3.

Research Methodology
3.1 Introduction
Saunders et al (2009) discuss an onion analogy for an approach to research
design in that the researcher reflects on their epistemology approach to the study
in question through peeling layers (as in an onion). The main layers are
described as research philosophy, research approach, strategy, choices, time
horizon, and methods of data collection. This layered, step-wise approach,
Saunders et al (2009), provided a framework for the approach in this research
project

The research questions and objectives are illustrated in the Introduction section 2
of this report.
3.2 Philosophy
Saunders et al (2009) discuss an onion analogy for an approach to research
design in that the researcher reflects on their epistemology approach to the study
in question through peeling layers (as in an onion). The main layers are
described as research philosophy, research approach, strategy, choices, time
horizon, and methods of data collection. This layered, step-wise approach,
Saunders et al (2009), provided a framework for the approach in this research
project.
Research philosophy is associated with the ideas around how information is
collected and processed, as discussed in Bryman and Bell (2011). Following on
from this the epistemology, in other words what is known about a particular
subject, surrounds the different research approaches. Various research
approaches are possible and further described in Bryman and Bell, such as
deductive/inductive study, positivism, interpretivism, and realism. All these have
advantages and disadvantages which are too complex to consider in detail for
this project. However, the choice of approach requires justification to illustrate
validity.

Because the objectives of the project encompass researching existing literature

and canvassing opinion of employees at Harlan, an overall qualitative approach
has been used combined with an interpretivism strategy, as described in Bryman
and Bell (2011). This was considered appropriate since interpretivism is the
overarching philosophy which lends itself to a qualitative approach. It allows
subjective interpretation in the human sense, eg by the author, and allows
consideration of differences of participants opinion. This is important for this
project from a qualitative aspect as other managers may have a different opinion
to myself and senior management. Differences of opinion can be taken into
account and evaluated when designing a new department.
3.3 Qualitative/quantitative data and strategy

Qualitative and quantitative data are described by Bryman and Bell (2011:
149,150) as:
Quantitative: A research strategy that emphasizes quantification and analysis
of data
Qualitative: A research strategy that emphasizes words rather than the
quantification and analysis of data
There is some controversy in this however as Bryman and Bell (2011) go on to
discuss that some writers consider the differences just in terms of compare and
contrast and that any differences are merely not relevant or useful. One could add
here however that in this research project, there is a distinction as clearly

organisational, factual data (eg quantitative) will be used alongside

questionnaires and opinions (eg qualitative).

In any case, a mixed method approach, as discussed in Creswell (2003), with a

lesser emphasis on quantitative rather than qualitative aspects was employed as
it was desirable to obtain research of both natures because of the need for
factual data and opinions. This was justifiable by the predominant need for
publicised literature reviews, interpretation of the opinions of employees and
since smaller, but equally important aspects of the research require quantitative
data. Examples of this are supporting facts, documents and financial data from
the organisation. Hence, the quantitative aspects provide factual support for the
opinions generated and to support the need for predicted intervention.
In addition, some material has come from personal in-depth knowledge and
understanding of the current situation at Harlan. This, in conjunction with the
research data, both qualitative and quantitative, has assisted in how intervention
can be made, that is as a process of inductive interpretivism. This personal
impact has been tempered essentially by limiting it to the background knowledge
that the author has in knowing that some form of intervention is necessary, and
that a new department could have benefits. Data to support how these are
achieved and in what way has come from research.
3.4 Design and further strategy
Interpretivism-allowingoptionsandconsiderationsofdifferentpointsofview Inductivemoving from observations to broader theories Case study/surveyprocessing,
preparing and presenting information Mixed methodincorporating quantitative
and qualitative data Cross sectionaldata collection from a subpopulation
The onion analogy, Saunders et al (2009) was discussed as an approach to this
research. This is discussed further after the initial step-wise approach is
identified, see below:
More specifically, the research adopted a case study/survey format as identified
by Soy (2006) in 6 steps:
1.

Determine and define the research questions

2.

Select the cases and determine data gathering and analysis techniques

3.

Prepare to collect the data

4.

Collect data in the field

5.

Evaluate and analyze the data

6.

Prepare the report

These 6 steps related to this research approach because they described the
natural progression to a solution of the identified problem. However, one could
argue that step 1 could coincide with step 2 as some cases were identified to
help define and refine the research questions.
A fixed, mixed method approach, that is the study is pre-determined and planned,
Cresswell (2009) was further justified because it was considered that it was the
best solution to understand the issues involved as literature and publication
reviews were required alongside company data and facts, as also discussed in
Bryman and Bell (2011). Some of the advantages of a mixed method research
approach include the ability to better understand particular phenomena
compared to a single method and that if well planned, more information that is
relevant to the study can be obtained, again discussed by Bryman and Bell
(2011). Conversely however Bryman and Bell (2011) discuss a drawback that if
the approach is not well planned and time constraints are not considered, the
depth and accuracy of the content can suffer. Any possible issues with time and
planning were mitigated with the use of a Microsoft Project planning file (see
Appendix 9.2). Finally, the sample study of literature selected was by a compare
and contrast approach. Relevance was considered based on the support or not
to the arguments in question by the author.

3.5 Data collection and selection

Objectives 1 and 2:
Authenticity, credibility, representation (that is reputable sources such as
publications and official internet sources) and meaning with relevance to the
research questions and objectives have been used. This evaluation method of
documents is discussed by Scott (1990). Specifically, advantages and
disadvantages that are relevant to service centralisation and methods and
concepts within change management have been researched. The data collection
method was a Literature review and comparison approach, using qualitative
content analysis, as discussed in Bryman and Bell (2011) mapped to themes,
where possible.
The reason for this choice was to provide a critical comparison of the literature
and publications. The sole participant of this process was the author.

The data collection procedure for the objectives detailed in the introduction
section are shown below:

To research and critically review current business literature and publications to

identify the advantages and disadvantages of centralised support services in
industry.

To analyse methods for change management in literature and publications and

identify key concepts that will guide the approach and support

Objective 3:

To canvas opinion regarding the feasibility of a new centralized department and

manner in which this intervention should be managed.
For this objective, opinions and facts, particularly associated with benefits
envisaged, past experiences, current problems, issues and delays, advantages
and disadvantages have been evaluated. Personal experience from previous
departmental manager roles has also been used.
The data collection method was using responses via e-mail to questionnaires
with a population of approximately 10 within an age range of approximately 35 to
60. The only criteria was that the person must be management level and hold
accountability for a department, because of reasons discussed in the previous
section. A Questionnaire with e-mail response was selected because advantages
such as the same questions could be asked to each participant. Additionally, email response was selected because it allowed managers who are busy to
answer, save and review drafts before completion. An online survey, eg Survey
Monkey, was not favoured because of the expected knowledge limitations with
regards to IT of some managers and the familiarity of e-mail. The main weakness
of this approach was that respondents had to type out their response however it
was personally considered that this forced them to think more critically about
their response. An online survey may have provided more Yes/No answers and
less detail for example. Importantly, the questions were selected so that
categorisation could be made, for example into benefits, advantages and
disadvantages, is intervention required etc. These factors align to the research
method approach better, that is inductive interpretivism.
An additional data source was the current instrumentation supervisor, age ~55.
This time the data collection method was face to face interview because this
person had first hand experience and more in-depth knowledge of the current

situation. Direct 1 to 1 feedback could be obtained, which potentially could yield

more information from this key source rather than a questionnaire. A
disadvantage of this approach was that notes were difficult to take quickly
enough and sound problems experienced with trial voice recording were
intermittent. This was overcome by the interviewee verifying recorded typed
material after the event with counter signature. Again this approach was selected
to align to the research approach.
Examples of the questionnaire and consent form are illustrated in Appendix 9.3.
The questions were determined to allow subsequent grouping of data, see
results section and to try to draw out key points in relation to the new department,
whether it was required, the benefits etc. Additionally, any change management
links or concepts were requested in relation to service centralisation.
In the case of both questionnaire and interview, these were conducted
throughout June 2014 and material is stored centrally on internal secure Harlan
servers in a designated drive, which has login/password protection.

Objective 4:

To outline specific recommendations as to the nature of intervention required

and how that intervention should be executed.
For this objective, evaluation of feedback from objective 3 interviews with
comparison against the authors ideas was performed. Company financial data
and any relevant literature sourced from objectives 1 and 2 was also used. The
data collection method was from research, literature and company intranet and
organisational documents content analysis, as discussed in Bryman and Bell
(2011) with the reason for choice being to bring together data and propose
action. This was performed by the author.
Finally, although not as a specific objective, the instrumentation supervisor and a
selection of local departmental managers based on availability will be consulted
for feedback for phase 3 report. This will be on a face to face basis and notes
recorded. This will help to establish any benefits as a result of the implementation
and support personal interpretation. However, personal interpretation will be
critical as I will be the manager for the new group implemented. Additionally, any
relevant company data, examples such as resource usage time, customer
feedback and QA feedback will be used to evaluate impact on efficiency/bottom
line performance.

3.6 Participants

Sampling is described as the process of selecting units from a population of

interest to generalize results back to the population chosen, Social Research
Methods (2006). The process ensures a manageable size of data and because it
is impossible to usually access all stakeholders, or all the population. However, it
was not possible to select an entirely random or probability sample in the
research for a number of reasons, such as cost, time constraints and the number
of departmental managers available. However, importantly the proposed
participants are considered the best employees to yield relevant information
because of their experiences and stakeholder status in the process. Bryman and
Bell (2011: 176) state that a sample that reflects the population accurately so
that it is a microcosm of the population is a key concept in sampling from a
population. Although the sampling population is limited, the above reasons justify
this. This sampling strategy is described as purposive and more accurately,
homogeneous purposive, in that it is a sample characteristic of a group or people
with similar occupation and therefore advantageous in this project. This sampling
strategy is discussed by Denscombe (2002).
Further specific reasons for these choices were:
Departmental managers: Stakeholders in the process. They possess the
knowledge of tasks that may be required by the new department and are aware
of current issues and problems. A selection of these will also be consulted for
phase 3 to understand any benefits achieved from the implementation.
Instrumentation supervisor: The choice of this person is appropriate since they
are currently coordinating much of the work at present. They have in-depth
knowledge of the current problems, issues and situation and are hands on at
present. Again this person will be consulted for phase 2 and phase 3 reports of
the BIS and this will be in the form of a face to face interview.
Interviews were conducted with local managers in a cross sectional approach.
This was justified because data was collected at one time point from a selection
of employees, designed to detect patterns of association, as discussed in

Bryman and Bell (2011). This will be applicable for any interviews held for phase
2 or phase 3 reports.