The mediating effect of the organizational socialization on leader-member relationship and job stress

This research aims to verify the mediating effect of organizational socialization on the relationship between leader-member relationship and job stress (physical and psychological health), targeting employees of the top 500 businesses in Taiwan ranked by Business Weekly Magazine in 2010. This research adopted the questionnaire survey method and a total of 303 valid questionnaires were returned. From the analysis, we found the following results: (1) organizational socialization has a mediating effect on the correlation between leader-member relationship and physical health, and (2) organizational socialization has a mediating effect on the correlation between leader-member relationship and psychological health. In conclusion, this research discussed the theoretical and practical connotations of the research results and proposed recommendations for follow-up studies.


INTRODUCTION
Along with the changes in the environment and advances in time, the patterns of work have also been changing.Such changes are causing more and more instability and uncertainty, and bringing the workers, heightened job distress.To relieve the stress on employees, some organizations place special emphasis on the leadermember relationship to help the employees successfully achieve organizational socialization, and stress incurring from work is reduced at the source of formation.
The statistics released by the Council of Labor Affairs (Taiwan) in 2007 indicated that the workers work an average of 2180 h per year, which comes to approximately 41.9 h per week (Chen, 2007).A survey conducted by the Taipei City Department of Health indicated that over 30% of workers gave expressions like "often" or "all the time" when asked if they feel serious stress at work (Na, 2008).These statistics give us a window of reference, from which we can infer the workers' fatigue index.When the fatigue index gets higher, the chances for the workers to make mistakes at *Corresponding author.E-mail: Jennifer1592001@seed.net.tw.Tel: 886-2-2557-3055.
work also gets higher and the low production value incurred from low efficiency is often accompanied by issues of stress.The Service Systems Technology Center, Industrial Technology Research Institute of Taiwan (ITRI) conducted the "Healthy Business Survey"; in which, as high as 83% of workers worry that job stress will make them sick.Among the factors that are considered to have the potential to affect the workers' health, overly long work hours or excessive workload was widely agreed (74.4%) as the number one cause.25% of the workers feel that they are "often" anxious about work and 72% feel that they "sometimes" feel anxious about work.This result shows that there is still room for the businesses to make improvements on the health of their workers.(Wang, 2007) When facing stress, it is necessary to find the causes first before the negative reactions from stress can be relieved.Organizational socialization is one of the factors that cause job stress.When a person enters a new workplace, everything, including job duties, work environment, and organizational culture is unfamiliar, and imperceptibly such unfamiliarity often brings uneasiness and anxiety, which in turn cause stress.Organizational socialization expedites the process for a person to merge into the organizational culture to reduce the stress likely to occur during the initiation period.Researchers found that the process of organizational socialization not only has a positive influence on the organization's members but also brings positive effects to the organization (Louis et al., 1983) and both short-and long-term influences/effects, such as job satisfaction, organizational commitment, and work performance, are generated (Wanberg and Kammeyer-Mueller, 2000).Therefore, based on the different cultures and needs, each organization adopts a different organizational socialization strategy that is most suitable to the organization, for example, the mentor system is one of the highly effective strategies.This strategy is a form of social support.The organization provides support through the managers and provides knowledge skills and the contents of the organization's socialization, such as internal goals, values, and regulations to the managers, so that they can help their employees to adapt to the organization and the job content.Many businesses now place high emphasis on the leader-member relationship, including Taiwan Semiconductor Corporation and Acer, and these businesses indeed have benefited from such strategies in terms of lowered employee turnover rate and minimized losses from employee errors (Ying-Qi, 2007).In addition, researchers have also found that the managers and the employees, as well as the organization, have benefited from organizational socialization strategies, as appeared in the increased job satisfaction rate, organizational commitment, and organizational citizenship behaviors (Allen et al., 2006;Chao, 1997;Donaldson et al., 2000;Dreher and Ash, 1990;Kram, 1985;Scandura, 1997).And leader-member relationship appears to be in a positive correlation with service oriented organizational citizenship behaviors (Weng et al., 2010).
Compared to the past researches, this research makes an attempt to achieve contributions in the following three aspects: first, this research attempts to construct a complete model for verification of whether organizational socialization has mediating effect to the relationship between leader-member relationship and job stress.Although pass studies have proved that leader-member relationship has negative correlation to job stress (Sosik and Godshalk, 2000), these studies have not explored how leader-member relationship affect job stress.Through literature review, this research proposes that organizational socialization is a potential mediating variable.Past researches have studied the correlations between leader-member relationship, organizational socialization, and job stress and found that leadermember relationship has a correlation with socialization, socialization has a correlation with job stress (Allen et al., 1999), and leader-member relationship has a negative correlation to job related stress (Sosik and Godshalk, 2000).However, an integrated model expressing the correlations among leader-member relationship, organizational socialization, and job stress has not been Lin et al. 10145 constructed in the past studies; therefore, the mediating effect of organizational socialization is not verified.Nonetheless, organizational socialization is an important process to employees upon entrance to an organization because socialization carries significant connotations to employee work behaviors and attitudes, (Reichers et al., 1994).How fast the employees situate into the organization affects employee organizational commitment, job satisfaction, turnover intention, and various in-roll/extraroll behaviors in the future (Lin et al., 2010;Lin and Lin, 2001).On a personal level, organizational socialization motivates the individuals to have positive attitudes towards their work or the organization, which is helpful to their personal development.Moreover, past studies have explored the strategies of organizational socialization or the cause/effect variables of organizational socialization, but very few have explored its role as a mediating variable.Therefore, this research configures "job stress" as the outcome variable and "organizational socialization" as the mediating variable to explore the correlations among "resources", "organizational socialization", and "job stress".
Secondly, this research explores the leader-member relationship from a different aspect to supplement the majority of the past studies that viewed the correlation from the positive aspect.Most of the past studies placed the focus on what positive effects can a positive leadermember relationship bring to the organization or the members of the organization, for example, job satisfaction, but very few have studied the subject from the negative aspect of how to effectively reduce the negative effects and influences through improvement in the leadermember relationship, for example, job stress.

Research aims and questions
In summary, this research aims to verify the mediating effect of organizational socialization on the correlation between leader-member relationship and job stress and propose practical strategies and recommendations based on the findings in this research.Based on stated research objectives, this thesis aims to study the following research questions: 1. What are the effects of leader-member relationship on organizational socialization? 2. What are the effects of organizational socialization on leader-member relationship and job stress?

The leader-member relationship
The leader-member relationship has multiple effects to the operations of an organization, for example, employees' job satisfaction, work motivation, work efficiency, and efficiency, and innovative ability (Adams et al., 1988;Albrecht and Hall, 1991;Stewart et al., 1986;Tjosvld, 1984).Fiedler et al. (1977) think that the leader-member relationship is the friendship, acceptance, and loyalty built up between people, as well as the level of their subordinates' confidence, trust, and respect in the leaders (Chen, 1989).Meanwhile, if the managers can win the support and trust from the subordinates, the managers' ability to influence will be much better than the unsupported managers (Fiedler et al., 1977).Graen and Cashman (1975) proposed that, in a leadermember relationship, subordinates play a detrimental part in the relationship, and Liao et al.'s (2008) research indicated that the quality of the leader-member relationship reveals the effectiveness of communication between the two parties, as well as a representation of how well the two parties work with one another.The behaviors conducted in the course of communication between the leader and the members affect the leadermember relationship.
In the Contingency Theory of Leadership proposed by Fieldler et al. (1977), the leader-member relationship is the most important factor among the control variables of a leadership situation.Although a good leader-member relationship does not necessarily mean high leadership effectiveness, among all situational control variables, the leader-member relationship has the most significant power to influence (Fiedler and Chemers, 1984).A manager's leadership and employees' work attitude have the power of influence to work performance that cannot be overlooked.In another word, if a company wishes to have high work performance and high job satisfaction, good leader-member relationship would be a prerequisite (Liao et al., 2008).Cobb (1976) advocated the social support theory.He thinks support of a society should include love, the feel of being loved and cared, dignity, a sense of value, mutual reciprocity, mutual understanding, sympathy and communication, and a connection, because these can relieve stress and minimize the chances of becoming ill.House (1981) pointed out that social support comes from several sources, including, partners, other related persons at work, friends, neighbors, supervisors, and coworkers.
Zheng-Sen and Shu-Yue (2005) proposed, from the perspective of organization, that the sources of social support include support from the organization, support from the supervisor, and support from the coworkers, in other words, the leader-member relationship.A leader is able to provide guidance, direction, support, training, education, advice, and feedback to the members and help the members to grow, advance in the career, progress in personal and professional development, expand career path, and improve work performance.

Organizational socialization
There are roughly three directions in the definition of organizational socialization: information contact (Applbaum et al., 1974), the process of adaptation to the new work (Van Maanen and Schein, 1979;Chao et al., 1994), and the process of knowledge sharing (Nonaka and Takeuchi, 1995).Many researchers also think that the process of organizational socialization is similar to the process of induction and orientation.This process enables the individuals to adapt to the new work and organization smoothly and, through which, the individuals gain the knowledge or skills needed to carry out the work.It is also a process of transition; in which, the key contents in the work and the organization's norms and values are conveyed to the members or employees (Chow, 2002;Kwesiga and Bell, 2004;Wu et al., 2008).This research defines organizational socialization as: the process a person takes to successfully become a member of an organization.In the process, the employee gains and learns the knowledge, skills, behaviors, and attitude to adapt to the new job, role, workgroup, and organizational culture, and this process reduces the feel of fuzziness faced by an employee after they join an organization and turns an outsider into an insider in an organization to enable them to effectively participate in the activities of the organization and adapt to the workplace (Alam, 2009).
Regarding the construct of "organizational socialization", Chao et al. (1994) proposed the six dimensions of socialization based on the concept of organizational socialization proposed by Schein (1971), Feldman (1981), andFisher (1986).Performance proficiency refers to expansion of the missions learned on the job.Personnel refer to the establishment of successful and satisfying work relationships with the organization's members.Politics refer to a person who successfully gained the information relating to an organization's internal power structure and the official/unofficial information relating to work relationships (Alam et al., 2010).Language refers to personal understanding of the professional technical language and a special language used in the organization.Organizational goal and value refer to learning of concrete organizational goal and value, including understanding of the organization's traditions, customs, stories, and behavioral standards.

Job stress
The concept of stress was first proposed by Hans Selyey in the Life Science Journal in 1936, borrowing from a concept in natural science.Derived from the Latin word "stringere", the word "stress" was used to refer to the feeling of being in a predicament, tense situation, adverse circumstance, or hardship in the 17 th century.In the 18 th and 19 th centuries, the word "stress" was used to refer to the strong effect manifested from coercion, pressure, tension, or things relating to a target or person.
The general concept of "stress" is often used by researchers in the field of psychology and social science (Pestonjee, 1998).
"Job stress" is often defined toward two directions, personal physical and psychological imbalance caused by the environment (Margolis and Kroes, 1974;Caplan et al., 1975) and the inability to coordinate with the works (Gui-Zhen and Mei-Xiang, 2003).This research defines "job stress" as the personal experience of discomfort incurred from personal inability to coordinate with the works or factors in the work environment that cause the reaction of physical/psychological imbalance, tension, uneasiness, anxiety, detest, and potential injuries.
This research uses the conservation of resource theory and the stress model to further illustrate the concept of job stress.First of all, the conservation of resource theory was proposed by Hobfoll (1989).Resource refers to the tangible resources deemed personally valuable and personal attributes/status/power, including advantages, dignity, resourcefulness in knowledge, social status, and occupation.The basic principle of the conservation of resource theory is: people work hard to maintain, protect, and build up resources, but when they face threats, they will potentially or actually lose these valuable resources.When an individual faces (1) the threat to lose resources, (2) actual loss of resources, (3) inability to gain returns from investment of resources, they will feel discomfort psychologically.
Secondly, the stress model proposed by Robbins (2001) includes three parts: the source of stress, the feel of stress, and the result of stress.The source of stress may include environmental, organizational, and personal factors.The result of stress may include physical, psychological, and behavioral symptoms.When the sources of stress from the external environment, organization, or personal issues emerge, the individual will feel stressed and physical, psychological, or behavioral symptoms will appear.Most of the studies done in the past expressed "job stress" from the aspects of physical health, psychological health, and behavioral symptoms (Dubrin and Hong, 1997;Steven et al., 2006).However, individual differences exist in behavioral symptoms, yet symptoms/reactions incurred from physical/psychological health apply universally to the majority of individuals; therefore, this research will explore "job stress" from the aspects of physical and psychological health.

Physical health
When facing stress, short-term physical changes and reactions, such as increased heart rate, blood pressure, blood sugar, and blood clotting (Karasek et al., 1981(Karasek et al., , 1988) ) headache, muscle aches, and back problems.If the issues of stress are not properly handled in time and the short-term physical symptoms persist, serious illness, such as heart problems, heart murmurs, heart pain, hypertension, increased cholesterol, ventricular hypertrophy (Schnall et al., 1990;Theorell et al., 1984), migraine, Lin et al. 10147 skin rashes, stomach pain, or even cancer, may develop.

Leader-member relationship and job stress
In a leader-member relationship, a member of the organization is led by a leader, who provides assistance and support to their work and psychological adaptation; therefore, job stress is less likely to happen under such situation.
The study of Allen et al. (1999) found that guidance provided by a leader helps a member to deal with stress.A person experiencing stress in an organization may benefit from good leader-member relationship (Kram and Hall, 1989).Seibert (1999) found that leader-member relationship is related to stress.According to the social support theory, members who are supported by their leaders, have clear directions for work, which helps the members to reduce the anxiety incurred from uncertainties in work and stay in a healthier physical and psychological state.Therefore, when a leader provides clear work instructions that gave the members the knowledge and skills to execute their assignments and has a clear idea of how much the job can be done, the members will be able to execute the assigned tasks with a clear mind.This will reduce the problems of stress caused by unclear job assignments and keep the individuals in a healthier physical state, and when physical health is optimized, psychological health will also develop towards positive terms.
Secondly, studies done in the past have also found that good leader-member relationship helps reduce jobrelated stress (Kram and Hall, 1989).When a person senses stress, psychosocio support reduces the stress one is experiencing and moderates the external source of stress (Manning et al., 1996).According to the social support theory, when a member receives emotional support and feels care and sympathy from a leader, the member gains comfort and encouragement that releases the tension and maintains physical and psychological health in an optimal state.Therefore, when a leader provides psychological support to, and establishes friendship with members, the members would be more willing to tell the leader problems they encounter at work and seek solutions from the leader.This effectively reduces the stress from work and keeps the individuals in a healthier physical state, and better physical health reflects more positive feelings.
Furthermore, according to the social learning theory, members learn from the behavioral presentation of the leader through observation to reduce occurrence of errors, and this learning process reduces stress to keep the members in a healthier physical and psychological state.A person who has the ability to be a leader must have something that is worth learning.Members learn the skills they need from observing the behaviors of the leader.This effectively reduces the stress from work and keeps the individuals in a healthier physical state, and better physical health reflects more positive feelings.

Organizational socialization has mediating effect on the co-relation between leader-member relationship and physical health
This research thinks the leader-member relationship is only effective in reducing job stress through the organizational socialization process.According to the social support theory, in the social environment, members gain the support and help they need from their leader, which helps them to gain job satisfaction and clear job goals and adapt to the demands of work and the organization.Studies of Chao et al. (1992) and Allen et al. (1999) found that leader-member relationship has significant correlation with socialization.In addition, according to the conservation of resource theory, through the process of organizational socialization, members gain the valuable resources owned by the organization internally or the members of the organization; through which, the members get to know the written or unwritten rules and present the behaviors expected by the organization to minimize occurrence of errors.Such process reduces stress and keeps the members in a healthier physical state.Other empirical studies relating to organizational socialization and job stress also pointed out that the higher the level of organizational socialization, the lower the level of stress would be (Haueter et al., 2003).Through the process of organizational socialization, new members gain the information they need, which reduces uncertainty and anxiety often occur at the initial stage of a new job (Ashforth and Saks, 1996;Jones, 1986) and reduces the occurrence of role conflict and ambiguity (King and Sethi, 1998).
Therefore, this research deduces that a leader who helps the members to learn the skills they need for work provides the members clear directions and information that helps the members to perform the content of socialization expected by the organization, so that they can successfully merge into the organization, and such process reduces job stress to keep the individuals in a healthier physical state.Based on the above, this research proposes the following hypothesis: H 1 : Organizational socialization has mediating effect on the co-relation between leader-member relationship and physical health.

Organizational socialization has mediating effect on the co-relation between leader-member relationship and psychological health
This research thinks the leader-member relationship is only effective in reducing job stress through the organizational socialization process.According to the social support theory, a leader who provides members emotional support, or encourages peer support, gives the members higher level of emotional support and care and often have better social interactive relationships with the members, which helps them merged into the organization quickly.The study of Allen et al. (1999) found that psychosocial support and politics have significant correlations to the factor of socialization in the dimension of performance proficiency.In addition, according to the conservation of resource theory, through the process of organizational socialization, members gain the valuable resources owned by the organization internally or the members of the organization; through which, the members get to know the written or unwritten rules and present the behaviors expected by the organization to minimize occurrence of errors.Such a process reduces stress and keeps members in a healthier psychological state.Other empirical studies relating to organizational socialization and job stress also pointed out that the higher the level of organizational socialization, the lower the level of stress would be (Haueter et al., 2003).Through the process of organizational socialization, new members gain the information they need, which reduces uncertainty and anxiety often occur at the initial stage of a new job (Ashforth and Saks, 1996;Jones, 1986) and reduces the occurrence of role conflict and ambiguity (King and Sethi, 1998).Socialization is related to stress incurred from work (Allen, 1999).
Therefore, this research deduces that a leader who provides members psychosocial support and establishes good interpersonal relationships with the members is capable of helping these members gain information relating to the organization and provides advice and consultation to enable them to understand the organization's rules and the content of socialization and follow the written or unwritten rules in the organization to become a member recognized by other members.This process effectively reduces the stress from work and keeps the individuals in a healthier physical state, and better physical health reflects more positive feelings.Based on the ongoing, this research proposes the following hypothesis:

Research subjects and sampling method
This research targeted on employees of the service industry and conducted sampling from the top 500 businesses ranked by Business Weekly Magazine in 2010.This research was conducted through the questionnaire survey method and subjects were selected through the convenient sampling method.The instructions given at the beginning of the questionnaire clearly stated that the respondents must be members led by a leader to ensure that all respondents are valid subjects for this research.Questionnaires were sent out through mail, totaling 500 copies and 323 questionnaires were returned, yielding a sample return rate of 64.6%.After deleting the incomplete samples, samples without answers for the reverse questions, and samples with the same answer to almost every question, which comes to a total of 20 copies, the valid samples come to a total of 303 copies with a valid return rate of 60.6%.
In the 303 valid samples, females take the majority with 158 persons (52.1%) and male, 145 persons (47.9%).The majority of the respondents are unmarried (174 persons, 57.5%); 122 respondents (40.2%) are married and seven respondents (2.3%) fall into the category of others.The majority of the respondents (172 persons, 56.6%) have a college level education or more, persons with a high school/occupational school level of education or less come to a total of 62 (20.6%), and 69 (22.8%) have a postgraduate education or more.The majority of the respondents fall into the age group of "over 20 and below 30" years (48.5%).Their average work years in the current organization fall in between the range of more than 2.5 and less than 5.5 years and the mean value of the total work years also fall in between the range of more than 2.5 and less than 5.5 years.

The reliability and validity test for the official questionnaire
This research first analyzed the reliability of each scale through analysis of the Cronbach's coefficient; the Cronbach's coefficient of leader-member relationship is 0.90.The Cronbach's coefficient for the organizational socialization scale is 0.92.Results of the job stress reliability analysis shows that the Cronbach's coefficient for physical and psychological health is 0.87 and 0.86, as shown in Table 1.Overall, the reliability level of each scale reaches the acceptable level of above 0.7.
To test the validity of the research concept, software AMOS 17.0 was used to conduct the confirmatory factor analysis after the questionnaires were returned and the purpose is to verify whether the latent variables were balanced.First, this research conducted confirmatory factor analysis on the variables, including leadermember relationship, organization of socialization, and job stress, to verify whether the overall pattern of each scale has goodness-offit.Results show that the overall patterns of reader-member relationship and job stress scales have acceptable goodness-of-fit and the overall pattern of the organization of socialization scale has poor goodness-of-fit.Therefore, this research will discuss the six dimensions of organizational socialization under a consolidated mode, as shown in Table 2.
Secondly, to confirm whether each dimension has a certain level of distinctiveness, this research constructed several measurements to measure competition and comparative models; the purpose is to ensure that the factor construct of the measurement models conform to the best data type models.Results are shown in Table 3. Table 3 shows that the base model of the four factors have better goodness-of-fit; among which 2 /df is 1.348; the GFI, NFI,CFI and AGFI values are 0.991, 0.987, 0.996 and 0.965 respectively.All of them have reached 0.90 -the acceptable level.The RMSEA vale is 0.035, which also reaches the acceptable standard (less than 0.08).The goodness-of-fit of each competition model is not as good as the four-factor model.This shows that each of the four dimensions-leader-member relationship, organizational socialization-, physical health, and psychological health has distinctiveness.

Treatment of common method variance
The data used in this research comes from the same source (members' viewpoints) and collected in the same time slot.Therefore, issues of common method variance (CMV) are likely to occur.This research takes reference from the management study of Tai-Guang, (2006) relating to common method variance to treat the problems through two strategies-preventive measures and post-hoc detection, as shown further.The four factors are leader-member relationship, organizational socialization, physical health, and psychological health; the three factors areleader-member relationship, organizational socialization, and job stress; the eight factors areleader-member relationship, performance proficiency, personnel, politics, language, organizational goals and values, history, and job stress.

Preventive measures
Targeting on the layout design of the questionnaire, the test is conducted through anonymous survey to reassure the respondents and minimize bias.In addition, the research objective and names of variables were concealed and the questions of different variables were arranged in random orders.Finally, the questions were designed in easily-understandable language, avoiding difficult technical management terms, asking two questions in the same sentence, and double negatives, and reverse questions were also designed in the questionnaire to minimize the chances of careless answers and potential consistency motivation.
In addition, this research added the "social approval scale" to test how much the respondents wish to conform to the expectations of social values; the purpose is to the CMV problems caused by social approval.This research used the social approval scale designed by Reynolds (1982).

Post-hoc test
This research used the three-factor method to test the CMV problems.Social approval is often seen as the source of CMV.Therefore, this research first tested the correlations between the dependant and independent variables before testing the correlations between independent variables and social approval/dependant variables and social approval.After the tests, the parts that have a high level of correlations to social approval are deleted for the purpose of eliminating the factors of social approval.The test results show that no questions are found to have high level of correlation with social approval (0.035 < r < 0.457); therefore, no items have been deleted.

Descriptive statistics
Table 4 is an analysis of the descriptive statistics.From Table 4, we can see the average values, standard errors, and relevant information of each variable.

Regression analysis
This research is an empirical test on the mediating effect of organizational socialization on the correlation between leader-member relationship and job stress (physical and psychological health).This research adopted the recommendation of Baron and Kenny (1986), which proposes that the regression model must comply with three major components during verification of the mediating variable: 1) independent variable and mediating variable both have significant relationships with the dependent variable; 2) a significant relationship exists between the independent variable and mediating variable; and 3) after placing in the meditating variable, the relationship between the independent variable and the dependent variable is weakened.Furthermore, full mediating effect refers to the phenomenon that adding of the mediating variable weakens the relationship between the original independent variable and dependent variable to the level of insignificance.Partial mediating effect refers to the phenomenon that the adding of the mediating variable weakens the relationship between the independent variable and dependent variable to a weaker or less significant level.First of all, this research verifies whether the regression model complies with the above 1 and 2; results of the analysis confirm their compliance.Following the verification, a two-stage linear regression model is constructed to verify the compliance with the three major components during verification of the mediating variable In the model used to verify the leader-member relationship and organizational socialization, as shown in Table 5, when physical health (job stress) is set as the dependent variable, a significant relationship exists between the original independent variable (leader-member relationship) and defendant variable (physical health) , but when a mediating variable is placed in, the relationship between the independent variable and  and dependent variable is weakened to a level of insignificance ( value drops from 0.323 to0.152, and P < 0.001 changes to P > 0.05).The mediating variable reaches the level of significance (P < 0.01 therefore, "organizational socialization" has full mediating effect to the correlation between leader-member relationship and physical health.When psychological health (job stress) is set as the dependent variable, a significant relationship exists between the original independent variable (leader-member relationship) and dependent variable (psychological health), but when a mediating variable is placed in, the relationship between the independent variable and dependent variable is weakened to a level of insignificance( value drops from 0.341 to 0.053 and P < 0.001 changes to P > 0.05).The mediating variable reaches the level of significance (P < 0.001); therefore, "organizational socialization" has full mediating effect to the correlation between leadermember relationship and psychological health.
From the regression analysis, we can see that organizational socialization has full mediating effect on the correlations between leader-member relationship and job stress shown in physical and psychological health; based on which, H 2 sustains.
From this result, we derive that through the process of organizational socialization, good leader-member relationship can effectively enhance members' physical and psychological health and lower job stress.This shows that organizational socialization is a very important process.A leader helps members to familiarize the skills they need for work perform the content of socialization expected by the organization, so that they can successfully merge into the organization and such process keep the healthier healthier physical state and reduce job stress.A leader provides members information relating to the organization and helps them to understand/follow the organization's rules the content of socialization, so that they can successfully merge into the organization, and such process keep the individuals in a healthier physical state and reduce job stress.
Compared to the previous studies, these tests on the mediating effects conform to the results of the study done by Nelson et al. (1991); that is, mental contract is effective in reducing stress for the new members in an organization.In other words, when a new employee (member) establishes a good leader-member relationship with a member of the organization (leader), both parties are able to understand and achieve the expectations set in between the leader and member.For example, a member expects the leader to helps them increase the depth of organizational socialization and the leader expects the member to have good performance after completion of the organizational socialization.This gives the leader the opportunity to be rewarded, and in turn, reduces job stress.

CONCLUSION
The majority of the past studies explored the correlations between the variables, such as leader-member relationship, coworkers' relationship, organizational socialization, job stress, job satisfaction, and organizational commitment (Liao et al., 2008;Lin et al., 2010;Lin and Chang, 2010;Lin et al., 2011).This research takes a further step to verify the mediating effect of organizational socialization on the correlation between leader-member relationship and job stress (physical and psychological health).We found the following results: organizational socialization has a mediating effect on the correlation between leader-member relationship and physical/ psychological health.
Compared to the past researches, this research makes an attempt to achieve contributions in the following three aspects: firstly, this research established an integrated model to verify whether organizational socialization has full mediating effect on the correlation between leadermember relationship and job stress.Although past studies have proved that leader-member relationship has negative correlation to job stress (Sosik and Godshalk, 2000), these studies have not explored how the leadermember relationship affect job stress.This research proposes that organizational socialization is an important mediating variable to supplement the gap in this research.Second, this research explores the leadermember relationship from a negative aspect (job stress) to supplement the majority of the past studies that viewed the correlation from the positive aspect, such as job satisfaction and organizational commitment (Liao et al., 2008;Lin et al., 2010Lin et al., , 2011)).Most of the past studies of leader-member, relationship placed focus on what positive effects can a positive leader-member relationship bring to the organization or members of the organization, but very few have studied the subject from the negative aspect of how to effectively reduce the negative effects and influences through improvement in the leadermember relationship.Finally, through verification of the mediating effect of organizational socialization, this research made recommendations for practical practice and suggestions to the organizations to place special emphasis on the factor of socialization in terms of building reinforced leader-member relationship.Academically, this research provides researchers who are interested in follow-up studies a basis for theoretic development.Results of this research serve as good references for further studies on the systems of leadermember relationship, organizational socialization, and job stress.

Connotations in theory
Results of this research verified the viewpoints on the social support and conservation of resource theories, as well as the stress model.First of all, the main viewpoint of the social support theory is that a person's stress is relieved, personal needs satisfied, and adaptation to the environment maximized through various sources of social support.Under such a viewpoint, the leader plays the role as the source of social support.This research found that timely support given by a leader helps members to achieve higher level of organizational socialization and relieve stress.Therefore, this research verified the viewpoints and the significance of the social support theory.
The basic principle of the conservation of resource theory is: people work hard to maintain, protect, and build up resources, but when they face threats, they will potentially or actually lose these valuable resources.Therefore, maintaining and protecting the resources becomes a significant issue.In the leader-member relationship, the leader is one of the significant figures who help members to maintain, protect, and build up resources.The leader has more resources to reduce the chances of the members from being threatened, that is, enabling members to maintain, conserve, and even build up their personal valuable resources.This research uses the conservation of resources theory to illustrate the leader-member relationship and the importance for the members to conserve and build up resources.Through the help of a leader who owns a range of resources to establish and increase resources for the members, both the individuals and the organization are benefited.
Furthermore, the concept of the stress model was built upon the phenomenon that when the source of stress emerges from external, organizational or personal factors, an individual will feel the stress and respond with physical, psychological, or behavioral symptoms.This research tested the effect of the stress source from the organization to physical and psychological health and found that the level of the members' organizational socialization has such an effect on job stress.That is, when the level of organizational socialization is higher, the physical and mental health of members is better.Therefore, organizational socialization can be seen as one of variables in the individual differences of the stress model.

Connotations in practice
This research proposed several recommendations for practical applications.First, this research found that when a leader provides more resources, their members have higher level of organizational socialization, and when their members have higher level of organizational socialization, their physical and mental health will improve and job stress lower.Therefore, how to motivate leaders to provide various resources to members is a very important issue.It is inevitable that people have private agendas, especially in a highly competitive job market.Sometimes, leaders worry that members would outperform them if they teach their members too much, so very often leaders are not willing to teach their members everything they know.Therefore, it is a good way to avoid conflict of interest and competition between the leaders and the members if the members' performance is also factored into the evaluation of the leaders.It is also viable to motivate the leaders through incentive systems; and various incentives are available to the leaders, leaders would be more willing to pass knowledge to the members without reservation.
Secondly, this research also found that organizational socialization is a very important process, since it connects the factors of leader-member relationship and job stress.Employees are important assets of an organization.Nowadays, organizations have begun to attach high importance to the value of human resources, but this does not mean that persons with good knowledge, skills, and abilities are necessarily of high economic value to an organization.The more important thing is whether the member knows the organization's goals, works towards the goals, uses his/her abilities to achieve the goals, and brings benefits to the organization.Therefore, the organization can utilize some socialization strategies, such as reinforcing the leader-member relationship and appoint experienced organizational members to lead new members, to maximize the level of employees' organizational socialization and keep employees in the best health state, so that they can perform better to bring higher production value and capacity; in such a way, the organization also benefits.Adapt the teamwork mode to enable the organization's members to jointly complete the task; under such mode, Lin et al. 10153 members exchange the content of socialization.Instead of singling out a member for individual education and training, place the new member into a team to enable them to learn and situate into the organization quickly and achieve organizational socialization.Or establish work logs and instructions, so when a new member takes over the tasks, they get a rough idea of the organization through the documents to shorten the time of organizational socialization.If an organization can effectively utilize various socialization strategies to upgrade the level of the employees' organizational socialization and ensures that new members are helped by senior members to understand the content of organizational socialization, they will be able to situate into the organization quickly and perform the behavior is expected by the organization.In such scenario, job stress will in turn be minimized.

LIMITATIONS AND RECOMMENDATIONS
Further, we will introduce the indications and relevant recommendations in this research.First of all, this research derived its samples from the employees of the top 500 businesses in Taiwan ranked by Business Weekly Magazine without differentiating the positions of the employees; therefore, we do not know whether variances exist in the ranks of the positions.In addition, the tested samples came from a single source (viewpoints of the employees).Although this research has implemented the preventive (questionnaire layout design) and post-hoc detection (the third factor test) measures to minimize the problem of common methods variance, we still suggest the researchers who are interested in followup studies to further study the subject through the leadermember pairing method.Furthermore, representativeness bias may exist in the samples of this research due to the limitations on data collection.Even so, results of this research still provide an overview on the overall status of job stress and verified that establishing good leader-member relationship is an effective way to promote organizational socialization and help the employees to reduce stress.
Secondly, this research is designed to study the crosssection of the issues; therefore, the cause-effect relationships between variables are not completed confirmed.In addition, job stress is measured by physical and psychological health in this research, yet the effects on mental health does not show instantaneously (it appears only after a period of time); therefore, we suggest researchers who are interested in follow-up study to explore the vertical section of the issues, in order to confirm the cause-effect relationships between the variable and monitor the dynamic changes in psychological health.Moreover, since job stress is measured by physical and psychological health in this research, the possibility of inter-influence between the factors of physical and psychological health.Problems in physical health, such as pains and aches, may give a pleasant or unpleasant feeling, and comparatively emotional fluctuation may cause health problems.Therefore, we recommend that follow up studies may take the inter-effects of physical and psychological factors into consideration or measure job stress in a different mode.In addition, an economic slowdown was ongoing during the time of the survey and many employees were being laid off.Such job stress cannot be resolved by any single system or measure.This shows that environmental factors may also affect the outcome of a research.Therefore, we recommend that this model can be retested under a different time to further confirm its validity.
Thirdly, extending from the previous leader-member pairing concept, this research thinks that further exploration may be made to the genders of the leaders and members for exploration of whether a leader and member pair of the same sex interact better or in the contrary or whether the differences in the level of education between a pair of leader and member affect the leader-member relationship.The reinforced subjects will enrich the content of future studies on the issues of leader-member relationship, job stress, and organizational socialization.
In all, the leader-member relationship is a form of social support.Members gain the knowledge and support needed from leaders to enhance their abilities to adapt to the environment; and organizational socialization is a key process.Through internal assimilation and adaptation, members established goals consistent to that of the organization.Therefore, a good leader-member relationship benefits both the organization and its members.Through the results of this empirical study, this research aims to provide valuable references to the industries.It is also our goal to advocate the importance of organizational socialization to the organizations, so that the organizations will act to reinforce the leadermember relationship and minimize job stress for workers, as well as lay down a foundation for follow-up studies, so that further exploration may be done on the relevant issues in the future.

Table 2 .
Leader-member relationship, organizational socialization, and job stress variable goodness-of-fit index.

Table 3 .
Comparative analysis on the competition of measurement models.

Table 5 .
The mediating effect of organizational socialization on the correlation between leader-member relationship and job stress.