3.

4 STUDY INSTRUMENT

3.4.1 Instrument

The instrument of this study was NIOSH Generic Job Stress Questionnaire. The
questionnaire was used to determine the perceived job stressors and its association with
individual factors, non-work factors, and buffer/social factors among the workers.

The NIOSH Generic Job Stress Questionnaire was comprised of 21 forms. But some
of the questions on the questionnaire have been excluded to match it for the study purposes,
i.e. perceived job stressors and its association with individual factors, non-work factors, and
buffer/social factors. Those questions that had been excluded were regarding to job
satisfaction, depression, somatic complaints, and behavioral acute reactions. After excluded,
the questionnaire consists of thirteen parts.

3.4.2 Translation of Research Instrument

NIOSH questionnaire was used in this study has sent to the Languages and Translation
Centre, to translate it into Bahasa Indonesia from its original language English and then it
was translated from Bahasa Indonesia to English to ensure that there is no change in the
meaning of the verse after the translation done.

3.4.3 Reliability

To test the reliability of the translated questionnaire, pilot study had been done among 30
respondents to find out the respondents’ understanding of the translated questionnaire. The
Cronbach’s alpha for the translated questionnaire was 0.84.

3.5 ANALYSIS

3.5.1 Variables Assessment

The assessment of NIOSH Generic Job Stress Questionnaire was by computing the scores
from each question.
a. Independent Variables

1. Individual Factors
a) Age of worker from his current birthday.
b) Marital status.
c) Number of children.
d) Job position.
e) Duration of employment.
f) Type of shift.
g) Self-esteem was assessed by using 10 questions on part XII with 5 point
scales, with reverse questions 2, 3, 6, 7, and 9. After the reverse score has been
calculated and adjusted becoming ordinary scores, the self-esteem is
considered low if the respondent’s total score is less than 31.

2. Buffer Factor/Social Support

Buffer factors/ social support in part XI questions 1 – 4 were broken into three
categories: support from supervisors, support from coworkers and support from
family and friends. Using the 5 point scales. The social support from supervisor,
co-worker, and family is low if the respondent’s total score is less than 13 for
each of social support.

3. The Non-Work Factors Or Non Work Activities

The non-work factors or non-work activities were measured by using seven
questions on form XIII, use yes or no responses to assess non-work-related
activities.

b. Dependent Variable

Job stressors were the dependent variables. The job stressors were present, if more
than 50% of the workers had scored less or more than the median value each of the
stressors depending on the scoring scale. For the example, a high score for mental
demand would suggest it was a stressor, while low score for physical environment
would suggest it was a stressor. The job stressors were:

a) Physical Environment
Physical environment on part III, questions 1 to 8 are assessed by computing
average of item, reverse score 1, 2, 4, 7, 8. After the reverse score has been
calculated and adjusted becoming ordinary scores, the physical environment is
considered lack if the respondent’s total score is less than 9.

b) Role Conflict And Role Ambiguity

Role conflict and role ambiguity on part IV were assessed by using fourteen
questions. These items used a seven-point scale where a score of 1 would be
indicative of role conflict and 7 would indicate no role conflict. Role
ambiguity is also assessed by using a seven-point scale, however the items
were reverse scored, 1 indicating no role ambiguity and 7 indicating role
ambiguity. Eight of the items measured role conflict; the remaining 6 items are
related to role ambiguity. After the reverse score has been calculated and
adjusted becoming ordinary scores, the role conflict is considered low if the
respondent’s total score is less than 57.

c) Interpersonal Conflict/Conflict at Work

The sixteen questions on part V assess group conflict, eight of the questions
address intergroup conflict and eight assess intragroup conflict. The group
conflict items used a five-point scale and both intergroup and intragroup
factors are comprised of a combination of regular and reverse-scored items.
After the reverse score has been calculated and adjusted becoming ordinary
scores, Interpersonal conflict/conflict at work is considered low if the
respondent’s total score is less than less than 49.

d) Job Future Ambiguity

Job future ambiguity was measured using four of the five questions on part VI.
Reverse scoring was used on all of the questions for Job Future Ambiguity.
After the reverse score has been calculated and adjusted becoming ordinary
 scores, the job future ambiguity is considered high if the respondent’s total
score is less than 16.

e) Job Control
Job control aspects are a combination of four different measures, all found on
part VII. All sixteen questions on part VII have a five-point scale and were not
reverse scored (1 is low, 5 is high). Job control is measured by using all
sixteen questions. Task control, decision control, physical environment and
resource control were then comprised of different combinations of these
sixteen items. Task control was consisted of seven items, decision control was
measured using four items and physical environment and resource control
were determined from two questions each. The job control is considered low if
the respondent’s total score is less than 49.

f) Perceived Employment Opportunity

Perceived employment opportunity was measured by using 4 questions in a
five point-scale. No reverse scoring. The perceived employment opportunity is
considered low if the respondent’s total score is more than 12.

g) Quantitative Workload
Quantitative workload was measured in two different manners on part IX. The
factor measures the amount of work the employee has to complete. In one
factor this was done by assessing how fast or slow one must work and how
much time he or she has to think about the job while performing tasks in
questions 1 to 4 in part one. The second factor examines how many
assignments or tasks the person has, how much time he or she has to think and
how often there is a lull in the work in questions 1 to 7, reverse score question
1, 2, 5, and 7. After the reverse score have been calculated and adjusted
becoming ordinary scores, the quantitative workload is considered high if the
respondent’s total score is more than 33.

h) Variance in Workload
Variance in workload is on part IX, consisting of questions 5, 6 and 7. The
items were scored on a 5-point scale and described marked changes in
 workload, the need to think quickly and changes in the level of concentration
required. The variance in workload is considered high if the respondent’s total
score is more than 9.

i) Responsibility for People

Responsibility for the people is on part IX, specifically questions 8, 9, 10 and
11. All of the items used a 5-point scale. The topic of the questions included
the worker’s responsibility for the future, job welfare, morale and welfare of
other people. The responsibility for people is considered high if the
respondent’s total score is more than 12.

j) Utilization of Abilities
Utilization of abilities is assessed by 8, 9, and 10 questions on part IX. These
items also used a five point-scale however all items were reverse-scored. After
the reverse score have been calculated and adjusted becoming ordinary scores,
the utilization of abilities is considered low if the respondent’s total score is
less than 10.

k) Mental Demand
Mental demand is assessed by questions 1-5 on part X. These items used a five
point-scale. Reverse scoring are on questions number 1, 2, and 3. After the
reverse score have been calculated and adjusted becoming ordinary scores, the
mental demand is considered high if the respondent’s score is more than 10.

3.5.2 Data Analysis Plan

A dependent factor would be considered a possible perceived job stressors if at least 50% of
the workers scored the factor as either above or below the median score for that factor. For
each dependent variable, the median score was determined. The proportion of respondent
who scored more or less than the median (depending on type of variable, see 3.5.1 a.) was
also ascertained. If the proportion of the respondents exceeded 50%, this dependent variable
was considered as a stressor. Subsequently, this stressor subjected to bivariate analysis as
well as logistic regression to investigate any association with various independent variables
(individual factors, buffer factors, and non-work factors). The associations were significant if
the p value was less than 0.05. It could consider as risk factors if prevalence odds ratio more
than 1 and protective factors if prevalence odds ratio more than zero but less than 1. This
whole process was repeated for each of 11 dependent variables.