Chapter 3. Methods and Materials
3.1 Quantitative questionnaire survey
3.1.1 Study subjects and design
The Council of Labor Affairs of Taiwan has conducted a nationwide survey of paid
employees “Survey of Perceptions of Safety and Health in the Work Environment”
every 3–5 years since 1988. The questionnaire survey was conducted along with the
“Human Recourses Survey” of Department of Statistics, Executive Yuan. Subjects for
this study were the participants of the survey conducted in 2013, while it was the first
time question items regarding employees’ collective power were added. The survey was
distributed between March 17th to March 22nd in 2013.
Participants were selected through a two-stage random sampling process. In the
first stage, all districts and villages throughout Taiwan were grouped into strata
according to their levels of urbanization. A random sample of districts and villages was
chosen from each stratum. In the second stage, a random sample of households was
selected within each district or village, and residents of the sampled households who
were currently working as paid employees at the time of survey were identified and
invited to participate in the survey. Self- administered questionnaires were delivered to
the selected households by trained interviewers. After 1 week, completed questionnaires were collected and onsite checking was performed by the same interviewer. The
questionnaire was issued to 28,677 employees, and 25,480 valid questionnaires were recovered with return rate of 88.9%.
My study subjects were participants aged between 25 to 65 years old, and those
who did not answer the question items regarding employees’ collective power were
excluded. The total number of study subjects was 16,449.
3.1.2 Operational definition of study variables
3.1.2.1 Independent variables
Collective power: measured by asking the study participants two questions about
their perception of collective power in their working environment, “(1) in my work
place, there is a labor union or a labor organization to defend the interest of the
employees”; “(2) in my work place, there is a labor union or a labor organization which
is able to influence the policy decision of the enterprise.” (Please refer to appendix
questionnaire section 2, question 9, No.22 and No.23.)
3.1.2.2 Dependent variables
Self-rated health: A single item assessing general self-rated health, “How is your
health at present”, was raised to the study participants. The same question was included
in the universally recognized Short-Form 36 Health Survey. Previous research has
shown that a simple, single item general self-rated health question is a strong predictor
of mortality and use of physician services.
Occupational injuries: A single item self-reported question was used to measure
occupational injuries, “in the past year, did you ever have any injuries or diseases
because of your work?”
Mental health: the Brief Symptom Rating Scale (BSRS-5) was used to measure the
mental health status, including five symptom items of anxiety, depression, hostility,
interpersonal sensitivity/inferiority, and insomnia. BSRS-5 is derived from the 50-item
Brief Symptom Rating Scale and has demonstrated good reliability and validity. A
cut-off score of 6+ for BSRS-5 was determined for psychiatric disorder, with accurate
classification rate of 76.3%.
3.1.2.3 Control variables
The demographic variables included sex, age, and education. The work
characteristics variable consisted of average work hours per week, employment contract,
pay system, industry, type and size of enterprise, employment grades, union
membership and psychosocial work characteristics such as job demands (psychological
and physical), job control, job security, and workplace justice.
Job demands and job control were measured with the Chinese version of Job
Content Questionnaire (C-JCQ). Nine items for the job control scale are about learning
new things, non-repetitive work, creative work, allowing own decision, high level of
skills, freedom to make decision, various tasks, influential opinions and develop one’s
abilities. Job demands includes seven psychological items related to fast work, hard
work, excessive work, insufficient time, concentration on job for long time, hectic work,
and insufficient manpower, as well as one physical item related to physically demanding
work. The JCQ was based on Karasek’s Demand-Control model, which claimed that
high job demands and low job control are one of the main factor for job stress.
The 9-item workplace justice scale consists of three items for distributive justice
(work duties and responsibilities arranged fairly, rewards and benefits arranged fairly
and performance evaluated fairly), two items for procedural justice (employees’
opinions influential and employees well informed in decision making process), two
items for informational justice (information not hidden, information reliable) and two
items for interpersonal justice (supervisors trust employees and supervisors treat
employees with respect).
Table 3-1. Operational definition of study variables
Variable Operational definition Point system/Grouping
Independent variables
Collective power (1) in my workplace, there is a labor union or a labor organization to defend the interest of the employees;
(2) in my workplace, there is a labor union or a labor organization which is able to influence the policy decision of the enterprise
1=strongly disagree
Education (1)illiteracy; (2)self-study; (3)elementary school; (4)middle school; (5)high school;
(6)vocational high school; (7)junior college; (8)university; (9)master; (10)doctor
Junior school or below =1, 2, 3, 4 High school=5, 6
College and graduate =7, 8, 9, 10 Work characteristics variable
Working hours per week How many hours did you work last week? <40; 40-48; >48 Employment contract (1)long-term employment for continuous work; (2)contract for temporary, short-term,
seasonal or specific work; (3)part-time, substitution; without specific employment period
fixed term=2, 3 non-fixed term=1 Pay system (1)fixed salary; (2)performance-based with basic pay; (3)piece-rated without basic pay;
(4)time-based without basic pay
1;2; (3and 4)
Table 3-1. Operational definition of study variables (con.)
Variable Operational definition Point system/Grouping
Industry What is your workplace and what is the main business? Manufacture; construction;
service, others Type and size of
enterprises
How many people are employed at your workplace?
1; 2-9; 10-29; 30-49; 50-99; 100-199; 200-499; >500; government agency
1-99; 100-499; >500; government agency
Employment grades What are your work department, position title, and response tasks? Managers; professionals; non-manual skilled; non-manual low-skilled;
manual skilled; manual low-skilled Union membership Do you join the union? If yes, which type of union is it: (1) craft union (2) industrial union
(3) corporate union?
(1) my job requires working very fast;
(2) my job requires working very hard;
(3) I am not asked to do an excessive amount of work;
(4) I have enough time to get the job done;
(5) my job requires long periods of intense concentration on the tasks;
(6) my job is hectic;
(7) there is no enough workforce at my workplace
1=strongly disagree
Table 3-1. Operational definition of study variables (con.)
Variable Operational definition Point system/Grouping
Physical job demands My job requires lots of physical effort 1=strongly disagree
2=disagree 3=agree
4=strongly agree no=1, 2; yes=3, 4 Job control (1) my job requires that I learn new things;
(2) my job involves a lot of repetitive work;
(3) my job requires me to be creative;
(9) I have an opportunity to develop my own special abilities
1=strongly disagree
Job security My job security is good 1=strongly disagree
2=disagree 3=agree
4=strongly agree no=1, 2; yes=3, 4
Table 3-1. Operational definition of study variables (con.)
Variable Operational definition Point system/Grouping
Workplace justice (1) my supervisor and management trust employees;
(2) information released by my supervisor and management is reliable;
(3) my supervisor and management hide important information from employees;
(4) in my company, employees’ opinions are influential in the company’s decision making;
(5) in my company, employees’ work duties and responsibilities are arranged fairly;
(6) in my company, employees’ monetary rewards, benefits and welfare are arranged fairly;
(7) in my company, employees’ performance is evaluated fairly;
(8) during the process of making important decisions, my supervisor and management inform employees and provide sufficient information;
(9) my supervisor and management treat employees with respect
1=strongly disagree
Self-rated health How is your health at present? 1=excellent
2=very good
Table 3-1. Operational definition of study variables (con.)
Variable Operational definition Point system/Grouping
Occupational injuries In the past year, did you ever have any injuries or disease because of your work? yes; no Mental health (1) I have troubles falling asleep;
(2) I am feeling tense;
(3) I am feeling easily annoyed or irritated;
(4) I am feeling blue;
(5) I am feeling inferior to others
0=not at all 1=a little bit 2=moderately 3=quite a bit 4=extremely
score=Q1+Q2+Q3+ Q4+Q5
3.1.3 Research framework
3.1.4 Research hypothesis
Employees with weaker collective power may report poorer health outcomes:
poorer self-rated health, higher occupational injury rate, and poorer mental health.
Collective power may influence health outcomes through its association with work
characteristics: working hours, job demands, job control, job security and workplace
justice.
3.1.5 Data analysis
Descriptive analyses of collective power by demographic, work characteristic and
health status were performed. Chi-square test was used to examine whether there is
significant difference of collective power among categories of nominal variables; while
t-test was used for continuous variables.
The associations of employees' collective power with poor self-rated health,
occupational injuries and poor mental health were examined by logistic regression
explore the underlying mechanism by which collective power influences these health
outcomes through some intermediate process (Baron & Kenny, 1986). The associations
of collective power and working hours, job demands, job control, job security and
workplace justice were tested with either linear regression model or logistic regression
model. Then collective power and working hours, job demands, job control, job security
and workplace justice were used simultaneously to predict the health outcomes. I
evaluated if the significant associations between collective power and health outcomes
were greatly reduced or became non-significant, when these work characteristics
variables were added into the regression model. Lastly, occupational injuries high-risk
subgroup: male construction employees was selected to compare the association with
the total study population.
SAS 9.4 edition (SAS Institute, Cary, North Carolina, US) was used for all of the
analyses.