Research Questions and Analysis Plan Guidance
Research Questions and Hypothesis
RQ1: What is the relationship between workplace bullying and turnover intention among corrections employees?
Ho1: There is no statistically significant relationship between workplace bullying and turnover intention among corrections employees?
Ha1: There is a statistically significant relationship between workplace bullying and turnover intention among corrections employees?
RQ2: What is the relationship between workplace bullying and organizational commitment among corrections employees?
Ho2: There is no statistically significant relationship between workplace bullying and organizational commitment among corrections employees?
Ha2: There is a statistically significant relationship between workplace bullying and organizational commitment among corrections employees?
RQ3: What is the relationship between workplace bullying and job satisfaction among corrections employees?
Ho3: There is no statistically significant relationship between workplace bullying and employee job satisfaction among corrections employees?
Ha3: There is a statistically significant relationship between workplace bullying and employee job satisfaction among corrections employees?
The instruments for data collection will comprise eight demographic questions and questions from four validated surveys: The NAQ-R, the Minnesota Satisfaction Questionnaire, the Organizational Commitment Scale, and the Turnover Intention Scale. The combination of the answers to these questions and the analysis thereafter will answer the research questions. The survey questions will be divided into sections: one for the demographic questions to collect the participants’ age, gender, tenure, and academic status, and the other three sections will be for the bullying. The bullying questions will be multiple-choice questions and have a 5-point Likert-type scale for answers on the survey for the respondents to select from (a) strongly agree, (b) agree, (c) undecided, (d) disagree, and (e) strongly disagree.
Negative Acts Questionnaire-Revised (NAQ-R) Scale
The NAQ-R is a 22-item questionnaire of negative acts to measure respondents’ perceptions of work-related bullying, person-related bullying, and physical intimidation on a 5-point Likert-type scale: (1) Never, (2) Now and then, (3) Monthly, (4) Weekly, and (5) Daily. The 22 questions in the instrument put in place a threshold of four or more affirmative answers to indicate that employees are being subjected to workplace bullying. Higher scores are indicative of greater levels of hostility in the workplace. The item numbers represent the question numbers in the survey. The work-related bullying subscale has 8 items (1, 2, 3, 14, 16, 18, 19, 21); the person-related bullying subscale has 11 items (4, 5, 6, 7, 10, 11, 12, 13, 15, 17, 20), and physically intimidating bullying has 3 items (8, 9, 22). The NAQ-R provided frequency data for each of the 22 negative behaviors and the overall score, which could range from 22 (respondent never experienced any of the 22 negative behaviors) to 110 (respondent experienced all of the 22 behaviors daily). For this study, I will use a 5-point Likert-type scale where 1 = strongly disagree, 2 = disagree, 3 = neither disagree nor agree, 4 = agree, and 5 = strongly agree to show how participants view bullying in the workplace. The average sum of the three factors of the NAQ-R will be computed and divided by the number of single items to identify bullying. The range value of 22-32 indicates Not Bullied, 33-44 range value indicates Sometimes Bullied, while the range value of 45 or more indicates a Victim of Workplace Bullying.
Minnesota Satisfaction Questionnaire
The MSQ’s short-form general satisfaction raw score will be calculated by adding all of the representative values from each question. Calculating the corresponding values for questions 1-4, 7-11, 15, 16, and 20 will yield the intrinsic and general raw scores, with the remaining questions 5, 6, 12, 13, 14 and 19 yielding the extrinsic raw scores. The items in questions 1-20 will yield the score for general satisfaction. The responses will be scored. Low scores indicate that respondents do not feel very satisfied with their job and/or work environment. A high score on the scale will indicate that respondents are satisfied with their job and/or work environment. After calculating the raw scores, the researcher transforms the raw score percentiles based on each participant’s corresponding norm group. The most meaningful scores to use in interpreting the MSQ are the Percentile scores for each scale obtained from the most appropriate norm group for the individual. Weiss et al (1967) explained that ordinarily, a percentile score of 75 or higher would be taken to represent a high degree of satisfaction; a percentile of 25 or lower would indicate a low level of satisfaction; and scores in the middle range of percentiles indicate average satisfaction.
Organizational Commitment Scale
The Organizational Commitment Scale is composed of, in its full length, 24 items, eight items in each of the following claimed dimensions: affective commitment scale (ACS), continuance commitment scale (CCS), and normative commitment scale (NCS). Each of these scales is scored individually and can be used to identify employee commitment levels within the organization. The items are responded to using a 5-point Likert-type scale, where 1 = strongly disagree, 2 = disagree, 3 = neither disagree nor agree, 4 = agree, 5 = strongly agree. The score for each of the 18 items will be calculated to obtain a sum score using the Likert scale. Then the overall score will also be divided by the 18 items of the model to address an individual’s organizational commitment where bullying prevails.
Turnover Intention Scale
The turnover Intention Scale (TIS-6) is a self-report inventory that measures turnover intention as a global construct on a 5-point Likert scale (1 strongly disagree to 5 strongly agree). Higher scores indicate a greater turnover intention. For scoring, add the item scores to get a total score. The midpoint of the scale is 18 (3 x 6). If the total score is below 18 then it indicates a desire to stay. If the scores are above 18 it indicates a desire to leave the organization. The minimum a person can get is 6 (6 x 1) and the maximum is 30 (5 x 6).
Data Analysis Plan
The collected data will be inputted into an Excel spreadsheet and uploaded to SPSS. In addressing the research questions, I will analyze the collected data using a 2×3 factorial MANOVA to test for the difference in means of the variables of job satisfaction, organizational commitment, and turnover intention. I will use the Box’s M test in SPSS to determine variances. Homogeneity tests. Homogeneity tests assume that the within-group variance matrices are equal, meaning the design is balanced and the robustness of the MANOVA tests is guaranteed. The primary purpose of the two-way MANOVA is to understand if there is an interaction between the two independent variables on the three dependent variables. The analysis will be followed by the multivariate analyses, which are applied to the two independent variables of vertical and horizontal workplace bullying. Levene’s test of equality of error variances will be next to test the equality of variances at the threshold of 5%. I will also test the estimated marginal means to examine the mean values.
To simultaneously test the equality of means from all the responses, there is the need to compare the p-values in the MANOVA test tables for each term to the significance level. The threshold significance level (denoted as α or alpha) of 0.05 usually works well. A significance level of 0.05 indicates a 5% risk of concluding that an association exists when there is no actual association. p-value ≤ α: The differences between the means are statistically significant. If the p-value is less than or equal to the significance level, it can be concluded that the differences between the means are statistically significant. P-value > α: The differences between the means are not statistically significant. If the p-value is greater than the significance level, then it cannot be concluded that the differences between the means are statistically significant. One may want to refit the model without the term.
Tips 1: General
1. Calculate the Percentages and range of the variables in the demographic questions.
2. Establish that bullying exists.
3. Analyze Turnover Intension data.
4. Analyze Organizational Commitment data.
5. Analyze Job Satisfaction data.
6. Answer RQ1
7. Answer RQ2
8. Answer RQ3
9. Carryout a 2×3 Factorial MANOVA (Lateral Bullying and Horizontal Bullying) x (Turnover Intension, Organizational Commitment, and Job Satisfaction) to determine the extent of the relationship, if any.
Note: Horizontal Bullying is bullying by colleagues in the work who are neither supervisors nor subordinates. Horizontal bullying is represented by response 1 in column AF of the excel spreadsheet. Lateral Bullying is the negative act perpetrated by either supervisors or subordinates to the victim. Lateral bullying is represented by responses 2 and 3 in column AF of the excel spreadsheet.
10. Produce transferable Tables and Graphs.
Tips 2: Explanation of the Spreadsheet
1. Columns A–H represent responses from Demographic Questions
A=Age; B=Gender; C=Race; D=Job Tenure; E=Education; F=Marital Status; G=Job Function Level; H=Facility Type
2. Columns I–AF represent responses from Negative Acts Questionnaire – Revised (18 questions).
3. Columns AG-AZ represent responses from Minnesota Satisfaction Questionnaire (20 questions).
4. Columns BA-BR represent responses from Organizational Commitment Questionnaire (18 questions).
5. Columns BS-BX represent responses from Turnover Intention Questionnaire (6 questions).
6. Rows 2-115 represent the valid survey responses that I received (114 in all).
7. Row 1 represent specific question items in the survey instruments.
Tips 3: SPSS Dataset
I performed and produced the SPSS Dataset after my online data collection and cleaning. The cleaning was done in excel spreadsheet and later exported to SPSS Version 28. The dataset I produced has the labels of the variables inputted already to make the job easier.