In this chapter the focus will be on the research methodology. It is composed of five sections: research framework, research procedure, data collection method, instrumentation and data analysis method.
Research Framework
Model was developed by Cheng-Ping Shih and Tatjana Tica (fugure 3.1.). TOKSIP model served as the research framework for this study (see figure 3.1.). In Appendix will be presented additional model, TOKSIP II (see Appendix B).
Figure 3.1. TOKSIP model, developed by Cheng-Ping Shih and Tatjana Tica
Research Hypothesis
The following null- hypothesis are based on the research questions, literature review and research framework:
H1: Transformational leadership has no effect on organizational trust.
H2: Organizational trust has no effect on customer relationship management (CRM).
H3: Customer relationship management has no effect on innovation capabilities.
H4: Innovation capabilities have no effect on organizational performance.
24
Research Procedure
The researcher reviewed some literature on customer relationship management and innovation capabilities. Furthermore, there are some articles that emphasize the importance of transformational leadership, organizational trust and business performance. The greatest advantage of this research is that it combines all the variables together and investigates the importance of transformational leadership on organizational trust, customer relationship management, innovation capabilities and business performance in the theoretical and practical field. Moreover, the focus of this research is the ICT industry in Bosnia & Herzegovina and Taiwan. This study will provide a more comprehensive perspective for innovation to improve the HR field in the ICT industry. The research study was conducted following the processes mentioned below (see figure 3.2.).
25 Figure 3.2. Research procedure
Conclusion and Suggestions Data Analysis
Data Coding Data Collection Review of Instrument
Conduct Pilot Study Proposal Meeting
Translation and Expert Review of Instrument Develop Instrument
Develop Research Method Develop Theoretical Framework Identify Research Questions and Hypothesis
Identify Problems Review of Literature Research Motivation
26
Research Method
The purpose of this study is to investigate the relationship and the effects of transformational leadership on organizational trust, customer relationship management, innovation capabilities and business performance. The data in this research were collected using a survey. Surveys represent the most used data-gathering technique in social science.
Therefore, quantitative measurement was implemented. The quantitative measurement was chosen because the researcher wants to gather numerical data and draw a conclusion, if there is a strong relationship between variables. In quantitative studies the data are standardized which mean it uses numbers.
Sample
The target population for this study consists of full-time employees who are working in ICT industries in Bosnia & Herzegovina and Taiwan. The pilot test (paper questionnaire) was done on 33 employees working in ICT Companies. After gathering and analyzing data for pilot study the questionnaire was used to collect data for the main study.
In order to answer the research questions of this study, data were collected first from a small group of employees who participated in a pilot study, and second from a larger sample of the target population in the main study.
Regarding the main study, there were 225 questionnaires collected from employees in both Taiwan and Bosnia & Herzegovina ICT companies. 123 questionnaires were collected from ICT companies in Bosnia & Herzegovina and 102 questionnaires from ICT companies in Taiwan. The participants who were employed in ICT companies in Taiwan filled in internet based questionnaires. On the other hand, some employees in Bosnia & Herzegovina filled in an internet based questionnaire and also a paper based questionnaires. Questionnaires were later scanned by HR working in the ICT company and sent via e-mail to the researcher.
Instrumentation
The questionnaire used for this study consists of 5 variables and 80 questions. The instrument was first tested on pilot study. After the results have been collected and statistically analyzed the validity of the instrument was proved. The questionnaire was divided in 6 parts. They were: part I-Organizational Trust; part II-Customer Relationship Management; part III-Business Performance; part IInnovation capabilities; part V-Transformational Leadership and part VI-Demographics.
27
The instructions were provided in the beginning of the questionnaire. The respondents had to choose on a scale from 1 (strongly disagree) to 5 (strongly agree). The 5 point Likert scale
“provides an ordinal-level measure of person’s attitude”. In the last part (Demographics) respondents were asked to choose one of different available options.
Part I, Organizational Trust (29 items) was measured by Organizational Trust Index (OTI) developed by Pamela Shockley-Zalaback, Donald Morley, Ruggero Cesaria and Kathleen Ellis. The measurement is composed of five dimensions: competence, openness/honesty, concern for employees, reliability and identification.
Part II-Customer Relationship Management (8 items): Adopted from Öztaysi et al.
(2011), the variables of customer relationship management (CRM) consist of win - back management (WbM), production/ service customization (P/SC), customer information management (CIM).
Part III-Business Performance (10 items): Adapted from the empirical research of Lee and Choi (2003), the variables of market leadership (Mark) and financial performance (FinP).
The validity and reliability of the items, as established in the authors’ original work, are suitable for this research.
Part IV-Innovation capabilities (9 items): Adopted from Liao et al. (2007), the variables of innovation capabilities consist of product innovation (ProdI) and process innovation (ProcI).
Part V-Transformational Leadership (20 items): Adopted from Multifactor Leadership Questionnaire (MLQ), the variables of intellectual consideration (TLIC), intellectual stimulation (TLIS), inspirational motivation (TLIM) and idealized influence (TLIIA,TLIIB).
Pilot test was conducted to ensure the validity of each item, before gathering data for the main study.
Construct Coding and Scales
The following tables show the items that measure variables of the research study. For every construct will be provided specific code and a questionnaire item. Every item has a code that will be used in the statistical analysis of data.
The last part of the questionnaire, part 6 represents demographic questions. The last four questions are concerned about gender, age, highest degree of education and working experience.
28 Table 3.1.
Items Measuring Organizational Trust
Construct Code Questionnaire Items
Competence (Com; 4 items)
Com1 I am highly satisfied with our company’s overall efficiency of operation.
Com2 I am highly satisfied with the overall quality of the financial products and services of our company.
Com3 I am highly satisfied with the capacity of our company to achieve its objectives.
Com4 I am highly satisfied with the capability of our company’s employees.
Construct Code Questionnaire Items
Openness/Honesty
Op/h3 I have a say in the decisions that affect my job.
Op/h4 My immediate supervisor keeps confidences.
Op/h5 I receive adequate information regarding how well I am doing in my job.
Op/h6 I receive adequate information regarding how I am being evaluated.
Op/h7 I receive adequate information regarding how my job-related problems are handled.
Op/h8 I receive adequate information regarding how organizational decisions that affect my job are made.
Op/h9 I receive adequate information regarding the long-term strategies of our company.
(continued)
29 Table 3.1. (continued)
Construct Code Questionnaire Items
Concern for employees (Conc; 7 items)
Conc1 My immediate supervisor listens to me.
Conc2 Top management is sincere in their efforts to communicate with their employees.
Conc3 Top management listens to their employees’
concerns.
Conc4 My immediate supervisor is concerned about my personal well-being.
Conc5 Top management is concerned about their employee’s well-being.
Conc6 My immediate supervisor is sincere in his/her efforts to communicate with team members.
Conc7 My immediate supervisor speaks positively about subordinates in front of others.
Construct Code Questionnaire Items
Reliability (Reli; 4 items)
Reli1 My immediate supervisor follows through with what he/she says.
Reli2 My immediate supervisor behaves in a consistent manner form day to day.
Reli3 Top management keeps their commitments to their employees.
Reli4 My immediate supervisor keeps his/her commitments to team members.
Construct Code Questionnaire Items
Identification (Iden; 5 items)
Iden1 I feel connected to my peers.
Iden2 I feel connected to my company.
Iden3 I feel connected to my immediate supervisor.
Iden4 My values are similar to the values of my peers.
Iden5 My values are similar to the values of my immediate supervisor.
30 Table 3.2.
Items Measuring Customer Relationship Management (CRM)
Construct Code Questionnaire Items
Win-back management (WbM; 3 items)
WbM1 Our department has processes (or tools) to win-back valued lost customer.
WbM2 Our department has processes (or tools) to determine the value of lost customers.
WbM3 Our department has processes (or tools) to evaluate the cost of wining back the lost customers.
Construct Code Questionnaire Items
Production/ Service customization (P/SC; 3 items)
P/SC1 Our department has processes (or tools) to differentiate the customer acquiring efforts based on their value.
P/SC2 Our department has processes (or tools) to provide customized product/service to customers based on their value.
P/SC3 Our department has processes (or tools) to manage expectations of high valued customers.
Construct Code Questionnaire Items
Customer information management (CIM; 2 items)
CIM1 Our department has processes (or tools) to get in connection with potential customers using various channels (e.g., e-mail, customer service center, phone, FAX, face-to-face, etc.).
CIM2 Our department has processes (or tools) to trace the status of our relationship with customers.
Table 3.3.
Items Measuring Business Performance
Construct Code Questionnaire Items
Market Leadership (ML; 5 items)
ML1 Market share is better than that of others in the financial industry.
(continued)
31 Table 3.3. (continued)
Construct Code Questionnaire Items
ML2 Future prospect is better than that of other financial institutions.
ML3 Industry market leadership is better than that of other financial institutions.
ML4 Overall financial product and service innovation
is better than that of other financial institutions.
ML5 Average sales growth has improved over the last three years.
Construct Code Questionnaire Items
Financial Performance (FP; 5 items)
FP1 After tax return on sales is better than that of other financial institutions.
FP2 Profit growth over the years is better than that of other companies.
FP3 After tax return on assets is better than that of other financial institutions.
FP4 Average profit in the recent three years is better than that of financial institutions.
FP5 Overall business performance is better than other financial institutions.
Table 3.4.
Items Measuring Innovation Capabilities
Construct Code Questionnaire Items
Product innovation (ProdI; 5 items)
ProdI1 Our company often develops new services well accepted by the market.
ProdI2 A great majority of our company’s profits are generated by the new services developed.
(continued)
32 Table 3.4. (continued)
Construct Code Questionnaire Items
ProdI3 The new services developed by our company always
Construct Code Questionnaire Items
Process innovation (ProcI; 4 items)
ProcI1 Our company often tries different operation procedures to hasten the realization of the company’s goals.
ProcI2 Our company always acquires new skills or equipment to improve the manufacturing operation process.
ProcI3 Our company can develop more efficient manufacturing process or operation procedure.
ProcI4 Our company can flexibly provide products and services according to the demands of the customers.
Table 3.5.
Items Measuring Transformational Leadership
Construct Code Questionnaire Items
Intellectual Consideration having different needs, abilities and aspirations from others.
TLIC4 My direct supervisor helps others to develop
their strengths.
(continued)
33 Table 3.5. (continued)
Construct Code Questionnaire Items
Intellectual Stimulation (TLIS; 4 items)
TLIS1 My direct supervisor re-examines critical assumptions to question whether they are appropriate.
TLIS2 My direct supervisor seeks differing perspectives when solving problems.
TLIS3 My direct supervisor gets others to look at problems from many different angles.
TLIS4 My direct supervisor suggests new ways of looking at how to complete assignments.
Construct Code Questionnaire Items
Intellectual Motivation (TLIM; 4 items)
TLIM1 My direct supervisor talks optimistically about the future.
TLIM2 My direct supervisor talks enthusiastically about what needs to be accomplished.
TLIM3 My direct supervisor articulates a compelling vision of the future.
TLIM4 My direct supervisor expresses confidence that goals will be achieved.
Construct Code Questionnaire Items
Idealized Influence Attributes (TLIIA, 4 items)
TLIIA1 My direct supervisor instills pride in me for being associated with me.
TLIIA2 My direct supervisor goes beyond self-interest for the good of the group.
TLIIA3 My direct supervisor acts in ways that builds my respect.
TLIIA4 My direct supervisor displays a sense of power and confidence.
(continued)
34 Table 3.5. (continued)
Construct Code Questionnaire Items
Idealized Influence Behavior (TLIIB, 4 items)
TLIIB1 My direct supervisor talks about their most important values and beliefs.
TLIIB2 My direct supervisor specifies the importance of having a strong sense of purpose.
TLIIB3 My direct supervisor considers the moral and ethical consequences of decisions
TLIIB4 My direct supervisor emphasizes the importance of having a collective sense of mission.
Validity and Reliability of Instrument
Reliability refers to a measure’s consistency in producing similar results in different occasions. Validity refers to whether the instrument measures what is really intended to measure. There are two types of validity: internal and external (Coolican, 2004).
Table 3.6.
Reliability of Instrument
Reliability Type Theoretical Meaning Reliability of Instrument Reliability of Analysis It is the dependability or
consistency of the measure of a variable (Neuman, 2014).
Cronbach’s Alpha is calculated for each variable.
All results above 0.70 are considered acceptable (Nunnnally, 1978).
Average variance extracted (AVE) is computed as well.
A cut – off value equal or above 0.50 is considered acceptable (Chin, 1998).
35 Table 3.7.
Validity of Instrument
Validity Type Theoretical Meaning Validity of Instrument Face Validity It is a type of measurement validity in
which an indicator ‘makes sense’ as a measure of a construct in the judgment of others, especially in the scientific community (Neuman, 2014).
The questionnaire is checked and translated by native
The questionnaire was examined by two peer reviews and expert reviews. Every questionnaire items refers back to the definition.
Criterion Validity It is a type of measurement validity that relies on some independent, outside verification (Neuman, 2014).
The questionnaire underwent pilot tests.
Face Validity
A questionnaire was available in English, Serbian and Chinese language. The original version of questionnaire is in English language. However, the research was done in two countries Taiwan and Bosnia & Herzegovina so there was a need to translate it in other languages. The researcher translated questionnaire from English to Serbian language. The bilingual expert checked the translation of the questionnaire. A questionnaire was translated from English to Chinese language by Chinese speaking researchers. The bilingual experts revised the translation of the questionnaire. The questionnaire was undergoing a back translation in order to preserve the meaning of each item.
36
Data Analysis Method
The data collected in the pilot study was analyzed using the following statistical analysis.
Descriptive Statistics
Descriptive Statistics refers to mathematical values that interpret the data obtained from the sample. It will help in better understanding data that was collected from the pilot study. Two most used descriptive statistics in this research will be mean and standard deviation. These will be calculated in both pilot and main study. Descriptive statistics shows the importance of the study.
Correlation Analysis
A Correlation Analysis was performed to show the direction and strength between independent and dependent variable. SPSS 16.00 was used to test descriptive statistics.
Coefficient of Determination (R2)
The coefficient of Determination (R2) is also called multiple correlation coefficient and is defined as “as the proportion of variance” explained by the regression model makes it useful as a measure of success of predicting the dependent variable from the independent variables’ (Nagelkerke, 1991). In other words, R2 is using a concrete number to explain and predict the future outcomes.
Structure Equation Modeling (SEM)
Structural equation model (SEM) involves 3 different tasks: a structure between exogenous and endogenous variables must be hypothesized, it should be decided how exogenous variables will be measured and there should be find a model to measure endogenous variables (Mueller, 1996). Furthermore, SEM should be perceived as research process not just as statistical technique (Mueller, 1996).
SEM is method that is can be seen in two different perspectives: an economic perspective and psychometric perspective. The advantages of SEM are: model relationships between multiple predictor and criterion variable, construct unobservable latent variables, model errors for observed variables and measurement assumptions against empirical data (Marcoulides, 1998).
SEM is used to explain the collected data and to determine if the model is relevant to a given situation. There are two reasons for using SEM: first, to perform CFA and second, to perform path coefficients. Furthermore, Partial least square (PLS) is used in order to calculate path coefficients.
37 Partial Least Square (PLS)
Partial least square (PLS) is a statistical technique that combines component analysis and multiple regression (Abdi, 2007). It is used to predict dependent variables (Y) based on independent variables (X). Partial least square was developed by the Swedish statistician Herman Wold in 1975. It originated from social sciences then became very popular in chemometrics. Nowadays, PLS regression is becoming an important technique in social sciences for non-experimental and experimental data (Mcintosh & Lobaugh, 2004; Worsley, 1997).
In this research study SmartPLS 3.0 software will be used in order to perform a confirmatory factor analysis and a path analysis.
Bootstrapping
Bootstrapping represents a statistical technique in PLS program to duplicate the sample and retrieve the t-value to test if the sample would be significant. Furthermore, bootstrapping sample allows the estimated coefficients in PLS-SEM program to be verified for their significance (Henseler, Ringle & Sinkovics 2009). According to Hair et al. (2011) emphasized that critical t-values from two-tailed test are 1.65, 1.96 and 2.58 and they represent weak, moderate and strong relations.
38