METHODOLOGY - Moderating Effect of Organizational Justice on the Relationship between Perceived

In this chapter, this study provided the research methodology of the study, including research framework, research hypotheses, research procedure, participant, data collection, measurement, and data analysis.

Research Framework

This research was conducted with three variables: perceived organizational support, organizational commitment, and organizational justice. This research is to exam significantly related to each variable. The research framework is showed below in figure 3.1.

Figure 3.1. Research framework

Research Hypotheses

Following hypotheses are developed based on the research questions and purpose of the study.

Hypothesis 1: Perceived organizational support is related to organizational commitment.

Hypothesis 2: Organizational justice moderates the relationship between perceived organizational support and organizational commitment.

Research Procedure

This section shows the research procedure which can be divided into these eight steps which this research went through. First, the statement of the research problem was to be distinguished. Then the author identified the research topic, then form the theoretical

Perceived organizational

framework and set the hypotheses via reviewing the literature. After concreting the topic, the author developed the questionnaire, also organize the group of participants. Then the researcher has conducted the pilot test before collecting and analyzes data through the online survey.

Lastly, the conclusions and implications were to be given from the statistical analysis in SPSS 23.0. The research process is illustrated in figure 3.2.

Set the statement of research problem

Review the literature

Form the hypothesis

Design the questionnaire

Conduct the pilot test

Collect the data

Analyze the data

Conclude from the findings

Figure 3.2. Research procedure

Participants

The samples of this study are Japanese tenure workers who work in Japanese companies and who are in their twenties to fifties. The study aims to resolve the early leave of the younger generation. However, in order to gain more in-depth insight into their work ethic and

willingness to contribute to the organization, this survey recruited the participants from a broader range of generations, as from the assumption that studying differences between younger generations and older generations could provide useful suggestions. During the pilot test, 50 participants joined, and 215 responses were accepted in the main survey.

Data Collection

A quantitative approach has been adopted to test and verify the relationship in order to measure the correlation between independent and dependent variables. The data was being collected using a survey questionnaire that was be conducted online and is self-administered instruments.

The reason why the author selected this method is with intentions to collect the data in forms of numbers, which will allow the research conduct the analyzation by using statistical methods to test and verify the relationship between three variables. An online self-reported survey was to be designed as the major research instrument to collect data. Lastly using online format are reasonable measures to maintain confidentiality.

In this research, snowball sampling was chosen to collect and select the participants with the support of the author's friends and colleagues. Participants received the invitation and an online survey form created by using Google Forms by E-mail, LINE and messenger of Facebook. Then, they responded to the survey between March 9, 2020 and April 9, 2020.

Firstly, collecting data for a pilot test was conducted between March 9 and March 21 and collected data on 50 participants. Then, the main survey was conducted from April 3 to April 9, and a total number of respondents is 215.

Questionnaire Design

The survey includes instructions and demographic questions followed by several instruments to assess his or her perceptions of commitment and perceived organizational support, organizational justice.

Also, although the mother tongue of the participants is Japanese, all existing scales to be used are initially formulated in English. Among three valuables, the questionnaires which are corresponding to POS and organizational justice was being adopted the Japanese versions which are developed from the original versions. Because those Japanese versions are already

ensured the validity and reliability of the previous researches (Shibaoka et al., 2010; Takada &

Kawamura, 2018). Moreover, the items of organizational commitment were be translated into Japanese initially by the author and then were be done reverse translation in order to obtain valid data.

Perceived Organizational Support

Previous studies (Rhoades & Eisenberger, 2002; Shanock & Eisenberger, 2006; Shore &

Tetrick, 1991) concluded that almost all studies which are related to POS had been conducted utilizing a survey of perceived organizational support (SPOS). Therefore, this study also followed their achievement.

SPOS has proven its reliability and validity in many previous studies (Eisenberger, Fasolo,

& Davis-LaMastro, 1990; Eisenberger & Stinglhamber, 2011; Lynch, Eisenberger, & Armeli, 1999; Shore & Tetrick, 1991). All of the researches were mainly conducted using a short form of SPOS. Since the Japanese version of SPOS which were formulated by Kato (1995) based on a shortened form of SPOS (Eisenberger et al., 1986).

Respondents rated their POS using 15 items comprising the short form of the POS Survey.

In continuing with the suggestions of Rhoades and Eisenberger (2002), this study used those items which covered the aspects of the definition of POS (e.g., estimation of employees’ role and concern about employees’ welfare). Sample items include “My organization really cares about my well-being” and “My organization is willing to help me if I need a special favor”.

Respondents selected the extent of agreement with each statement on a 5-point scale (5 = Strongly agree to 1 = Strongly disagree).

Organizational Commitment

This area used three dimensions’ scale, which was the achievement of Allen and Meyer (1990). Measures consist of 24 items: there are eight items for affective commitment, eight items for continuous commitment, and eight items for normative commitment. The scale comprises five items which are exacted on a five-point scale (5 = Strongly agree to 1 = Strongly disagree).

Although it has high validity and reliability which has been examined in many empirical studies, organizational commitment questionnaire (OCQ) was developed by Porter, Steers,

Mowday, and Boulian (1974) and was not be utilized in this research. Because this scale has several disadvantages which are mentioned, mainly it was noted that OCQ was not always appropriate to measure the commitment of Japanese employees to the organization.

For the above reasons, the scale developed by Meyer and Allen (1991) was to be utilized by translating into Japanese. Sample items include “it would be very hard for me to leave my organization right now, even if I wanted to with” and “things were better in the days when people stayed with one organization for most of their careers”.

Organizational Justice

This study adopted a scale which was developed by Colquitt (2001). Because this scale allowed the study to assess all sub-components of OJ: distributive justice (four items), procedural justice (seven items), interactional justice (nine items, which are divided into two sub-dimensions: interpersonal justice and informational justice), add that, a Japanese version of this scale was developed, and its reliability and validity have been already established.

Sample items include “Have you been able to express your views and feelings during those procedures” and “Does your (outcome) reflect what you have contributed to the organization?”.

Respondents selected the extent of agreement with each statement on a five-point scale (5 = Strongly agree to 1 = Strongly disagree).

Control Variables

This section contains age, job tenure, industry classification and gender. Age was divided into seven groups in a five years intervals. Job tenure was divided into five groups in a five years intervals. Industry classification was divided into 20groups. Gender was divided into male and female. Those variables were used as control variables for all statistical analyses for the sake of providing a descriptive analysis of survey participants.

Please refer to Appendix A and B for the completed questionnaires.

Pilot Test

In this study, a pilot test was carried out in advance for collecting data in order to test the accuracy and the qualities of questionnaires and for the sake of ensuring the reliability and the internal consistency among each element (Mozafari et al., 2008). Collecting data for a pilot test

was conducted between March 9 and March 21 and received 50 responses. The test results showed that the Cronbach's alpha values for each item: POS is 0.962, organizational commitment is 0.803, and Organizational Support for Development is 0.949. Based on the achievement of Peterson (1994), which showed the score of the Cronbach's Alpha must be more than .70, these figures indicated that sufficient reliability had been ensured.

Table 3.1.

Reliability Results of Cronbach's Alpha (N=50)

Variables Number of Items Cronbach's Alpha

To summarize the numerical results and show them clearly and concisely (Healey, 2014), this research conducted descriptive statistics by testing the central tendency and dispersion.

The descriptive statistics were included the means and the standard deviation of the data.

Correlation Analysis

In this study, correlation analysis conducted to judge the existence of correlation and assess the strength of the relationships among the independent variables, the dependent variable, and the moderator. The author examined Pearson correlation coefficient with the purposes to assess the strength and the direction of the relationship among variables. Controls variables, except industry, were also utilized in the correlation analysis. Since industry was a nominal variable, it was excluded from the subject.

Regression Analysis

Regression analysis is a method for explaining, predicting, and examining the relationship between the variation of one variable and the variation of another variable. In this study, two types of regression analyses were executed. Firstly, the influence of POS on organizational commitment was investigated by simple linear regression analysis. This procedure explored the causal effect of the amount of change in the POS on organizational commitment. Next, the role of organizational justice as moderator examined by hierarchical regression analysis. This approach was one of the multiple regressions and was used to investigate how the moderator effect among variables and which extent it effects to other variables (Gray & Kinnear, 2012).

Overall, the author tested the relationship between POS and organizational commitment, and then add organizational justice to explore how moderator effects the relationship between independent variables and dependent variables.

Comparison among Groups

In order to explore the differences among the groups, independent t-test analysis and one way analysis of variance (one-way ANOVA) were conducted. Independent t-test analysis is the test to determine whether there was a significant difference between the means of two groups. In this study, this method was adopted for testing the difference between the groups divided by gender and by the position, which has a managerial position and non-managerial position. One-way ANOVA is a method of analysis to test the difference in means between unpaired groups. In this study, this method was adopted for testing the difference between the groups divided by ages, tenure, and industries which participants belong in.

在文檔中 Moderating Effect of Organizational Justice on the Relationship between Perceived Organizational Support and Organizational Commitment in Japan (頁 29-37)