Assignment 1: Job Insecurity and Well-being Research Analysis

This assignment analyzes a research study by De Witte et al. (2010) that investigated the associations between quantitative and qualitative job insecurity and employee well-being in the Belgian banking sector. The study explores how employees' concerns about their work-related future impact job satisfaction and psychological distress. The research involved a survey of 15,000 employees from 63 banks, utilizing a simple random sampling method. The assignment delves into the study's methodology, including sample size, sampling techniques, measures of variables (reliability and validity), data collection methods (self-administered questionnaires and interview guides), and the mixed methods research design. The analysis highlights key aspects such as the importance of representative samples, reliability and validity of measurement instruments, and the use of control variables. The assignment also discusses the strengths and weaknesses of the research design and data collection methods, providing a comprehensive overview of the study's approach and findings.

BUACC5931 2019 Semester 1, Assignment 1
Associations between Quantitative and Qualitative Job Insecurity and Well-being
Research questions
De Witte et al. (2010) investigated the association of employee’s perception of quantitative and
qualitative job insecurity with job satisfaction, and psychological distress in the Belgium banking
sector.
Job insecurity is defined as the employees’ concerns about their work-related future. There are
two kinds of job insecurities, the quantitative job insecurity, and the qualitative job insecurity.
The quantitative job insecurity is about the threat to the continuation of the job in the future. The
qualitative job insecurity is about the threat to the various valued aspects of the job, such as job
content or working conditions.
Data collection and respondents
In total, there were 69,000 employees working in the 63 Belgian banks affiliated to the sector’s
joint industrial committee in 2001. As for questioning all employees would be too expensive, the
researchers decided to survey a sample of 15,000 employees (roughly 21%).
All 63 banks participated in the survey. About 21% of employees in each bank were invited to
participate in the survey. Within each bank, the respondents were selected at random with no
particular quota for gender, age or employee level. The survey was based on addresses which
had been provided by the banks (name, language, address) and each randomly selected employee
received a personalized envelope through regular mail, sent to him/her by the employer. The
completed questionnaire needed to be returned (free of charge) through the internal post within
each bank. The researchers travelled to each bank to collect the completed survey.
The sample was representative for employees in the banking sector, however, not for the total
working population. More men (58.5 percent) than women (41.5 percent) participated. About
two in three respondents were between 35 and 44 years old or between 45 and 54 years old,
while about one in four was between 25 and 34 years old. Only a minority (4 percent) was
younger than 24 or older than 55. Most respondents had an education beyond high school (63.9
percent), had partnered with an income and children (72.4 percent), and worked full-time (85
percent). There were about as many white-collar workers (54.4 percent) as executives (45.6
percent).
Measures
Quantitative job insecurity was measured with four items developed by De Witte (2000) on a
scale from 1 (strongly disagree) to 4 (strongly agree). Sample items were “I feel insecure about
the future of my job”’ and “I am sure that I will be able to keep my job” (reverse coded).
Reliability (Cronbach’s alpha) equalled .89.
Qualitative job insecurity was measured with ten items from the 17 item measure that was
originally proposed by Ashford, Lee, and Bobko (1989). These job features concerned four broad
dimensions previously distinguished to describe the various characteristics of a job: job content
(autonomy, skill utilization, and specific tasks), working conditions (workload and quality of
Page 1 of 6

BUACC5931 2019 Semester 1, Assignment 1
working conditions), employment conditions (wage, working hours, and opportunities for
promotion), and social relations at work (relations with colleagues and supervisors, respectively).
Respondents had to indicate whether each of the job features would likely improve or deteriorate
in the near future (1 = strongly deteriorate; 5 = strongly improve). We recoded the items so that a
high score reflected qualitative job insecurity. Cronbach’s alpha equals to .87.
Job satisfaction was measured with one item: “Overall, how satisfied are you with your current
job?” (1 = very dissatisfied; 5 = very satisfied).
Psychological distress was measured with the 12-item version of the General Health
Questionnaire (Goldberg, 1978). A sample item was “Have you recently lost much sleep over
worry?” Responses varied from 1 (“less than usual”) to 4 (“much more than usual“). Reliability
(Cronbach’s alpha) was .89.
Control variables. The following social demographics and work-related factors were included:
gender (0 = men; 1 = women), age (1 = 18–24; 2 = 25–34; 3 = 35–44; 4 = 45–54; 5 = 55+),
education (0 = no education beyond high school; 1 = education beyond high school), extra
income (0 = no partner with extra income; 1 = partner with extra income), children (0 = no
children; 1 = children), occupational position (0 = white-collar worker; 1 = executive), working
hours (0 = part-time; 1 = full-time). The demographics were used as control variables in data
analysis.
Q1: Sample size
Basically, a sample size of less than 30% is not representative enough in social science
and as a result, the findings cannot be generalized to other settings away from the study area,
(Best, & Kahn, 2016). It is an essential component in a scientific process. Due to sampling the
researcher can take what they have learned on a small scale and related it to the entire
population. Generalization is a good thing as it helps save time and money and paints a picture
that represents the entire population. On the other hand, it’s a bad idea as in generalization the
most frequent answer is assumed to be the position of the entire population and it might not be
the case. Additionally, if there was bias in the sampling for example use of snowballing or
volunteers, they may not be representing the views of the entire population.
It was anticipated that the study would utilize a sequential explanatory strategy, starting
with the online emails, which would then be followed by a series of individual interviews with a
sample of the survey respondents. However, this process proved to be an extremely lengthy
endeavor. The employees were sampled for the original dataset.
Unfortunately, these efforts also yielded disappointing results with limited understanding
of job insecurity among employees within the region. As the sampling strategy was conducted in
two overlapping phases, it took some time to elicit the responses from the interviews.
Q2: Sampling method
A sample refers to a small portion of the entire population that will participate in the
study. The findings from this sample are generalized to the entire target population as the sample
is a true representation of the total population. Sampling is therefore, the process where
individuals who will participate in the study are selected. It is not possible to conduct a study in
the entire population especially for a large population, due to constraints both financially and
time factor, therefore a sample is used. For this study, the participants were selected through the
simple random sampling method. This is to give each participant an equal chance to be selected
Page 2 of 6

BUACC5931 2019 Semester 1, Assignment 1
to participate in the study. It will also ensure the validity of the research findings for them to be
generalized to the entire population as a true reflection of the situation in the university. It will
form a heterogeneous sample of employees from different banks.
Furthermore, the simple random sampling technique is advantageous in the sense that the
performed data analysis has a reduced risk of producing error measures now that the sampling
process in one way or the other normally takes given boundaries, hence representing the entire
population. In addition, this type of sampling is not knowledge extensive during research
undertaking since the participant can easily respond to questions within their place of stay
despite the researcher not coming from the same place with the participants. For instance,
employees can easily respond to questions on job insecurity in Belgium even if the researcher
does not come from Belgium.
Moreover, this sampling method demonstrates simple data collection requirements such
as recording answers to the designed questionnaires. However, it may not give the best outcomes
especially if the sampled population is not representative enough to the entire population.
Through this method of sampling, the participants can easily be sampled by giving them yes or
no options. Those who pick yes can be allowed to participate in the study. Therefore, data
collection processes take a short time, unlike other sampling techniques where sampling
processes are too long.
On the other hand, this method of sampling is disadvantageous in the sense that it is rigid
when it comes to exploring knowledge given the fact that the researcher in one way or the other
can be tempted to choose the specific area where they believed that they can get the desired
results thus leading to intentional bias. Due to qualitative interviews, there is need for the
researcher to have experience and professional skills to conduct accurate and quality interviews.
At times, it may not be obvious that the collected data is a representation of the entire population
hence limits its generalizations within the populations.
Q3: Measures of variables
Reliability is the measure of the extent to which the instruments used in research will
give consistent results if they will be subjected to repeat trials, (Koo, & Li, 2016). Usually,
random error influences any reliability in research where an increase in random error results to a
decrease in reliability as well. The error may arise from ambiguous instructions to the subjects,
inaccurate coding, interviewer`s fatigue, interviewee`s fatigue, interviewer`s bias among others.
Again, reliability can be considered as the consistency of an instrument to yield the same results
at different times.
Reliability of the instruments was determined through Cronbach test where questions on
the test were chosen randomly and comparing the result with other half and will have a
significant positive correlation between two halves, then the results were deemed reliable. This
study found a correlation coefficient of 0.8 and greater than 0.6 thus it is considered appropriate.
Hence the measure of the variables had very high reliability indicating very high consistency in
measuring instruments used.
The validity, on the other hand, is the measure of the degree that the empirical measure
represents the concepts that are being studied. In other words, validity is the degree to which
result obtained from the analysis of the data actually represents the phenomenon under study. To
determine the validity of the instruments, normally a pilot study is used in sampled banks that
were not part of the study population. However, this was not done for this study. Therefore, the
items that failed to measure the variable intended were not modified and others discarded. The
researcher failed to use expert advice from supervisors and other lecturers who deal with
Page 3 of 6

BUACC5931 2019 Semester 1, Assignment 1
business matters, by examining the data collection tools thus giving their professional advice.
Therefore, failure to modify and improve the questionnaires were found necessary.
Q4: Collection of data on social demographics
Data collection refers to the empirical evidence gathering to answer the questions under
the study, (Ary, Jacobs, Irvine, & Walker, 2018). The research used self-administered
questionnaires as an instrument for data collection. In this method, the participants themselves
fill in answers to the provided questions, and the researcher comes later to pick the
questionnaires upon completion. A self-administered questionnaire is a way to collect
information on perceptions, beliefs, and attitudes of the population without biases.
Questionnaires are preferred because of no matter the research problem; the participants are free
to give their honest opinions. Interview guides were also used as another instrument for data
collection. Casual inference between variables is always needed as it needs to establish control
over the experiment by use of a random assignment to the sample. This process additionally
helps to control other variables that may have a positive or negative effect on the results. This
additionally ensures that an experiment only one variable will dominate the result. On the other
hand, this can be defined as the process of drawing conclusions about the causal connections
based on the conditions of the effects that occurred.
Q5: Research design
Basically, a mixed methods purpose statement encompasses research methodology that is
commonly used given the fact that it allows integration which is systematic or mixing of data
which is both quantitative and qualitative for either one or multiple investigations or sustained
program of inquiry.
Mixed methods research has been in the past well utilized in the social sciences.
However, its utilization has recently expanded into other disciplines including the health and
medical sciences (nursing, family medicine, social work, mental health, pharmacy, allied health,
and others). According to (Creswell and Plano Clark, 2017), mixed methods research procedures
have been developed and refined to suit a wide variety of research questions and hypotheses.
The main principle of this design is the fact that through the integration, the researcher in
one way or the other can utilize separate quantitative and qualitative data collection and analysis.
Furthermore, the rationale for combining both qualitative and quantitative data is since it will
help in the collection and analysis of both quantitative (closed-ended) and qualitative (open-
ended) data. As a result, this will provide feasible, information-rich data that can enhance
traditional quantitative research approaches.
An example is the utilization of the experimental design, qualitative data, and content
analysis in a study that provide an ideal opportunity for mixed methods studies to contribute to
learning about best practices in how to implement a job security and enhance its effectiveness in
achieving the triple aim outcomes of cost, quality, and employee experience. In other words,
quantitative research will be on the learning about best practices in how to implement
quantitative job security while qualitative research will be based on the employment
effectiveness in achieving the triple aim outcomes of satisfaction, quality, and employee
experience.
The use of a convergent mixed-methods design was deemed to be advantageous in the
context of the evaluation of a small pilot program which outcomes we need to appraise in a short
period of time, and with limited data points. The results of a validated quantitative scale present
estimated value of the program, and the qualitative focus groups and interviews enabled us to
probe participant’s experiences to guide future improvements to the program. The interpretation
Page 4 of 6

BUACC5931 2019 Semester 1, Assignment 1
of the integrated results of both phases allowed the linking of themes regarding the program
reported by participants.
Qualitative methods
The in-depth patient interview involved initial recruitment of a target group by purposive
maximum variety sampling method by age, gender, ethnic group, and family structure variables.
Collection of data entailed an interview with open-ended questions for instance, “which aspect of
the program you found most appealing?”
A unique number is designed for each interview and focus group, which was used to
identify all notes taken, audio files and transcript documents. All notes, consent forms, contact
forms, and audio files were kept with the research investigators always, in a locked room or on a
password-protected computer. The content of the notes, audio recordings or content of
discussions and interviews were not revealed to any person outside the study team.
Transcription
Transcription of the audio files was checked by the research coordinator by listening to
sections of the audio recordings and cross-checked with the transcription.
Quantitative data analysis
After data collection, the questionnaires were coded then data entered into the computer
for analysis. Microsoft Office Excel was used to process and analyze data. Quantitative data
were analyzed using descriptive analysis and inferential statistics. The descriptive statistics were
used to describe and summarize the data inform of tables, frequencies, and percentages. The
inferential statistics were used to help make inferences and draw conclusions. Statistical test
including multinomial logistic regression was used to test the hypotheses. All tests of
significance were computed at α = 0.05.
Qualitative data analysis
Data from interviews were analyzed by using the thematic analysis. During content
analysis, a deductive-inductive approach was used for data analysis of the transcriptions of the
interviews and of the focus groups, as well as the written short answers from the questionnaires.
following the approach described in Green & Thorogood’ “Qualitative Methods for Health
Research”. A deductive coding frame was developed to group the data according to different
parts of the program, such as navigator involvement, first medical appointments, follow-up
visits, etc. Grouped data was then used to develop inductive codes to extract recurrent themes.
These codes were periodically updated as new themes and ideas emerge. The first few interviews
were coded by two members of the research team and analyzed for agreement. The researchers
compared their coding, resolved disagreements and then coded all remaining transcripts of
interviews, focus groups, and key informant interviews.
Integration
After completing both the quantitative and qualitative analysis, a joint display was
created to illustrate how the results of the qualitative phase might explain the results of the
quantitative survey, as well as to highlight some incongruences between both data sets.
References
Ary, D., Jacobs, L. C., Irvine, C. K. S., & Walker, D. (2018). Introduction to research in
education. Cengage Learning.
Page 5 of 6

BUACC5931 2019 Semester 1, Assignment 1
Ashford, S. J., Lee, C., & Bobko, P. (1989). CONTENT, CAUSE, AND CONSEQUENCES OF
JOB INSECURITY: A THEORY-BASED MEASURE AND SUBSTANTIVE TEST.
Academy of Management Journal, 32(4), 803-829. doi:10.2307/256569
Best, J. W., & Kahn, J. V. (2016). Research in education. Pearson Education India.
Creswell, J. W., & Clark, V. L. P. (2017). Designing and conducting mixed methods research.
Sage publications.
De Witte, H. (2000). Arbeidsethos en jobonzekerheid: meting en gevolgen voor welzijn,
tevredenheid en inzet op het werk (Work Ethic and Job Insecurity: Measurement and
Consequences for Well-Being, Satisfaction, and Performance at Work). In van group naar
gemeenschap, ed. R. Bouwen, K. De Witte, H. De Witte, and T. Taillieu, 325–350.
Leuven: Garant. Goldberg, D. P. (1978). Manual of the General Health Questionnaire.
Windsor, UK: NFER-Nelson.
De Witte, H., De Cuyper, N., Handaja, Y., Sverke, M., Näswall, K., & Hellgren, J. (2010).
Associations between quantitative and qualitative job insecurity and well-being: A test in
Belgian banks. International Studies of Management & Organization, 40(1), 40-56.
doi:10.2753/IMO0020-8825400103
Koo, T. K., & Li, M. Y. (2016). A guideline of selecting and reporting intraclass correlation
coefficients for reliability research. Journal of chiropractic medicine, 15(2), 155-163.
Page 6 of 6