RESEARCH METHODOLOGY - Perception of Managerial Trust and the Effect on the Job Attitudes of E

This chapter refers to the research methodology which gives a detailed report on the methods used to carry out this study. This chapter contains the research framework, research procedure, research methods, sampling, ethical considerations, instrumentation, data collection methods, and finally data analysis methods.

Research Framework

The research framework as shown in figure 3.1 below was created based on the review of literature. The literature showed that that harmony, reliability and concern are used as dimensions in measuring interpersonal trust and that is based on the instrument created by Tzafrir and Dolan (2004) in measuring managerial-subordinate trust. The literature review also shows that trust has a relationship with job attitudes particularly with job satisfaction and organization commitment. The control variables used in this model was suggested by Straiter (2005) that personality in particular positive and negative affect would possibly have some influence on job satisfaction outside of trust and tenure would possibly have an effect on affective organizational commitment outside of trust. Control variables are used in other to rule it out as a different explanation to the findings of a research. Control variables should not be restricted to the method section of a study but for the role of the variables to be truly understood it is essential that they are addressed in the hypothesis, results, and the discussions (Atinc, Simmering, & Kroll, 2011; Becker, 2005; Spector & Brannick, 2010). Hence this research will include two added hypothesis to facilitate the use of the control variables tenure, negative affect, and positive affect as control variables in this research.

Figure 3.1 Research framework

Hypothesis of the Study

The three main hypothesis of this study were H3, H4, and H5. To accommodate for the use of control variables controlling for the effect of tenure, positive, and negative affect; two additional hypotheses were included H1 and H2.

H_a: µ > 0

1. H_a: There is a significant relationship between tenure and affective organizational commitment.

2. H_a: There is a significant relationship between positive and negative affect and job satisfaction.

3. H_a: There is a significant and positive relationship between perceived managerial trust in subordinates and job involvement.

4. Ha: There is a significant and positive relationship between perceived managerial trust in subordinates and affective organizational commitment when controlling tenure.

Job

5. H_a: There is a significant and positive relationship between perceived managerial trust in subordinates and job satisfaction when controlling positive and negative affect.

Research Procedure

This research followed a quantitative research layout and entails the following elements as suggested by Coughlan, Cronin, and Ryan (2007). The authors proposed that the following elements influence the forcefulness of quantitative research hence the researcher will follow each of steps explained below and figure 2.2 shows a graphical summary of the research procedure.

1. Research purpose/research problem: This section of the study identified why research in managerial trust in their subordinates is important in the organization.

2. Aims/objectives/research questions: This section shows what the researcher accomplished by doing this research, the questions were selected based on the purpose of the research and finally the Operational definitions: The researcher provided clear definitions of all the terms and concepts that are mentioned in the study.

3. Literature Review: The researcher provided a careful analysis of literature on organizational trust, job attitudes namely: Job satisfaction and organizational commitment (affective organizational commitment) and job involvement mainly from books, journals, and articles.

4. Conceptual framework/Hypothesis formulation: A conceptual framework was established based on the research purpose and the variables which are measured in this research. The hypothesis was formulated as a platform for testable predictions in the study.

5. Sample: Convenience sampling used in this research, employees from the financial sector in St. Lucia was targeted.

6. Ethical Considerations: Participants were guaranteed confidentiality with information provided to the researcher.

7. Research Methods: This section includes the study design, and instruments that were used. This study took a quantitative approach. The instrument was reduced and slightly adjusted to suit the purposes this research. A pilot study of 30 participants was carried

out to assess the feasibility of using this particular instrument. Principal Component Analysis revealed several items that were faulty which were removed from the final instrument.

8. Data Analysis/Results: This research used SPSS to analyze for data collected through the survey method. The results were communicated and significance of results was provided.

9. Discussions: The link between the literature and findings were discussed, the researcher also discussed whether hypothesis were supported or not.

10. Conclusion/Recommendation: Study limitations and strengths were discussed and recommendations for future research provided and also recommendations for the financial sector in St. Lucia.

11. References: Detailed references for all materials including books, journals, and articles used were provided.

Figure 3.2 on the next page shows a graphical summary of the research procedure of steps one to eleven as outlined above.

Figure 3.2 Research procedure

Research Design

This study examined the relationship between employees’ perception of management trust and its effect on employee job satisfaction and affective organizational commitment. Due to the nature of this research, where the researcher was trying to establish the relationship among variables, a quantitative approach was found to be the most appropriate. In a quantitative research method, the researcher explored the relationship among variables using numerical data.

Because the aim was to empirically seek precise measurement on how different variables impacted on each other; in particular the relationship of managerial trust with job attitudes, then according to Neill (2007), this research was justified in using the quantitative method to conduct this study.

Sampling

Purposive or Judgmental sampling was used in this research. This is where the researcher chose participants who in his or her opinion were best suitable to provide the required information and who would also be willing to share it, Kumar (1996). The population of this study was employees in the financial sector in St. Lucia. The sample were employees because the researcher aim was to measure subordinates’ perception of managers’ trust in their employees and the effect it would have on their job satisfaction, affective commitment, and job involvement. In this case the outcome would be dependent on what the employees perceive and not on whether or not managers believe that they trust their employees.

Employees from the financial sector were chosen because it provided a wide pool of workers making it more convenient to get participants for the survey. Employees had to be in their positions for 6 months or more since this could be enough time to build an employee-employer trust relationship. Each participants were asked if they had been working in the organization for six months or more before they were given the questionnaires to ascertain that the criteria was met. Also in the questionnaires the participants were asked the number of years that they have been working for their organization, hence if any did not meet the criterion of 6 months or more, the questionnaire would be considered invalid. The financial sector included banks, insurance companies and the credit unions.

Sampling Characteristics

After conducting a pilot study of 30 participants and running factor analysis as a data reduction method and also to refine the instrument as explained later on in this chapter, the main study was conducted. There were six items in the demographic information which were explored in this study: Gender, age, education, tenure, years of experience, and position held at the organization.

The descriptive analysis revealed that 54% of the respondents were female and 46% were male. The most frequent respondents were within the age range of 21-25 accounting for 33% of the sample, and the least frequent were within the age range of 50 or more accounting for 4% of the sample. Most of the respondents have an Associate degree (51%) and majority have been working at respective organizations for 0-4 years (35%). The majority of the respondents also have work experience of 5-9 years (33%). Finally the most frequent job position was bank teller accounting for 28% of the respondents. Table A1, A2, and A3 with the coding system used in SPSS for demographic variables, descriptive statistics for job positions, and other demographics is provided in the Appendix section of this research.

Ethical Considerations

The data collected for this research followed the ethical guidelines suggested for researchers. The main ethical issues this research will follow are taken from Miles and Huberman (1994, as cited in Punch, 2000) they are namely:

1. Informed consent of participants: This means that participants will be informed about research. The researcher will ensure non-coercive consent of each participant.

2. Privacy: Information was safeguarded to ensure the identities of participants were protected.

3. Ownership of data and conclusions: Since the researcher is affiliated with the National Taiwan Normal University and this research was done as a part of a thesis, the research results will be published and be available at the NTNU library for reference.

4. Uses of Results: The researcher’s aimed that the findings in this research be used appropriately and not be misused in anyway.

Instrumentation

This research employed the use of questionnaire survey as the primary means of data collection to answer the questions of the study. Babbie (2011) suggests that survey research can be used for descriptive, explanatory, and exploratory purposes, and it maybe the best method for the social researcher who is interested in collecting new data to describe a population that is too large to be observed directly. Babbie (1990) proposed survey research as being logical, deterministic, and specific. The major advantages of survey research is that a large amount of information can be collected within a short period of time and it takes less time to analyze the data because questions are usually structured. One major disadvantage however is that one cannot collect in depth information as is the case for interviews and focus groups.

In search of the most appropriate measurement to use for this study, careful study of literature found the trust questionnaire created by Tzafrir and Dolan (2004) to be best suited for the purpose of this research because most of the other instruments found in literature focused on subordinates trust in leadership and trust in the organization. This questionnaire measures management trust in their employees which is the aim of this research. The questionnaire was designed to be filled out by either the managers or the employees. A feasibility study was conducted to test the instrument and gather information.

The trust questionnaire was designed based on the Tzafrir and Dolan (2004) assumption that trust is three dimensional including harmony, reliability, and concern. The original measurement consisted of attitude as a fourth dimension but the authors stated that based on a factor analysis on trust, only three factors were produced indicating strong similarities between harmony and ability. The questionnaire uses a 16 item scale with 5 items measuring harmony, 5 items measuring reliability, and 6 items measuring concern. The scale follows a 5-point Likert scale format with 1 representing disagree strongly and 5 representing agree strongly. The Cronbach’s alpha for the 16-item instrument was 0.91which is suggestive of very good internal reliability. The Cronbach’s alpha for the subscales of the instrument ranged from 0.70 to 0.88.

To measure Job Satisfaction, the researcher used the Global Job Satisfaction questionnaire. This questionnaire was developed by Wann, Cook, and Wall (1979 as cited in Fields, 2002). The Global Job Satisfaction used 15-item scale. Previous research using this measurement showed a coefficient alpha ranging from 0.80 to 0.91. The instrument follows a

7-31

point Likert scale format but the researcher used a 5-point Likert scale with 1 representing very dissatisfied and 5 representing very satisfied to follow the same format with other instruments which was be used in this research.

Since personality affectivity is known to have an effect on employee job satisfaction, the researcher used positive and negative affect as control variables as suggested by Straiter (2005).

The shortened form of the Positive and negative affectivity scale developed by Kercher (1992 as cited in Danhauer, Carlson, & Andrkowski, 2005) was used in this research with alphas of 0.75 for positive affect and 0.81 for negative affect.

Organizational commitment was measured using the shortened organizational commitment questionnaire (OCQ). The shortened OCQ is a 9-item shortened version of the original 15-item (Mowday et al., 1982, as cited in Fields, 2002). This questionnaire measures attitudinal or affective commitment. Different studies using the shortened OCQ reported coefficient alpha ranging from 0.74 to 0.92 (Fields, 2002). The questionnaire uses a 7-point Likert scale with one representing strongly disagree and 7 representing strongly agree again the researcher proposed to use a 5-point Likert scale to correspond with the other measurements used in this research.

For job involvement, this research used the 20-item scale designed and developed by Lodahl and Kejner (1965 as cited in Reeve & Smith 2001) which is often used in research on job involvement. The questionnaire uses a 5-point Likert scale ranging from 1(strongly disagree), 2(disagree), 3(neutral), 4(agree), and 5(strongly agree).

For the purpose of this research, the questionnaire utilized the items mentioned above to assess the main variables which are management trust of subordinates as the independent variable with sub-independent variables harmony, reliability, and concern. Job involvement, Job satisfaction and affective organizational commitment will be the three independent variables.

Finally, tenure, positive affect, and negative affect were used as control variables controlling for their possible effect on affective organizational commitment, and job satisfaction.

The questionnaire was a combination of each of the measurements mentioned above.

Additional demographic data were included in the questionnaire for the purpose of comparison.

Variables included: Gender, age, tenure, experience, education level, and employee position.

Tenure was used as a control variable to control for organizational commitment. Straiter (2005)

suggest that tenure has been known to affect levels organizational commitment outside of trust.

Tenure will be measured by the number of years one has been with the organization.

Data Collection

Since the data was being collected in a different country than where the researcher currently resides, the fitting method of data collection was that of online survey. Unfortunately, this method was not effective and being pressed for time, a printed form of the questionnaire was created. A trusted friend helped in distributing the questionnaires to the different institutions which was no easy task seeing that participants were very suspicious concerning the use of the questionnaires and were very skeptical to participate. Most of the data was collected from the major banks, and insurances in the capital city of St. Lucia (Castries); data was also collected from other bank branches and credit unions in the southern part of the island (Micoud and Vieux-Fort). The researcher contacted several key persons at the banks and insurances to facilitate their aid and as expected they too had some concerns. 130 copies of the questionnaires were distributed out of which 110 were returned making it a return rate of 84.62%; 10 were excluded because they were incomplete hence the sample size used was 100.

Data Analysis

This research used SPSS as the dominant tool for data analysis. The data provided in the survey were coded to perform data analysis. The research used:

Descriptive Statistics

Descriptive statistic which gave frequencies, measures of central tendency (mean). The standard deviations were also provided in the descriptive statistic. This helped the researcher group the participants in different categories for example age, gender etc.

Correlation Analysis

Correlation analysis was also used to measure associations between two variables. The Pearson’s correlation was used to measure the strength of linear dependence between the X and Y variables. The Pearson’s correlation returns values between a positive and negative one,

correlation analysis reveals how much one variable tends to change when the other one does.

When r is zero it means that there is no relationship. When r is greater than zero it means that when X goes up then Y would also go up. When r is less than zero it means that when X goes up then Y would down.

Multiple Regression Analysis

The researcher used the null hypothesis method where the alternative hypothesis assumed that there is a positive correlation between variables X and Y. To find evidence against the null hypothesis a test of significance needed to be carried out which was done in the following steps:

 Stating the hypothesis for the study which is the alternative hypothesis

 Selecting a probability of error level (alpha level) P<0.05

 Select and compute the test for statistical significance using multiple regression analysis

 Interpreting results

Multiple Regression Analysis was used because it allowed the researcher to separately enter additional factors in the analysis to estimate the effect of each. This is important because it quantifies the impact of more than one independent variable on a single dependent variable. The hierarchical multiple regression analysis was used since control variables were used in this research.

Pilot Study

A pilot study is also called a feasibility study which is a small scale initial work done before the main study. The purpose of the pilot study was to help the researcher decide whether the instrument was feasible to use or not. The pilot study also provided the researcher with the chance to evaluate the appropriateness of the data collection and other procedures and to make necessary changes if needed.

This research used the pilot study for validity and reliability testing of the instrument, since the questionnaire was adapted from the original to suit the needs of this research. The

questionnaire was filled out by 30 participants in the financial sector in St. Lucia. This is the same sample group of the main study. The process of face validity, construct validity, and construct reliability is explained below.

Face Validity

Face validity was done on the initial questionnaire which contained 70 questions which were reduced to conduct the pilot study. One HRD academic professional along with five colleagues agreed that some of the questions were redundant or difficult to understand hence were removed. A feasibility or pilot study was done on the remaining items which were 50 in total. A small sample of 30 subjects filled out the questionnaires and principle component analysis was done for the purpose of data reduction and determining construct validity.

Cronbach’s Alpha was also used to ascertain reliability of instrument.

Construct Validity

Before establishing construct validity, the researcher needed to follow some basic steps in preparing the data for Principal Components Analysis (PCA, for short) which is a variable-reduction technique that shares many similarities to exploratory factor analysis. Its aim is to reduce a larger set of variables into a smaller set of variables (called principal components) that account for most of the variance in the original variables. Although principal components analysis is theoretically different to factor analysis it is often used interchangeably with factor analysis in practice and is included within the Factor procedure in SPSS. Different researchers suggest different sample size for a data set to be factorable. The rule of thumb is that cases to variable be at least 5:1 or a sample size of 100 or more ; however a study conducted by Preacher and McCallum (2002) suggested that small sample size should not be an issue if communalities are high and the number of expected factor is small; the instrument met this criteria. Factor analysis was run for separate construct in the questionnaire for this study and the ratio for cases to variables ranged from 3:1 to 7.5:1. It should be noted that principle component analysis was not run on positive and negative affect since it was not the intention of the researcher to alter or

在文檔中 Perception of Managerial Trust and the Effect on the Job Attitudes of Employees in the Financial Sector in St. Lucia (頁 32-56)