CHAPTER-3

RESEARCH METHODOLOGY

The research methodology in any research work answers the simple question of ‗how‘,
i.e., how the researcher designed and conducted the study, how the data was collected
and analyzed, how the results were derived, how the validity and reliability of research
was ensured and how the conclusions were reached? It also includes justification for
the choices made in the study on various stages and shows why the chosen methods and
techniques were the most suitable for the study. It ensures realizing the objectives of
the study while providing valid and reliable results.

Any study on occupational stress focuses on the extra and intra organisational stressors
that cause stress to the employees and the impact of stress on other aspects of their
work and life. Extra organizational stressors include factors like, technological changes,
race and social class, family relocation, consumer behaviour, neighbours and society,
standard of living, etc. Intra organizational stressors can be categorised into four
categories, namely, working conditions, organizational design, organizational processes
and administrative strategies. This study focuses on intra-organisational stressors and
the question in focus here is what is happening inside the work-places that is causing
stress to nurses? The relationship it shares with work-life balance and turnover
intentions among nurses. How does it impact their work life balance and turnover
intentions? And how the impact of stress variated among nurses working in various
jobs?

This section describes the research procedure which will include overall research
design, the sampling procedure, a research framework of key variables, the data
collection methods and instruments, the techniques of data analysis and limitations of
the study.

3.1 Research Problem and Aim of the Study

Nursing is a highly demanding occupation which has some stressors that are peculiar to
the nursing profession, e.g. dealing with patients with critical health conditions on a
regular basis, risks of catching contagious diseases, radiations and other health hazards,
handling difficult patients, any mistakes can turn fatal for the patients, handling

59
grieving and worried families, etc. All these conditions lead to additional occupational
stress among nursing staff which causes negative impact on their performance as well
as their organization‘s performance. These stressors and the shift system they have to
work in cause work life balance problems among them. These in totality raise
intentions to leave their jobs, organisations and even the professions among them.
Therefore the aim of the study is to,

Analyse the level of stress, work life balance and turnover intentions among nursing
professionals. Measure the impact of stress on work life balance and turnover intentions
among them. Examine the inter relationship between these three variables and study the
variations in impact of stress among different nursing jobs.

Objectives of the study:

In the light of the aim of the study, the objectives of the study chalked out to be:

1. To examine the relationship between occupational stress and work-life balance

of nursing staff.

2. To analyse the impact of occupational stress on turnover intentions of nursing

staff.

3. To explore the inter-relationship between stress, work-life balance and turnover

intentions of nursing staff.

4. To compare the impact of occupational stress on various jobs in nursing

profession.

3.2 Research Design

The research design is the blue print that helps in making decisions regarding the
collection and analysis of data. It is the overall framework of the study that answers
questions like what information is to be collected, from which sources and by what
procedure. Keeping in view the major objectives and need of the study, this study will
follow an experimental cum diagnostic research design. This study is quantitative in
nature as it analyses the relationship between different variables through hypothesis
testing.

60
Population and Sample design

The universe population of the study is whole nursing staff and the sample population
of this study consists of all the nursing staff working in the Haryana state of India,
irrespective of their demographic or social profiles. Since every research study has its
own constraints, whole sample population cannot be included in this study due to the
time and resource constraints. A representative sample is drawn from the population to
collect data for analysis and inferences are drawn on the basis of the sample results.
The sampling technique used to select elements for sampling is judgemental and
snowball sampling. The questionnaires were distributed among a mix of private and
government hospitals in various districts of Haryana and the hospitals with the work
force of at least 20 nurses were included in the study. Questionnaires were distributed
electronically after some time due to wide spread of COVID-19, therefore no particular
or rigid sampling technique was used later in the study.

Sample Size

On the basis of the sample size formula given by Malhotra and Dash (2011) using the
proportion method, (1- )*z2 / D2, the sample size came out to be of 385 units
(rounded off for 384.16 units) with maximum value of = 0.5 to adjust for variation in
population. A total of 450 questionnaires were distributed at first among various
hospitals in Haryana, out of which 280 came back filled, giving a response rate of
62.2%. After the initial screening of filled questionnaires only 257 were deemed fit to
be included in the study and the rest were rejected due to inconsistency, incompleteness
or other errors with the responses. The number of responses was less than the target of
385 but further distribution of questionnaires by visiting hospitals was deemed risky
due to wide spread of corona. Electronic questionnaires in the form of Google forms
were distributed after some time as the researchers had to take protective measures. 125
responses were received in the digital form and the filled questionnaires were screened
again for suitability, and 111 responses were deemed fit to be included in the study,
thus making the respondent base of 368 participants, which was still less than the target
respondent base for the study. According to Bartlett, Kotrlik and Higgins (2001), if the
study requires regression analysis techniques and other such statistical tests, than the
sample size should be equal to or greater than five times the number of the statements.

61
The number of statements in the present study is 39, which depicts that the sample size
of at least 195 units should be enough. Since this study uses correlation and regression
techniques for data analysis, the respondent base of 195 units should be enough for this
study. Hence, the data analysis process was started with the respondent base of 368
units, and no further data was collected due to the precautionary measures required for
health and safety. The overall response rate (both paper and digital) for the
questionnaires and the geographical Diaspora of the responses collected as a whole
could not be determined as some of the responses could not be picked back up from the
hospitals and a large chunk of questionnaires (unknown number) was also sent through
digital platform due to inability to reach the nursing staff physically in hospitals.

Variables:

Three variables are considered to be studied in this research. These are occupational
stress, employee‘s work-life balance and turnover intentions. In this study the impact of
occupational stress on the other two variables and their counter impact on stress are
studied. The interrelationship of these three variables with each other and the variations
in these variables on the basis of demographic characteristics of respondents is also
studied.

Employee Occupational Stress:

Occupational stress is one of the variables in this study. Occupational stress is caused to
employees due to excessively demanding work routines and practices and limited
resources that affects their job performance and satisfaction. The occupational stress of
nursing staff in Haryana is considered in this study.

Employees Work-Life Balance:

Work-life balance refers to maintaining a balance between employee‘s personal and

professional life, so the employee enjoys both his work and his personal family time.
The work life balance status of nursing staff of Haryana is considered in this study.

Employees Turnover Intentions:

Employee turnover intentions mean the intentions of employees to leave the job and the
organisation in near future. The turnover intentions of nursing staff in Haryana are
considered in this study.

62
Research Model

Occupational Stress

Employees‘ work-life Balance Employees‘ Turnover Intentions

Based on the objectives of the study the above research model will be used to conduct
the study. According to the model the impact of occupational stress on employee‘s
work life balance and turnover intentions will be studied. The counter impact of work-
life balance and turnover intentions on occupational stress and on each other will also
be studied along with the inter-relationship and inter-impact of the three variables on
each other.

Hypothesis:

The objectives of the study helped in hypothesis formulation and testing in the study
for data analysis and reaching conclusions. The major hypotheses for the study are:

1. Hypothesis 1 - The first hypothesis of the study helps in reaching conclusions

regarding the relationship between occupational stress and work life balance
among nurses.

H01: There is no significant relationship between occupational stress and work

life balance among nursing staff.

H11: There is significant relationship between occupational stress and work life
balance among nursing staff.

2. Hypothesis 2 – The second hypothesis of the study is regarding the impact of

occupational stress on turnover intentions among nursing staff.

H02: There is no significant impact of occupational stress on turnover intentions

of nursing staff.

H12: There is significant impact of occupational stress on turnover intentions of

nursing staff.

63
 3. Hypothesis 3 – The third hypothesis of the study tests the interrelationship
between occupational stress, work life balance and turnover intentions among
nursing staff.

H03: There is no significant relationship among occupational stress, work life

balance and turnover intentions among nursing staff.

H13: There is significant relationship among nursing stress, work life balance
and turnover intentions.

4. Hypothesis 4 – The fourth hypothesis compares the impact of occupational

stress among various jobs in nursing profession.

H04: There is no significant variation regarding occupational stress among

various nursing jobs.

H14: There is significant variation regarding occupational stress among various

nursing jobs.

3.4 Data collection, Processing and Analysis

Data collection is required in this study since this research is quantitative in nature.
Primary data are collected to make inferences regarding the variables under study and a
questionnaire is used to collect data from target respondents. Data processing is
necessary to make sense of the data, analyse it and reach conclusions to generalize the
findings of the study. The data in its raw form are hard to read and understand and
therefore it is necessary to process the data by data screening, data cleaning, removing
incomplete responses, tabulation, calculating percentages, etc. to squeeze useful
information out of it.

Data collection Instrument

This research work is based on primary data therefore primary data was collected using
a questionnaire to make any inferences regarding the variables in this study. A
structured questionnaire is constructed to collect data with the help of standard
questionnaires already existing in nursing research literature. The questionnaire is
divided into four parts with first part documenting the demographic characteristics of
the respondents and the other three parts consisting of three scales measuring stress,
work life balance and turnover intentions among nursing staff using a five point

64
Likert‘s scale to measure data with strongly disagree, disagree, neutral, agree and
strongly agree as measure dimensions. The first part, Part-A, of the questionnaire
collects demographic information of the respondents through 8 questions including
name (optional), type of organization, gender, age group, family status, educational
qualifications, experience (in years) and ward/unit of work.

The second part, Part-B, consists of the stress scale used to collect data on stress from
nursing staff. The standard questionnaires used to provide basis to construct stress scale
for this study were the 34 item nurses stress scale by Gary Toft and Anderson (1981),
the english translation of 41 item KEZKAK questionnaire developed by Zupiria et al.
(2003b) used by Xabier Zupiria Gorostidi et al. (2007), the 57 item ENSS by Susan
Elizabeth French et al. (2000) and the 10 item Workplace stress survey developed in
1998 by the American Institute of Stress. A 25 item stress scale was developed and
used in this study to collect data on stress, constructed on the basis of questionnaires
mentioned above. Since the nursing professionals mostly work on a time crunch the
questionnaire was required to be shortened in length to increase the probability of
response and response rate among nurses. The researcher tried to include all the
important dimensions of nursing stress while reducing the number of questions to their
best possible judgement. Some umbrella questions were used to cover a range of
questions while some other questions were either merged or the questions that seemed
to be of repetitive nature in view of the researchers were omitted or replaced to keep the
questionnaire short and concise while still retaining the validity and reliability of the
instrument. The same goes for work life balance and turnover intentions scales.

According to Grzywacz and Carlson (2007), work life balance can also be regarded as
the absence of work life conflict. Since shortening the questionnaire was a major
concern while designing questionnaire for this study, for collecting work life balance
data an 8 item work life conflict scale was developed to measure work life balance with
the help and on the basis of scales developed by Jeremy Hayman (2005), Kakul Agha et
al. (2017), Netemeyer et al. (1996) and Carlson et al. (2000). The positive relationships
with work life conflict scale are considered as negative relationships with work life
balance and vice-a-versa. For collecting data on turnover intentions, a 6 item turnover
intentions scale was constructed with the help of the Bothma and Roodt (2013) TIS-6
scale and 8 item turnover intentions scale by Tanaya Nayak (2016).

65
Data analysis and Statistical Tools

Appropriate statistical tools and latest and appropriate software packages are used for
data analysis based on the objectives of the study. Both descriptive and inferential
statistics are used to process and analyse the data to give data a readable form and reach
conclusions. The data are analyzed through software package IBM SPSS Statistics
2021 and the p-values of less than 0.01 are considered significant in this study. Mean,
variance, standard deviation and alpha values are calculated for descriptive analysis and
other advanced tools are used for inferential analysis in this study to get findings. The
following statistical tools are used to analyse data, interpret the results and reach
conclusions regarding the variables under study.

Cronbach’s alpha: Cronbach‘s alpha helps in establishing reliability of the data

collection instrument. A data collection instrument with an alpha value of more than
0.7 is considered reliable for conducting research.

Correlation Analysis: Correlation analysis is used in this study to establish reliability

and validity of the measuring instrument by calculating inter item and item to total
correlations and making inferences on the basis of the results obtained.

Factor analysis: Confirmatory factor analysis has been used in this study to establish
validity of the questionnaire. If any statement or item in the questionnaire does not load
significantly on any factor than that item is omitted from the questionnaire to ascertain
validity of the questionnaire.

Pearson Correlation: A Pearson correlation coefficient test is used to determine the

strength and direction of relationship between any two variables. It helps to detect if a
linear relationship exists between two variables without indicating any cause and effect
association between the variables. Pearson correlation coefficient analysis is used in
this study to determine inter relationships between the three variables under study.

Regression: Regression analysis is used to determine causal relationships between two

or more variables, i.e., it is used to depict if a dependent variable can be predicted with
the help of independent variables and to what extent. It helps in ascertaining the change
in dependent variable caused by a change in independent variables. Both linear and

66
multiple Regression analysis are used in this study to diagnose the inter-relationships
between the three variables.

ANOVA: ANOVA analysis helps in finding out the variation caused in one variable
based on the characteristics of another variable. ANOVA stands for analysis of
variance and as is clear by name helps in studying variances among variables. It helps
in ascertaining whether the variation in dependent variable based on the segments of
independent variable is significant or not. ANOVA analysis is used in this study to
analyse the significance of variances in impact of stress among various nursing jobs.

Post hoc analysis: Once the significance of variance is established through ANOVA
analysis, a post hoc analysis is used to determine the significance of variation among
different independent factors. In this study, post hoc analysis is used to determine the
significance of variation among different nursing jobs regarding impact of stress in
groups of six. It provides specification and further clarification regarding variances
established through ANOVA test.

3.5 Ethical Considerations

Every research work that has live beings as target respondents should pay attention to
the ethical issues concerning research studies. The researchers should abide by the
ethical considerations while collecting and analyzing the data and should refrain from
doing any such things that breach these ethical conditions related to a study. Unethical
measures of collecting data or any acts on the part of the researcher that can adversely
affect the respondents or the organizations they work for and harm their reputation
should not be entertained while conducting research.

The researcher assures that none of the ethical guidelines prescribed for conducting
research has been violated in this study. The identity of the respondents or the
organizations they work for is not disclosed in any part of the study. The purpose and
objectives of the study were clearly communicated to the respondents before collecting
any responses from them and none of the respondents were forced to provide
information without their will. The anonymity and confidentiality of the respondents
and the data has been maintained soundly, appropriate credit has been given where it
was due and the researcher has made sure to keep the reputation and integrity of the

67
respondents as well as their organizations safe to the best of efforts by strictly adhering
to the ethical guidelines.

3.6 Limitations of the Study

One of the major limitation of this study is the wide spread of Corona Virus around the
world mid-study. The impact of COVID-19 was largest on the health care sector and
the nursing professionals are the target respondents for this study, which made it
impossible to reach nurses for data collection as even performing their routine work
activities became stressful for nurses. The physical collection of data was halted due to
the added risk of Corona and government imposed lockdowns around the country and
digital methods of data collection were used. It resulted in compromising the sampling
design and authenticity of responses of the study. Furthermore the geographical
concentration or distribution of the respondents could not be ascertained due to online
data collection which also affects the findings of the study.

Like any other study, this study also suffers due to time and resource constraints on the
part of the researcher. Another limitation is the process of getting responses from
government and private hospital nurses. The government hospital nurses were not very
willing or were wary to provide honest responses while the private hospital nurses
could not be contacted directly but through the HR Departments of these hospitals. This
can result in instilling biasness in nurses‘ responses, the respondents included in the
study and compromise the credibility of the study.

Since this study is based on Haryana and is limited in its geographical reach, and is
regarding nursing staff only, the study results may not be generalized to other work
settings or professions. Another limitation could be the sample size under study. No
matter how large the sample is extracted, it becomes small in respect to the population
under study and that impacts the inferences made under study.

Lastly, there are three factors that are considered to make any kind of studies on stress
difficult to conduct, interpret and generalize. Firstly, there is no clarity and precision
regarding the concept and definition of stress making it difficult to provide a clear
direction to the study or compare it with other studies. The second one is the choice of a
perspective that‘s appropriate for research. Stress studies can be conducted from the
perspectives of individuals, organizations or a mix of the two. For example the

68
psychological researchers use individual perspective and base their study on personal
traits to study stress while the organizational researchers study the environmental
factors of job stress rather than the individual ones. A mixed approach uses a balance of
both the perspectives to conduct studies and reach conclusions. The last one is the
methodological problems that occur in stress studies. These three problems are inherent
in all kinds of studies on stress however their proportions may vary from study to
study. Efforts had been made by the researcher to overcome these three limitations but
some effects of these may still be there on this study too.

69