This chapter provides information about the research methodology. It explains the research framework, states the research hypothesis, explains the research procedures, and research design. Furthermore, it will provide information about the statistical analysis methods that will be used to analyze the data.
Research Framework
The research framework of this study, see figure 3.1, is based on the research purpose and research questions explained in Chapter I. The research is aimed to explore how the training and development is influencing organizational commitment and how both of these variables are influencing employees’ turnover intention. In addition, the research is also aimed to investigate if the organizational commitment mediating role on the relationship between training and development and turnover intention.
H4
H1 H3
H2
Figure 3.1. Research framework Training and Development Training Motivation
Perceived Availability of Training Training Benefits
Management Training Support Perceived Support from Colleagues
Organizational Commitment Affective Commitment Normative Commitment Continuance Commitment
Turnover Intention
Research Hypothesis
Based on the research questions and purpose of the study, the following hypotheses were formulated:
H1: Training and development has a significantly positive influence on organizational commitment.
H2: Training and development has a significantly negative influence on turnover intention.
H3: Organizational commitment has a significantly negative influence on turnover intention.
H4: Organizational commitment has mediating effect on the relationship between the training and development and turnover intention.
Research Method
This study followed a quantitative approach to measure the relationship between training and development, organizational commitment and turnover intention. The data was collected using both online and paper survey questionnaires. In order to increase effectiveness and efficiency of the data collection the Internet was used as a medium to approach the sample population. The data was also collected by paper questionnaires approaching the sample population on social gatherings and sending survey to companies. The questionnaire was selected because the researcher intends to collect data in form of numbers and it can be used because it allow the researcher to gather information on the backgrounds, beliefs, or attitudes of a large number of people (Neuman, 2014) which will be analyzed using statistical methods to test and verify the relationship between the variables. The advantages of using survey questionnaire are internal and external validity, efficiency, surveys can cover geographically spread samples, they have ethical advantages and are flexible (Mathers, Fox & Hunn, 2007).
As the literature explains, a quantitative study tries to verify or falsify a relationship or hypothesis, and it focuses on outcomes or effects between variables (Neuman, 2014) therefore, the quantitative approach is more effective for the purpose of this research. In addition, the survey questionnaire is structured around key item and it ensures anonymity of respondents.
Instrument
The research utilized the self-reported online and paper survey questionnaires to collect the required data and to test the established hypothesis. The research instrument was designed
and its purposes and giving clear instructions. The survey questionnaire was bilingual and it was offered in both English and Chinese languages. This instrument consisted of 3 variables and had 51 questions which were divided into the following sections:
I. Training and development II. Organizational commitment III. Turnover intention
This study used different scales taken from the work of previous researchers which have been proved to have a good reliability and validity. From section I to section III this survey questionnaire used a 5-point Likert’s scale and the participants will be asked to choose from the different available demographic options. The survey was anonymous and it also included 7 questions on demographics of the participants including participant’s gender, age, marital status, industry, education, position and tenure. The full survey questionnaire can be seen in the Appendix A. Part of the data was gathered via online survey. Sample of design of the online survey can be seen in the Appendix B.
Training and Development
This variable was measured using a 23-items scale (Bartlett, 2001; Newman et al., 2011).
This survey comprises five dimensions and it used a 5-point Likert’s scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree) and an illustrative item is “I am willing to invest effort to improve skills and competences related to my current job”.
The 5 dimensions are:
1. Training motivation (Tharenou & Conroy, 1994) 4-items (α= .87)
2. Perceived availability of training (Newman et al., 2011) 5-items (α= .928) 3. Training benefits (Noe & Wilk, 1993) 5-items. (α= .82)
4. Management training support (Noe & Schmitt, 1986) 6-items. (α= .96)
5. Perceived support of training from colleagues (Noe & Wilk, 1993) 3- items (α= .83)
The Cronbach's alpha scores listed above were taken from the original researches as a proof of good internal consistency therefore they were suitable for this research. In the survey questionnaire the questions 25 till 28 were on training motivation, questions 29 till 33 were on perceived availability of training, questions 34 till 38 were on training benefits, questions 39 till 44 were on management training support and questions 45 till 47 were on perceived support of training from colleagues. All the questions can be seen in the Appendix A.
Organizational Commitment
In order to measure this variable, the organizational commitment scale originally developed by Meyer and Allen (1990) was taken. The scale has three sub-dimensions:
affective, normative and continuance commitment and each of the sub-dimensions were measured by 8 questions. It used a 5-point Likert’s scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree) and in total it consisted of 24 questions. In addition, an illustrative item of this part of the questionnaire is “One of the major reasons I continue to work for this organization is that leaving would require considerable personal sacrifice- another company may not match the overall benefits I have here”. Moreover, in this part of the questionnaire there were several reversed questions originally developed in the study by Meyer and Allen (1990). Questions 1 till 8 measured affective commitment, questions 9 till 16 measured continuance commitment and questions 17 till 24 measured normative commitment. All of the questions can be seen in the Appendix A.
Turnover Intention
Turnover intention was measured by using a 4-item scale (Newman et al., 2011). The Cronbach's alpha scores in the research of Newman et al. (2011) were above 0.7 (α= .909) proving that it has good internal consistency. A representative item can be seen as follows: “I often think of quitting my present job”. In addition, one reversed question was used to measure this variable. In the survey questionnaire, that can be seen the Attachment A, the questions 48 till 51 are measuring turnover intention.
Personal and Demographic Data
The last section provides a descriptive analysis of the survey participants. Examples of demographic characteristics include age, race, gender, ethnicity, religion, income, education, and many other variables. This section contained information such as the participant’s gender, age, marital status, industry, education, position and tenure.
Participants Population
According to the topic and purpose of the study, the target population of this study were full-time employees currently working in Taiwan. The employees were had to be employed in service or manufacturing industries. According to the Business Dictionary, service industry is part of business that primarily earn revenue through providing intangible goods and services and are not involved in the manufacturing of the goods, and, on the other hand, the
semi-finished goods using machines, tools and labor. For the research sample to be more reputable for representing the whole population, the surveys were gathered from various companies in Taiwan.
Sample
The researcher employed convenience sampling. Employees from various service and manufacturing companies in Taiwan were contacted and asked to fill in paper or online survey questionnaire. As it was already mentioned earlier, before the actual study there was a pilot study that analyzed the data from 40 participants. The pilot test was employed in order reduce bias in this study. After that, in total 277 people took part in this study. The selected participants had to be currently employed in a service or manufacturing industry in Taiwan.
Data Analysis
According to the purpose and hypothesis test of the study, the researcher used SPSS 22.0 statistics software for statistical analysis. The following statistics were used to analyze the data:
Descriptive Statistics
Descriptive statistics summarize the numerical results and show them in a clear and concise manner (Healey, 2008). The results were shown in forms of graphs, and tables to provide the reader a better picture of the data (Neuman, 2014). The descriptive statistics include the means and the standard deviation of the data and was applied to all the variables included in this research.
Reliability Analysis
For the reliability analysis Cronbach’s alpha analysis were used. Cronbach’s alpha is used to provide a measure of the internal consistency of a test or scale and it is expressed as a number between 0 and 1 (Tavakol & Dennick, 2011).
Correlation
The correlation measures the direction and strength of the linear relationship between two quantitative variables (Moore, McCabe, Duckworth, & Alwan, 2008). The correlation analysis shows initial relationship between the variables.
Hierarchical Regression
This analysis is used when there are two or more independent variables; and the purpose of this regression is to test causation (Gray & Kinnear, 2012). The research is aimed to test the direction and effect the independent variable, i.e., training and development, has on the mediator, i.e., organizational commitment, and the effect the mediator has on the depend
variable, i.e., turnover intention, as well as, the effect the independent variable has on the dependent variable.
Research Procedure
First the researcher identified the research topic, and the theoretical framework was created. Based on the theoretical framework the researcher reviewed the available literature on this subject. After the researcher had conducted the literature review, the sample population was identified. The sample population were people currently employed in service or manufacturing industries in Taiwan. Later the researcher searched and adopted the scales to measure each of the variables used in the research. Due to the chosen sample of this study the survey questionnaire was translated into Chinese. It was offered in both English and Chinese languages. Later on the data was collected and analyzed. The researcher employed convenience sampling. The questionnaires were distributed in the researcher’s social network such as friends, language exchange, classmates and social gatherings. The paper questionnaires were sent to three companies by mail. The survey questionnaire also had printed QR codes that were distributed in the researcher’s social network and the participants after scanning the code were directed to the online version of the questionnaire. The link to the online questionnaire was sent in social media and forwarded by friends of the researcher to service and manufacturing industry employees. In total there were 157 people who have submitted valid online questionnaires and 120 people who filled in the paper version of the survey questionnaire. Later the researcher gathered the results and discussed them. All the gathered data was analyzed in SPSS software. After that the conclusions and suggestions were offered. The research process is illustrated as figure 3.2.
Figure 3.2. Research procedure
Reliability and Validity Reliability of Instrument
The survey questionnaire items were adapted from previous research papers that have good reliability scores that are described in more details in the section ‘Instrument’.
Reliability analysis is the dependability or consistency of the measure of a variable and Cronbach’s alpha value above .7 is generally considered as an acceptable result (Neuman, 2014). The survey that will be used for this research has 51 questions and all items have Cronbach’s alpha above .7.
Validity of Instrument
The questionnaire was offered in both English and Chinese. In order to ensure that the scales and the translations are accurate and consistent and have good face validity, the peer review, expert review and back translation of the scales were conducted. Face validity is a type of validity measurement that4 examines if measuring device measures what it intends to measure (Neuman, 2014). The expert and people in charge of peer review were native Chinese speakers with a high proficiency level of English. The feedback regarding the content and the translation of the scales and the suggestions will be taken into an account and the necessary changes were made. In addition, in order to ensure the content and construct validity every questionnaire items were examined by two peer reviews and expert reviews.
The expert and the persons in charge of doing peer review were HR professionals. In order to achieve criterion validity, the questionnaire items were adopted from previously validated measure and the pilot test was conducted.
Pilot Test
According to McBurney and White (2010) a pilot study helps to detect any problems or deficiencies in the design of a research instrument before proceeding to the formal study. The pilot test consisted of 40 participants. In order to measure internal consistency Cronbach's alpha tests were performed. As it was already mentioned before, result of 0.7 or above is an acceptable reliability coefficient. The tests revealed that the reliability for the whole questionnaire of 51 item was 0.88. The pilot test showed that the Cronbach's alpha value for the variable of training and development was 0.85 and for turnover intention it was 0.89. The Cronbach’s alpha value of organizational commitment was 0.89. The researcher also made sure that the questions are easy to understand and that the translations of the questions are accurate. Based on the results of the pilot study the researcher made the necessary changes.
questions were not changed but the researcher in order to make the questionnaire more user friendly modified the layout of it. The font and spacing of the questionnaire was formatted that it would take less space on the paper, the questions were put in a table, the NTNU logo and QR code was added to the questionnaire and the introduction to the questionnaire was slightly modified. After the pilot test the translation of three questions also was improved.
Table 3.1.
Reliability Test (Pilot Test, N= 40)
Constructs No. of Items Cronbach’s alpha Value
TD 23 0.85
OC 24 0.89
TI 4 0.85
Note. N= 40; TD= Training and Development; OC= Organizational Commitment;
TI= Turnover Intention
Data Coding System
Before the analysis of the pilot test, all of the data was coded to facilitate the processing of information in the SPSS software. The 51 items of the questionnaire were coded using a 5-point Likert’s scale (1= Completely Disagree; 5= Completely Agree). In addition, the survey questionnaire also contained 7 questions on demographics. The demographic variables had multiple choice answers and the participants had to select which demographic group is the most suitable for them. The demographic variables were also coded. The coding system for all the variables and demographic questions are included in Table 3.2.
Table 3.2.
Coding System Used in SPSS Data Analysis
Variables Codes
Organizational Commitment (24 questions)
1= Completely Disagree 2= Disagree
3= Neutral 4= Agree
5= Completely Agree
(continued)
Table 3.2. (continued)