensions in this study were all well above 0.50.

Findings and Discussion
436 questionnaires returned, out of the 700 distributed, which gave a 62.3% overall response rate.However, only 398 questionnaires were usable.The respondents' demographic characteristics are shown in Table 1.Table 2 depicts that the direct eff ect coeffi cient for organizational commitment on intention to leave which is about 0.137 (negative) and p value is 0.040.The fi gures (coeffi cient value was negative and p value < 0.05) clearly show that organizational commitment has a signifi cantly negative relationship with intention to leave.Thus, the results supported Hypothesis 1 (organizational commitment is negatively related to intention to leave).The result is also consistent with many previous results (eg : Rahman, Naqvi & Ramay, 2008;Labatmediene et al., 2007;Slatt ery & Selvarajan, 2005;Nogueras, 2006;Angle & Pery, 1981).Bline et al. (1991) also found a signifi cant relationship between organizational commitment and intention to leave.With those arguments and statistical fi ndings, it is confi rmed that organizational commitment has a signifi cant negative relationship with the intention to leave feeling among nurses who are working in public hospitals in this country.
The results for estimating the direct eff ect for all the three dimensions of organizational commitment (aff ective commitment, continuance commitment and normative commitment) found that only one dimension of organizational commitment has a signifi cantly negative relationship with intention to leave which is aff ective commitment.
Normative commitment also shows a negative relationship with intention to leave but is not signifi cant (coeffi cient = -0.0121,t = -0.249,p = 0.804).The continuance dimension shows a positive relationship with intention to leave, even though it is not signifi cant (t = 1.557, p > 0.05).The results supported hypothesis H1b; however they did not support H1a and H1c.
Table 3 depicts the results of estimating the indirect eff ect of organizational commitment (total/overall and dimensions of organizational commitment) on intention to leave through moral obligation as a mediator.The IJMS 19 (2), 31-46 (2012) table shows that the indirect eff ect coeffi cient for total/overall organizational commitment on intention to leave through moral obligation is 0.192 (negative).Testing the signifi cance indicates that the true indirect eff ect was estimated to lie between -0.289 and -0.124 with 95% confi dence interval.Because zero is not in the 95% confi dence intervals, it then is concluded that the indirect eff ect is signifi cant at p < 0.05.Thus, moral obligation is a mediator for the relationship between organizational commitment and intention to leave.The results of estimating the indirect eff ect of the three dimensions of organizational commitment, found that the relationship of two dimensions of organizational commitment -aff ective commitment and normative commitment -with intention to leave was mediated by moral obligation.Meanwhile, continuance commitment was found to be not mediated by moral obligation.Thus, the results supported H2a and H2c but did not support H2b.

Conclusion
The results of the analysis show that organizational commitment is signifi cantly and inversely related to intention to leave while the macro results show that moral obligation is only a partial mediator in the relationship between organizational commitment and intention to leave.The results of the study have not only fulfi lled the objectives but have also fi lled the research gap, especially with respect to moral obligation as a mediating variable in the study.In fact, the inclusion of moral obligation as a mediating variable in examining the relationship between organizational commitment and turnover intention is considered a contribution of this study.
The fi ndings of the study regarding organizational commitment mirrored the previous fi ndings in that it revealed a negative relationship with intention to leave.The fi ndings also suggest that the eff ects of moral obligation as a mediating variable in the relationship between organizational commitment and intention to leave among nurses who work in Malaysian public hospitals is a crucial consideration enabling management and employers att ain ample understanding and guidelines to draft the retention strategies.

Table 2
Results of Estimating the Direct Eff ect of Organizational Commitment on Intention to Leave