This study used a quantitative research paradigm to measure the effects of the independent variables towards the dependent variable. Pilot study and expert review were performed to confirm the factorial validity, face validity and construct reliability of the constructs. This chapter includes six sections, specifically research framework, research procedure, data collection technique, instrumentation, and data analysis technique.
Furthermore, it describes the different constructs that were use and their associated reliability estimates, showing that they meet the requirement for academic purpose.
Research Framework
Relying on the review of literature, research purposes and questions, the research framework for the study was developed (see Figure 3.1, p.37). This research analysed the role of psychological capital and employee job satisfaction on the relationship between training and development and organizational commitment.
Firgure 3.1. Reseach framework H2
H3 H4
H1
H5 Training and
Development
Psychological capital
Organizational Commitment Employee Job
Satisfaction
38
Research Hypotheses
Based on the framework above and to answer the research questions, the following hypotheses were formulated:
Table 3.1
The Research Hypotheses
Hypothesis content
H1: Training and development is positively related to employee job satisfaction.
(Employees with a higher orientation to training and development are more likely to exhibit stronger propensities of job satisfaction)
H2: Training and development is positively related to psychological capital. (Employees with a higher orientation to training and development are more likely to possess higher psychological capital).
H3: Training and development is positively related to organizational commitment.
(Employees with a higher orientation to training and development are more likely to exhibit stronger propensity of organisational commitment behaviour)
H4: Employee Job satisfaction positively mediates the relationship between training and development and organizational commitment.(on average, when employees perceive a higher degree of job satisfaction, the extent to which employees training and development contribute to organisational commitment is more likely stronger)
H5: Psychological capital mediates the relationship between training and development and organizational commitment. (On average, when employees have a high psychological capital, the extent to which employees training and development contribute to organisational commitment is more likely stronger)
Research Procedure
In carrying out this research, the process was done in three majors path, identification, execution and completion. The research procedure path is illustrated in Figure 3.2 and described in detail as follows:
Identification of Research Topic and Research Population
The research topic was identified based on the literature review conducted and was carried out on the basis of the researcher‘s interest which is centered on public organizations
39
and on various key topics which included training and development, organizational commitment, employee job satisfaction and psychological capital. Given that the topic is not well researched in public enterprises in the Gambia, all the constructs were identified and described. Moreover, the relationships among the constructs was also addressed so as to form a clear idea about the research objectives and the questions to have a comprehensive research framework.
The research population/ sample was selected on the basis of limited literature on the population. The researcher use convenience and purposive sampling to select participants.
Convenience sampling were used as the researcher was unable to reach all the employees in all branches of GRA, thus samples were selected based on accessibility and willingness to participate in the survey. Purposive sampling was adopted given the fact that the targeted participants for the study have important information which can answer the research questions.
Data Collection, Research Questionnaire and Data Analysis
Once the research population and sampling was identified, a survey methodology was use as it is an effective tool for data collection. In order to ease the burden of the survey participants and maintain their interest while completing the survey, considerable attention was paid to the design of the survey. The measurement for all research constructs were extracted solely from the work of previous researchers that have carried satisfactory reliability and validity analyzes on the selected constructs.
Once the survey questionnaire was collected, participants‘ answered was input in excel and analyzed in SPSS 22.0. The current study used a two-stage process for analyzing data.
First, the adequacy of the measurement instrument was assessed and in the second stage, the descriptive statistics for each construct and the relationships among constructs was examined.
Research Results , Discussions and Conclusion
Finally after analyzing the data and assessing the implications, the researcher draw the conclusions for the research and provide some recommendations for GRA and future researchers. After the results was then used to make appropriate conclusions that answer the
40
reseach questions posed in the study and put forward recommendations drawn from the conclusions. The Simplied research procedure is shown below
Figure 3.2 Research Procedures
Research Design
This research employed a quantitative approach that followed a self-reported survey methodology to gather the data and examined the relationship among the variables. A case study method was used to enable the researcher to closely examine the data within a specific context which is, in this case, a public organization in The Gambia. In addition, the strength
Identification of Research Population and
Literature Review
Research Results and Discussions
Data Collection for Analysis and Interpretation
Transform Data for Analysis and Interpretation
Analyze the Data and Assess their Interpretation
Draw Conclusion and Provide Recommendations Research Topic
Identification
41
of a quantitative case study is found in its capacity to describe, illustrate, and explore a given phenomenon (Yin, 2003). Survey was employed for several reasons. First, a major advantage is that they are readily analyze and interpreted, as they are structured around key items and topics. Second, they ensured confidentiality for the respondents who may be more inclined to provide honest feedback if their identity remains undisclosed. Third, the quantitative results can also be compared across the periods and group (Debowski, 2006, p.278). Fourth, they are efficient data collection method when the researcher has a precise idea of what is required and how to measure the variables (Sekaran, 2003). Since the researcher is interested in determining the relationship between training and development and organizational commitment, PsyCap and employee job satisfaction, self-reported measures was used as it allow respondents to record their own perceptions of reality. Finally, the method represents the most suitable methodology for the study of individuals‘ behaviour (Howard, 1994;
Spector 1994).
Lending support to this design is the fact that previous studies highlighted in the literature review chapter used the same research design. This quantitative study, focus on a single case public organization in The Gambia. Whist case studies lead to results that cannot necessarily be generalized, but can be a basis of developing a relationship between variables (Colorado State University, 2011)
Research Instrument
The research instrument used consists of a self-reported survey questionnaire to collect the required data and to test the established hypotheses. The research instrument adopted was developed by applying good practices and design principles as follow:
I. A cover letter was used to introduce the study and its purposes are clearly stated.
II. The questions are grouped into sections on the basis of their relevance and relationship.
III. Simple instructions were provided at the beginning of the questionnaire and at each section.
IV. The questions are free from subjective tones in order to prevent bias.
The questions are applicable and answerable by participants.
V. Items were mixed together to reduced common bias methods
On the basis of the research purposes, the questionnaire consists of four sections: A) training and development; B) organizational commitment; C) employee job satisfaction and
42
D) psychological capital. The participant background section was added as a fifth section in order to provide a descriptive analysis of the survey respondents. This section contains demographic information such as participants‘ age, gender, tenure, education and formal and informal training receive. Given that this section was included merely for informative purpose, the related questions are located at the end of the research instrument. This allowed the researcher to begin the questionnaire with items which are the primary focus of the research.
This survey questionnaire were measured using a 5–point Likert scale questionnaire over the 6–point due to the following reason. Survey respondent might truly feel neutral about a particular topic, presenting to these respondents a scale without a neutral midpoint can introduce a bias as they are forced to choose a more positive or negative response. Some researchers point out that in many cases, respondents accentuate the negative in an experience (Gwinner, 2010). With a 5 point Likert scale is important for a legitimate opinion that exists among respondents, and finally the scale has a midpoint which indicates neutrality or mixed opinion. When calculating the mean weighted average, there is a standard point of comparison. One will know that an average rating of 3.4 is above neural and a 2.8 is below (Gwinner, 2010).
Measures
Table 3.2 gives the items that were used to measure the variables in this study. Each item was given a code that is used in the statistical analysis of the data. The operational measures of each construct are presented in (p.44). The texts of items for all scales are provided in Appendix A.
Training and Development
Training and Development is the independent variable and divided into four sub variables to reflect the complex role and influences on HRD practice. A scale were adapted from Bartlett (2001) all variables included in the study are selected from well-established and validated scales. These include four items to determine training motivation ((a=.76) (Tharenou & Conroy, 1994), a shortened three–item scale from the perceived support of training from colleagues scale (a=.83) and the six–item management training support scale
43
(a=.87) developed by Noe and Schmitt (1986,), and the five–item ―Training Benefits‖ scale (a=.82) (Noe & Wilk, 1993).
Organizational Commitment
The study measured organizational commitment variable using Allen and Meyer‘s (1990) questionnaire. The instrument was developed to measure organizational commitment as a multidimensional concept which includes affective commitment, normative commitment and continuous commitment (Meyer & Allen, 1997). This questionnaire comprises 24 structured questions with 8 items in each dimension. The researchers established the reliability estimates (internal consistencies) for this instrument as follows: .85 for affective, .79 for continuance and .73 for normative, and its validity and reliability is evidenced in international samples (Coleman, Irving, & Cooper. 2000; Yousef, 2000). Employees were instructed to rate , on the basis of their personal observation, the extent to which their organisational commitment regarding the described conditions on a 5 point Likert –type scale with scale anchors ranging from ‗‘strongly agree to strongly disagree.
Psychological Capital
To measure psychological capital, the researcher adopts the measurement scale developed by Harms and Luthans (2013). There are 24 items which includes three reverse-coded items divided into four dimensions which are hope, resilience, optimism, and efficacy. Employees were instructed to rate, on the basis of their psychological capital regarding the described condition on 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree).
The high score of scale indicates that people have higher level of psychological capital (PsyCap). The Cronbach‘s alpha reliability of the PsyCap instrument is 0.87.
Employee Job Satisfaction
In measuring employee job satisfaction, attitude surveys are often used as they have proven to be more effective and accurate since they provide a picture of how satisfied employees are as well as explore the effectiveness of major job changes in attitude (Colquitt, 2010). Employee job satisfaction is used as a dependent and mediating variable in this study.
It was measured using the 15-item shortened version of the Job Diagnostic Survey (JDS) (Hackman & Oldham, 1976). The JDS was originally developed for the social service sector but it was shown to be adequate to use in other sectors as well (Roelen, Koopmans, &
44
Groothoff, 2008). The scale consists of five facets with three items in each facet. A 5-point Likert type response was use and this format ranged from 1 (strongly disagree) to 5 (strongly agree). According to Loher, Noe, Moeller., & Fitzgerald, (1985), JDS is among the job satisfaction measures that meet the quality criteria for reliability and validity. This is also evident from the internal consistencies or coefficient alphas drawn from a sample of 3,067 individuals who completed the JDS. The coefficient alphas ranged from satisfaction of skill variety to feedback of results subscale to 0.91 for the total scale (Hackman 1974).
Table 3.2.
Established Reliability of Variables
Dimension Factor Cronbach’s α Source
Training and Development
Motivation of training .76 Tharenou and Conroy (1994) Management Training
45
Reliability and Validity of the Instrument
This research adopted the same validity standard as the previous researchers on the subject under study. The cultural setting in which the research is conducted also led this researcher to adopt additional validity standards. The standard reliability methods adopted for this research are given in Table 3.3, while validity methods are shown in Table 3.4.
Table 3.3.
Reliability Methods for this Research
Reliability Type Meaning How the Study Achieve Reliability Reliability Test Refers to the probability that
an item will consistently perform a required function without failure under stated conditions (Field, 2006)
Cronbach‘s alpha was computed for each of the set of variables to provide evidence for reliability. According to Nunnally (1959) above 0.70 is generally considered acceptable which both the pilot and formal study attained. .
Table 3.4.
46
Data Collection Sampling
A target population is the group of people, objects, or events that the researcher wishes to study (Cavana et al., 2001). The data collected from the Gambia Revenue Authority was based on convenience sampling. The present study adopts the sampling strategy because the researcher anticipated that theoretical and empirical linkages between, training and development, employee job satisfaction, PsyCap, and organisational commitment would remain robust, although in the case, considering the factor and organization type.
In addition, convenience sampling was used in order to retrieve a larger number of completed responses quickly and efficiently as convenience sampling is a non-probability method. This means that respondents are chosen in a non-random manner, and some members of the population have no chance of being included. When time or cost is a factor, convenience sampling becomes a useful sampling method (Kahl, 2010),
Numerous studies regarding organisational commitment conducted in a context of various industries but this research is limited to The Gambia Revenue Authority and the study was not conducted before. Against this backdrop, The Gambia is a suitable choice for this research because of limited study conducted on these variables in public organizations.
Another reason is the remarkable changes public institutions are faced with. Thus, there is a great need to identify and elucidates how organizations can handle the different processes that occur within them.
Sample for the Study: The Gambia Revenue Authority
Specifically this research target one case organization, the Gambia Revenue Authority (GRA). GRA is a semi-autonomous body corporate with a partial 'de jure' and de factor' autonomy, providing its legal mandate to perform its inherent responsibilities, as a national agency for the assessment, collection and enforcement of revenue collection due to government in a fair and transparent manner. By and large, GRA was created by an Act of the National Assembly called the Gambia Revenue Authority Act 2004. Distinctively, this national institution is a combination of two renowned and distinct revenue departments, i.e.
the Customs and Excise Department (CED) and the Domestic Taxes Department (DTD), formerly Central Revenue Department (CRD) - Income Tax Division, who are merged
47
together under one unified platform with a common purpose geared towards enhancing, and channelizing effective domestic revenue mobilisation for a sustainable economic growth and development. The GRA, as of 2014, the office has 615 employees on its payroll. GRA has a training section under the HR unit and has been functional since 2007. This organization, because of its accessibility and size, is a suitable representation of the target population.
Sample Size and Criteria Selection
The population for this study consist of full-time employees working at GRA for the following reasons. Employees are the most important source of organization knowledge because they possess the expertise necessary for managing organization successfully. The sample size targeted for this study was 300 employees. To be eligible for this research the sample had to meet general criteria for sampling which are;
I. Respondents must be fully employed within GRA. This means that part-time employees and interns are not included in sample population.
II. Secondly, respondents must have at least one (I) year experience and receive training (formal or on-the-job training).
These criteria are chosen because work experience is a good predictor of job knowledge (Longoria, 1997).
Data Collection Procedure
The data used to conduct pilot and formal study were collected using various stages through hand deliver and email survey. The researcher provided gift (pen, notebook and headscarf) to enhance response rate. Data collection was carried out in various stages. In the first stage, the researcher sends a letter to the participants explaining the purpose of the study and requesting permission to be interview. In the second stage, after acceptance to conduct the survey, paper-based surveys was hand-delivered to the case organization by the researcher. The researcher was able to count with the assistance of a Human Resource employee designated by the organization in helping with the distribution of the questionnaire.
In addition in order to facilitate access to all employees, the survey was also email to the employees during the formal study
48
The email first assured confidentially of the information, followed by instruction of how to participate in the survey. The next phase, participants were asked to complete the questionnaire items for assessing training and development (TD), followed by psychological capital (PsyCap), employee job satisfaction (JS), organizational commitment (OC) and demographic characteristics by underlining or highlighting the response chosen.
The pilot study data was first collected from July through August and the formal study data collected follow after establishing the instrument suitability.
Method of Data Analysis
The data analyses were carried out by using Statistical Package for Social Sciences (SPSS) version 22.0 based on the following steps:
Descriptive Statistics
The primary function of descriptive statistics is to present research results clearly and concisely (Healey, 2008). Thus, the researcher conducted descriptive statistics in order to organize the data into a more presentable and interpretable form by summarizing sets of numbers such as the means and standard deviation of the data. Furthermore, the descriptive statistics assisted to demonstrate the relevance and the importance of the study
Confirmatory Factor Analysis.
The method of SPSS was used to perform confirmatory factors analysis (CFA) for testing the factorial validity of the variables. The CFA conducted using Amos 2.2 was, based on the covariance structure analysis, were one item per construct was fixed at 1 .0 and a simple structure maintained. The validity was confirmed by showing that the fit indexes fell within an acceptable range. The criteria are explained in p.63.
Correlation Analysis
Correlation analysis was conducted to examine the relationship between the independent, mediator and dependent variables respectively. It was also carried out because it is the primary test procedure required by Baron and Kenny (1986) in handling hierarchical regression analysis. Based on their guidelines, the correlations among the independent
49
variables (i.e. training and development) should be significantly associated to the mediator variables (i.e. psychological capital and employee job satisfaction).
Regressions Analysis
In order to analyze the mediating role of psychological capital and job satisfaction on the relationship between training and development and organizational commitment, a hierarchical regression approach was adopted. The hierarchical regression approach is a three-step process as described by Baron and Kenny (1986). In the first step, the mediating variables are regressed on the independent variables; second, the dependent variables are regressed on the independent variables; and then the dependent variables are simultaneously regressed on both the independent and mediator variables. On this basis, psychological capital and job satisfaction should be considered as significant mediators when the following conditions are met:
I. The independent variables (i.e. training and development) are significantly related to the mediating variables (psychological capital and job satisfaction) in the first test;
II. The independent variables are significantly related to the dependent variables (i.e.
organizational commitment) in the second test;
III. The mediating variables (i.e. psychological capital and job satisfaction) are significantly related to the dependent variables (i.e. organizational commitment).
Full mediation applies when an independent variable becomes insignificant
Full mediation applies when an independent variable becomes insignificant