MGM4105 Final exam format: ● Does employee tenure with a

company influence productivity?

Part A: 40%; 40 MCQ (all chapters: 1-14)
Scientific Method
Part B: 60%; Case study with 6 short essay questions ● The way researchers go about using
(chapters 4, 5, 10, 11 and research proposal knowledge (from literature) and evidence
development process(chap 2?)) (data collected from target respondent) to
reach objective conclusions about the real
world.
Chapter 1 ● The analysis and interpretation of empirical
Business research is the application of the scientific evidence (facts from observation or
method in searching for the truth about business experimentation) to confirm or disprove prior
phenomena. conceptions

The process includes:

● idea and theory development
● problem definition
● searching for and collecting information
● analyzing data
● communicating the findings and their
implications

This definition suggests that business research

information is:
● not intuitive or haphazardly gathered
● accurate and objective
● relevant to all aspects of the business
● limited by one’s definition of business

Not-for-profit organizations and governmental

agencies can use research in much the same was as
managers in for-profit organizations.

Applied and Basic Business Research

The decision-making process associated with the
Applied business research conducted to address a
development and implementation of a business
specific business decision for a specific firm or
strategy involves four interrelated stages:
organization.
Example:
1. Identifying problems and opportunities
● Should McDonald’s add Italian
2. Diagnosing and assessing problems or
pasta dinners to its menu?
opportunities
● Which health insurance plan should
3. Selecting and implementing a course of
a business provide for its
action
employees?
4. Evaluating the course of action
- Evaluation Research
Basic business research/ pure research conducted
The formal, objective measurement and
without a specific decision in mind that usually does
appraisal of the extent a given activity,
not address the needs of a specific organization.
project, or program has achieved its
● Attempts to expand the limits of knowledge
objectives.
in general.
● Not aimed at solving a pragmatic problem.
- Performance Monitoring
Example:
Research
● Do consumers experience cognitive
Research that regularly, sometimes
dissonance in low-involvement
routinely, provides feedback for evaluation
situations?
and control of business activity
The determination of the need for research /When account of the research, the research report
is business research needed? or thesis.
1. Time constraints ● You will be able to refer back to the research
2. The availability of data diary for inspiration and ideas for the writing
3. The nature of the decision to be made and this will save you a lot of time in the
4. Benefits versus costs (the value of the writing process, as well as radically
research information in relation to costs) improving the formal written account of the
● Will the payoff or rate of return be research.
worth the investment? ● The research diary is a notebook which you
● Will the information improve the have designated your research diary and set
quality of the managerial decision aside for use solely as your research diary.
enough to warrant the expenditure? ● Any hardback notebook will do for the
● Is the expenditure the best use of research diary.
the available funds? ● It needs to be a hardback notebook because
it will be used a lot throughout the research
project and you don’t want it to fall apart and
Business Research in the 21st Century you certainly don’t want to lose any pages
● Communication Technologies from it.
- Always “connected”, time, place,
and distance are irrelevant. The Test of Research-ability
- Decreases in information ● In order to assess whether or not you could
acquisition, storage, access, and undertake and complete your research
transmission costs. project, it is a good idea to apply ‘the test of
● Global Business Research research-ability’. In this simple test you
- Business research is increasingly examine whether or not you have the
global. resources to complete the research project.
- Must understand the nature of ● The time needed to conduct the research: to
particular markets. design the project, carry out the field work,
- Cross-validation; Verifying that the analyse the data, write up your findings,
empirical findings from one culture draw conclusions and make
also exist and behave similarly in recommendations
another culture. ● The money needed to conduct the research.
In general with small-scale research projects
Business Research Defined if any money is needed it is the amount of
● Business research is the application of social money required for the fieldwork, for
science research methods in the process of example, for posting questionnaires, for
examining business phenomenon. travelling to interviews and for organising
● The term ‘research’ connotes patient study focus groups;
and scientific investigation. The researcher ● Access to data. Many students
carefully examines data to discover all that is underestimate the difficulties that
known on the topic. researchers can encounter in accessing
● Business research is an essential tool for data, in attempting to access data, in
management in virtually all problem-solving securing access to data, and in maintaining
and decision-making activities. access to data over the time period required
to complete the fieldwork.

Keeping a research diary

● It is important to begin from the start of the
research project to record your thoughts,
ideas, inspirations, references and resources
in your research diary.
● This written, often scribbled, record of your
thoughts and decisions will be invaluable to
you when you are writing the formal written
Chapter 2: Research Process, Research Proposal - Techniques or procedures used to analyze
and Ethical Consideration empirical evidence in an attempt to confirm
or disprove prior conceptions.
● Suggested steps:
1. Assess relevant existing knowledge of
phenomenon
2. Formulate concepts and propositions
3. State hypotheses
4. Design research to test the hypotheses
5. Acquire empirical data
6. Analyze and evaluate data
7. Propose an explanation of the phenomenon
and state new problems raised by the
research

Research Process
● The research process follows a pattern, and The Scientific Approach
that pattern does not vary. The model of the ● Decisions need to be based on organised,
research process, detailed in earlier slide, systematic, data-based, critical, objective
illustrates this pattern. and scientific enquiry for specific solutions.
● The structure of the research project too ● Scientific focus on solving problems or
follows a pattern or a convention, and that finding opportunities through a step-by-step
pattern or convention doesn’t change. The logical, organised and rigorous method to
steps are sequential, as indicated in the identify the problems or opportunities, gather
model of the research process. data, analyse them and draw valid
● Research projects are very organic, they are conclusions from them.
living and growing entities and they change ● Helps researchers to state their findings with
all the time. Generally the changes happen accuracy and confidence.
in tiny shifts in emphasis, rather than giant ● Tends to be more objective than subjective
changes or u-turns. In any case, any and all and helps managers and business people
changes affect the entire project, and so the highlight the most critical factors at their
researcher is constantly engaged in a organisations that need specific attention so
process of editing the thesis or report as it as to avoid, minimise or identify and solve
develops, in order to ensure that all the problems.
changes made are properly incorporated into ● Scientific investigation and managerial
the thesis and all of the changes made ‘fit’ decision-making are integral aspects of
properly into the thesis. effective problem-solving.
● The thesis is the written record of the ● The term scientific research applies to both
research project. It is a synopsis of all of the basic and applied research.
work carried out. As the word count is ● Applied research may or may not be
always quite constrained, it is not possible generalizable to other organisations,
for this written record to be anything more depending on the extent to which differences
than a synopsis of all of the work that went exist in such things as size, nature of work,
into the research project. characteristics of the people and or structure
of the entire organisation.
The Scientific Method ● Has to be an organised and systematic
● Scientific Method process where problems and opportunities
- A set of prescribed procedures for are carefully identified, data, scientifically
establishing and connecting theoretical gathered and analysed, and conclusions are
statements about events, for analyzing drawn in an objective manner for effective
empirical evidence, and for predicting events decision-making.
yet unknown.
When a Researcher Does Not Apply the Scientific 3. Testability - Developing a set of research
Approach questions or hypotheses that can be
● Does not warrant investment of these scientifically tested
resources to do thorough, rigorous scientific - Research should be based on testable
research. assumptions/hypotheses developed after a
● When a decision needs to be made quickly, careful study of the problems involved.
lack of knowledge and other factors can - The scientific research should enable the
prompt businesses to try and solve problems testing of logically developed hypotheses to
based on hunches. see whether or not the data collected
● Rash decisions on critical matters may result support the hypotheses developed.
in failed implementation of plans.

The 8 Hallmarks of Scientific Research 4. Replicability – Results can be supported

1. Purposiveness – Solid aim or purpose again and again when the same type of
The research is conducted with a purpose. It research is repeated
has a focus. - Research findings would command more
- The purpose of the research should be faith and credence if the same results are
clearly mentioned in an understandable and evolved on different set of data.
unambiguous manner. - The results of the test hypothesis should be
- The statement of the decision problem supported again and again when the same
should include its scope, its limitations and type of research is repeated in other similar
the precise meaning of all words and terms circumstances.
significant to the research. - This will ensures the scientific nature of the
- Failure to mention the purpose clearly will research conducted and more confidence
raise doubts in the minds of stakeholders of could be placed in the research findings.
the research as to whether the researcher - It also eliminates the doubt that the
has sufficient understanding of the problem. hypotheses are supported by chance and
- Factors involved in purchasing socially ensures that the findings reflect the true
responsible products → Increased state of affairs
purchase intention for socially
responsible product → increased sales 5. Accuracy – Precision and confidence
of socially responsible products - In management research the findings are
seldom definitive due to the fact that the
2. Rigor - Based on a solid theoretical base universe of items, events or population are
and sound methodological design using not taken as such but based on sample
appropriate instruments to meet the drawn from universe. There is a probability
stated objectives that the sample may not reflect the universe.
- Rigor means carefulness, Measurement errors and other problems are
scrupulousness and the degree of bound to introduce an element of error in the
exactness in research investigation. findings. However the research design
- In order to make a meaningful and should ensure that the findings are as close
worthwhile contribution to the field to the reality as possible so that one can
of knowledge, research must be have confidence in the findings.
carried out rigorously. - Precision refers to the closeness of the
- Conducting a rigorous research finding to ‘reality’ based on sample. It reflects
requires a good theoretical the degree of accuracy or exactitude of the
knowledge and a clearly laid out results on the basis of the sample to what
methodology. exactly is in the universe.
- This will eliminate the bias; facilitate - The confidence interval in statistics is
proper data collection and analysis, referred here as precision.
which in turn would lead to sound - Confidence refers to the probability that the
and reliable research findings. estimation made in the research findings are
correct. It is not enough if the results are
precise but it is also important to claim that
 95% of the time the results would be true
and there is only a 5% chance of the results 8. Parsimony - Simplicity explaining
being wrong. This is known as confidence complex situations
level. If the precision and confidence levels - Research needs to be conducted in a
of the research findings are higher then the parsimonious i.e. simple and economical
findings of the research study would be more manner. Simplicity in explaining the
scientific and useful. Precision and problems and generalizing solutions for the
confidence can be attained through problems is preferred to a complex research
appropriate scientific sampling design. framework.
- Economy in research models can be
achieved by way of considering less number
of variables leading to greater variance
6. Objectivity - Conclusions drawn based on rather than considering more number of
the interpretation of results and facts variables leading to less variance.
based on findings derived from the - Clear understanding regarding the problem
results of analyzing actual data and the factors influencing the same will lead
- Research finding should be factual, data to parsimony in research activities. The
based and free from bias. sound understanding can be achieved
- The conclusion drawn should be based on through structured and unstructured
the facts of the findings derived form the interview with the concerned people and by
actual data and not on the basis of undertaking a study of related literature in
subjective or emotional values. the problem area.
- Business organizations will suffer a greater
extent of damage if a non-data-based or Social Science Research
misleading conclusion drawn from the The scientific research in management area cannot
research is implemented. fulfill all the above-discussed hallmarks to the fullest
- Scientific approach ensures objectivity of extent. In management research it is not always
research. possible to conduct investigations that are 100%
scientific like in physical science as it is difficult to
collect and measure the data regarding feelings,
emotions, attitudes and perception. It is also difficult
7. Generalisability - Scope of applicability of to obtain representative sample; these aspects restrict
research findings in one organizational the generalizability of the findings. Though it is not
setting to another organisational setting possible to meet all the above said characteristics of
- It refers to the scope of applying the the scientific research, to the extent possible the
research findings of one organizational research activities should be pursued in the scientific
setting to other settings of almost similar manner
nature. The research will be more useful if
the solutions are applicable to a wider range. Problems in Conducting Research in Business
- The more generalizable the research, the Using the Scientific Approach
greater will be its usefulness usefulness and ● The inability to meet all eight hallmarks of
value. However it is not always possible to scientific research.
generalize the research findings to all other ● Difficulties in the measurement and
settings, situations or organizations. collection of data when trying to quantify
- For achieving generalizability the sampling human behaviour and perceptions.
design has to be logically developed and ● Obtaining a truly representative sample.
data collection method needs to be very ● Comparability, consistency and wide
sound. This may increase the cost of generalisability are often difficult to achieve.
conducting the research. Inmost of the cases
though the research findings would be based
on scientific methods it is applicable onlyto a
particular organization, settings or situations.

Ethics and Business Research

● Research is conducted morally without
compromising the rights of any parties
(especially the respondents of the research).
● Researcher has a key role to play.
● The researcher has to exhibit and practice
good conduct throughout the research
endeavor.
● The ethical researcher is informed. S/he
must be organised and systematic.
● S/he is sensitive to the people involved in the
investigation and engages with them, and
with the entire research process, in an
ethical manner.
● The quality of every research project is
dependent upon the integrity of the
researcher.
● The researcher in reporting the research
details what s/he set out to do, how s/he did
it, what s/he found, and what that means.
Obviously, the account of the research must
be accurate and honest. It must give enough
detail for to allow the reader to evaluate the
work. The literature review must be
comprehensive, complete and up-to-date.
The data must be properly gathered,
properly managed, and analysed
appropriately. The conclusions drawn from
the research must be drawn from the
findings of the research, and these findings
must emerge from the data gathered.
● Every researcher has responsibilities to a
number of constituencies. These
constituencies include the institution within
which and/or under whose auspices the
research is being carried out; and all
individuals, groups, institutions and
organisations participating in the
research.
● The researcher has a duty of care towards
all participants, and must, above all, do no
harm.
● All participation in your research should be
on an informed, voluntary basis.
Chapter 3 Theories, Concepts and Variables ○ A concept that is not directly
observable or measurable, but can
What is a Theory? be estimated through proxy
Theory: A formal, logical explanation of some events measures.
that includes predictions or how things relate to one
another.

Goals of Theory
- Understanding
- Predicting

Research Concepts
Concept (or construct): A generalized idea about a
class of objects, attributes, occurrences or process
that has been given a name.
Examples:
● leadership
● morale
● gross domestic product
● assets Research Propositions and Hypotheses
● customer satisfaction ● Propositions
● market share ○ Statements explaining the logical
linkage among certain concepts by
Ladder of Abstraction asserting a universal connection
● Ladder of Abstraction between concepts.
○ Organization of concepts in ○ Example: Treating employees
sequence from the most concrete better will make them more loyal
and individual to the most general. employees.
● Hypothesis
○ Formal statement of an unproven
proposition that is empirically
testable.
○ Example: Giving employees one
Friday off each month will result in
lower employee turnover.

○
● Abstract Level
○ The level of knowledge expressing
a concept that exists only as an
idea or a quality apart from an
object.
● Empirical Level ●
○ The level of knowledge that is
verifiable by experience or
observation.
● Latent Construct
Empirical Testing
● Empirical Testing
○ Examining a research hypothesis
against reality using data.
● Variables
○ Anything that may assume different
numerical values.
○ The empirical assessment of a
concept.
● Operationalizing
○ The process of identifying the
actual measurement scales to
assess the variables of interest.

Constructs
● Variables are derived from constructs.
● Constructs are ideas that are abstract in
nature which are used to understand
theories. The constructs are converted to
Theory Building variables in order to make them measurable.
● Deductive Reasoning ● Therefore, it can be seen that constructs
○ The logical process of deriving a help to explain theories while variables help
conclusion about a specific instance to make the constructs measurable.
based on a known general premise
or something known to be true. Main Types of Variable
○ Deductive Approach - take ● Dependent variable
universal arguments (Theory) and ○ Dependent variables are variables
arrive at a more micro conclusion. which the researcher is concerned
○ The Deductive Approach Proess about in the study. The aim of the
= Theory → Develop hypothesis researcher is to determine if the set
→ Collect and analyze data → of independent variables are able to
Reject/ not reject hypothesis better explain the dependent
● Inductive Reasoning variables.
○ The logical process of establishing ● Independent variable
a general proposition on the basis ○ Independent variables are variables
of observation of particular facts. which the researcher can
○ Inductive Approach - build a manipulate or control.
conclusion based on smaller truths ● Moderating variable
(observation or data) or a set of ○ Moderating variables are those
beliefs and formulate a universal variables that either strengthen or
conclusion. weaken existing relationships
○ The inductive Approach between independent variable s
Process= Observe Phenomena and dependent variables.
→ Analyze patters and themes ● Mediating variable
→ formulate relationship → ○ Mediating variables, as the name
develop theory denotes, act “in between” the
independent and the dependent
variables.
Dependent variable (DV)
- Is of primary interest to the researcher. The
goal of the research project is to understand,
predict or explain the variability of this
variable.

Independent variable (IV)

- Influences the DV in either positive or
negative way. The variance in the DV is
accounted for by the IV.

Moderating variable
- Moderator is qualitative (e.g., gender, race,
class) or quantitative (e.g., level of reward)
variable that affects the direction and/or
strength of relation between independent
and dependent variable.

Mediating Variable
- surfaces between the time the independent
variables start operating to influence the
dependent variable and the time their impact
is felt on it.

Need for a Theoretical/cenceptual Framework

● A theoretical framework is essential in
research.
● This is because the theoretical framework
provides a schematic view of the constructs
and variables under investigation and the
interrelationships among them.

Hypothesis
● Good hypothesis:
○ Must be adequate for its purpose
○ Must be testable
○ Must be better than its rivals
● Can be:
○ Directional
○ Non-directional
Chapter 4 Research Problem, Objectives,
Significance of Study and Hypothesis Developing an effective research question
Development Three questions must be answered
1. What is the research topic
The Problem Definition Process - General area (innovative behavior)
● First step in ensuring research is relevant 2. What is the research question
● If not done correctly, remainder of research - Specific focus of the research
process will be useless “Garbage in, - One-sentence question (how do personal
garbage out” value types affect unemployment rates?)
● Must begin with a clear understanding of the 3. What is the research hypothesis
reason for undertaking research – focus on a - Proposed answer to your question
decision – no decision no research (Individuals emphasizing stimulation and
● Should not proceed with research until self-direction value types are more likely to
problem defined engage in innovative behavior)
● Identification of research problem opens up
new questions and possibilities. How to start?
● It is an opportunity to test new ideas or re- 1. Choose general topic
interpret old problems from a fresh 2. Start reading the literature
perspectives. ● What has been done
● What questions remain
● Are there contradictions in the literature
How to identify research problems? There are 3. Problem statement: gaps in the literature
THREE important sources; your study intends to address
1. Experience and observation (e.g. However, no known similar research has been
● Abnormal conditions or behavior conducted in Malaysia to identify value types related
● Complaints, protests, etc. to innovative behaviors among Malaysians.
● Deviation from plans/targets 4. Research Question: problem statement in
● Communication with leaders, question form
employees, customers, etc. (e.g. What are the personal value types that motivate
● Research findings and reports innovative behaviors among Malaysians?)
● New trends and policies
How to find gaps in the literature?
● Introduction to articles
2. Deductions from theory Every article will identify the research question and
● Questioning the general principles how it differs from other research in the introduction
● Questioning the assumptions section
● Translating into hypotheses (declarative; ● Conclusion of articles
represents the act or state as an objective Every article will list its weaknesses and areas for
fact ) further study in the conclusion section
● From hypotheses to problem statements or
questions (interrogative; Asking a question) Problem Statement
● Problem specified in terms of how, what,
which, where, when & why type question(s)
3. Literature review ● Adds clarity to the problem & also focuses
● The problem is perceived by the scientific on what issues that have to be resolved
community as relevant and significant ● An internally consistent logical argument
● The need to replicate the problems and having structure, sequence and rationale
confirm previous findings ● A problem statement necessarily leads into
● Questions applicable to other situations at least one central research question or
(countries) objective from which numerous research
● Identifying important variables (dependent questions and/or hypotheses could be
and independent variables) generated.
● Improving previous problems with greater ● You need to be able to clearly answer the
precision and clarity (refining) question:
 - What is the problem”?
- Why is this problem worth my A testable hypothesis
attention ● There exists an acceptable technique and
● The problem statement limits scope by necessary instruments to conduct the test
focusing on some variables and not others. - require quantitative data
● It provides an opportunity for you to - test statistically based on empirical
demonstrate why these variables are data collected through the research
important process
● The explanation is plausible given the
existing body of knowledge
Characteristics of a good problem statement
● Clarify exactly what is to be solved Usefulness of Hypotheses
● Limit the scope of the study to a specific ● Provide direction, propose explanations and
question extend knowledge
● One that is researchable ● Allow relational statements to be tested
● Guide collection and interpretation of data
Example ● Provide framework for organizing and
Topic: Customer satisfaction dimensions of a reporting the conclusions of the study
bank in Malaysia
What a good hypothesis means
General Research Questions ● Must have explanatory power, an ability to
- Are there differences in various aspects of explain relationships
customer satisfaction dimension across ● Must state expected relationships between
groups? variables
● Must be testable, relating variables that are
Specific Research Questions measurable
- Is there a satisfaction difference between ● Consistent with the existing body of
male and female customers? knowledge, not contradicting established
- Is there a satisfaction difference across theories and laws of nature
ethnic groups? ● Must be concise and clear
- Are more educated customers experiencing ● Probabilistic in nature, accept or reject, but
lesser level of satisfaction? never to prove to be true

Significance of Study
Research Objectives ● Justification & Importance of your Study
● Precise statements of what the research - Identification of perceived needs and
project will attempt to achieve constraints
● Specific statements of information required - Set the scene by explaining the current
to address/answer research problem situation & the importance of your study in
● Should be limited to a manageable number the context of the target areas / groups
(2-3) - List of target groups / beneficiaries
● If too many, the study is too large to - eg: parents and teachers in education
complete within a given time economics, central bankers in monetary
Research Hypothesis policy, and residents and businesses
● A suggested solution to a problem, an affected by pollution
intellectual guess
● Unproven statement of a research problem ● What is NEW in your study?
in a testable format ● How can you close the gap of any lacking in
● More specific, & declarative and closer to the the current literatures?
ground than problem statements ● Who will be benefited and How?
● A research tool to relate theory to ● Theoretical/ Practical
observation and observation to theory
● It states relationships among variables which
are operationally defined and observable
Chapter 5 Reviewing the Literature The purpose of Literature Review
● To know what is already known about your
What is a Literature Review? research area
● Provides a context for the proposed study ● to avoid repetition of research that have
and demonstrate why it is important and been done by other researchers.
timely. ● A comprehensive and complete literature
● This chapter needs to clarify the relationship review will
between the proposed study and previous ● help to identify important & relevant variables
work conducted on the topic. for the research
● a review of the most relevant, recent and ● Help in the selection & evaluation of
scholarly work in the subject / topic area. appropriate theories, hypotheses & methods.
● a piece of writing that supports, compare, ● Literature review can provide guidance to the
evaluates and critiques your research topic. formation of new research, it may suggest
further research questions.
Why Review the Literature? ● It may help you to develop an analytic
● When you review the literature, you are framework.
trying to create your own sense of what is ● Gives you the idea on specific terms to be
known and what is not known about the used in a research and how it is clearly
subject. defined.
● Example
- A student wanted to explain what In short, the purpose of a literature review is to:
factors determine contributions to ● establish a theoretical framework for your
charities. topic / subject area.
- Understandably, he initially thought ● define key terms, definitions and
that little or no management terminology.
research would have been done on ● identify studies, models, case studies etc
this topic. that supporting your topic.
- See what you can come up with in ● define / establish your area of study, i.e your
just a few minutes via a quick on research topic.
line journal database search.
If someone has done a similar analysis
● Educate yourself ● explain how yours is different.
● Educate your readers ● Explain their findings, and then follow up with
● Demonstrate that your knowledge is current. what you expect to find in your own
● Be able to show how your research fits in the research, and compare.
broader scheme of knowledge ● Highlight your contribution, in what way
● You will use your literature review to yours complement or close the gap on the
familiarize yourself with : existing ones.
- common research questions in the subject
(introduction)
- management models used to answer related Search preparation & planning
research questions, 1. Define the topic
- empirical methodologies common in the field 2. Think about the limits of the topic
(empirical methodology), 3. Identify the main reference tools
- Types of data you may use in your analysis 4. Think about housekeeping
(qualitative versus quantitative data 5. Plan the sources to be searched & start
description), search
- how to report your results (empirical
analysis), and
- how to identify your contribution relating to
the research question/subject you are trying
to address (conclusion/analysis).
 The sources of literature search: ● Conceptual Paper- Creates or extends a
1. Papers published in refereed journals theory and derives implications No test of
2. Master’s & doctoral theses model
3. Books (academic) ● Empirical- Testing or quantifying a model or
4. Government, industry or occasional reports theoretical analysis. Majority of paper.
5. Non-refereed journals, conference papers
and undergraduate project paper, working The Usual Set-up for Empirical Research
papers 1. Introductory material- Problem statement,
6. Magazines, pamphlets and newspaper motivation, very brief lit review
clippings (must be avoided) 2. Analysis of problem- Apply a
model/theoretical framework to understand
● As the rule, peer-reviewed journal problem
publications are of higher quality. 3. Empirical analysis- Test the
● for the academic dissertation, the main model/theoretical framework
source should be journal. 4. Concluding discussion- Explain results,
● This is because the most up to date research put in context, develop implications
& debates only found in the journals.
● In an academic research, no newspaper 4 Steps in Reading research articles
cuttings or magazines citations are allowed Step 1 - reading with & for a purpose
in literature review chapter/section. - Reading with and for a purpose helps you to
● This is because, newspaper & magazines quickly locate, evaluate and read relevant
reports they are not subjected to rigorous research.
peer review. - There are 2 stages in this step.
- First stage
Popular vs Scholar Publications - Look at the structure of the text, e.g
the Table of Contents, the Abstract,
Popular publications Scholarly Publications
headings and sub-headings, to see
if the text is suitable for your
are addressed to a general are addressed to a specialized
audience, namely, experts in purpose.
audience
the field - This will help you to locate relevant
research quickly.
are almost always secondary are often primary sources of - Second stage
sources of information information - If the text appears to be relevant for
your purposes, then you can read it
articles in Time magazine are original publications of in-depth to find specific research to
summarizing articles in the research studies; support your literature review.
Academy of Management articles in the Academy of
Review Management Review - These techniques enable you to identify
appropriate material by reading widely and to
gain a broad understanding of the available
Examples of Scholarly Journals
literature on your topic.
● Academy of Management Review
- Objectivity, you must be objective in your
● Administrative Science Quarterly
research.
● Harvard Business Review
- Remember the reason you are
● Journal of Management
reading is to be able to give an
● Journal of Business
evaluation and critique of the
● Journal of Management Studies
literature chosen.
● British Journal of Management
- Do not just select the parts of the
● Tourism Management
literature that agree with what you
think is right.
Types of Research Articles
● Meta analysis articles - like literature
Step 2 - evaluating your readings…
reviews, e.g. it reviews 50 articles research
done within a topic for past 10 years,
 ● It is important to critically evaluate your - It is advisable to have some references that
readings to establish their relevance and are current (e.g. 2015 or later)
credibility for your research topic. → Coverage
● you develop your theoretical framework and - Is the information complete?
your literature review on what you have read. - Based on your research so far, does the
● Here is a check list of critical questions to information appear to cover the area being
keep in-mind when you evaluate your studied?
readings: - What is the sample size? Is it adequate?
→ Authority - Is there any further research that has not
- Who conducted the research? been mentioned or deliberately omitted from
- Is the author an authority in their field of the findings?
study?
- What evidence is there to support this? Step 3 - summarizing, analyzing & organizing your
- cannot simply assume that all information / readings
research available on the Internet - It is important that you make notes as you
→ Validity read.
- Where has this research come from? - You should think about and include the
- Is it from a valid source, e.g an educational following in your notes:
institution? - What are the main points / theories / key
- Is it peer reviewed or been passed by an issues/key findings raised in the book or
editorial panel, e.g is it in a refereed journal? article?
- If it is from a website, does it contain details - Take details of any quotes, or page
of author, is it from a .edu site, does it have a references that you think may be good to
publication date? use in your literature review.
→ Accuracy - You can compile articles you read in a table
- What is the literature about? (author, year, variables, country, findings)
- Is the literature accurate and how do you and present them in Literature Review
know? One way to find out is to check if the chapter as “selected studies on…”
same research is referred to in other - Make sure you keep track of all bibliographic
sources, or is it inconsistent with other information,
findings? - e.g author, date, title of book, publisher /
- What makes the literature believable? Is the journal, page numbers etc.
literature from credible sources, see validity. - Note the way the author has used the
→ Objectivity original material.
- Is there evidence of bias in the article? - If you have copied the author's words
- For example, would you trust research from directly, make sure you place them in
a cigarette manufacturer claiming that quotation marks and cite the page number,
smoking does not damage your health? BEWARE OF PLAGIARISM CHARGE
- Do the statistics match those in other - What is the author's stated or implied
publications? purpose?
- If not, is the argument (method, research - What conclusions has the author made?
design etc) on which they are based - What points support the conclusions?
convincing? - It is also useful to write down your own
- How do you know the data is true? thoughts on / about the readings.
- What other supporting data is there? - These are useful when you revisit the notes
→ Currency and / or use them in your writings.
- What is the publication date of the material?
- Is it likely that more current information is
available?
- Have you found any more recent research
that casts doubt upon or refutes some
findings?
- Have you checked for more recent
information / research?
Step 4 - analyzing & organizing your summaries
and notes
- Use your summaries and notes to identify
relationships and links in the research
literature.
- You should now be able to identify:
- similarities & differences between
various authors & their research
- what research agrees & disagrees?
- what major questions remain
unanswered?
- what are the possible directions for
future research?
- To organize your research, cluster similar
research together,
- e.g what information is similar or
different.
- A useful technique for doing this is to draw a
mind map and organize the research into
major points under each theme.
- e.g. classify variables into individual
characteristics (e.g. value priority, perceived
capability), situational factors (co-worker
support), organizational factors (reward
systems, empowerment practices)
Chapter 8 Research Design ○ Temporal sequence—the
appropriate causal order of events.
Research design is the blueprint for fulfilling ○ Concomitant variation—two
research objectives and answering investigative phenomena vary together.
questions. ○ Nonspurious association—an
Its essentials include absence of alternative plausible
1) an activity and time-based plan, explanations.
2) a plan based on the research questions,
3) a guide for selecting sources and types of RESEARCH STRATEGIES
information, ● Research strategies that are commonly used
4) a framework for specifying the relationships are mainly based on their own uniqueness
among the study’s variables, and and prevalence; these include experiments,
5) a procedural outline for every research survey research, observation, case studies,
activity. grounded theory, action research, and mixed
methods.
Research Design Content ● Experimental research is an endeavour by
● Types of Research: Exploratory, Descriptive the researcher to keep control over all
or Causal elements that may influence the
● Target Respondent: consequence of an experiment.
● Sampling Design: ● Survey research is a frequently used
● Measurement Scales technique for systematically gathering data
● Instrument development about a population of interest. It is used for
generating quantitative descriptors or
Exploratory Research descriptive and analytic statistics of the
● Conducted to clarify ambiguous situations or attributes of a population based on selected
discover ideas that may be potential samples.
business opportunities. ● Participant observation is a variant of the
● Initial research conducted to clarify and observations where the researcher joins and
define the nature of a problem. becomes actively involved in the research
○ Does not provide conclusive process, aiming to assess the profound
evidence knowledge, understanding and evaluative
○ Subsequent research expected insight into participants’ situations and lives.
● Case study research is a research strategy
Descriptive Research that can help in analysing a complex issue or
● Describes characteristics of objects, people, question and can stretch out experience or
groups, organizations, or environments. add quality to what is already known through
○ Addresses who, what, when, past studies. For research involving
where, why, and how questions. questions like “how” and “why” about which
○ Considerable understanding of the the researcher has minimal control over the
nature of the problem exists. event, and when the attention is on a
○ Does not provide direct evidence of contemporary issue within a real-life context,
causality. a case study may be an ideal choice of
● Diagnostic analysis research strategy.
○ Seeks to diagnose reasons for ● Grounded theory is a qualitative exploration
market outcomes and focuses approach that was created by Glaser and
specifically on the beliefs and Strauss in the 1960s. The self-characterised
feelings consumers have about and aim for grounded theory is to create
toward competing products. hypotheses about a research interest. It is
not simply abstract conjecturing; rather, the
Causal Research hypothesis should be grounded or
● Research conducted to identify cause and established in observation.
effect relationships (inferences). ● Action research consists of a group of
● Evidence of causality: research approaches which seek action and
research results simultaneously, and which
 usually involves resembling a consultancy CROSS-SECTIONAL VERSUS LONGITUDINAL
project or change agency and field research. STUDIES
● Mixed methods research is more particular ● The cross-sectional research design is the
in that it incorporates the blending of most commonly used research design to
qualitative and quantitative information, gather data and information from a
strategies, approaches, and/or ideal models population at one point in time, and it is
in a research study. identical to a survey design. A single cross-
sectional design includes one wave or round
of information-gathering – data and
UNIT OF ANALYSIS information are gathered from a sample on
The unit of analysis is the most crucial aspect of any one occasion.
research project, as the wrong identification of a unit
of analysis in research will lead to invalid conclusions. ● Longitudinal research includes gathering
The unit of analysis is the real entity that a study is data and information from the same sample
analysing. It could be individuals, groups, artefacts, (of people or associations, for instance) on
geographical units (town, census tract, state) and more than one occasion for recognition of
social interactions (dyadic relations, divorces, the pattern of change. The changes could be
arrests). social, demographic, political, economic,
cultural or technological.
 Chapter 9 Measurement of Variables measure, or gather information about the standard
through our senses).
Measurement
The process of describing some property of a Levels of Scale Measurement
phenomenon, usually by assigning numbers in a reliable ● Nominal
and valid way. ○ Assigns a value to an object for
identification or classification purposes.
Concept ○ Most elementary level of measurement.
A generalized idea about a class of objects that has been ● Ordinal
given a name, an abstraction of reality that is the basic ○ Ranking scales allowing things to be
unit for theory development. Every discipline and theory arranged based on how much of some
is made up of concepts, e.g. key ideas, key words, key concept they possess
phrases. ● Interval
○ Interval scales have both nominal and
Operationalization ordinal properties.
The process of identifying scales that correspond to ○ But they also capture information about
variance in a concept involved in a research process. differences in quantities of a concept.
● Ratio
Scales ○ Highest form of measurement.
A device providing a range of values that correspond to ○ Have all the properties of interval scales
different values in a concept being measured. with the additional attribute of representing
absolute quantities.
Correspondence rules ○ Absolute zero.
Indicate the way that a certain value on a scale
corresponds to some true value of a concept.

Constructs
A term used for concepts that are measured with multiple
variables.

Review of Terms
Concept: a bundle of meanings or characteristics
associated with certain events, objects, conditions,
situations, or behaviors (market share, income, perceived
price fairness, total sales, tenure, education,
occupation) .

Construct: an image or idea specifically invented for a

given research and/or theory-building purpose (social
class, service quality, Job commitments).
We build constructs (eg social class) by combining the
simpler, more concrete concepts (income, education,
occupation), , especially when the idea or image we
intend to convey is not subject to direct observation

Variable: an event, act, characteristic, trait, or attribute

that can be measured and to which we assign numerals
or values; a synonym for the construct or the property
being studied.

Operational definition: a definition for a construct stated

in terms of specific criteria for testing or measurement;
refers to an empirical standard (we must be able to count,
 Index Measures
Scale Characteristic
Empirical Operations ● Attributes
Type s
○ Single characteristics or
Classification fundamental features that pertain to
(mutually •Count (frequency an object, person, situation, or
exclusive and distribution); mode as issue.
collectively central tendency; No ● Index Measures (10-item soft-skill index
Nominal exhaustive measure of dispersion score – formative scale)
categories), but •Used with other variables ○ An index assigns a value based on
no order, to discern patterns, reveal how much of the concept being
distance, or relationships measured is associated with an
natural origin observation.
○ Indexes often are formed by putting
•Determination of greater several variables together.
Classification or lesser value ● Composite Measures (return on asset =
and order, but •Count (frequency profit before tax/total asset; 3-item intention
Ordinal
no distance or distribution); median as mean or summated score – reflective scale)
natural origin central tendency; ○ Assign a value to an observation
nonparametric statistics based on a mathematical derivation
of multiple variables.
•Determination of equality
of intervals or differences Computing Scale Values
Classification,
•Count (frequency ● Summated Scale or Mean scale
order, and
distribution); mean or ○ A scale created by simply summing
distance (equal
Interval median as measure of (or mean) the response to each
intervals), but
central tendency; measure item making up the composite
no natural
of dispersion is standard measure.
origin
deviation or interquartile ● Reverse Coding
range; parametric tests ○ A method of making sure all items
forming a composite scale are
•Determination of equality
scored in the same direction.
of ratios
Negative items can be recoded,
Classification, •Any of the above
changing the value of a response to
order, distance, statistical operations, plus
Ratio a scale so it is opposite of the
and natural multiplication and division;
original value. Done so that
origin mean as central tendency;
negative items in a scale are scored
coefficients of variation as
in the same direction as positive
measure of dispersion
items.

Mathematical and Statistical Analysis of Scales

● Discrete Measures
○ Measures that can take on only one
of a finite number of values.
● Continuous Measures
○ Measures that reflect the intensity
of a concept by assigning values
that can take on any value along
some scale range.
Reversed Items Sensitivity : A measurement instrument’s ability to
Transformational leadership (Chumg et al, 2016) accurately measure variability in stimuli or responses.
- Generally increased by adding more
1. I feel quite confident that my leader will always try response points or adding scale items.
to treat me fairly.
Practicality ; economy, convenience, and
2. My manager would not try to gain an advantage by
interpretability. There is generally a trade-off between
deceiving workers.
the ideal research project and the budget.
3. I have complete faith in the integrity of my
manager/supervisor. ● A measuring device passes the convenience
test if it is easy to administer.
4. I feel a strong loyalty to my leader. ● The interpretability aspect of practicality is
relevant when persons other than the test
5. I would support my leader in almost any
emergency. designers must interpret the results. In such
cases, the designer of the data collection
6. I have a divided sense of loyalty toward my leader.* instrument provides several key pieces of
information to make interpretation possible.
* Reversed Item 1) A statement of the functions the instrument
was designed to measure and the
procedures by which it was developed;
Reliability: The degree to which measures are free 2) Detailed instructions for administration;
from random error and therefore yield consistent 3) Scoring keys and instructions;
results. 4) Norms for appropriate reference groups;
5) Evidence of reliability;
Internal Consistency 6) Evidence regarding the intercorrelations of
Represents a measure’s homogeneity or the extent to subscores;
which each indicator of a concept converges on some 7) Evidence regarding the relationship of the
common meaning. test to other measures; and
8) Guides for test use.
Validity refers to whether we are measuring what we
want to measure.
Can be evaluated by comparing definition of the
Construct and measurement items used to measure
the construct.
 Chapter 10 QUESTIONNAIRE SURVEY AND measurement scales for two variables:
ERRORS uncertainty avoidance and the intention to
visit New Zealand.
● Validity is measuring the concept it is
Survey Questionnaire
supposed to measure.
A research strategy A data-collection method ● Reliability is whether the measurement item
shows consistent responses.
Collects standardised Respondents are asked to ● If the measurement items are taken from the
primary data from a large respond to the same set of scale without any change of words, the
number of respondents. questions in a predetermined researcher is said to adopt the scale. If some
way and answer the of the words in the measurement items are
questions on their own.
changed (e.g., changing from “manager” to
It is often referred to as Most popular and most “teacher”), the researcher is said to adapt
quantitative research. common data-collection the scale.
technique used in survey
work.
Measurement Validity
Other data-collection methods that can be used in surveys ● Content validity (Face Validity) is
include face-to-face interviews and telephone interviews. established if your measuring
scale/instrument covers all aspects of the
The Advantages Of Questionnaires concept.
● Less expensive ● Construct validity refers to the quality of a
● Greater anonymity measurement scale in terms of whether it
● Easy and fast for the respondent to complete truthfully represents a unique concept (that
● Familiar to most respondents is, not a replication of an existing concept).
● Greater geographical area - It requires two statistical
● Higher generalisablity procedures: convergent validity and
● Structured questions in questionnaire allow divergent validity.
quantitative analysis ● Predictive validity (Criterion Validity)
● Respondent convenience refers to the ability of the measurement scale
● Ability to replicate to make accurate predictions.
● Reduced bias
● Less intrusive Measurement Reliability
● Reliability is the consistency of the
Questionnaire Administration Method measurement scale.
● Postal ● In order to reduce measurement error and
● Online increase the reliability of the measurement
● Delivery and collection scale, use a multi-item scale (instead of a
● Collective administration single-item scale).
Steps in Developing Questionnaire ● Test-retest reliability is the degree to which
1. Identify the measurement scales scores are consistent over time.
2. Justify choice of measurement scale ● Internal Consistency Reliability refers to
3. Develop measurement table determining how well all items (or questions)
4. Convert scales into questionnaire on the measurement scale are related to
5. Identify demographics varables each other.
6. Prepare cover letter
7. Pilot test the questionnaire Step 2: Measurement scale for the variables of
interest.
Step 1: Review literature carefully to identify a A few criteria that are usually used in justifying scale
measurement scale that can be used to measure selection are:
variables of interest. 1. Reliability of the scale reported by earlier
● Example: “H1: Uncertainty-avoidant people authors;
will have a less positive intention of visiting
New Zealand” - requires finding appropriate
 - evidence that the scale is reliable
in terms of having high correlation
between items.
- This is indicated by Cronbach’s
Alpha of greater than 0.6 (Hair et
al., 1998).
2. Validity of the scale reported by earlier
authors;
- evidence that the scale is really
capturing the concept it is supposed
to measure; that all subcomponents
of the concept were included in the
measure (e.g., content validity); that
Step 5: Pick up from the literature the
it is significantly correlated to other
characteristics of the population that are usually
similar concepts (e.g., convergent
measured in earlier research to describe
validity); and that it is weakly
respondents’ backgrounds.
correlated to different concepts
The questions’ response format needs to fulfil two
(e.g., divergent validly).
criteria:
3. Study context used , If the study context in
● Exhaustive - there is a response category
terms of the respondent group is about the
option for all respondents to tick.
same, this justifies adoption of the scale.
● Mutually exclusive - There is no overlap
- For example, measurement that
among response categories.
measures firm level performance
may not be appropriate for use in
understanding individuals’ job
performance. Thus the researcher
would need to find a measurement
scale that is designed for the right
context.

Step 3: To ensure essential data are collected

A measurement table can be a checklist.

Step 6: Prepare a cover letter that explains the

purpose of the study, assures anonymity of the
participants and persuades participation.

Step 7: Pilot-test the questionnaire with about 10

– 30 respondents to check if the questions or
items are understandable and whether the
instructions are clear.
- This pilot-test data should NOT be included
Step 4: Convert the measurement items in the actual data set that will be used to test
(questions or statements) into questionnaire the hypothesis.
format, complete with instructions.
Overall, a pilot test serves the following purpose: 2. Cognitive interviewing
● To determine the time taken to complete the “Cognitive interviews” are a good way to really
questionnaire. understand what is going on the minds of your
● To check whether respondents understand respondents when they are answering your questions.
and whether they are able to follow the These are typically performed face-to-face with a
instructions on the questionnaire. small sample of 5–15 respondents. As the
● To check whether respondents understand respondents answer each survey question, they are
the wording of the individual questions and asked to “think aloud,” which can include
whether there are unclear or ambiguous paraphrasing, providing retrospective thinking or
questions. providing judgments of their confidence in what each
● To check whether respondents have question means.
problems in answering the questions and
whether response categories are exhaustive
and mutually-exclusive. 3. Expert evaluation
● To provide an idea of the reliability of the Your survey can be dramatically improved by
questions by checking responses of feedback from two types of experts: 1) topic experts
questions relating to a concept. that have deep knowledge and expertise about the
● To check that the questionnaire layout is subject matter of your survey, and 2) survey
clear and easy to follow. methodologists that have expertise in how to collect
the most accurate data for your research question.
These expert evaluations can help shape the content
and form of your survey and result in better data
quality and more valuable insights.
6 WAYS TO PRETEST YOUR SURVEY BEFORE
YOU SEND IT 4. Focus groups
- Time is money, and often researchers want In the preliminary phases of questionnaire
to send their surveys to respondents as development, it can be very helpful to ask a focus
quickly as possible. But rushing to distribute group discuss your survey. These discussions, which
a survey can often result in unforeseen are usually semi-structured discussions between 7–
problems with the data collection. 15 people led by a moderator, are particularly helpful
for clarifying basic concepts in the survey and
- Running your survey through a series of evaluating perceptions of respondent burden or topic
tests to check for potential problems can sensitivity.
save you a lot of headaches down the road
and ensure that you get the data that you
want. In this post we discuss six different 5. Experiments
strategies for testing surveys before starting Splitting a pre-test sample of respondents into groups
your data collection: and testing different variations of your survey design
and content can be very powerful for understanding
1. Respondent debriefing the results you will get when you field your main
It is a good idea to run your survey on a small number survey. These experiments are particularly useful for
of respondents prior to sending it out to your entire understanding how changes in question wording,
sample to get feedback on your survey. This is called questionnaire design, visual layout, question order,
“respondent debriefing.” For this approach, you will and many other methodological factors may influence
need to add several evaluation questions to the end the data you collect.
of your survey for the respondents to answer. These
can be open-ended or closed-ended questions and 6. Pilot studies
usually focus on assessing respondent Testing the final version of your survey on a small
comprehension and interpretation of survey sample of your target population is critical—it can give
questions. It should also include overall evaluations of you a sense of the kind of responses you will receive
the survey content, time, satisfaction and difficulty. and any issues that may arise during the real survey
period. Pilot studies often serve as a ‘dry run’ and are
typically done just before fielding the survey to the
entire sample. It is usually a good idea to include
some evaluative questions, such as respondent
perceptions of the length or difficulty of the
questionnaire, satisfaction with taking the survey, etc.
At Qualtrics we typically recommend that our
customers use a sample of about 50 respondents for
these pilot studies, or ‘soft launches,’ unless you need
to do additional testing across different demographics.

ENCOURAGE PARTICIPATION BY
● Giving advance notification
● Checking the appearance of the
questionnaire
● Using a stamped return envelope
● A personalised letter addressed to a specific
person—this shows the respondent that he
or she is important.
● Creating interesting questions
● Explaining the benefits that the respondent
Sources Of Error In Questionnaire Data
will enjoy as a result of participating in the
● Random Sampling Error
survey
○ Statistical error that occurs in the
● Targeting and collecting
sampling selection process.
● Explaining that they are valued respondents
○ Can be reduced by increasing
● Offering incentives for participation
sample size to 400.
● Having a prize draw for participation
● Systematic Error
● Assuring confidentiality
○ Caused by mistakes in the
● Explaining how the findings will be used
execution of the research.
● Asking nicely when handing out the
○ Can be divided into:
questionnaire
○ •Respondent error refers to
● Keeping the questionnaire short – a
sample bias due to respondents’
maximum of 6 pages
lack of cooperation (non-response
● Following up with respondents
bias and response bias).
○ Administrative errors involve
3 steps should be taken before entering the data.
improper execution of the research
plan, e.g., data entry error,
1. Editing – This is the process of checking the
interviewer error, sample selection
completeness, consistency and legibility
error or interviewer cheating.
(readability) of data. Keep only good data;
bad data should be identified and excluded
Nonresponse Error
from keying in the data set.
The statistical differences between a survey that
2. Coding – is the process of assigning
includes only those who responded and a perfect
numbers to response categories. Example:
survey that would also include those who failed to
Strongly Disagree is coded as 1 and
respond.
Strongly Agree is coded as 7.
3. Data Entry – is the process of entering data
Nonrespondents
into the data set. Before data are entered, be
People who are not contacted or who refuse to
sure to label each questionnaire with
cooperate in the research.
identification numbers such as 001, 002 and
No contacts: people who are not at home or who are
so on.
otherwise inaccessible on the first and second
contact.
Refusals: People who are unwilling to participate in a
research project.
 - The tendency of some Individuals
Self-Selection Bias to use extremes when responding
A bias that occurs because people who feel strongly to questions.
about a subject are more likely to respond to survey ● Interviewer Bias
questions than people who feel indifferent about it. - The presence of the interviewer
influences respondents’ answers.
Administrative Error ● Social Desirability Bias
An error caused by the improper administration or - Bias in responses caused by
execution of the research task. respondents’ desire, either
● Data-processing error: incorrect data entry, conscious or unconscious, to gain
incorrect computer programming, or other prestige or appear in a different
procedural errors during data analysis. social role.
● Sample selection error: improper sample Auspices Bias – Providing answers acceptable by
design or sampling procedure execution. the organization sponsoring the research, eg. WWF
● Interviewer error: mistakes made by survey.
interviewers failing to record survey
responses correctly.
● Interviewer cheating: filling in fake answers
or falsifying questionnaires by an interviewer.

Response Bias
Deliberate Falsification
● Occasionally people deliberately give
false answers.
- Misrepresent answers to appear
intelligent
- Conceal personal information
- Avoid embarrassment
● Average-person hypothesis:
- Individuals may prefer to be viewed
as average, so they alter their
responses to conform more closely
to their perception of the average
person.
Unconscious Misrepresentation
● When a respondent is consciously trying to
be truthful and cooperative, response bias
can arise from the question format, the
question content, or some other stimulus that
affects their response to a question.
● Sources of misrepresentation:
- Misunderstanding the question
- Unable to recall details
- Unprepared response to an
unexpected question
- Inability to translate feelings into
words
- After-event underreporting

Types of Response Bias

● Acquiescence Bias
- A tendency to agree with all or most
questions.
● Extremity Bias
Chapter 11 Sampling Strategies Steps in Sampling
1. Identify target population
Population And Population Elements 2. Determine sampling frame
Population in research refers to a complete group of 3. Determine sampling strategy
people (e.g., students, employees, teachers, 4. Determine sampling technique
managers, patients or customers) or institutions (e.g., 5. Determine sample size
households, stores, schools, hospitals or firms) that 6. Select actual sampling unit
share some common set of characteristics.
Step 1 IDENTIFY THE TARGET POPULATION
Population element refers to an individual member ● The target population of a research topic is
in the population. Example: If Universiti Putra defined by the researcher and needs to be
Malaysia (UPM) students are the population of the clearly identified at the beginning of a study.
study, the population element would be the individual ● The study should be based on a clear
student. understanding of who or what is of interest,
as well as the type of information required
Usually for a population size of fewer than 50, it is from that population.
more appropriate to collect data from the entire
population, so no sampling is required. Step 2 DETERMINE THE SAMPLING FRAME
● A sampling frame is a list that records all
When Is Sampling Used? population elements.
● When we hardly know who makes up the ● Once a target population is chosen,
entire population. sampling frame availability can be
● When there are cost and time constraints. determined. Example: the sampling frame
● When there is a lot of error to control and for a study that assesses UPM students’
monitor. satisfaction level would be a list of students
● When lists are rarely up to date. provided by the registrar of UPM. If the
● When there is a destruction of the Sampling population size of UPM students is 25,000,
Unit the sampling frame should list 25,000
● When the sample data are sufficient for student names and contact details.
decision-making. ● There are also times when a sampling frame
is not available. Example: international
tourists are the target population, but a
sampling frame that consists of the names of
all international tourists coming to Malaysia
is highly confidential (for security reasons)
and therefore not available to the researcher.
Sampling For Quantitative And Qualitative Studies
Step 3 DETERMINE SAMPLING STRATEGY
The purpose of a sampling strategy is to select
sampling units (e.g., female consumers) as a sample
(e.g., 500 female consumers) from the population
(e.g., 10 million).
Probability Sampling Non-probability Sampling

A sampling technique in which The probability of any

every member of the particular member of the
population has a known, population being chosen is
nonzero probability of unknown.
selection.

Use when there is a need to A sampling technique in which

answer research questions units of the sample are
that require statistical selected on the basis of
estimation on the personal subjective judgment
characteristics of the or convenience. Result not
population from the sample (to generalisable to the
generalise results to the population.
population).

Often associated with Often associated with

quantitative studies where a qualitative studies where an in-
questionnaire data-collection depth-interview data-collection
approach is used. approach is used.

Can be used only when a Use when a sampling frame is

sampling frame is available. not available.
Probability Sampling Techniques
1. Simple Random Sampling Technique
● This follows a lucky draw procedure, where
every unit in the population has an equal
chance of being selected. First, you assign a
running number to each unit in the sampling
frame. Then, you select a unit randomly
accordingly to the number suggested by a
random table.

4. Cluster Random Sampling Technique

● Population is broken into heterogeneous
groups or clusters. Each cluster consists of
units with very different characteristics.
● For example, Malaysia has 13 states,
representing 13 clusters where each cluster
has male and female teachers. Within each
2. Systematic Random Sampling Technique
cluster, a unit is selected randomly (following
● In this procedure, the first unit is selected at
random numbers).
random (following random numbers). Then,
the subsequent unit is picked with kth
interval.
● Say k=10 and the first random number is 7
(from a random table). You simply pick
sampling unit numbers 7, 17, 27, 37 and so
on.

3. Stratified Random Sampling Technique

The purpose of this technique is to provide
fair representation of subgroups. First, the
population is broken into strata
(homogeneous groups). Elements in a strata
share similar characteristics. Then, within
each strata list, a sampling unit is picked
randomly.
Non-probability Sampling Techniques 2. Follow statistical rule
1. Convenient Sampling Technique - the desired ratio of sample size to
● This technique is completely based on the construct numbers should be 20:1
convenience of the researcher. Thus, the (Hair, Anderson, Tatham and Black
selection is of those who happen to be at the 1998).
data-collection venue. Units are selected - When this ratio is observed, the
conveniently. Elements not at the data- result should be generalizable if the
collection venue have no chance of being sample is representative.
selected. 3. minimum sample size of 30 is required
for statistical analysis of each category
within the overall sample.
- This is because only sample size of
30 or more produce a sampling
distribution that is very close to a
normal distribution that allows
statistical analysis to be performed
(Stutely, 2003).
- Thus, in the case of a sub-category
in the population that has less than
2. Judgemental Sampling Technique
30 units, all units should be
● This technique imposes judgements such as,
included as sample.
for example, Malaysians who have travelled
4. the desired data analysis method plays a
to 10 countries or more, as they are
role in determining sample size.
presumed to have rich information on issues
- If the researcher desire to get
of interest (e.g., travel experience). Thus, it
reasonable result, at least 200
approaches units that meet the criteria
sample size is required (Hair et. al,
conveniently.
1998).
3. Quota Sampling Technique
Determine Sample Size (QUALITATIVE)
● Its purpose is to have fair representation of
● The sample size requirements usually follow
subgroups in the sample.
the rule of data saturation.
● It is very similar to Stratified Sampling. The
● The theoretical sample size is a general
difference is that Stratified Sampling selects
guideline. Qualitative research that aims to
units from each strata randomly, while with
understand a fairly homogeneous population
this technique units within each subgroup
requires 12 in-depth interviews (Guest et al.,
are selected conveniently until the quota is
2006) while for a fairly heterogonous
achieved.
population, 25-30 interviews are needed
4. Snowball Sampling Technique
(Creswell, 2007).
● The initial respondents are identified
conveniently through friends and any
Select The Actual Sampling Unit
subsequent respondent is referred by a first
● Select the individual unit for the sample
respondent. This is most commonly used
based on the sampling procedure of the
when dealing with rare populations (e.g.,
chosen sampling technique.
cancer survivors).
● Then collect data from the sampling unit
(e.g., employees, tourists or teachers) using
Determine Sample Size (QUANTITATIVE)
the methods, such as questionnaires,
1. use sample size calculator that is
interviews or observations, for primary data
available online
collection.
- Key in population size, confidence
interval 5% and confidence level of
95%, sample size required will be
generated.
- For instance, for a population size
of 25,000, sample size generated is
378.
Chapter 12 Qualitative Research and Analysis ● Get people’s opinion; learn how they look at
the world
Qualitative Research is research that addresses the ● Collect information
research objectives through the technique that allow
the researcher to provide elaborate interpretations of Plan your interview process
a phenomenon without depending on the numerical ● What do you want to learn – what is your
measurement. purpose for conducting the interviews?
● Who do you intend to interview?
○ Exactly who will you interview?
○ How many?
○ How will they be selected?
Purposive sampling – select those
with rich information
● Where will the interview be held?
○ How long will each interview take?
● Who will do the interviewing?
○ You?
○ What do you need to prepare?
● Will there be an incentive for participants?
○ Food may be helpful!
○ Or a small gift

Develop the interview questions

● What do you want to know?
○ Review the purpose of your
evaluation and why you are
conducting these interviews
● Decide on the questions you will ask
○ Brainstorm questions
○ Prioritize and identify the key
questions
○ Identify possible probing questions
○ Anticipate how long each question
will take to answer and keep within
your timeframe
● Sequence the questions
○ Opening questions, transition
questions, ending questions
● Write your questions down in an interview
guide that lists the topics or questions to be
covered
○ Remember: the type and specificity
of your questions depends upon
whether you are conducting a
structured, semistructured or
unstructured interview

Purpose of Interviews
● Generate ideas
● Develop hypothesis (prediction)
● Gain insight into complex issues
● Seek expert opinion
Pilot test the questions Capturing what people say during the interview:
● With a colleague, friend, or family member Options
and/or ● Use paper and write down what the person
● With a few people like those you intend to says as she talks
interview ● Jot down notes during the interview and fill in
● Take the comments from pilot test immediately after the interview
respondents and Improve the “Interview ● Type responses into a computer
Schedule” ● Tape record the interview (audio or video
tape) – for transcription
Contact the respondent ● Work in pairs: one interviews and the other
● By telephone or in-person records the responses
● Personalize the request
● Explain: Analyzing interview data
○ Purpose of the interview – how data ● Review notes, listen to tapes or review
will be used transcripts
○ How long it will take ● Organize interviews by question and
○ How it will be conducted – when, summarize across all interviews
where, etc. ● Identify themes, patterns and divergences
○ Solicit participation and arrange ● Highlight key points and notable quotes
date/time for the interview ● Depending upon the amount of interview
● Send the interview questions in advance as data you have, you may do the data
appropriate processing by hand or by computer (Word,
Excel, qualitative software programs)
Introduction
● Create an open, respectful environment Review (Interview) Research Stages
● Ensure the person is comfortable ● Thematizing: why and what of the research
● Review the purpose of the interview ● Designing: plan the study (including
● Assure Human subjects protection interview guide)
○ Distribute consent form ● Interviewing: conduct based on a guide (best
○ “May I proceed?” “Is it okay to format for research)
begin” ● Transcribing: prepare material for analysis,
time-consuming
Use probing to gain insight ● Analyzing: looks for themes, key points
● Probing is the process of asking follow-up related to research questions,
questions to dig deeper in order to obtain compare/contrast across interviews
useful, meaningful information ● Verifying: assess validity of findings (multiple
● Interview question: sources of information, another researcher to
“What did you like best about this program?” assess appropriateness of the coding :inter-
Response: “I liked everything.” rater reliability)
Probe 1: “What one thing stood out?” ● Reporting: write up report using
Response: “Being with my friends.” information/analysis from interviews
Probe 2: “What about the program
activities?”
Response : “I liked it when we worked as a
team.”
Probe 3: “How come?”
Response : “It was neat to hear each other’s
perspectives. I heard some things I hadn’t
considered before.”
Probe 4: “What is one thing that you
learned?”
A Simple Thematic Analysis analysis, in the written account of the
● One simple approach to qualitative data analysis.
analysis is to read through all of the data, ● data analysis the researcher is above all
and while reading, make a list of all of the concerned with the key concepts, the words
themes that occur in the data. The themes or key phrases in the data. It is important to
are the key ideas or the key concepts that remember that concepts are the building
strike the researcher identifies as important blocks of theory. It is from the work of
and relevant to the research as s/he reads connecting concepts together that theories
through the data transcripts. are developed, extended or changed.
● The researcher continues reading the ● The researcher engages in a process of
transcripts and recording themes until no attempting to group concepts together
new themes emerge (saturation point). around key or core concepts.
● Then the researcher takes the complete list ● Then the researcher groups the key or core
of themes and explores them with a view to concepts together in themes.
collapsing themes together. In other words, ● Finally the researcher groups the themes
the researcher tries to condense the list of together around key themes.
themes by fitting themes together that seem ● This process of analysis takes considerable
to logically fit together. In this way, the list of reflection on and engagement with the raw
themes becomes shorter, and more data. Eventually, the researcher presents an
manageable, and more abstract. account of the data analysis that s/he has
● When the researcher collapses themes engaged in, using the key themes that
together s/he needs to conceptualise a new emerged in data analysis.
theme, a theme which will encompass all the
themes collapsed together. This process is a RELIABILITY
process of abstraction. ● Reliability involves demonstrating that the
● Through this process, the researcher takes a operations of a study, such as data collection
step away from the raw data, and a step procedures, can be repeated with the same
towards an abstract or abstracted results (i.e. would a different researcher,
understanding of the data. The researcher using the same methods, reach the same
through this process of abstraction comes conclusions?)
closer to the meaning of the data, the ● To allay concerns, one documents the steps
meaning of the data in relation to overall aim undertaken and keeps proper records – for
of the research. instance, transcribing interviews and
● The researcher moves to deeper levels of explaining how one “coded” the results.
abstraction in relation to the data, moving However, no matter how “linear” one tries to
further and further away from the raw data, be, there is always a certain amount of
as s/he continues to collapse themes “doubling back.”
together.
● The researcher can colour-codes the
themes; in colour-coding the themes, s/he
assigns, arbitrarily, for example, the colour
red to the first theme, the colour yellow to the
second theme, the colour green to the third
theme, and do on.
● The researcher then reads through the
transcripts assigning the colours to the parts
of the data that represent each of the
themes.
● In this way the researcher can see, at a
glance, each of the themes as they are
represented in the data.
● The researcher then decides how to tell the
story of the data, based on this thematic
Chapter 13 Quantitative Data Analysis Two Basic Rules for Coding Categories:
● They should be exhaustive, meaning that a
coding category should exist for all possible
responses.
● They should be mutually exclusive and
independent, meaning that there should be
no overlap among the categories to ensure
that a subject or response can be placed in
only one category.

● Data integrity is essential to successful

research and decision making.
● What limits data integrity?
○ Researcher making up data
○ Nonresponse
○ Poor editing or coding
● Consistent coding is important for secondary
data.
Raw Data
The unedited responses from a respondent exactly as 1. Parametric Statistical Test
indicated by that respondent. ● In the literal meaning of the terms, a
parametric statistical test is one that makes
Editing assumptions about the parameters (defining
The process of checking the completeness, properties) of the population distribution(s)
consistency, and legibility of data and making the data from which one‘s data are drawn.
ready for coding and transfer to storage. ● For practical purposes, you can think of
"parametric" as referring to tests, such as t-
Checking for Consistency tests and the analysis of variance, that
Respondents match defined population assume the underlying source population(s)
Check for consistency within the data collection to be normally distributed;
framework ● they generally also assume that one's
measures derive from an equal-interval scale
Item Nonresponse (Interval or Ratio variables)
The technical term for an unanswered question on an ● And you can think of “non-parametric” as
otherwise complete questionnaire resulting in missing referring to tests that do not make on these
data. particular assumptions.

Coding
The process of assigning a numerical score or other
character symbol to previously edited data.

Codes
Rules for interpreting, classifying, and recording data
in the coding process.
The actual numerical or other character symbols
assigned to raw data.

Data File
The way a data set is stored electronically in
spreadsheet-like form in which the rows represent
sampling units and the columns represent variables.
 3. State the Hypotheses
● State the null and alternative hypotheses
based on underlying theory.
● The choice of the significance level is
decided, either 0.01, 0.05, or 0.10.
● Collect data and compute the p-value using
an accurate statistical test.
● Compare the p-value (generated by
software) with the significance value (0.01,
0.05, or 0.10, determined by researcher).
● Conclusion.
2. Basic Assumption of Parametric Test –
Normally distributed Data Set 4. Testing of Means
● Mean, Median, and Mode are equal. State the Hypotheses
● A standard deviation close to zero . a) T-Test uses to assess hypotheses involving a
● Skewness and Kurtosis close to zero or single sample, paired sample, or two independent
within the range of +1 to -1. samples.
● Shapiro-Wilk’s W or Kolmogorov-Smirnov D
test is not significant. b) Three techniques of t-test procedures: one-sample
● A histogram of a variable shows rough t-test, paired-sample t-test, and independent-samples
normality and takes the form of a symmetric t-test.
bell-shaped curve.
● A straighter line is formed by Q-Q plot. c) One-way Analysis of Variance (ANOVA) uses to
● Boxplot shows little outliers; median in the compare the mean of a variable between two or more
center of the box; all four quartiles about independent groups.
equally ranged.
One sample t-test: Used to compare the mean
between a variable and a standard mean.

Paired-sample t-test: Used to compare the mean of

a variable for two related samples or to compare two
means from the same sample.

Independent-sample t-test: Used to compare the

mean of a variable between two unrelated groups.

5. Measures of Association

a. Pearson Correlation Coefficient uses to

measure the strength of a linear association between
two variables.
b. Linear regression analysis uses to predict
changes in dependent variable based on the value of
independent variable(s) or predictor(s).
c. Logistic regression analysis uses to
predict changes in categorical dependent variable
based on the value of independent variable(s) or
predictor(s).
7. Structural Equation Modelling (SEM)
Structural equation modeling (SEM) uses when a
researcher is faced with a set of interrelated variables,
yet none of the multivariate techniques allow the
researcher to address the issues. SEM is widely used
for following:
a. Confirmatory Factor Analysis (CFA) uses
to explore the patterns of relationships
among a number of variables and specify
which variable load onto which factors.
b. Estimating a path model uses to show path
diagram, decomposing covariances and
correlations, and direct, indirect, and total
effects.