Organizational citizenship behavior and employee turnover intention an empirical study in pharmaceutical companies in ho chi minh city

Abstract The study investigates the influence of relations-oriented leadership behavior, perception of fairness and intrinsic motivation on organization citizenship behavior OCB and the

UNIVERSITY OF ECONOMICS HO CHI MINH CITY

International School of Business

HO CHI MINH CITY

MASTER OF BUSINESS (Honours)

Ho Chi Minh City – 2017

UNIVERSITY OF ECONOMICS HO CHI MINH CITY

International School of Business

HO CHI MINH CITY

ID: 22140053

MASTER OF BUSINESS (Honours)

SUPERVISOR: Dr TRAN HA MINH QUAN

Dr NGUYEN THI NGUYET QUE

Ho Chi Minh City – 2017

Acknowledgment

I would like to express my deep gratitude to my supervisor Dr Tran Ha Minh

Quan and Dr Nguyen Thi Nguỵet Que for their precious guidance, valuable suggestions

and advice as well as their immense knowledge to support me throughout the process

In addition, I would like to thank my family, my friends and my colleagues for

their constant encouragement and immediate helps without which the study would not

have been possible

Abstract

The study investigates the influence of relations-oriented leadership behavior, perception

of fairness and intrinsic motivation on organization citizenship behavior (OCB) and the

influence of OCB on employee turnover intention in the context of Vietnam

pharmaceutical industry The model was tested with a sample of 312 employees working

for pharmaceutical companies in Ho Chi Minh City The result reveals a good fit between

collected data and measurement scales which were introduced and developed in the

Western contexts The findings show that relations-oriented leadership behavior,

perception of fairness and intrinsic motivation has positive impacts on OCB, and OCB

has a negative impact on employee turnover intention The discovery highlights the

importance of documenting factors affecting OCB in reducing employee turnover

intention towards organization as well as draws the management attention to their

recruiting, training and long term development policies

Key words: Organizational Citizenship Behavior, turnover intention, relations-oriented

leadership behavior, perception of fairness, intrinsic motivation

Table of contents

Acknowledgement

Abstract

Table of Contents

1 Introduction 9

1.1 Research background 9

1.2 Research problem 10

1.3 Research questions and research objectives 12

2 Theoretical Background and Hypotheses 13

2.1 Organizational citizenship behavior 13

2.2 Employee turnover intention 14

2.3 Relations-oriented leadership behavior 14

2.4 Perception of fairness 16

2.5 Intrinsic motivation 17

2.6 OCB and employee turnover intention 19

3 Research methodology 22

3.1 Procedure and Sample 22

3.2 Measurement scales 23

3.3 Data analysis 25

4 Results 26

4.1 Descriptive analysis 26

4.2 Reliability test 27

4.5 Structural Equation Modeling (SEM) 38

4.6 Discussion 42

5 Conclusion, implications and limitation 43

5.1 Conclusion 43

5.2 Managerial Implications 44

5.3 Limitations and directions for future research 46

References 47

APPENDIX 67

Appendix 1- Questionnaire – English 67

Appendix 2 - Questionnaire – Vietnamese 73

Appendix 3- Description of study sample 78

Appendix 4- Reliability analysis - relations-oriented leadership behavior 81

Appendix 5- Reliability analysis - Perception of fairness 82

Appendix 6- Reliability analysis – Intrinsic motivation 83

Appendix 7- Reliability analysis – Organizational Citizenship Behavior 84

Appendix 8- Reliability analysis – Turnover intention 85

Appendix 9- Exploratory Factor Analysis (EFA) 86

Appendix 10 – Confirmatory Factor Analysis (CFA) 91

Appendix 11- Structural Equation Modeling (SEM) 97

List of figures

Figure 1 Conceptual Model 21 Figure 2 CFA measurement model 376 Figure 3 SEM measurement model 378

List of tables

Table 1 Source of data collection 26

Table 2 Overall reliability of the constructs and standardized loadings of indicators 29

Table 3 EFA results for measurement scales 35

Table 4 Factor colleration 35

Table 5 Fit indices for the test model result (CFA) 37

Table 6 Fit indices for the test model result (SEM) 39

Table 7 Standardized structural paths in the model 40

1 Introduction

Employee is considered as one of the most valuable assets and this becomes competitive

advantage of every organization in modern economy (Verma & Dewe, 2008) According

to Dobre (2012), human capital has a direct relationship to organizational performance in

terms of company’s profit and targets Other studies found that maintaining a stable labor

force is an optimal way to obtain effective cost management and enhance quality of

service in a firm (Jago & Deery, 2002) as well as achieve sustainable economic

development of a nation (Bousrih, 2013) However, keeping well trained and skilled

employees is getting more difficult for any organization (Wiggins, 1998) and it is

becoming an important challenge for the firms (Reda & Dyer, 2010) as employee will

switch to the next job if they are not satisfied with the current employer or the job and

employer will be left with no good employees in organization (Laddha, Singh, Gabbad &

Gidwani, 2012) Furthermore, employer has to suffer many losses such as investment in

recruitment, selection, training and promotion is wasted (Abrams, Ando & Hinkle, 1997),

negative influences on work flow, quantity and quality of production as well (Guthrie,

2000) if valued workers leave.So, keeping suitable and well-trained employees are

becoming the priority in every organization (Mitchell, Holtom & Lee, 2001)

There are many reasons why people leave organizations: family matter, unfair

treatment observation (Mitchell et al., 2001), salary level, job satisfaction (Lee, Huang &

Zhao, 2012), excess workloads, poor colleague relationships (Conklin & Desselle, 2007),

perceived organizational support, or employment external opportunities perception

(Hofaidhllaoui & Chhinzer, 2014) Another factor that can be taken into consideration is

discretionary work behavior, or Organizational Citizenship Behavior (OCB) OCB is

defined as employee’s willingness to contribute extra efforts for the achievement of

organizational outcomes, willingness to help coworkers, or willingness to obey extra

rules (Katz, as cited in Jahangir,Akbar & Haq, 2014) and it is not in formal reward

system of organization (Teh, Boerhannoeddin & Ismail, 2012; Konovsky & Pugh, 1994)

If employee has high OCB, their turnover intention is likely to reduce so OCB can be

considered as one of the important impacts on employee turnover intention

Organizational Citizenship Behavior (OCB) has been studied widely over many years by

many researchers Its outcomes are also stated including increased employees’

satisfaction (Aamir & Zafar, 2006), enhanced employees retention (Podsakoff &

Mackenzi, as cited inChahal & Mehta, 2011) which means reduced turnover (Aamir &

Zafar, 2006) Studies have shown that organizations with high OCB levels can have

positive outcomes such as lower levels of turnover and higher levels of efficiency

(Podsakoff, Whiting, Podsakoff & Blume, 2009) The benefits of OCB can be great, and

the antecedents of OCB haven been studied widely so that managers can provide an

appropriate environment to foster OCB’s in their employees (Lilly, 2015) Impacts of

OCB on employee turnover intention have been widely studied in terms of job

satisfaction (Bolon, 1997; Alotaibi, 2001; Lee et al., 2012; Hofaidhllaoui & Chhinzer,

2014), organizational commitment (Bolon, 1997; Alotaibi, 2001; Lee et al., 2012;

Conklin & Desselle, 2007), organizational support (Kamani & Namdari, 2012; Pohl,

Battistelli & Librecht, 2013) or leadership style (Jiao, Richards & Zhang, 2011; Suliman

& Obaidli, 2013; Lilly, 2015; Shah, Hamid, Memon & Mirani, 2016) but the main field

of investigation is in hotel (Lee et al., 2012), engineer (Hofaidhllaoui & Chhinzer, 2014)

telecom (Ibrahim & Aslinda, 2014), hospital (Pohl et al., 2013) or finance - banking industry (Jiao et al., 2011; Kamani & Namdari, 2012; Suliman & Obaidli, 2013; Shah et

al., 2016) in the U.S, Pakistan, Indonesia Whereas in Vietnam, turnover rate in

pharmaceutical companies is still increasing year by year and it continues to be in top

three out of 15 industries, even though salary is increased (Towers Watson Vietnam

,2014; Talentnet – Mercer Total Remuneration Survey, 2015) According to Towers

Watson Vietnam (2014), its Total Rewards Survey 2014 highlighted that the general

industry average staff turnover rate in Vietnam is at 12.7%, higher than the median of

11% By sector, the survey showed that the pharmaceutical industry is the highest not

only in terms of overall salary increases with 11.2%, but also in terms of staff turnover

rate at 15.9% Furthermore, Talentnet – Mercer Total Remuneration Survey (2015) made

a survey with participant list of 520 well-established Multinational Corporations and

leading local companies from 15 industries In this year, Pharmaceutical is still in top

three industries providing the highest salary increase rate in the market with 10.1% but

staff turnover rate is also in top three accounting for 16.7% It can be seen that employee

turnover rate in increasing and becoming big concern for organization, especially in

pharmaceutical industry where increased salary still cannot reduce high turnover rate

Employee turnover will inflict numerous costs such as losing valuable knowledge and

experience; affecting the normal workflow of the companies, the efficiency of business

and relationships with suppliers and customers (Mitchell et al., 2001) Consequently, how

to explore the OCB factors that can reduce employee turnover intention becomes the big

problem for management to face and deal with in order to run the business well

The result of the research is important to bring the awareness to the managers in

Pharmaceutical industry in Ho Chi Minh City about antecedents of OCB that affect

employee turnover intention Management can make appropriate policies for employees

to reduce their turnover intention and increase their working efficiency

- The research can be useful the reference or guidance for the others to conduct further

studies in related topics and other organizations regards to similar problems

- The research can be believed to add more value to the literature on OCB and employee

turnover intention, especially in Vietnam setting

According to the above mentioned research background and research problem, this

research aims to investigate the following questions:

RQ1: Identify the antecedents of Organizational Citizenship Behavior and how does each

antecedent affect Organizational Citizenship Behavior?

RQ2: How does Organizational Citizenship Behavior affect employees’ turnover

intention?

The main objectives of this research are to:

- Identify factors affecting Organizational Citizenship Behavior including

relations-oriented leadership behavior, perception of fairness, intrinsic motivation and the extent to

which each factor affects Organizational Citizenship Behavior

- Analyze the impacts of Organizational Citizenship Behavior on employees’ turnover

intention

2 Theoretical Background and Hypotheses

Organizational citizenship behavior (OCB) can be interpreted within the framework of

social exchange theory: employees are willing to perform extra role behaviors to satisfy

and reward relationships with specific individuals or groups in the workplace

(Cropanzano & Mitchell, 2005) According to Konovsky and Pugh (1994), OCB is

“employee behavior that is above and beyond the call of duty and therefore is

discretionary” (p 656) and OCB is not rewarded in organizational formal structure Lilly

(2015) defined OCB as positive employee behavior that is not required for the job in

order to help the organization or to protect the organization from arising problems

Successful organizations will be organizations that have employees willing to go

beyond their normal job responsibilities, give their time and energy to complete the

assigned job successfully, and contribute to organizational smooth operations As the

result, organizations cannot operate well without having good OCB behaved employees

and understanding OCB has been a high priority for organizational management

(Jahangir et al., 2014).

Turnover intention has been widely studied in academic community According to Cotton

and Tuttle (1986), turnover intention is an individual’s perceived probability of staying or

leaving a current organization while Sousa-Poza and Henneberger (2004) stated that

“Turnover intention reflects the (subjective) probability that an individual will change his

or her job within a certain time period” (p 113) and Tett and Meyer (1993) defined this

term as employee’s willingness to seek for other opportunities in other organizations.

Turnover intention has always been an important issue for management for many

years and it is a major problem for organizations until now (Chen, Lin & Lien, 2011)

Researchers also stated that organizations are giving increased attention to turnover

intention problem because they know that it will affect the cost and the performance of

organizations

According to Lilly (2015), leadership behavior is classified into two basic types: task

focused behavior and people behavior focused on people, also called relations-oriented

leadership behavior In general, fast work speed, good quality, high quantity, and close

supervision are focused on task focused behavior On the other hand, nurturing

relationships, less close supervision, stress reduction, trust and respect people in the

workplace are factors that will be concentrated in relations-oriented leadership behavior

Tucker and Russell (2004) argued that transformational leadership (similar to

relations-oriented leadership) adds to the quality of life in the employee and the

organization Jiao et al (2011) pointed out that under influence of relations-oriented

leaders; employees will recognize and accept the values of the organization and the

leader As the result, they do not only perform their job for obligation or reward, they

actually transform their attitudes, beliefs, and values (Bono & Judge, 2003; Rafferty &

Griffin, 2004).Krishnan (2012) stated that transformational leadership is positively

related to employees’ meaning of life and subjective well-being Due to impact of

relations-oriented leadership on personal and organizational outcomes, it is needed in all

organizations (Tucker & Russell, 2004)

Leadership always plays an important role in every business organization, it is

sought to identify the behaviors that can improve the performance of individual and

organization (Yukl, Gordon & Taber, 2002) Employees can be motivated or demotivated

according to their working and cooperation with their leaders Yan and Yan (2013)

pointed out that relations-oriented leaders tend to increase cooperation, teamwork and job

satisfaction Examples can be seen as looking out for welfare of employees, being willing

to accept employees’ suggestions (Yan & Yan, 2013), helping employees to perform the

task better or giving praise when they perform the tasks effectively (Yukl et al., 2002)

Relations oriented leadership behavior includes supporting, developing,

recognizing, consulting and empowering with individuals (Yukl et al., 2002) and it can be

seen to enhance team spirit and influence on the employee’s engagement in OCB (Chahal

& Mehta, 2011; Jahangir et al., 2014) Furthermore, many researchers have found that supportiveness can help employees cope with stress and reduce interpersonal conflicts

(Amason 1996; Jehn 1995; Whetten & Cameron 1991, as cited in Yan & Yan, 2013)

Netemeyer, wBoles, McKee & McMurrian (1997) also said that leadership support

behaviors can increase job satisfaction which effects positively on employee’s OCB

Podsakoff et al (as cited Yan & Yan, 2013) showed that supportive leadership perception

is positively associated with OCB All these suggest that employee’s OCB can be

enhanced by relations-oriented leadership behavior Thus, the following hypothesis is

developed:

H1: Relations-oriented leadership behavior has a positive effect on employee’s OCB

Fairness or justice perception is defined as whether or not employees feel organizational

decisions are made fairly and in the same way with the necessary employee input (usually

called procedural justice) and whether or not employees feel that they are fairly treated

rewarded (called distributive justice) (Jahangir et al., 2014)

Organ (as cited in Messer & White, 2006) argued that fairness perception is

important as employees who believe that they are fairly treated will hold positive

attitudes about their work and outcomes

Williams, Pitre and Zainuba (2002) stated that employees are always concerned

fairness treatment in the organization Buluc (2015) stated that one of the most significant

influences on employees is employees’ perception of fairness in organization

Employees’ fairness perception is defined as perception that they have been fairly treated

or not (Moorman, as cited in Messer & White, 2006) Concerning fairness perception

effects; Ertürk, Yilmaz, and Ceylan (2004) posited that employee perceptions of

managerial fairness will be a more important factor of OCB than other factors such as job

satisfaction or organizational commitment

Organ (as cited in Messer & White, 2006) believed that employees will have

positive working attitudes if they are fairly treated; he also argued that employees’

perception of fairness plays an important role in creating OCB Employees will increase

their OCB if they receive and perceive that they are fairly treated Furthermore, Williams

et al (2002) also proved that employee’s OCB intentions increase when their fairness

perception becomes positive Chahal and Mehta (2011) pointed out that fairness

perception may increase cohesive employees within organization and turn to increase

OCB Based on these studies, the following hypothesis is developed:

H2: Perceptions of fairness has a positive effect on employee’s OCB

Intrinsic motivation is defined in Cognitive evaluation/self-determination theory

(CET/SDT) as the willingness to pursue an activity based on one’s interest and personal

capacity to do so (Ryan & Deci, 2000) The researchers also pointed out that intrinsic

motivation refers to willing to do something because it is interesting or enjoyable

Teleonomic theory of self (TTS) describes intrinsic motivation as “deep

absorption and full involvement in an activity” (Keller &Bless, as cited in

Yogarabindranath, 2013)

Intrinsic motivation is defined as the willingness to perform an activity for the

interest or enjoyment of the activity itself, it does not mean to another reward (Isen &

Reeve, 2005) Researchers have shown that intrinsic motivation has a positive effect on

employees’ creativity (Ryan & Deci, 2000), job satisfaction and well-being (Ilardi,

Leone, Kasser, & Ryan, 1993), affective commitment (Eby, Freeman, Rush, & Lance,

1999) and job performance (Guo, Liao, Liao & Zhang, 2014)

Chahal and Mehta (2011) found that in order to enhance OCB, motives play an

important part The finding of Rodriguez (as cited in Ibrahim & Aslinda, 2014) showed

that intrinsic motivation relating with employees’ OCB is stronger than extrinsic

motivation He also confirmed that variable of intrinsic motivation has a great effect on

employee’s OCB According to Ying-Wen (2012), intrinsic work values enhance intrinsic

career success such as satisfaction with employee’s job and with his/her career Erdogan,

Kraimer and Liden (2014) pointed out that employees may remain committed and work

productively when they have intrinsic values They also tend to protect company property

and comply with informal rules that are created for work efficiency When employees see

their intrinsic values and career, they will have intrinsic motivation; they will

automatically engage in the job as they find interesting and enjoyable when performing

their job, they are also willing to help each other to complete their jobs efficiently When

intrinsic motivation is high, employees consider citizenship behaviors as free choice and

beneficial, they enjoy working and helping others (Gagné & Deci, 2005) Therefore,

intrinsic motivation may have a significant positive impact on OCBs Furthermore,

Lazauskaite-Zabielske, Urbanaviciute & Bagdziuniene (2015) stated that the more

intrinsic motivation the employees have, the more they enjoy their work and care about

benefits and good results, the more willing they will help and respect their colleagues as

well as follow company regulations and participate in company activities As the result,

the employees will be more engaged in OCB and be active members of the company

The following hypothesis is proposed according to the above mentioned studies:

H3: Intrinsic motivation has a positive effect on employee’s OCB

According to Priyadharshini and Mahadevan (2014), OCB is defined as the extent in

which employees feel they belong to the organization and OCB can enhance the ability to

retain employees in organization (Podsakoff & Mackenzie, 1997, as cited in

Priyadharshini & Mahadevan, 2014) Chen, Hui and Sego (1998) proved that lower OCB

levels relate to unwillingness to involve and to stay in the organization If employees

experience a positive working environment in the organization, they tend to be more

committed and motivated (Cohen, 2006), and these motivated and committed employees

are negatively related to turnover In addition, employees who are satisfied, committed,

and with organizational citizenship behavior are likely to stay in their current

organizations Paillé, Raineri and Valeau (2015) also proved that the more OCB

employees demonstrate, the less they want to leave their organization Other researchers

also argued that OCB is negatively related to employee turnover (Nathan, Steven, Philip

& Brian, 2009; Paillé, 2013; Yafang & Shih, 2010) Thus, the following hypothesis is

proposed:

H4: OCB has a negative effect on employee turnover intention

Figure 1 Conceptual Model

Employee turnover intention

H1 (+)

H2 (+)

H3 (+)

H4 (-)

3 Research methodology

The research was conducted in Ho Chi Minh City Two phases of study, a pilot study and

a main survey, were undertaken to collect data for testing the proposed model The

respondents were employees at many levels in various types of pharmaceutical

companies in Ho Chi Minh City

Although most of the measures of constructs in the model are available in the

literature, it is important to ensure they are appropriate for the Vietnamese context,

especially for pharmaceutical companies in Ho Chi Minh City In-depth interviews with 7

employees in pharmaceutical companies were conducted to modify the measures by

examining how they describe relations-oriented leadership behavior, perception of

fairness, intrinsic motivation, OCB, and turnover intention This study is very important

to make them appropriate for the pharmaceutical field context of this study

The structured questionnaire was created by Google Form The questionnaire was

sent to a convenience sample of 2,496 respondents The author was working in HR

Department of a pharmaceutical company (Altus Pharmaceutical Co., Ltd) while doing

this research There was an HR database of all candidates who applied for all positions of

Altus Co., Full names, phone numbers and emails of all candidates are stored in this

database There were 2,496 candidates in the database including people who were

currently working in pharmaceutical companies, just left from pharmaceutical companies,

just applied to pharmaceutical companies or were working in non-pharmaceutical

companies All of them could be appropriate respondents and the reliability of the

research can be high if large amount of them in 2,496 people sent their feedbacks That’s

why the author copied the link in the email and stated clearly in the email that the

respondents should help by sending feedback if they were currently working in

pharmaceutical companies In these 2,496 respondents, 312 of them were satisfied this

condition and sent their feedbacks The feedbacks then were used to assess the scale and

test the proposed hypotheses by means of Cronbach’s alpha and exploratory factor

analysis (EFA)

The purpose of this main study is to validate the measures and to test the

structured model The data collected from the pilot study was used to refine the measures

by means of Cronbach’s alpha reliability and exploratory factor analysis (EFA) After

that, confirmatory factor analysis (CFA) was used to test the measures validation and to

test the structural model by means of structural equation modeling (SEM)

Five constructs were examined: Relation-oriented leadership behavior, perception of

fairness, intrinsic motivation, organizational citizenship behavior, and employee turnover

intention Relation-oriented leadership behavior was measured by six items borrowed

from Avolio, Bass and Jung (1999), Suliman and Obaidli (2013),Hemsworth, Muterera,

and Baregheh (2013) and based on results of pilot study Perception of fairness was

measured by four items adapted from Moorman (1991), Niehoff1 and Moorman (1993),

Jafari and Bidarian (2012) Intrinsic motivation was measured by five items borrowed

and adapted from Rioux and Penner (2001), Finkelstein (2011) Turnover intention was

measured by three items, adapted from Lichtenstein, Alexander, McCarthy and Wells

(2004), Paillé and Grima (2011), Paillé et al (2015) Although Organ’s five-dimension

framework has been widely used in many previous research ((Moorman, 1991;

Podsakoff, Mackenzie, Moorman & Fetter, 1990), current researchers have not agreed

upon specific dimensions for OCB LePine, Erez and Johnson (2002) conducted a

meta-analysis to examine the relationship among these dimensions and suggested that

researchers should consider OCB at the aggregate level and distinguish OCB by the

target towards whom such behavior is directed Thus, this study used a two-factor model:

OCB directed at individuals (OCBI) and directed at organization (OCBO)

Organizational Citizenship Behavior Questionnaire (Lee & Allen, 2002) developed the

scale with 16 statements including eight items for OCBO and eight items for OCBI

McNeely and Meglino (1994), Williams and Anderson (1991) stated that using these two

subscales to distinguish may be problematic because they will be overlapped In these

studies, the items that are beneficial to individuals and to the organization were selected

to avoid possible overlap Eight items reflecting OCBI and OCBO were selected (see the

Appendix)

Five-point Likert-type scale ranged from 1 (strongly disagree) and 5 (strongly

agree) was used for all items in this study A questionnaire of the survey included 21

main questions, the remaining part contained questions related to gender, age, position,

education…to collect biodata The questionnaire was initially prepared in English and

then translated into Vietnamese This procedure was organized because English is not

well understood by all employees in Vietnam

Cronbach’s alpha reliability and Exploratory Factor Analysis (EFA) were used to refine

the measures of constructs based on the data collected in the quantitative pilot study (n =

312)

Cronbach’s alpha calculation was used to examine the inter-item consistency

reliability of global variables and their components After that, EFA with the "Principal

Axis Factoring" extraction method with the "Promax" rotation is used in factor analysis

Next, model proposed in Figure 1 will be evaluated withStructural Equation Modeling

(SEM) to test the general fit of the proposed model and to test the hypotheses After that,

all within-construct discriminant validity and convergent validity were examined if they

are satisfied for Confirmatory Factor Analysis (CFA) and Structure Equation Modeling

(SEM) Finally, CFA and SEM were examined if the measurement model received an

acceptable fit

As can be seen from Table 1, the majority of the survey respondents were males

(230) accounted for 73.7% of the samples, the number of participated females was 82

which accounted for 26.3% Besides, while more than a half of respondents aged under

25 dominated the survey with 55.5%, pharmaceutical staffs aged from 25 to 30 and aged

over 30 just occupied approximately 29.8% and 14.7% respectively Most of the

respondents were working in pharmaceutical industry from one to under three years

(43.9% and from three to under five years (41%), while just two respondents were

working under one year (0.6%) and 45 of them were working for more than five years

accounted for 14.4% Most of the respondents working in pharmaceutical sector had

college degree about 178 out of 312 which accounted for 57.1%, whereas this figure of

bachelor degree and master degree were approximately the same which were 22.1% and

20.8% respectively Finally, their monthly income were mostly fall in under 12 million

VND (76.9%), 10.3% of the respondents had income from 12 million VND to 15 million

VND, 7.1% of them earned from 16 million VND to 20 million VND and 5.8 % of them

could earn more than 21 million VND per month The respondents are from both genders,

from various departments with different positions and income; they are also from many

pharmaceutical companies So their responses can reflect general results of most

pharmaceutical companies

4.2 Reliability test

According to Hatcher (1994), reliability refers to the consistency of measurement, while

validity refers to the extentto which an instrumentreally measures what it claims to

measure Cronbach’s alpha, average variance extracted, composite reliability and

standardized loadings which demonstrated internal consistency reliabilities for the scales

are reported in Table 2

Hair et al (2009) stated that average variance extracted and composite reliability of

each construct should be greater than 0.5 Another factor that should be taken into

consideration is standardized loadings of all observed variables of each construct

(Schumacker & Lomax, 2006) and all of them should be greater than 0.6 (Nunnally &

Burnstein, 1994) As we can see in Table 2, all constructs in this study satisfied these

requirements

Construct Indicators Average

variance extracted

Cronbach’s Alpha

Composite reliability

Standardized loadings

To test the reliability of the sample data, Cronbach’s Alpha is popularly used as a

measure of the internal consistency of a test score for a sample of examinees The value

of Cronbach’s alpha accepted should be over 0.7 and any variables which the Item-total

correlations are smaller than 0.3 will be deleted (Nunnally & Peterson, 1994; Peterson,

1994)

Organizational Citizenship behavior (OCB):

From table 2 and Appendix 7, the value of Cronbach’s alpha of organizational citizenship

behavior was 0.893 which was higher than 0.7 The values of Item-total correlation of all

three items were greater than 0.3 This result showed that the measurement scales of

intrinsic motivation were well designed and trustworthy, and all three items from OCB1

to OCB8 should be kept in this research

Relation-oriented leadership behavior (ROLB):

From table 2 and Appendix 4, the value of Cronbach’s alpha of relation-oriented

leadership behavior was 0.892 which was higher than 0.7 The values of Item-total

correlation of all six items were greater than 0.3 This result showed that the

measurement scales of relation-oriented leadership behavior were well designed and

trustworthy, and all six items from ROLB1 to ROLB6 should be kept in this research

Perception of fairness (PCOF):

From table 2 and Appendix 5, the value of Cronbach’s alpha of perception of fairness

was 0.874 which was higher than 0.7 The values of Item-total correlation of all four

items were greater than 0.3 This result showed that the measurement scales of perception

of fairness were well designed and trustworthy, and all six items from POCF1 to POCF4

should be kept in this research

Intrinsic motivation (ITMT):

From table 2 and Appendix 6, the value of Cronbach’s alpha of intrinsic motivation was

0.849 which was higher than 0.7 The values of Item-total correlation of all five items

were greater than 0.3 This result showed that the measurement scales of intrinsic

motivation were well designed and trustworthy, and all six items from ITMT1 to ITMT5

should be kept in this research

Turnover intention (TOIT):

From table 2 and Appendix 8, the value of Cronbach’s alpha of turnover intention was

0.842 which was higher than 0.7 The values of Item-total correlation of all three items

were greater than 0.3 This result showed that the measurement scales of turnover

intention were well designed and trustworthy, and all three items from TOIT1 to TOIT3

should be kept in this research

The results showed that all Cronbach’s alphas of the scales were greater than 0.7,

and all corrected item - total correlations are greater than 0.3, satisfying the requirement

for internal consistency (Nunnally & Peterson, 1994; Peterson, 1994) A more detailed

Cronbach Alpha analysis performance is presented in appendix of the study The

measurement model achieved a good fit to the data

Exploratory Factor Analysis is a technique to evaluate interdependence It studies all

interrelationships without defining variables to be dependent or independent In this

research, exploratory factor analysis (EFA) is applied to test the model with KMO and

Barltlett’s combined with Promax rotation which are highly recommended for structural

equation modeling (SEM) EFA (Principal Axis Factoring with promax rotation)

extracted five factors (relations-oriented leadership behavior, perception of fairness,

intrinsic motivation, OCB and employee turnover intention), corresponding with the five

constructs used in the theoretical model

Measure for Factor Loading with sample size from 100 to less than 350 should be

equal or greater than 0.5 (Hair, Black, Babin & Anderson, 2009) All factor loadings

satisfied this requirement which means that EFA can be used in this study

According to Othman and Owen (2001), EFA only can be used when KMO

(Kaiser-Meyer-Olkin) is greater than 0.5 KMO in this study is 0.932 which means that

EFA can be used in this study

Bartlett’s test of Sphericity considers the hypothesis that the correlation between

the variables equals 0 in the population When this test was statistically significant (Sig ≤

0.05), the correlation between variables in the population was sufficient for conducting

EFA analysis (Hair, Black, Babin, & Anderson, 2009) In this study, Sig = 0.00 and

satisfied this requirement

Total variance extracted was greater than 50 percent (Nunnally & Peterson, 1994;

Peterson, 1994; Jabnoun & Al-Tamimi, 2003) In this study, total variance extracted was

58.780% which means it satisfied the requirement These EFA results indicated that all

the measures of constructs satisfied the requirements for the factors extracted, factor

loadings and total variance extracted As a consequence, these measures were used in the

main survey Consistent with the refinement of pilot study, EFA analysis (Principal Axis

Factoring with promax rotation) of main survey resulted in a similar assessment which is

shown in Appendices

In table 3, it can be seen that the loadings for all the constructs are above 0.5 In

addition, all composite scale reliabilities shown in Tables 2 exceed the recommended

cutoff of 0.7 and average variances extracted (AVE) by the scales exceeded the

recommended cutoff of 0.5 In table 3, the difference of maximum factor loading and

any other factor loading of each item equal or greater than 0.3, no item load highly (>

0.40 or 0.50) on more than one factor and eigenvalue of each factor must above 1 to be

accepted As shown in Table 3 and 4, the square root of the average variance extracted

(AVE) for the construct exceeds its correlation of that construct with other constructs All

standardized regression weights are greater than 0.5 and all regression weights were

significant with 95% confidence interval, factor correlations were significantly below

unity (p < 0.001) so all within-construct discriminant validity and convergent validity

were satisfied (Steenkamp & van Trijp, 1991; Anderson & Gerbing, 1988)

Table 3 EFA results for measurement scales

OCB LEADERSHIP MOTIVATION PERCEPTION TURNOVER

4.4 Confirmatory Factor Analysis (CFA)

Figure 2 CFA measurement model

All indices of the model are reliability for the next analyzing steps if GFI, TLI and

CFI are equal or above 0.9 (Bentler & Bonett, 1980); even though the values for GFI

does not exceed 0.9 (the threshold value), they still met the requirement suggested by

Baumgartner and Homburg (1995), and Doll, Xia, and Torkzadeh (1994): the value is

acceptable if above 0.8; Chi-square/df is equal or lower 2 (Bagozzi & Yi, 1988); and

RMSEA is equal or lower 0.08 (RMSEA ≤ 0.05 is excellent) (Steiger, 1990)

The CFA results indicated that the measurement model received an acceptable fit A

modification index was adopted to select indicator variables (Joreskog & Sorbom, as

cited inChen, Lin & Lien, 2011).Chi-square significant p = 0.000, Chi-square/df= 1.915,

GFI = 0.881, TLI = 0.937, CFI = 0.945, and RMSEA = 0.054 The findings indicated that

there was a satisfactory fit between the proposed model and the data

Model testing result

Result

Chi-square significant (CMIN)

P <0.05 P = 000 Fitted

Root Mean Square Error of Approximation (RMSEA)

Table 5 Fit indices for the test model result (CFA)

4.5 Structural Equation Modeling (SEM)

Figure 3 Structural Equation Modeling (SEM)

SEM was used to test the theoretical model and hypotheses The results indicate

that the model received an acceptable fit: Chi-square significant p = 0.000, Chi-square/df

=1.918, GFI = 0.879, TLI = 0.937, CFI = 0.944 and RMSEA = 0.054 The findings

indicated that the fit between the initial proposed model and the data was satisfactory

(Bagozzi & Yi, 1988) The structural coefficients were shown in Table 3

Model testing result

Result

Chi-square significant (CMIN)

P <0.05 P = 000 Fitted

Root Mean Square Error of Approximation (RMSEA)

Table 6 Fit indices for the test model result (SEM)

intention

Table 7 Standardized structural paths in the model