Research Methodology

Unit – II

Hypothesis:

A supposition or explanation (theory) that is provisionally accepted in order to

interpret certain events or phenomena, and to provide guidance for further
investigation. A hypothesis may be proven correct or wrong, and must be capable
of refutation. If it remains unrefuted by facts, it is said to be verified or
corroborated.

Statistics: An assumption about certain characteristics of a population. If it specifies

values for every parameter of a population, it is called a simple hypothesis; if not, a
composite hypothesis. If it attempts to nullify the difference between two sample
means (by suggesting that the difference is of no statistical significance), it is called
a null hypothesis.

Main Sources of Hypothesis in Research

A hypothesis may be formulated through a number of is sources. Following are the
main sources of hypothesis.

1. Personal Experience
2. Imagination & Thinking
3. Observation
4. Scientific Theory
5. Previous Study
6. Culture
Previous Study
Previous study is also a source of developing a concrete hypothesis. If a researcher
uses previous knowledge about a phenomenon for a particular place, then another
researcher followed his techniques and formulates his own. For example increase in
fertilizers and irrigation leads to higher production in agriculture in District Mardan.
Now another researcher studies his work and applies it to another District
Nowshera.

Personal Experience
On the basis of his personal experience he uses his mind and suggests some points
for the eradication of a social problem through developing a good hypothesis.
Greater the researcher experience lead to higher degree of formation.

Imagination & Thinking

Creative thinking and imagination of a researcher sometimes help in formulating a
good hypothesis. Personal ideas and the thinking capabilities of a researcher would
lead to greater number of hypothesis formulation as well as control over the
problem.

Observation
In consideration and undertaking a research problem, observation is necessary. The
collection of previous facts and current facts related to the problem lead to the
formulation of a good hypothesis.

1
Scientific Theory
Theory is capable in explaining all the facts relating to the problem. Scientific
theory is a fertile source of hypothesis formulation. The theory which is used by a
researcher may satisfied the needs of making it, because theory explains the
known facts.

Culture
Culture is the accumulation of ways of behaving and adoption in a particular place
and time. While formulating a hypothesis for a problem, culture should be studied.
If we want to study trends towards female education in a particular area, for this
purpose we will study, traditions, family system, Norms, Values, region and
education system of that area.

Types of Hypothesis, Null, Empirical, Complex & Logical Hypothesis

Below are some of the important types of hypothesis

1. Simple Hypothesis
2. Complex Hypothesis
3. Empirical Hypothesis
4. Null Hypothesis
5. Alternative Hypothesis
6. Logical Hypothesis
7. Statistical Hypothesis

Simple Hypothesis
Simple hypothesis is that one in which there exists relationship between two
variables one is called independent variable or cause and other is dependent
variable or effect. For example
1. Smoking leads to Cancer
2. The higher ratio of unemployment leads to crimes.

Complex Hypothesis
Complex hypothesis is that one in which as relationship among variables exists. I
recommend you should read characteristics of a good research hypothesis. In this
type dependent as well as independent variables are more than two. For example
1. Smoking and other drugs leads to cancer, tension chest infections etc.
2. The higher ration of unemployment poverty, illiteracy leads to crimes like
dacoit, Robbery, Rape, prostitution & killing etc.

Empirical Hypothesis
Working hypothesis is that one which is applied to a field. During the formulation it
is an assumption only but when it is pat to a test become an empirical or working
hypothesis.

Null Hypothesis
Null hypothesis is contrary to the positive statement of a working hypothesis.
According to null hypothesis there is no relationship between dependent and
independent variable. It is denoted by ‘HO”.

Alternative Hypothesis
2
Firstly many hypotheses are selected then among them select one which is more
workable and most efficient. That hypothesis is introduced latter on due to changes
in the old formulated hypothesis. It is denote by “HI”.

Logical Hypothesis
It is that type in which hypothesis is verified logically. J.S. Mill has given four
cannons of these hypothesis e.g. agreement, disagreement, difference and residue.

Statistical Hypothesis
A hypothesis which can be verified statistically called statistical hypothesis. The
statement would be logical or illogical but if statistic verifies it, it will be statistical
hypothesis.

Role of Hypothesis in Social Research!

In any scientific investigation, the role of hypothesis is indispensable as it always

guides and gives direction to scientific research. Research remains unfocused
without a hypothesis. Without it, the scientist is not in position to decide as to what
to observe and how to observe. He may at best beat around the bush. In the words
of Northrop, “The function of hypothesis is to direct our search for order among
facts, the suggestions formulated in any hypothesis may be solution to the problem,
whether they are, is the task of the enquiry”.

Several near consequences are provided in the process of deductive development

of hypothesis. In the process of conducting experiments for confirming the
hypothesis, scores of new facts develop and expand the horizon of knowledge of
the scientist. Since hypothesis is concerned with explaining facts, the rejection of
hypothesis is not futile.

Rather, it is worthwhile in the sense that it can be of great service in pointing out
the way to true hypothesis. Even a false hypothesis is capable of showing the
direction of inquiry. Realizing the indispensability of hypothesis in a scientific
investigation, Cohen and Nagel observe, ‘Hypotheses are required at every stage of
an inquiry. It must not be forgotten that what are called general principles or laws
can be applied to a present, still un-terminated inquiry only with some risk. For they
may not in fact be applicable.

The general laws of any science function as hypothesis, which guide the inquiry in
all its phases”. Thus, there is little doubt that the importance of hypothesis in the
field of scientific research is tremendous. At least five reasons may be advanced for
justifying hypothesis as a significant device for scientific research.

First, it is an operating tool of theory. It can be deduced from other hypotheses and
theories. If it is correctly drawn and scientifically formulated, it enables the
researcher to proceed on correct line of study. Due to this progress, the investigator
becomes capable of drawing proper conclusions.

In the words of Goode and Hatt, “without hypothesis the research is unfocussed, a
random empirical wandering. The results cannot be studied as facts with clear
meaning. Hypothesis is a necessary link between theory and investigation which
leads to discovery and addition to knowledge.

3
Secondly, the hypothesis acts as a pointer to enquiry. Scientific research has to
proceed in certain definite lines and through hypothesis the researcher becomes
capable of knowing specifically what he has to find out by determining the direction
provided by the hypothesis. Hypotheses acts like a pole star or a compass to a
sailor with the help of which he is able to head in the proper direction.

Thirdly, the hypothesis enables us to select relevant and pertinent facts and makes
our task easier. Once, the direction and points are identified, the researcher is in a
position to eliminate the irrelevant facts and concentrate only on the relevant facts.
Highlighting the role of hypothesis in providing pertinent facts, P.V. Young has
stated, “The use of hypothesis prevents a blind research and indiscriminate
gathering of masses of data which may later prove irrelevant to the problem under
study”.

For example, if the researcher is interested in examining the relationship between

broken home and juvenile delinquency, he can easily proceed in the proper
direction and collect pertinent information succeeded only when he has succeed in
formulating a useful hypothesis.

Fourthly, the hypothesis provides guidance by way of providing the direction,

pointing to enquiry, enabling to select pertinent facts and helping to draw specific
conclusions. It saves the researcher from the botheration of ‘trial and error’ which
causes loss of money, energy and time.

Finally, the hypothesis plays a significant role in facilitating advancement of

knowledge beyond one’s value and opinions. In real terms, the science is
incomplete without hypotheses.

Qualitative Research

1. Qualitative research is “an informal, subjective, semi systematic research

approaches that usually emphasizes words rather than numbers in the
collection and analysis of data” and that is inductive in nature. In other words
it generates theory. General Aims Of Qualitative Research: 1- Views of
participants 2- It does not claim to generalize 3- Theory generation SUPERIOR
GROUP OF COLLEGES

2. THE MAIN PREOCCUPATIONS OF QUALITATIVE RESEARCHER These

preoccupations reflect epistemologically grounded believes about what
constitutes acceptable knowledge. The qualitative researchers are more
influenced by inter pretivism. In this section five different pre-occupations
among qualitative researchers will be outlined and examined 1. See through
the eye of the people been studied 2. Emphasis on process of social life 3.
Description and the emphasis on contexts 4. Flexibility and limited structure
5. Concept and theory grounded in data

Introduction to Qualitative Research

What is qualitative research?

Qualitative research is a type of scientific research. In general terms,
scientific research consists of an investigation that:
• seeks answers to a question
• systematically uses a predefined set of procedures to answer the question
4
• collects evidence
• produces findings that were not determined in advance
• produces findings that are applicable beyond the immediate boundaries of
the study

Qualitative research shares these characteristics. Additionally, it seeks to

understand a given research problem or topic from the perspectives of the
local population it involves. Qualitative research is especially effective in
obtaining culturally specific information about the values, opinions, behaviors,
and social contexts of particular populations.

What can we learn from qualitative research?

The strength of qualitative research is its ability to provide complex textual
descriptions of how people experience a given research issue. It provides
information about the “human” side of an issue – that is, the often
contradictory behaviors, beliefs, opinions, emotions, and relationships of
individuals. Qualitative methods are also effective in identifying intangible
factors, such as social norms, socioeconomic status, gender roles, ethnicity,
and religion, whose role in the research Overview 1 OVERVIEW 2 issue may
not be readily apparent. When used along with quantitative methods,
qualitative research can help us to interpret and better understand the
complex reality of a given situation and the implications of quantitative data.

Although findings from qualitative data can often be extended to people with
characteristics similar to those in the study population, gaining a rich and
complex understanding of a specific social context or phenomenon typically
takes precedence over eliciting data that can be generalized to other
geographical areas or populations. In this sense, qualitative research differs
slightly from scientific research in general.

What are some qualitative research methods?

The three most common qualitative methods, explained in detail in their
respective modules, are participant observation, in-depth interviews, and
focus groups. Each method is particularly suited for obtaining a specific type
of data.
• Participant observation is appropriate for collecting data on naturally
occurring behaviors in their usual contexts.
• In-depth interviews are optimal for collecting data on individuals’ personal
histories, perspectives, and experiences, particularly when sensitive topics
are being explored.
• Focus groups are effective in eliciting data on the cultural norms of a group
and in generating broad overviews of issues of concern to the cultural groups
or subgroups represented.

What forms do qualitative data take?

The types of data these three methods generate are field notes, audio (and
sometimes video) recordings, and transcripts.
Comparing Quantitative and Qualitative Research What are the basic
differences between quantitative and qualitative research methods?
Quantitative and qualitative research methods differ primarily in:

• their analytical objectives

5
• the types of questions they pose
• the types of data collection instruments they use
• the forms of data they produce
• the degree of flexibility built into study design Table 1, page 3, briefly
outlines these major differences.

For a more in-depth theoretical treatment of the differences between

qualitative and quantitative research, we refer the reader to the suggested
readings listed at the end of this chapter, especially Bernard 1995.

What is the most important difference between quantitative and

qualitative methods?
The key difference between quantitative and qualitative methods is their
flexibility. Generally, quantitative methods are fairly inflexible. With
quantitative methods such as surveys and questionnaires, for example,
researchers ask all participants identical questions in the same order. The
response categories from which participants may choose are “closed-ended”
or fixed. The advantage of this inflexibility is that it allows for meaningful
comparison of responses across participants and study sites. However, it
requires a thorough understanding of the important questions to ask, the best
way to ask them, and the range of possible responses. Overview Table 1.

Comparison of quantitative and qualitative research approaches OVERVIEW

General framework Analytical objectives Question format Data format
Flexibility in study design Quantitative Seek to confirm hypotheses about
phenomena Instruments use more rigid style of eliciting and categorizing
responses to questions Use highly structured methods such as
questionnaires, surveys, and structured observation To quantify variation To
predict causal relationships To describe characteristics of a population Closed-
ended Numerical (obtained by assigning numerical values to responses)
Study design is stable from beginning to end Participant responses do not
influence or determine how and which questions researchers ask next Study
design is subject to statistical assumptions and conditions
Qualitative Seek to explore phenomena Instruments use more flexible,
iterative style of eliciting and categorizing responses to questions Use semi-
structured methods such as in-depth interviews, focus groups, and participant
observation To describe variation To describe and explain relationships To
describe individual experiences To describe group norms Open-ended Textual
(obtained from audiotapes, videotapes, and field notes) Some aspects of the
study are flexible (for example, the addition, exclusion, or wording of
particular interview questions) Participant responses affect how and which
questions researchers ask next Study design is iterative, that is, data
collection and research questions are adjusted according to what is learned 4

Qualitative methods are typically more flexible – that is, they allow greater
spontaneity and adaptation of the interaction between the researcher and the
study participant. For example, qualitative methods ask mostly “open-ended”
questions that are not necessarily worded in exactly the same way with each
participant. With open-ended questions, participants are free to respond in
their own words, and these responses tend to be more complex than simply
“yes” or “no.” In addition, with qualitative methods, the relationship between

6
the researcher and the participant is often less formal than in quantitative
research.

Participants have the opportunity to respond more elaborately and in greater

detail than is typically the case with quantitative methods. In turn,
researchers have the opportunity to respond immediately to what participants
say by tailoring subsequent questions to information the participant has
provided. It is important to note, however, that there is a range of flexibility
among methods used in both quantitative and qualitative research and that
flexibility is not an indication of how scientifically rigorous a method is.
Rather, the degree of flexibility reflects the kind of understanding of the
problem that is being pursued using the method.

What are the advantages of qualitative methods for exploratory

research?
One advantage of qualitative methods in exploratory research is that use of
open-ended questions and probing gives participants the opportunity to
respond in their own words, rather than forcing them to choose from fixed
responses, as quantitative methods do. Open-ended questions have the
ability to evoke responses that are: • meaningful and culturally salient to the
participant • unanticipated by the researcher • rich and explanatory in nature
another advantage of qualitative methods is that they allow the researcher
the flexibility to probe initial participant responses – that is, to ask why or
how. The researcher must listen carefully to what participants say, engage
with them according to their individual personalities and styles, and use
“probes” to encourage them to elaborate on their answers. (See the modules
on In-Depth Interviews and Focus Groups, pages 42-43 and 64-65
respectively, for discussions of probes.)

Is my quantitative experience applicable to qualitative research?

Although the objectives of quantitative and qualitative research are not
mutually exclusive, their approaches to deciphering the world involve distinct
research techniques and thus separate skill sets. This guide is intended to
train researchers in the skill set required for qualitative research. Experience
in quantitative methods is not required, but neither is it a disadvantage.
Essential for our purposes, rather, is that all qualitative data collectors have a
clear understanding of the differences between qualitative and quantitative
research, in order to avoid confusing qualitative and quantitative techniques.
Whatever a researcher’s experience in either approach, a general grasp of
the premises and objectives motivating each helps develop and improve
competence in the qualitative data collection techniques detailed in this
guide.

Qualitative Research Methods: A Data Collector’s Field Guide 5 Sampling

in Qualitative Research Even if it were possible, it is not necessary to collect
data from everyone in a community in order to get valid findings. In
qualitative research, only a sample (that is, a subset) of a population is
selected for any given study. The study’s research objectives and the
characteristics of the study population (such as size and diversity) determine
which and how many people to select. In this section, we briefly describe
three of the most common sampling methods used in qualitative research:
purposive sampling, quota sampling, and snowball sampling. As data
7
collectors, you will not be responsible for selecting the sampling method. The
explanations below are meant to help you understand the reasons for using
each method.

What is purposive sampling?

Purposive sampling, one of the most common sampling strategies, groups
participants according to preselected criteria relevant to a particular research
question (for example, HIV-positive women in Capital City). Sample sizes,
which may or may not be fixed prior to data collection, depend on the
resources and time available, as well as the study’s objectives. Purposive
sample sizes are often determined on the basis of theoretical saturation (the
point in data collection when new data no longer bring additional insights to
the research questions). Purposive sampling is therefore most successful
when data review and analysis are done in conjunction with data collection.

What is quota sampling?

Quota sampling, sometimes considered a type of purposive sampling, is also
common. In quota sampling, we decide while designing the study how many
people with which characteristics to include as participants. Characteristics
might include age, place of residence, gender, class, profession, marital
status, use of a particular contraceptive method, HIV status, etc. The criteria
we choose allow us to focus on people we think would be most likely to
experience, know about, or have insights into the research topic. Then we go
into the community and – using recruitment strategies appropriate to the
location, culture, and study population – find people who fit these criteria,
until we meet the prescribed quotas. (See the section in this module on
Recruitment in Qualitative Research, page 6.)

How do purposive and quota sampling differ?

Purposive and quota sampling are similar in that they both seek to identify
participants based on selected criteria. However, quota sampling is more
specific with respect to sizes and proportions of subsamples, with subgroups
chosen to reflect corresponding proportions in the population. If, for example,
gender is a variable of interest in how people experience HIV infection, a
quota sample would seek an equal balance of HIV-positive men and HIV-
positive women in a given city, assuming a 1:1 gender ratio in the population.
Studies employ purposive rather than quota sampling when the number of
participants is more of a target than a steadfast requirement – that is, an
approximate rather than a strict quota.

What is snowball sampling?

A third type of sampling, snowballing – also known as chain referral sampling
– is considered a type of purposive sampling. In this method, participants or
informants with whom contact has already been made use their social
networks to refer the researcher to other people who could potentially
Overview OVERVIEW 6 participate in or contribute to the study. Snowball
sampling is often used to find and recruit “hidden populations,” that is,
groups not easily accessible to researchers through other sampling strategies.

Recruitment in Qualitative Research

A recruitment strategy is a project-specific plan for identifying and enrolling
people to participate in a research study. The plan should specify criteria for
8
screening potential participants, the number of people to be recruited, the
location, and the approach to be used. In this section, we address some of the
questions that may come up during the recruitment of participants.

How are recruitment strategies decided?

Ideally, the local principal investigator and qualitative research team
members work together, in close consultation with community leaders and
gatekeepers (that is, community members in positions of official or unofficial
authority), to develop a plan to identify and recruit potential participants for
each site. Recruitment strategies are determined by the type and number of
data collection activities in the study and by the characteristics of the study
population. They are typically flexible and can be modified if new topics,
research questions, or subpopulations emerge as important to the study, or if
initial strategies do not result in the desired number of recruits. The criteria
for selection can also be changed if certain data collection activities or
subpopulations of people prove not to be useful in answering the research
questions, as discussed in greater detail below.

What if we disagree with recommendations from local leaders’?

It is important for the research team to be respectful of and responsive to the
guidance and advice of local experts and community leaders. Remember that
they have had more opportunity to establish rapport with the local
community and they will also have to maintain that rapport after the study is
complete. Remember also that community members may hold community
leaders and local organizations accountable for any misunderstandings or
other problems resulting from the behavior of the field staff.

What should we say to people when we try to recruit them?

Each research team develops guidelines for the introductory comments staff
make to potential participants at each site. These guidelines need to be
sensitive to the social and cultural contexts from which participants will be
recruited. They should also reflect the researchers’ awareness that willingness
to participate in an interview or focus group will depend on how well the
participants understand what the study is about, what will be expected of
them if they participate, and how their privacy will be respected. In
developing recruitment guidelines, it is important to take special care to avoid
saying anything that could be interpreted as coercive. The voluntary nature of
participation in research studies should always be emphasized.

What if the recruitment strategy is not working?

After data collection is under way, the local principal investigator and field
staff may find that the recruitment strategy is not working as well as
anticipated. Because qualitative research is an iterative process, it is
permissible to change the recruitment strategy, as long as the proper
approvals are obtained. For example, it may be necessary to develop a new
recruitment strategy because following the original plan has resulted in
inadequate enrollment or because researchers determine that they need
participants who meet a different set of criteria. After meeting to discuss
alternatives, the research team should write down reasons why the strategy
was not working or needs to be changed and outline how they would like to
change it.
9
 Proposed changes in the recruitment strategy must be submitted to the
sponsoring organization, and some will require submission of a protocol
amendment for approval by the ethics committees that initially approved the
research. If new criteria for participation are proposed, for instance, they must
be approved by relevant ethics committees before the new phase of
recruitment can begin. Similarly, increasing the number of recruits would also
require ethics committee approval. Because of the limited time frame for data
collection, it is important that the field staff work closely with the site
principal investigator and community gatekeepers to identify and recruit the
new set of research participants. Overview OVERVIEW 8 Ethical Guidelines in
Qualitative Research

This section briefly summarizes ethical issues relevant to qualitative

research. It is intended to provide a context for discussion in subsequent
modules of procedures for safeguarding research participants’ interests.
Qualitative researchers, like anyone conducting research with people, should
undergo formal research ethics training. The material presented here is not a
substitute for training on research ethics.

Research ethics deals primarily with the interaction between researchers and
the people they study. Professional ethics deals with additional issues such as
collaborative relationships among researchers, mentoring relationships,
intellectual property, fabrication of data, and plagiarism, among others. While
we do not explicitly discuss professional ethics here, they are obviously as
important for qualitative research as for any other endeavor. Most
professional organizations, such as the American Anthropological Association,
the Society for Applied Anthropology, the American Sociological Association,
and the American Public Health Association, have developed broad
statements of professional ethics that are easily accessible via the Internet.

Why is research ethics important in qualitative research?

The history and development of international research ethics guidance is

strongly reflective of abuses and mistakes made in the course of biomedical
research. This has led some qualitative researchers to conclude that their
research is unlikely to benefit from such guidance or even that they are not at
risk of perpetrating abuses or making mistakes of real consequence for the
people they study.

Conversely, biomedical and public health researchers who use qualitative

approaches without having the benefit of formal training in the social sciences
may attempt to rigidly enforce bioethics practices without considering
whether they are appropriate for qualitative research. Between these two
extremes lies a balanced approach founded on established principles for
ethical research that are appropriately interpreted for and applied to the
qualitative research context. Agreed-upon standards for research ethics help
ensure that as researchers we explicitly consider the needs and concerns of
the people we study, that appropriate oversight for the conduct of research
takes place, and that a basis for trust is established between researchers and
study participants. Whenever we conduct research on people, the well-being
of research participants must be our top priority.
10
The research question is always of secondary importance. This means that if
a choice must be made between doing harm to a participant and doing harm
to the research, it is the research that is sacrificed. Fortunately, choices of
that magnitude rarely need to be made in qualitative research! But the
principle must not be dismissed as irrelevant, or we can find ourselves
making decisions that eventually bring us to the point where our work
threatens to disrupt the lives of the people we are researching.

Qualitative Research Methods:

A Data Collector’s Field Guide 9 What are the fundamental research ethics
principles? Three core principles, originally articulated in The Belmont
Report,1 form the universally accepted basis for research ethics. Respect for
persons requires a commitment to ensuring the autonomy of research
participants, and, where autonomy may be diminished, to protect people from
exploitation of their vulnerability. The dignity of all research participants must
be respected. Adherence to this principle ensures that people will not be used
simply as a means to achieve research objectives. Beneficence requires a
commitment to minimizing the risks associated with research, including
psychological and social risks, and maximizing the benefits that accrue to
research participants.

Researchers must articulate specific ways this will be achieved. Justice

requires a commitment to ensuring a fair distribution of the risks and benefits
resulting from research. Those who take on the burdens of research
participation should share in the benefits of the knowledge gained. Or, to put
it another way, the people who are expected to benefit from the knowledge
should be the ones who are asked to participate. In addition to these
established principles, some bioethicists have suggested that a fourth
principle, respect for communities, should be added.

Respect for communities “confers on the researcher an obligation to respect

the values and interests of the community in research and, wherever
possible, to protect the community from harm.”2 We believe that this
principle is, in fact, fundamental for research when community-wide
knowledge, values, and relationships are critical to research success and may
in turn be affected by the research process or its outcomes.

What is informed consent?

Informed consent is a mechanism for ensuring that people understand what
it means to participate in a particular research study so they can decide in a
conscious, deliberate way whether they want to participate. Informed consent
is one of the most important tools for ensuring respect for persons during
research. Many people think of informed consent primarily as a form, that is,
a piece of paper that describes in detail what the research is about, including
the risks and benefits. This form generally goes through ethics committee
approval procedures, includes legalistic language, and is signed by the
participant, the researcher, and possibly a witness.

Such informed consent forms are appropriate for biomedical and other
research – including qualitative – when the risks faced by participants may be

11
substantial. They may also be necessary for minimal risk research when the
foundation for trust between researchers and participants is weak.

Quantitative Research:

Quantitative research is a structured way of collecting and analyzing data

obtained from different sources. Quantitative research involves the use of
computational, statistical, and mathematical tools to derive results. It is
conclusive in its purpose as it tries to quantify the problem and understand
how prevalent it is by looking for projectable results to a larger population

Companies who use quantitative research rather than qualitative are typically
looking to measure extent and looking for statistical results that are
interpreted objectively. While the results of qualitative research can vary
according to the skills of observer, the results of quantitative research are
interpreted in an almost similar manner by all experts.

Both types of research vary widely in not only their results, but all other
aspects as well. While qualitative data provides a subjective overview of
marketing problems, quantitative defines a structured cause-and-effect
relationship between the problem and factors.
One of the major differences in the two types of researches is the difference
in the data collection method. Data collection is one of the most important
aspects of the quantitative research process. Data collection involves having
the researcher to prepare and obtain the required information from the target
audience.

Data preparation includes determining the objective of data collection,

methods of obtaining information, and the sequence of data collection
activities. One of the most important aspect in this process is selecting the
right sample for data collection. The data is then carefully collected from only
those people who are most relevant to the objectives of the study. Known as
a target segment, this sample is a group of people who are similar across a
variety of variables.

The data collection tools for a quantitative research are surveys and
experiments. Experiments can provide specific results regarding the cause-
and-effect relationship of several independent or interdependent factors
related to a particular problem.

The most common approach to doing quantitative market research is a

survey or questionnaire. Surveys can include interviews, which can be carried
out using several different methodologies including face-to-face, telephone,
online or computer assisted interviews.

After data collection, another step is the data analysis process. The analysis
of statistical data requires systematic tools and processes to be conducted.
Many analytical tools exist such as independent sample t-tests, correlated t-
tests, variance calculations, and regression analysis that can be used to
derive results from the data.

Quantitative Scales of Measurement

12
 This module discusses the basics of measurement and scales of
measurement commonly used in quantitative research.

Learning Objectives:
 List and describe the four types of scales of measurement used in
quantitative research.
 Provide examples of uses of the four scales of measurement.
 Determine the appropriate measurement scale for a research problem.

Quantitative research requires that measurements be both accurate and

reliable. Researchers commonly assign numbers or values to the attributes of
people, objects, events, perceptions, or concepts. This process is referred to
as measurement. The variables that are measured are commonly classified as
being measured on a nominal, ordinal, interval or ratio scale. The following
discussion defines and provides examples of each of the four levels of
measurement.

Nominal Scale: The nominal scales is essentially a type of coding that

simply puts people, events, perceptions, objects or attributes into categories
based on a common trait or characteristic. The coding can be accomplished
by using numbers, letters, colors, labels or any symbol that can distinguish
between the groups. The nominal scale is the lowest form of a measurement
because it is used simply to categorize and not to capture additional
information. Other features of a nominal scale are that each participant or
object measured is placed exclusively into one category and there is no
relative ordering of the categories. Some examples include distinguishing
between smokers and nonsmokers, males and females, types of religious
affiliations, blondes vs. brunettes and so on. In a study related to smoking,
smokers may be assigned a value of 1 and nonsmokers may be assigned a
value of 2. The assignment of the number is purely arbitrary and at the
researcher’s discretion.

Ordinal Scale: The ordinal scale differs from the nominal scale in that it
ranks the data from lowest to highest and provides information regarding
where the data points lie in relation to one another. An ordinal scale typically
uses non-numerical categories such as low, medium and high to demonstrate
the relationships between the data points. The disadvantage of the ordinal
scale is that it does not provide information regarding the magnitude of the
difference between the data points or rankings. An example of the use of an
ordinal scale would be a study that examines the smoking rates of teenagers.
The data collected may indicate that the teenage smokers in the study
smoked anywhere from 15 to 40 cigarettes per day. The data could be
arranged in order and examined in terms of the number of smokers at each
level.
Interval Scale: An interval scale is one in which the actual distances, or
intervals between the categories or points on the scale can be compared. The
distance between the numbers or units on the scale are equal across the
scale. An example would be a temperature scale, such as the Farenheit scale.
The distance between 20 degrees and 40 degrees is the same as between 60
degrees and 80 degrees. A distinguishing feature of interval scales is that

13
there is no absolute zero point because the key is simply the consistent
distance or interval between categories or data points.
Ratio Scale: The ratio scale contains the most information about the values
in a study. It contains all of the information of the other three categories
because it categorizes the data, places the data along a continuum so that
researchers can examine categories or data points in relation to each other,
and the data points or categories are equal distances or intervals apart.
However, the difference is the ratio scale also contains a non-arbitrary
absolute zero point. The lowest data point collected serves as a meaningful
absolute zero point which allows for interpretation of ratio comparisons.
Time is one example of the use of a ration measurement scale in a study
because it is divided into equal intervals and a ratio comparison can be made.
For example, 20 minutes is twice as long as 10 minutes.
The following Slideshare presentation, Measurement and Scales, provides an
excellent overview of measurement terminology, the four scales of
measurements, a concept map of when to use each type of scale, specific
examples and information concerning the development of a scale.
When and How to Use Qualitative Research
Qualitative research is by definition exploratory, and it is used when we don’t
know what to expect, to define the problem or develop an approach to the
problem.
It’s also used to go deeper into issues of interest and explore nuances related
to the problem at hand. Common data collection methods used in qualitative
research are:
 Focus groups
 Triads
 Dyads
 In-depth interviews
 Uninterrupted observation
 Bulletin boards
 Ethnographic participation/observation.

The Best Times for Quantitative Research

Quantitative research is conclusive in its purpose, as it tries to quantify a
problem and understand how prevalent it is by looking for projectable results
to a larger population.
For this type of study we collect data through:
 Surveys (online, phone, paper)
 Audits
 Points of purchase (purchase transactions)
 Click-streams.

Guidelines For Using Both Types of Research

Ideally, if budget allows, we should use both qualitative and quantitative
research since they provide different perspectives and usually complement
each other.
Advanced survey software should give you the option to integrate video and
chat sessions with your surveys, which can give you the best of both
quantitative and qualitative research.
14
This methodological approach is a cost-effective alternative to the
combination of in-person focus groups and a separate quantitative study.

It allows us to save on facility rental, recruitment costs, incentives and travel

usually associated with focus groups, and clients still are able to monitor the
sessions remotely from the convenience of their desktops and ask questions
to respondents through the moderator.

If you still want to go with traditional methods and can only afford one or the
other, make sure you select the approach that best fits the research
objectives and be aware of its caveats.
Never assume that doing more focus groups is a substitute for quantitative
research or that a long survey will give you all the in-depth information you
could get through qualitative research methods.

Different Types of Errors in Measurement and Measurement Error

Calculation

The measurement of an amount is based on some international standards

which are completely accurate compared with others. Generally,
measurement of any quantity is done by comparing it with derived standards
with which they are not completely accurate. Thus, the errors in
measurement are not only due to error in methods, but are also due to
derivation being not done perfectly well. So, 100% measurement error is not
possible with any methods.

It is very important for the operator to take proper care of the experiment
while performing on industrial instruments so that the error in measurement
can be reduced. Some of the errors are constant in nature due to the
unknown reasons, some will be random in nature, and the other will be due
to gross blunder on the part of the experimenter.

Errors in Measurement System

An error may be defined as the difference between the measured value and
the actual value. For example, if the two operators use the same device or
instrument for finding the errors in measurement, it is not necessary that they
may get the similar results. There may be a difference between both
measurements. The difference that occurs between both the measurements
is referred to as an ERROR.
Sequentially, to understand the concept of errors in measurement, you should
know the two terms that define the error. They are true value and measured
value. The true value is impossible to find out the truth of quantity by
experimental means. It may be defined as the average value of an infinite
number of measured values. Measured value can be defined as the estimated
value of true value that can be found by taking several measured values
during an experiment.
Types of Errors in Measurement System
Generally errors are classified into three types: systematic errors, random
errors and blunders.
1) Gross Errors
2) Blunders
3) Measurement Errors
15
 Systematic Errors
 Instrumental Errors
 Environmental Errors
 Observational Errors
 Theoretical Errors

Random Errors

Types of Errors in Measurement System

1) Gross Errors
Gross errors are caused by mistake in using instruments or meters,
calculating measurement and recording data results. The best example of
these errors is a person or operator reading pressure gage 1.01N/m2 as
1.10N/m2. It may be due to the person’s bad habit of not properly
remembering data at the time of taking down reading, writing and
calculating, and then presenting the wrong data at a later time. This may be
the reason for gross errors in the reported data, and such errors may end up
in calculation of the final results, thus deviating results.
2) Blunders
Blunders are final source of errors and these errors are caused by faulty
recording or due to a wrong value while recording a measurement, or
misreading a scale or forgetting a digit while reading a scale. These blunders
should stick out like sore thumbs if one person checks the work of another
person. It should not be comprised in the analysis of data.
3) Measurement Error
The measurement error is the result of the variation of a measurement of the
true value. Usually, Measurement error consists of a random error and
systematic error. The best example of the measurement error is, if electronic
scales are loaded with 1kg standard weight and the reading is 10002grams,
then
The measurement error is = (1002grams-1000grams) =2grams
Measurement Errors are classified into two types: systematic error and
random errors
Systematic Errors
The Systematic errors that occur due to fault in the measuring device are
known as systematic errors. Usually they are called as Zero Error – a positive
or negative error. These errors can be detached by correcting
the measurement device. These errors may be classified into different
categories.
16
 In order to understand the concept of systematic errors, let us classify the
errors as:
 Instrumental Errors
 Environmental Errors
 Observational Errors
 Theoritical
Instrumental Errors
Instrumental errors occur due to wrong construction of the measuring
instruments. These errors may occur due to hysteresis or friction. These types
of errors include loading effect and misuse of the instruments. In order to
reduce the gross errors in measurement, different correction factors must be
applied and in the extreme condition instrument must be recalibrated
carefully.

Environmental Errors
The environmental errors occur due to some external conditions of the
instrument. External conditions mainly include pressure, temperature,
humidity or due to magnetic fields. In order to reduce the environmental
errors

 Try to maintain the humidity and temperature constant in the laboratory by

making some arrangements.
 Ensure that there shall not be any external electrostatic or magnetic field
around the instrument.

Observational Errors
As the name suggests, these types of errors occurs due to wrong
observations or reading in the instruments particularly in case of energy
meter reading. The wrong observations may be due to PARALLAX. In order to
reduce the PARALLAX error highly accurate meters are needed: meters
provided with mirror scales.

Theoretical Errors
Theoretical errors are caused by simplification of the model system. For
example, a theory states that the temperature of the system surrounding will

17
not change the readings taken when it actually does, then this factor will
begin a source of error in measurement.

Random Errors
Random errors are caused by the sudden change in experimental conditions
and noise and tiredness in the working persons. These errors are either
positive or negative. An example of the random errors is during changes in
humidity, unexpected change in temperature and fluctuation in voltage.
These errors may be reduced by taking the average of a large number of
readings.

Measurement Error Calculation

There are several ways to make a reasonable measurement error calculation
such as estimating random errors and estimating systematic errors.

Estimating Random Errors

There are a number of ways to make a reasonable estimate of the random
error in a particular measurement. The best way is to make a series of
measurements of a given quantity (say, x) and calculate the mean and
standard deviation (x ̅ & σ_x ) from this data.
The mean x ̅ is defined as

Where, Xi is the result of the i th measurements

‘N’ is the number of measurements
The standard variation is given by

If a measurement is repeated many times, then 68% of the measured valves

will drop in the range x ̅ ± σ_x
We become more positive that , is an accurate representation of the true
value of the quantity x ̅ . The standard deviation of the mean σ_x is defined
as
σ_(x ̅ )=σ_x⁄√N

18
The quantity σ_x is a good estimate of our uncertainty in x ̅. Notice that the
measurement precision increases in proportion to √N as we increase the
number of measurements. The following example will clarify these ideas.
Assume you made the following five measurements of a length:

Error Calculations
Therefore, the result is 22.84±.08mm
In some cases, it is hardly useful to repeat a measurement many times. In
that situation, you can estimate frequently the error by taking account of the
smallest division of the measuring instrument.
For example, when using a meter stick, one can measure, perhaps a half or
sometimes even a fifth of a millimeter. So, the absolute error would be
estimated to be around 0.5 mm or 0.2 mm.
Thus, this is all about the various types of errors in measurement and error
measurement calculation. We hope you are satisfied with this article. We
express our gratitude to all the readers. Please share your suggestions and
comments in the comment section below.

Tests of Sound Measurement

While evaluating a measurement tool, three major considerations must be

taken into account: validity, reliability and practicality. A sound measurement
should fulfill all of these tests.
Test of Validity
It is the most important criterion. It indicates the degree to which an
instrument measures what it is supposed to measure. There are three types
of validity: Content validity, Criterion-related validity, and Construct validity.
Content validity refers to the extent to which a measuring instrument
adequately covers the topic under study. Its determination is mainly
judgmental and intuitive. It cannot be expressed in numerical terms. It can
also be determined by a panel of persons who judge the extent of the
measuring instruments standards.

Criterion-related validity refers to our ability to predict or estimate the

existence of a current condition. It reflects the success of measures used for
empirical estimating purposes. Criterion-related validity is expressed as the
coefficient of correlation between the test scores. Here, the concerned
criterion must possess the following characteristics:
19
 Relevance: When a criterion is defined in terms judged to be the proper
measures, it is known to be relevant.
 Unbiased: When the criterion provides each subject an equal opportunity to
score, it is unbiased.
 Reliability: When a criterion is stable or reproducible, it is considered as
reliable.
 Availability: The information specified by the criterion should be easily
available.
Construct validity is most complex and abstract. It is the extent up to which
the scores can be accounted for by the explanatory constructs of a sound
theory. Its determination requires association of a set of other propositions
with the results received from using the measurement instrument. If the
measurements correlate with the other propositions as per our predictions, it
can be concluded that there is some degree of construct validity.
If the above criteria are met, we may conclude that our measuring instrument
is valid and provides correct measurement; if not, we may have to look for
more information and/or depend on judgment.

(or)

Sound measurement must meet the tests of validity, reliability and

practicality. In fact, these are the three major considerations one should use
in evaluating a measurement tool. “Validity refers to the extent to which a
test measures what we actually wish to measure. Reliability has to do with
the accuracy and precision of a measurement procedure ... Practicality is
concerned with a wide range of factors of economy, convenience, and
interpretability ...” We briefly take up the relevant details concerning these
tests of sound measurement.
1. Test of Validity*
Validity is the most critical criterion and indicates the degree to which an
instrument measures what it is supposed to measure. Validity can also be
thought of as utility. In other words, validity is the extent to which differences
found with a measuring instrument reflect true differences among those
being tested. But the question arises: how can one determine validity without
direct confirming knowledge? The answer may be that we seek other relevant
evidence that confirms the answers we have found with our measuring tool.
What is relevant, evidence often depends upon the nature of the research
problem and the judgement of the researcher. But one can certainly consider
three types of validity in this connection: (i) Content validity; (ii) Criterion-
related validity and (iii) Construct validity.
Content validity is the extent to which a measuring instrument provides
adequate coverage of the topic under study. If the instrument contains a
representative sample of the universe, the content validity is good. Its
determination is primarily judgemental and intuitive. It can also be
determined by using a panel of persons who shall judge how well the
measuring instrument meets the standards, but there is no numerical way to
express it.
Criterion-related validity relates to our ability to predict some outcome or
estimate the existence of some current condition. This form of validity reflects
the success of measures used for some empirical estimating purpose. The
concerned criterion must possess the following qualities: Relevance: (A
20
 criterion is relevant if it is defined in terms we judge to be the proper
measure.)
Freedom from bias: (Freedom from bias is attained when the criterion gives
each subject an equal opportunity to score well.)
Reliability: (A reliable criterion is stable or reproducible.)
Availability: (The information specified by the criterion must be available.)
In fact, a Criterion-related validity is a broad term that actually refers to (i)
Predictive validity and Concurrent validity. The former refers to the usefulness
of a test in predicting some future performance whereas the latter refers to
the usefulness of a test in closely relating to other measures of known
validity. Criterion-related validity is expressed as the coefficient of correlation
between test scores and some measure of future performance or between
test scores and scores on another measure of known validity.
Construct validity is the most complex and abstract. A measure is said to
possess construct validity to the degree that it confirms to predicted
correlations with other theoretical propositions. Construct validity is the
degree to which scores on a test can be accounted for by the explanatory
constructs of a sound theory. For determining construct validity, we associate
a set of other propositions with the results received from using our
measurement instrument. If measurements on our devised scale correlate in
a predicted way with these other propositions, we can conclude that there is
some construct validity.
If the above stated criteria and tests are met with, we may state that our
measuring instrument is valid and will result in correct measurement;
otherwise we shall have to look for more information and /or resort to exercise
of judgement.
2. Test of Reliability
The test of reliability is another important test of sound measurement. A
measuring instrument is reliable if it provides consistent results. Reliable
measuring instrument does contribute to validity, but a reliable instrument
need not be a valid instrument. For instance, a scale that consistently
overweighs objects by five kgs., is a reliable scale, but it does not give a valid
measure of weight. But the other way is not true i.e., a valid instrument is
always reliable. Accordingly reliability is not as valuable as validity, but it is
easier to assess reliability in comparison to validity. If the quality of reliability
is satisfied by an instrument, then while using it we can be confident that the
transient and situational factors are not interfering.
Two aspects of reliability viz., stability and equivalence deserve special
mention. The stability aspect is concerned with securing consistent results
with repeated measurements of the same person and with the same
instrument. We usually determine the degree of stability by comparing the
results of repeated measurements. The equivalence aspect considers how
much error may get introduced by different investigators or different samples
of the items being studied. A good way to test for the equivalence of
measurements by two investigators is to compare their observations of the
same events. Reliability can be improved in the following two ways:
 By standardising the conditions under which the measurement takes
place i.e., we must ensure that external sources of variation such as
boredom, fatigue, etc., are minimised to the extent possible. That will
improve stability aspect.
 By carefully designed directions for measurement with no variation
from group to group, by using trained and motivated persons to
21
 conduct the research and also by broadening the sample of items used.
This will improve equivalence aspect.
2 Test of Practicality
The practicality characteristic of a measuring instrument can be judged in
terms of economy, convenience and interpretability. From the operational
point of view, the measuring instrument ought to be practical i.e., it should be
economical, convenient and interpretable. Economy consideration suggests
that some trade-off is needed between the ideal research project and that
which the budget can afford. The length of measuring instrument is an
important area where economic pressures are quickly felt. Although more
items give greater reliability as stated earlier, but in the interest of limiting
the interview or observation time, we have to take only few items for our
study purpose. Similarly, data-collection methods to be used are also
dependent at times upon economic factors. Convenience test suggests that
the measuring instrument should be easy to administer. For this purpose one
should give due attention to the proper layout of the measuring instrument.
For instance, a questionnaire, with clear instructions (illustrated by
examples), is certainly more effective and easier to complete than one which
lacks these features. Interpretability consideration is specially important when
persons other than the designers of the test are to interpret the results. The
measuring instrument, in order to be interpretable, must be supplemented by
(a) detailed instructions for administering the test; (b) scoring keys; (c)
evidence about the reliability and (d) guides for using the test and for
interpreting results.

The technique of developing measurement tools involves a four-stage

process, consisting of the following:
 Concept development;
 Specification of concept dimensions;
 Selection of indicators; and
 Formation of index.
The first and foremost step is that of concept development which means that
the researcher should arrive at an understanding of the major concepts
pertaining to his study. This step of concept development is more apparent in
theoretical studies than in the more pragmatic research, where the
fundamental concepts are often already established.

The second step requires the researcher to specify the dimensions of the
concepts that he developed in the first stage. This task may either be
accomplished by deduction i.e., by adopting a more or less intuitive approach
or by empirical correlation of the individual dimensions with the total concept
and/or the other concepts. For instance, one may think of several dimensions
such as product reputation, customer treatment, corporate leadership,
concern for individuals, sense of social responsibility and so forth when one is
thinking about the image of a certain company.

Once the dimensions of a concept have been specified, the researcher must
develop indicators for measuring each concept element. Indicators are
specific questions, scales, or other devices by which respondent’s knowledge,
opinion, expectation, etc., are measured. As there is seldom a perfect
measure of a concept, the researcher should consider several alternatives for
22
 the purpose. The use of more than one indicator gives stability to the scores
and it also improves their validity.

The last step is that of combining the various indicators into an index, i.e.,
formation of an index. When we have several dimensions of a concept or
different measurements of a dimension, we may need to combine them into a
single index. One simple way for getting an overall index is to provide scale
values to the responses and then sum up the corresponding scores. Such an
overall index would provide a better measurement tool than a single indicator
because of the fact that an “individual indicator has only a probability relation
to what we really want to know.” This way we must obtain an overall index for
the various concepts concerning the research study.

Scaling Techniques
Definition: Scaling is the process of generating the continuum, a continuous
sequence of values, upon which the measured objects are placed.

In Marketing Research, several scaling techniques are employed to study the

relationship between the objects. The most commonly used techniques can
be classified as:

Comparative Scales: In comparative scaling there is a direct comparison of

stimulus object. For example, the respondent might be asked directly about
his preference between the ink pen and gel pen. The comparative data can
only be interpreted in relative terms and hence possess the ordinal or rank-
order properties. This is the reason why the comparative scaling is also called
as nonmetric scaling. The Comparative Scaling includes the following
techniques:
 Paired Comparison Scaling
 Rank Order Scaling
 Constant Sum Scaling
 Q-Sort Scaling

Noncomparative Scales: The noncomparative scale, also called as monadic

or metric scale is a scale in which each object is scaled independently of the
other objects in the stimulus set under study. Generally, the resulting data are
assumed to be an interval and ratio scaled. For example, a respondent may
be asked to rate their preference for the gel pen on a preference scale (1 =
not at all preferred, 6 = greatly preferred). The noncomparative scale includes
the following techniques:
 Continuous Rating Scale
 Itemized Rating Scale
Thus, the researcher can apply any of the scaling techniques to determine the
characteristic of an individual and then locating him on the scale that best fits
the defined characteristics.

1. The level of measurement refers to the relationship among the values that
are assigned to the attributes, feelings or opinions for a variable. Typically,
there are four levels of measurement scales or methods of assigning
numbers: (a) Nominal scale, (b) Ordinal scale, (c) Interval scale, and (d) Ratio
scale.

23
2. Nominal Scale is the crudest among all measurement scales but it is also
the simplest scale. In this scale the different scores on a measurement simply
indicate different categories. The nominal scale does not express any values
or relationships between variables. The nominal scale is often referred to as a
categorical scale. The assigned numbers have no arithmetic properties and
act only as labels. The only statistical operation that can be performed on
nominal scales is a frequency count. We cannot determine an average except
mode. For example: labeling men as ‘1’ and women as ‘2’ which is the most
common way of labeling gender for data recording purpose does not mean
women are ‘twice something or other’ than men. Nor it suggests that men are
somehow ‘better’ than women.

3. Ordinal Scale involves the ranking of items along the continuum of the
characteristic being scaled. In this scale, the items are classified according to
whether they have more or less of a characteristic. The main characteristic of
the ordinal scale is that the categories have a logical or ordered relationship.
This type of scale permits the measurement of degrees of difference, (i.e.
‘more’ or ‘less’) but not the specific amount of differences (i.e. how much
‘more’ or ‘less’). This scale is very common in marketing, satisfaction and
attitudinal research. Using ordinal scale data, we can perform statistical
analysis like Median and Mode, but not the Mean. For example, a fast food
home delivery shop may wish to ask its customers: How would you rate the
service of our staff? (1) Excellent • (2) Very Good • (3) Good • (4) Poor • (5)
Worst •

4. Interval Scale is a scale in which the numbers are used to rank attributes
such that numerically equal distances on the scale represent equal distance
in the characteristic being measured. An interval scale contains all the
information of an ordinal scale, but it also one allows to compare the
difference/distance between attributes. Interval scales may be either in
numeric or semantic formats. The interval scales allow the calculation of
averages like Mean, Median and Mode and dispersion like Range and
Standard Deviation. For example, the difference between ‘1’ and ‘2’ is equal
to the difference between ‘3’ and ‘4’. Further, the difference between ‘2’ and
‘4’ is twice the difference between ‘1’ and ‘2’. Measuring temperature is an
example of interval scale. But, we cannot say 40°C is twice as hot as 20°C.

5. Ratio Scale is the highest level of measurement scales. This has the
properties of an interval scale together with a fixed (absolute) zero point. The
absolute zero point allows us to construct a meaningful ratio. Ratio scales
permit the researcher to compare both differences in scores and relative
magnitude of scores. Examples of ratio scales include weights, lengths and
times. For example, the number of customers of a bank’s ATM in the last
three months is a ratio scale. This is because you can compare this with
previous three months. For example, the difference between 10 and 15
minutes is the same as the difference between 25 and 30 minutes and 30
minutes is twice as long as 15 minutes
6. In comparative scaling, the respondent is asked to compare one object
with another. The comparative scales can further be divided into the following
four types of scaling techniques:
(a) Paired Comparison Scale,
(b) Rank Order Scale,
24
 (c) Constant Sum Scale, and
(d) Q-sort Scale.

7. Paired Comparison Scale: This is a comparative scaling technique in which

a respondent is presented with two objects at a time and asked to select one
object according to some criterion. The data obtained are ordinal in nature.
For example, there are four types of cold drinks Coke, Pepsi, Sprite, and
Limca. The respondents can prefer Pepsi to Coke or Coke to Sprite, etc.

8. Rank Order Scale: This is another type of comparative scaling technique in

which respondents are presented with several items simultaneously and
asked to rank them in the order of priority. This is an ordinal scale that
describes the favoured and unfavoured objects, but does not reveal the
distance between the objects. The resultant data in rank order is ordinal data.
This yields better results when direct comparison are required between the
given objects. The major disadvantage of this technique is that only ordinal
data can be generated.

9. Constant Sum Scale: In this scale, the respondents are asked to allocate a
constant sum of units such as points, rupees, or chips among a set of
stimulus objects with respect to some criterion. For example, you may wish to
determine how important the attributes of price, fragrance, packaging,
cleaning power, and lather of a detergent are to consumers. Respondents
might be asked to divide a constant sum to indicate the relative importance
of the attributes. The advantage of this technique is saving time. However,
main disadvantages are the respondents may allocate more or fewer points
than those specified. The second problem is respondents might be confused.

10. Q-Sort Scale: This is a comparative scale that uses a rank order
procedure to sort objects based on similarity with respect to some criterion.
The important characteristic of this methodology is that it is more important
to make comparisons among different responses of a respondent than the
responses between different respondents. Therefore, it is a comparative
method of scaling rather than an absolute rating scale. In this method the
respondent is given statements in a large number for describing the
characteristics of a product or a large number of brands of a product.

11. In non-comparative scaling respondents need only evaluate a single

object. Their evaluation is independent of the other object which the
researcher is studying. The non-comparative scaling techniques can be
further divided into: (a)Continuous Rating Scale, and (b)Itemized Rating Scale.

12. Continuous Rating Scales : It is very simple and highly useful. In

continuous rating scale, the respondent’s rate the objects by placing a mark
at the appropriate position on a continuous line that runs from one extreme of
the criterion variable to the other. Example : Question: How would you rate
the TV advertisement as a guide for buying?

13. Itemized Rating Scales : Itemized rating scale is a scale having

numbers or brief descriptions associated with each category. The categories
are ordered in terms of scale position and the respondents are required to
select one of the limited number of categories that best describes the
25
 product, brand, company, or product attribute being rated. Itemized rating
scales are widely used in marketing research. Itemised rating scales is further
divided into three parts, namely (a) Likert scale, (b) Semantic Differential
Scale, and (c) Stapel Scale.

The itemised rating scales can be in the form of : (a) graphic, (b) verbal, or (c)
numeric as shown below :

14. Likert Scale: Likert, is extremely popular for measuring attitudes,

because, the method is simple to administer. With the Likert scale, the
respondents indicate their own attitudes by checking how strongly they agree
or disagree with carefully worded statements that range from very positive to
very negative towards the attitudinal object. Respondents generally choose
from five alternatives (say strongly agree, agree, neither agree nor disagree,
disagree, strongly disagree). A Likert scale may include a number of items or
statements. Disadvantage of Likert Scale is that it takes longer time to
complete than other itemised rating scales because respondents have to read
each statement. Despite the above disadvantages, this scale has several
advantages. It is easy to construct, administer and use.

15. Semantic Differential Scale: This is a seven point rating scale with
end points associated with bipolar labels (such as good and bad, complex and
simple) that have semantic meaning. It can be used to find whether a
respondent has a positive or negative attitude towards an object. It has been
widely used in comparing brands, products and company images. It has also
been used to develop advertising and promotion strategies and in a new
product development study.

16. Staple Scale: The Staple scale was originally developed to measure
the direction and intensity of an attitude simultaneously. Modern versions of
the Staple scale place a single adjective as a substitute for the Semantic
differential when it is difficult to create pairs of bipolar adjectives. The
modified Staple scale places a single adjective in the centre of an even
number of numerical Values.

A number of issues decide the choice of scaling technique. Some significant

issues are: 1) Problem Definition and Statistical Analysis, 2) The Choice
between Comparative and Non-comparative Scales, 3) Type of Category
Labels, 4) Number of Categories, 5) Balanced versus Unbalanced Scale, and
6) Forced versus Non-forced Categories

17. There are four levels of measurements: nominal, ordinal, interval,

and ratio. The measurement scales, commonly used in marketing research,
can be divided into two types; comparative and non-comparative scales. A
number of scaling techniques are available for measurement of attitudes.
There is no unique way that you can use to select a particular scaling
technique for your research study.

26
 III – UNIT

Sampling Design
A sample design is the framework, or road map, that serves as the basis for
the selection of a survey sample and affects many other important aspects of
a survey as well. In a broad context, survey researchers are interested in
obtaining some type of information through a survey for some population, or
universe, of interest. One must define a sampling frame that represents the
population of interest, from which a sample is to be drawn. The sampling
frame may be identical to the population, or it may be only part of it and is
therefore subject to some under coverage, or it may have an indirect
relationship to the population (e. g. the population is preschool children and
the frame is a listing of preschools). ...

DEFINITIONS
Population-totality of the objects or individuals regarding inferences are made
in a sampling study. Sample-smaller representation of a large whole.
Sampling- is a process of selecting a subset of randomized number of the
members of the population of a study

Sampling frame /Source list -complete list of all the members/ units of the
population from which each sampling unit Sample design / sample plan-is a
definite plan for obtaining a sample from a given population. Sampling unit-is
a geographical one (state, district) Sample size-number of items selected for
the study Sampling Error-is the difference between population value and
sample value. Sampling distribution-is the relative frequency distribution of
samples.

CENSUS/SAMPLING
Census-collection of data from whole population. Sampling is taking any
portion of a population or universe as representative of that population.
Sampling method has been using in social science research since 1754 by
A.L.BOWLEY

Indispensable of sampling in Research Saves lot of time Provides accuracy

Controls unlimited data Studies individual Reduces cost Gives greater
speed /helps to complete in stipulated time Assists to collect intensive and
exhaustive data Organizes conveniences

Steps in Sampling Process / Procedures

 Define the population (element, units, extent and time)
 Specify sampling frame(Telephone directory)
 Specify sampling unit (retailers, our product ,students, unemployed)
 Specify sampling method/technique
27
  Determine sampling size
 Specify sampling size-(optimum sample)
 Specify sampling plan
 Select the sample

PRINCIPLES OF SAMPLING
Two important principles: Principles of Statistical Regularity-random
(sufficient representative of the sample),Principles of Large Numbers-
(steadiness , stability and consistency)Principles are referred to as the laws of
sampling

Good sampling:
The sample should be true representative of universe. No bias in selecting
sample Quality of the sample should be same Regulating conditions should be
same for all individual. Sampling needs to be adequate Estimate the
sampling error Sample study should be applicable to all items

Preparing a sampling design Type of universe (set of objects) Finite/Non-finite

Sampling unit (district, school, products) Sampling frame Sampling size
Sampling technique

Methods of sampling Bloomers and Lindquist Probability Non Probability

Random/simple Quota Stratified random Purposive Cluster Accidental
Systematic Incidental Multistage Proportionate Snow ball

DIFFERENT TYPES OF SAMPLE DESIGNS

There are different types of sample designs based on two factors:

The representation basis.

Element selection technique.

On the representation basis, the sample may be probability sampling or it

may be non-probability sampling.·

On the element selection basis the sample maybe restricted or unrestricted.·

Sample designing are basically of two types.

1. Probability sampling.

2. Non probability sampling.

Probability

28
Probability sampling technique is one in which every unit in the population
has a chance of being selected in the sample This probability can be
accurately determined.

Non probability sampling

Non probability sampling is any sampling method where some elements of
the population have no chance of selection (these are sometimes referred to
as out of coverage/under covered), or where the probability of selection can’t
be accurately determined. It involves the selection of elements based on
assumptions regarding the population of interest, which forms the criteria for
selection. The selection of elements is non random.

Simple random sampling

In a simple random sample (SRS) of a given size, all such subsets of the
frame are given an equal probability. Method of chance selection. Lottery
method, Tippet’ stable, Kendall and Babington smith, Fisher and Yate’s
numbers. Simple random sampling with replacement:- equal probability
selection of each unit=1/N (Monte-Carlo simulation) Simple random without
replacement -varying probability selection of each. First unit=1/N, Second
unit=1/N-1,Probality of selection of the nth unit=1/N-(n-1)(Monte-Carlo
simulation

Systematic sampling
Systematic sampling involves a random start and then proceeds with the
selection of every kth element from then on wards. In this case,
k=(population size/sample size). It is important that the starting point is not
automatically the first in the list, but is instead randomly chosen from within
the first to the kth element in the list Sampling interval
width=I=N/n=800/40=20

Stratified or Mixed sampling

Where the population embraces a number of distinct categories, the frame
can be organized by these categories into separate "strata." Each stratum is
then sampled as an independent sub-population, out of which individual
elements can be randomly selected. (Homogenous group)Two types-
Proportionate (equal number of unit from each stratum proportion to size of
the strata) and Disproportionate (not equal number of unit from each stratum
proportion to size of the strata)

Cluster sampling
Cluster sampling is an example of two-stage sampling or multistage
sampling/Multi phase sampling in the first stage a sample of areas is chosen
in the second stage a sample of respondents within those areas is selected.
(several stages)- State level, Dist level, Village level, Hose hold level.

Cluster Sampling
This stepwise process is useful for those who know little about the population
they’re studying. First, the researcher would divide the population into
clusters (usually geographic boundaries). Then, the researcher randomly
samples the clusters. Finally, the researcher must measure all units within the
sampled clusters. Researchers use this method when economy of
administration is important.
29
Sequential sampling
Single sampling
Double Sampling
Multiple sampling

Non probability
Non probability sampling does not involve random selection and probability
sampling does .

Multistage sampling
Multistage sampling is a complex form of cluster sampling in which two or
more levels of units are embedded one in the other. The first stage consists of
constructing the clusters that will be used to sample frame. In the second
stage, a sample of primary units is randomly selected from each cluster
(rather than using all units contained in all selected clusters).In following
stages, in each of those selected clusters, additional samples of units are
selected and so on. All ultimate units (individuals, for instance) selected at
the last step of this procedure are surveyed.

Purposive/Judgment Sampling
In purposive sampling, selecting sample with a purpose in mind Purposive
sampling can be very useful for situations where we need to reach a targeted
sample quickly and where sampling for proportionality is not the primary
concern. It is for pilot study Questions / questionnaires may be tested.

Quota sampling
Quota sampling, the population is first segmented into mutually exclusive
sub-groups, just as in stratified sampling. Then judgment is used to select the
subjects or units from each segment based on a specified proportion. For
example, an interviewer may be told to sample 200 females and 300 males
between the age of 45 and 60.Proportional quota sampling Non proportional
quota sampling It is very popular for market survey and opinion poll.

Snowball Sampling
Identifying someone who meets the criteria for inclusion in the study.
Snowball sampling is especially useful when we are trying to reach
populations that are inaccessible or hard to find This method would hardly
lead to representative samples Initially certain members and add few
members latter

Convenience sampling
Convenience sampling (sometimes known as grab or opportunity sampling)is
a type of non probability sampling which involves the sample being drawn
from that part of the population which is close to hand

Accidental Sampling The researcher can select any sample in any place;
can collect the data from pedestrian also. It can be used for exploratory
studies It has sample error. It has less accuracy

Combination of Probability sampling and Non Probability sampling If sampling

is carried out in series of stages, it is possible to combine probability and non-

30
probability sampling in one design Users of particular product in one street for
the particular group of people. Utility of the particular product in the town.

Sampling Errors
The errors which arise due to the use of sampling surveys are known as the
sampling errors. Two types of sampling errors-Biased Errors, Unbiased Errors
Biased Errors-Which arise due to selection of sampling techniques.-size of the
sample Unbiased Errors / Random sampling errors-arise due to chance
differences between the members of the population included in the sample
and not included.

Methods of reducing Sampling Errors Specific problem selection Systematic

documentation of related research Effective enumeration Effective pre testing
Controlling methodological bias Selection of appropriate sampling techniques.

Non-sampling Errors Non-sampling errors refers to biases and mistakes in

selection of sample.

CAUSES FOR NON-SAMPLING ERRORS sampling operations Inadequate of

response Misunderstanding the concept Lack of knowledge Concealment of
the truth. Loaded questions processing errors Sample size

Factors related to Sample size The nature of population Complexity of

tabulation Problems relating to collection of data Selection of sampling
techniques Limitation of accuracy Calculating sample size=(SZ / T)2S-
preliminary SD of the universe Z-number of standard errors T-errors to be
tolerated.

Data Collection
Data collection is a process of collecting information from all the relevant
sources to find answers to the research problem, test the hypothesis and
evaluate the outcomes.

Data collection is the systematic approach to gathering and measuring

information from a variety of sources to get a complete and accurate picture
of an area of interest. The data collection component of research is common
to all fields of study including physical and social sciences, humanities,
business, etc. While methods vary by discipline, the emphasis on ensuring
accurate and honest collection remains the same.

Data collection methods can be divided into two categories: secondary

methods of data collection and primary methods of data collection.

Data collection methods

Surveys, interviews and focus groups are primary instruments for collecting
information. Today, with help from Web and analytics tools, organizations are
also able to collect data from mobile devices, website traffic, server activity
and other relevant sources, depending on the project.

Primary Data Collection Methods

Primary data collection methods can be divided into two groups: quantitative
and qualitative.
31
Quantitative data collection methods are based in mathematical
calculations in various formats. Methods of quantitative data collection and
analysis include questionnaires with closed-ended questions, methods of
correlation and regression, mean, mode and median and others.

Quantitative methods are cheaper to apply and they can be applied within
shorter duration of time compared to qualitative methods. Moreover, due to a
high level of standardization of quantitative methods, it is easy to make
comparisons of findings.
Qualitative research methods, on the contrary, do not involve numbers or
mathematical calculations. Qualitative research is closely associated with
words, sounds, feeling, emotions, colors and other elements that are non-
quantifiable.

Qualitative studies aim to ensure greater level of depth of understanding and

qualitative data collection methods include interviews, questionnaires with
open-ended questions, focus groups, observation, game or role-playing, case
studies etc.

Secondary Data Collection Methods

Secondary data is a type of data that has already been published in books,
newspapers, magazines, journals, online portals etc. There is an abundance
of data available in these sources about your research area in business
studies, almost regardless of the nature of the research area. Therefore,
application of appropriate set of criteria to select secondary data to be used
in the study plays an important role in terms of increasing the levels
of research validity and reliability.

These criteria include, but not limited to date of publication, credential of the
author, reliability of the source, quality of discussions, depth of analyses, the
extent of contribution of the text to the development of the research area etc.

Sources of Data Collection

Normally we can gather data from two sources namely primary and
secondary. Data gathered through perception or questionnaire review in a
characteristic setting are illustrations of data obtained in an uncontrolled
situation. Secondary data is the data acquired from optional sources like
magazines, books, documents, journals, reports, the web and more. The chart
below describes the flow of the sources of data collection.

32
 Sources of Primary Data Collection
Primary data will be the data that you gather particularly with the end goal of
your research venture. Leverage of Primary data is that it is particularly
customized to your analysis needs. A drawback is that it is costly to get hold
of. Primary data is otherwise called raw information; the information gathered
from the first source in a controlled or an uncontrolled situation. Cases of a
controlled domain are experimental studies where certain variables are being
controlled by the analyst.

The source of primary data is the populace test from which you gather the
information. The initial phase in the process is deciding your target populace.
For instance, if you are looking into the attractiveness of another washing
machine, your target populace may be newly-weds.
Clearly, it’s impracticable to gather information from everybody, so you will
need to focus on the sample size and kind of sample. The specimen ought to
be arbitrary and a stratified random sample is frequently sensible. In our
washing machine illustration, sub populations may incorporate adolescent
couples, moderately aged couples, old couples, and previously wedded
couples.

Sources of Secondary Data Collection

You can break the sources of secondary data into internal as well as external
sources. Inner sources incorporate data that exists and is stored in your
organization. External data refers to the data that is gathered by other
individuals or associations from your association’s outer environment.
Examples of inner sources of data incorporate, but are not restricted only to,
the following:
 Statement of the profit and loss
 Balance sheets
 Sales figures
 Inventory records
 Previous marketing studies

If the secondary data you have gathered from internal sources is not
sufficient, you can turn to outside sources of data collection, some outside
sources of data collection include:
 Universities

33
 Government sources
 Foundations
 Media, including telecast, print and Internet
 Trade, business and expert affiliations
 Corporate filings
 Commercial information administrations, which are organizations that find the
data for you

Tools of Data Collection

Case Studies
A case study is usually an in-depth description of a process, experience, or
structure at a single institution. In order to answer a combination of ‘what’
and ‘why’ questions, case studies generally involve a mix of quantitative (i.e.,
surveys, usage statistics, etc.) and qualitative (i.e., interviews, focus groups,
extant document analysis, etc.) data collection techniques. Most often, the
researcher will analyze quantitative data first and then use qualitative
strategies to look deeper into the meaning of the trends identified in the
numerical data.
Checklists
Checklists structure a person’s observation or evaluation of a performance or
artifact. They can be simple lists of criteria that can be marked as present or
absent, or can provide space for observer comments. These tools can provide
consistency over time or between observers. Checklists can be used for
evaluating databases, virtual IM service, the use of library space, or for
structuring peer observations of instruction sessions.
Interviews
In-Depth Interviews include both individual interviews (e.g., one-on-one) as
well as “group” interviews (including focus groups). The data can be recorded
in a wide variety of ways including stenography, audio recording, video
recording or written notes. In depth interviews differ from direct observation
primarily in the nature of the interaction. In interviews it is assumed that
there is a questioner and one or more interviewees. The purpose of the
interview is to probe the ideas of the interviewees about the phenomenon of
interest.
Observation
Sometimes, the best way to collect data through observation. This can be
done directly or indirectly with the subject knowing or unaware that you are
observing them. You may choose to collect data through continuous
observation or via set time periods depending on your project. You may
interpret data you gather using the following mechanisms:
1. Descriptive observations: you simply write down what you observe
2. Inferential observations: you may write down an observation that is
inferred by the subject’s body language and behavior.
3. Evaluative observation:
You may make an inference and therefore a judgment from the behavior.
Make sure you can replicate these findings.

Surveys or Questionnaires
Surveys or questionnaires are instruments used for collecting data in survey
research. They usually include a set of standardized questions that explore a
specific topic and collect information about demographics, opinions, attitudes,
or behaviors. Three popular programs that allow you to create online
34
surveys are Google Forms, Survey Monkey, and Poll Everywhere. A great
(low-cost) tool for reference statistics is Gimlet.

STEPS IN QUESTIONNAIRE CONSTRUCTION

Discuss STEPS IN QUESTIONNAIRE CONSTRUCTION within the Marketing
Research forums, part of the Resolve Your Query - Get Help and discuss
Projects category; STEPS IN QUESTIONNAIRE CONSTRUCTION A Questionnaire
is often the heart of a survey operation.

STEPS IN QUESTIONNAIRE CONSTRUCTION

A Questionnaire is often the heart of a survey operation. If the heart is not

properly set up then the whole operation is bound to fail. Thus studying the
main objective of the questionnaire is important. There are two main
objectives in designing a questionnaire:

1. To maximize the proportion of subjects answering our questionnaire that is,

the response rate: To maximize our response rate, we have to consider
carefully how we administer the questionnaire, establish rapport, and explain
the purpose of the survey. The length of the questionnaire should be
appropriate.

2. To obtain accurate relevant information for our survey: In order to obtain

accurate relevant information, we have to give some thought to what
questions we ask, how we ask them, the order we ask them in, and the
general layout of the questionnaire.

Thus the most important parameters in questionnaire designing can be

described as:

1. Question Content

2. Question Phrasing

3. Question Sequencing

4. Question Layout

1. Question content:

For each question in the questionnaire, we should pay attention to how well it
addresses the content we are trying to get at.

Deciding what to ask there are three potential types of information:

Information we are primarily interested in that is, dependent variables.
Information which might explain the dependent variables-that is, independent
variables.

Other factors related to both dependent and independent factors, which may
distort the results and have to be adjusted for - that is, confounding
variables.
35
Thus while forming the question content the following question must be
answered appropriately.

1. Is the Question Necessary/Useful?

Examine each question to see if there is a need to ask it at all and if you need
to ask it at the level of detail you currently have.

2. Do Respondents Have the Needed Information?

Look at each question to see whether the respondent is likely to have the
necessary information to be able to answer the question.

3. Does the Question Need to be More Specific?

Sometimes the questions are too general and the information we obtain is
more difficult to interpret.
4. Is Question Biased or Loaded?

One danger in question writing is that your own biases and blind spots may
affect the wording.

5. Will Respondents Answer Truthfully?

For each question see whether the respondent will have any difficulty
answering the question truthfully. If there is some reason why they may not,
consider rewording the question.

2. Question phrasing:

The way questions are phrased is important and there are some general rules
for constructing good questions in a questionnaire.

Use short and simple sentences

Short, simple sentences are generally less confusing and ambiguous than
long, complex ones. As a rule of thumb, most sentences should contain one or
two clauses.

Ask for only one piece of information at a time

For example, "Please rate the lecture in terms of its content and presentation"
asks for two pieces of information at the same time.

It should be divided into two parts: "Please rate the lecture in terms of

(a) its content,

(b) its presentation."

Avoid negatives if possible

36
Negatives should be used only sparingly.

For example, instead of asking students whether they agree with the
statement, "Small group teaching should not be abolished," the statement
should be rephrased as, "Small group teaching should continue." Double
negatives should always be avoided.

Ask precise questions

Questions may be ambiguous because a word or term may have a different

meaning.

3. Question sequencing: In order to make the questionnaire effective and to

ensure quality to the replies received, a researcher must pay attention to the
question-sequence in preparing the questionnaire.

• A proper question sequence reduces the chances of the questions being

misunderstood

• The question sequence must be clear and smooth- moving, with questions
that are easiest to answer being put in the beginning.

• The first few questions are particularly important because they are likely to
influence the attitude of the respondent and in seeking his desired
cooperation.

• Following the opening questions are the question that are rally vital to the
research problem and a connecting thread should run through successive
questions.

• Relatively difficult questions must be relegated towards the end so that

even if the respondent decides not to answer such questions, considerable
information would have been obtained.

• The order of the questions is also important. Some general rules are:

-Go from general to particular.

-Go from easy to difficult.

-Go from factual to abstract.

-Start with closed format questions.

-Start with questions relevant to the main subject.

-Do not start with demographic and personal questions.

4.Question layout:

37
 • Questions should form a logical part of a well thought out tabulation plan.

• Questions should basically meet the following standards

-Should be easily understood

-Should be simple

-Should be concrete and should conform as much as possible to the

respondent’s way of thinking.

• Items on a questionnaire should be grouped into logically coherent sections.

Grouping questions that are similar will make the questionnaire easier to
complete, and the respondent will feel more comfortable. Questions that use
the same response formats, or those that cover a specific topic should appear
together.

• Each question should follow comfortably from the previous question. Writing
a questionnaire is similar to writing anything else. Transitions between
questions should be smooth.

Questionnaires that jump from one unrelated topic to another feel disjointed
and are not likely to produce high response rates.

Reliability is the degree to which an assessment tool produces stable and

consistent results.

Types of Reliability

1. Test-retest reliability is a measure of reliability obtained by administering

the same test twice over a period of time to a group of individuals. The
scores from Time 1 and Time 2 can then be correlated in order to evaluate the
test for stability over time.

Example: A test designed to assess student learning in psychology could be

given to a group of students twice, with the second administration perhaps
coming a week after the first. The obtained correlation coefficient would
indicate the stability of the scores.

2. Parallel forms reliability is a measure of reliability obtained by

administering different versions of an assessment tool (both versions must
contain items that probe the same construct, skill, knowledge base, etc.) to
the same group of individuals. The scores from the two versions can then be
correlated in order to evaluate the consistency of results across alternate
versions.

Example: If you wanted to evaluate the reliability of a critical thinking

assessment, you might create a large set of items that all pertain to critical
thinking and then randomly split the questions up into two sets, which would
represent the parallel forms.
38
3. Inter-rater reliability is a measure of reliability used to assess the degree
to which different judges or raters agree in their assessment decisions. Inter-
rater reliability is useful because human observers will not necessarily
interpret answers the same way; raters may disagree as to how well certain
responses or material demonstrate knowledge of the construct or skill being
assessed.

Example: Inter-rater reliability might be employed when different judges are

evaluating the degree to which art portfolios meet certain standards. Inter-
rater reliability is especially useful when judgments can be considered
relatively subjective. Thus, the use of this type of reliability would probably
be more likely when evaluating artwork as opposed to math problems.

4. Internal consistency reliability is a measure of reliability used to evaluate

the degree to which different test items that probe the same construct
produce similar results.

A. Average inter-item correlation is a subtype of internal consistency

reliability. It is obtained by taking all of the items on a test that probe
the same construct (e.g., reading comprehension), determining the
correlation coefficient for each pair of items, and finally taking the
average of all of these correlation coefficients. This final step yields the
average inter-item correlation.

B. Split-half reliability is another subtype of internal consistency

reliability. The process of obtaining split-half reliability is begun by
“splitting in half” all items of a test that are intended to probe the same
area of knowledge (e.g., World War II) in order to form two “sets” of
items. The entire test is administered to a group of individuals, the
total score for each “set” is computed, and finally the split-half
reliability is obtained by determining the correlation between the two
total “set” scores.

Validity refers to how well a test measures what it is purported to measure.

Why is it necessary?
While reliability is necessary, it alone is not sufficient. For a test to be
reliable, it also needs to be valid. For example, if your scale is off by 5 lbs, it
reads your weight every day with an excess of 5lbs. The scale is reliable
because it consistently reports the same weight every day, but it is not valid
because it adds 5lbs to your true weight. It is not a valid measure of your
weight.

Types of Validity

1. Face Validity ascertains that the measure appears to be assessing the

intended construct under study. The stakeholders can easily assess face
validity. Although this is not a very “scientific” type of validity, it may be an
essential component in enlisting motivation of stakeholders. If the
stakeholders do not believe the measure is an accurate assessment of the
ability, they may become disengaged with the task.
39
Example: If a measure of art appreciation is created all of the items should be
related to the different components and types of art. If the questions are
regarding historical time periods, with no reference to any artistic movement,
stakeholders may not be motivated to give their best effort or invest in this
measure because they do not believe it is a true assessment of art
appreciation.

2. Construct Validity is used to ensure that the measure is actually measure

what it is intended to measure (i.e. the construct), and not other variables.
Using a panel of “experts” familiar with the construct is a way in which this
type of validity can be assessed. The experts can examine the items and
decide what that specific item is intended to measure. Students can be
involved in this process to obtain their feedback.

Example: A women’s studies program may design a cumulative assessment

of learning throughout the major. The questions are written with complicated
wording and phrasing. This can cause the test inadvertently becoming a test
of reading comprehension, rather than a test of women’s studies. It is
important that the measure is actually assessing the intended construct,
rather than an extraneous factor.

3. Criterion-Related Validity is used to predict future or current

performance - it correlates test results with another criterion of interest.

Example: If a physics program designed a measure to assess cumulative

student learning throughout the major. The new measure could be correlated
with a standardized measure of ability in this discipline, such as an ETS field
test or the GRE subject test. The higher the correlation between the
established measure and new measure, the more faith stakeholders can have
in the new assessment tool.
4. Formative Validity when applied to outcomes assessment it is used to
assess how well a measure is able to provide information to help improve the
program under study.

Example: When designing a rubric for history one could assess student’s
knowledge across the discipline. If the measure can provide information that
students are lacking knowledge in a certain area, for instance the Civil Rights
Movement, then that assessment tool is providing meaningful information
that can be used to improve the course or program requirements.

5. Sampling Validity (similar to content validity) ensures that the measure

covers the broad range of areas within the concept under study. Not
everything can be covered, so items need to be sampled from all of the
domains. This may need to be completed using a panel of “experts” to
ensure that the content area is adequately sampled. Additionally, a panel can
help limit “expert” bias (i.e. a test reflecting what an individual personally
feels are the most important or relevant areas).

Example: When designing an assessment of learning in the theatre

department, it would not be sufficient to only cover issues related to acting.
Other areas of theatre such as lighting, sound, functions of stage managers
40
 should all be included. The assessment should reflect the content area in its
entirety.

What are some ways to improve validity?

1. Make sure your goals and objectives are clearly defined and operationalized.
Expectations of students should be written down.
2. Match your assessment measure to your goals and objectives. Additionally,
have the test reviewed by faculty at other schools to obtain feedback from an
outside party who is less invested in the instrument.
3. Get students involved; have the students look over the assessment for
troublesome wording, or other difficulties.
4. If possible, compare your measure with other measures, or data that may be
available.

What is a Pilot Study?

Pilot studies can play a very important role prior to conducting a full-scale
research project
Pilot studies are small-scale, preliminary studies which aim to investigate
whether crucial components of a main study – usually a randomized
controlled trial (RCT) – will be feasible. For example, they may be used in
attempt to predict an appropriate sample size for the full-scale project and/or
to improve upon various aspects of the study design. Often RCTs require a lot
of time and money to be carried out, so it is crucial that the researchers have
confidence in the key steps they will take when conducting this type of study
to avoid wasting time and resources.
Thus, a pilot study must answer a simple question: “Can the full-scale study
be conducted in the way that has been planned or should some component(s)
be altered?”
The reporting of pilot studies must be of high quality to allow readers to
interpret the results and implications correctly. This blog will highlight some
key things for readers to consider when they are appraising a pilot study.

What are the main reasons to conduct a pilot study?

Pilot studies are conducted to evaluate the feasibility of some crucial
component(s) of the full-scale study. Typically, these can be divided into 4
main aspects:
 Process: where the feasibility of the key steps in the main study is assessed
(e.g. recruitment rate; retention levels and eligibility criteria)
 Resources: assessing problems with time and resources that may occur
during the main study (e.g. how much time the main study will take to be
completed; whether use of some equipment will be feasible or whether the
form(s) of evaluation selected for the main study are as good as possible)
 Management: problems with data management and with the team involved
in the study (e.g. whether there were problems with collecting all the data
needed for future analysis; whether the collected data are highly variable and
whether data from different institutions can be analyzed together).
 Reasons for not conducting a pilot study: A study should not simply be
labelled a ‘pilot study’ by researchers hoping to justify a small sample size.
Pilot studies should always have their objectives linked with feasibility and
should inform researchers about the best way to conduct the future, full-scale
project.
41
 How to interpret a pilot study
Readers must interpret pilot studies carefully. Below are some key things to
consider when assessing a pilot study:
 The objectives of pilot studies must always be linked with feasibility and the
crucial component that will be tested must always be stated.
 The method section must present the criteria for success. For example: “the
main study will be feasible if the retention rate of the pilot study exceeds
90%”. Sample size may vary in pilot studies (different articles present
different sample size calculations) but the pilot study population, from which
the sample is formed, must be the same as the main study. However, the
participants in the pilot study should not be entered into the full-scale study.
This is because participants may change their later behaviour if they had
previously been involved in the research.
 The pilot study may or may not be a randomized trial (depending on the
nature of the study). If the researchers do randomize the sample in the pilot
study, it is important that the process for randomization is kept the same in
the full-scale project. If the authors decide to test the randomization
feasibility through a pilot study, different kinds of randomization procedures
could be used.
 As well as the method section, the results of the pilot studies should be read
carefully. Although pilot studies often present results related to the
effectiveness of the interventions, these results should be interpreted as
“potential effectiveness”. The focus in the results of pilot studies should
always be on feasibility, rather than statistical significance. However, results
of the pilot studies should nonetheless be provided with measures of
variability (such as confidence intervals), particularly as the sample size of
these studies is usually relatively small, and this might produce biased
results.
After an interpretation of results, pilot studies should conclude with one of the
following:
(1) the main study is not feasible;
(2) the main study is feasible, with changes to the protocol;
(3) the main study is feasible without changes to the protocol OR
(4) the main study is feasible with close monitoring.
Any recommended changes to the protocol should be clearly outlined.
Take home message
 A pilot study must provide information about whether a full-scale study is
feasible and list any recommended amendments to the design of the future
study.
Data Preprocessing
Definition - What does Data Preprocessing mean?
Data preprocessing is a data mining technique that involves transforming raw
data into an understandable format. Real-world data is often incomplete,
inconsistent, and/or lacking in certain behaviors or trends, and is likely to
contain many errors. Data preprocessing is a proven method of resolving such
issues. Data preprocessing prepares raw data for further processing.

Data preprocessing is used database-driven applications such as customer

relationship management and rule-based applications (like neural networks).
Data goes through a series of steps during preprocessing:

42
 Data Cleaning: Data is cleansed through processes such as filling in missing
values, smoothing the noisy data, or resolving the inconsistencies in the data.
 Data Integration: Data with different representations are put together and
conflicts within the data are resolved.
 Data Transformation: Data is normalized, aggregated and generalized.
 Data Reduction: This step aims to present a reduced representation of the
data in a data warehouse.
 Data Discretization: Involves the reduction of a number of values of a
continuous attribute by dividing the range of attribute intervals.

Coding
Coding is the process of classifying responses into meaningful categories It
involves combining / pooling detailed information into a limited number of
categories that enable simple description of the data and allow for
statistical analysis Coding refers to the process of assigning numerals or other
symbols to answers that responses can be put into a limited number
of categories or classes.
Pre-coding and Post-coding

Basic Principles of Coding:

Link to Theory
Exhaustiveness:
Single / Married / Divorced
Mutual Exclusiveness:
Overlap – Hindus, Muslims, Shaikhs,Christians, CatholicsMixing of dimensions
– White Collar,Blue Collar, Government, Private Emp.
Detail:
Coding should be in detail. It is preferableto include too many categories
rather thantoo few.This would be easy to reduce the number of categories
later than slitting the alreadycoded rou of resonses.

On the whole, the number of categories should be small Categories that

include very few cases cannot be retained in the coding scheme

Code Book Construction:

Code design /Coding Frame – It describes the locations of variables and lists
of code assignments to the attributes composing those variables. It serves
two essential functions: It is primary guide used in the coding process. It is
the guide for locating variables and interpreting the columns in data file
during analysis
Coding Process:
Transfer sheets / Code Sheets / Master Sheets Edge Coding Punching directly
/ Direct data entry

Errors in Coding:
While preparation of the Code Design –Not following basic principles At the
time of transcription –Over writing, No clear writing, Misunderstanding of
codes, etc. Verification Procedures: Spot Checking – Sample or Census basis
Manual or Computer Checking2 Ways: Valid Coding Range Logical Checks

Editing

43
Editing is a process of examining the collected raw data (especially in
surveys) to detect errors, options, inadequate entries and to correct these
when possible.
Basic Principles of Editing:
Checking of the no. of Schedules / Questionnaire) Completeness (Completed
in filling of questions) Legibility To avoid Inconstancies in answers To Maintain
Degree of Uniformity To Eliminate Irrelevant Responses

Two Types: 1. Field Editing 2. Central Editing

Field Editing:
While the data collection is in the process, the supervisor in responsible to
collect the correct information.

Normally, it will be done at the same day or next day Here, 2 types
of problems may be avoided: Recording the Answer to Questions and
Illegible Hand Writing. At times Translation also.

Central Editing:
At the Office mostly after data collection or at times during data collection.
Editor must keep the following points in mind while editing the data Familiar
with instructions given to the interviewer and coder While crossing out the
original entry, there should be just one line, so as the original is visible Entries
should be made in a distinguished color and that too in a standared form
Editor should initial all answers which they change or supply Editor initials
and date of editing be marked on each completed form.

Exploratory Data Analysis

As mentioned in Chapter 1, exploratory data analysis or “EDA” is a critical first
step in analyzing the data from an experiment. Here are the main reasons we
use EDA: • detection of mistakes • checking of assumptions • preliminary
selection of appropriate models • determining relationships among the
explanatory variables, and • assessing the direction and rough size of
relationships between explanatory and outcome variables. Loosely speaking,
any method of looking at data that does not include formal statistical
modeling and inference falls under the term exploratory data analysis.

The four types of EDA are uni variate non-graphical, multivariate non
graphical, univariate graphical, and multivariate graphical.

Unit – IV

Tests of Significance
Once sample data has been gathered through an observational study or
experiment, statistical inference allows analysts to assess evidence in favor
or some claim about the population from which the sample has been drawn.
The methods of inference used to support or reject claims based on sample
data are known as tests of significance.
Every test of significance begins with a null hypothesis H0. H0 represents a
theory that has been put forward, either because it is believed to be true or
because it is to be used as a basis for argument, but has not been proved. For
example, in a clinical trial of a new drug, the null hypothesis might be that the
44
 new drug is no better, on average, than the current drug. We would write H0:
there is no difference between the two drugs on average.
The alternative hypothesis, Ha, is a statement of what a statistical
hypothesis test is set up to establish. For example, in a clinical trial of a new
drug, the alternative hypothesis might be that the new drug has a different
effect, on average, compared to that of the current drug. We would write Ha:
the two drugs have different effects, on average. The alternative hypothesis
might also be that the new drug is better, on average, than the current drug.
In this case we would write Ha: the new drug is better than the current drug,
on average.
The final conclusion once the test has been carried out is always given in
terms of the null hypothesis. We either "reject H0 in favor of Ha" or "do not
reject H0"; we never conclude "reject Ha", or even "accept Ha".
If we conclude "do not reject H0", this does not necessarily mean that the null
hypothesis is true, it only suggests that there is not sufficient evidence
against H0 in favor of Ha; rejecting the null hypothesis then, suggests that the
alternative hypothesis may be true.

What are the parametric assumptions?

In the literal meaning of the terms, a parametric statistical test is one that
makes assumptions about the parameters (defining properties) of the
population distribution(s) from which one's data are drawn, while a non-
parametric test is one that makes no such assumptions.

What is a parametric hypothesis test?

A parametric hypothesis test make assumptions about the underlying
distribution of the population from which the sample is being drawn, and
which is being investigated. This is typically that the population conforms to a
normal distribution. ...Hypothesis testing is covered in the MiC Quality
Advanced Statistics Course.

What is parametric and nonparametric tests?

Nonparametric tests are also called distribution-free tests because they
don't assume that your data follow a specific distribution. You may have
heard that you should use nonparametric tests when your data don't meet
the assumptions of theparametric test, especially the assumption about
normally distributed.

What are the assumptions of statistical tests?

In statistical analysis, all parametric tests assume some certain
characteristic about the data, also known as assumptions. Violation of these
assumptions changes the conclusion of the research and interpretation of the
results.
What are the assumptions of parametric tests?
Typical assumptions are:
 Normality: Data have a normal distribution (or at least is symmetric)
 Homogeneity of variances: Data from multiple groups have the same
variance.
 Linearity: Data have a linear relationship.
 Independence: Data are independent.

Is Anova a parametric or nonparametric test?

45
ANOVA is available for score or interval data as parametric ANOVA. This is
the type of ANOVA you do from the standard menu options in a statistical
package. Thenon-parametric version is usually found under the heading
"Nonparametric test". It is used when you have rank or ordered data.

Is a chi square test parametric or nonparametric?

When the assumptions of parametric tests can be met, parametric tests
should be used because they are the most powerful tests available. There
are, however, certain advantages of nonparametric techniques such as Chi
Square (X2). For one thing, nonparametric tests are usually much easier to
compute.

What is a parametric method?

Parametric statistics is a branch of statistics which assumes that sample
data comes from a population that follows a probability distribution based on
a fixed set of parameters. Most well-known elementary
statistical methods are parametric.

What is a non parametric tests?

These include, among others: distribution free methods, which do not rely on
assumptions that the data are drawn from a given probability distribution. As
such it is the opposite of parametric statistics. It
includes nonparametric descriptive statistics, statistical models, inference
and statistical tests.

What is a 'T-Test' ?
A t-test is an analysis of two populations means through the use of statistical
examination; a t-test with two samples is commonly used with small sample
sizes, testing the difference between the samples when the variances of two
normal distributions are not known.

The t test is one type of inferential statistics. It is used to determine whether

there is a significant difference between the means of two groups.

What is a T test?

The t-distribution, used for the t-test. Image: Carnegie Mellon.

The t test (also called Student’s T Test) compares two averages (means) and
tells you if they are different from each other. The t test also tells you
how significant the differences are; In other words it lets you know if those
differences could have happened by chance.

46
 What is a Two Tailed Test?

Area under a normal distribution curve–two tails.

A two tailed test tells you that you’re finding the area in the middle of a
distribution. In other words, your rejection region (the place where you
would reject the null hypothesis) is in both tails.
For example, let’s say you were running a z test with an alpha level of 5%
(0.05). In a one tailed test, the entire 5% would be in a single tail. But with a
two tailed test, that 5% is split between the two tails, giving you 2.5% (0.025)
in each tail.

wo Tailed T Test

Image: ETSU.edu

You may want to compare a sample mean to a given value of x with a t test.
Let’s say your null hypothesis is that the mean is equal to 10 (μ = 10). A two
tailed t test will test:
 Is the mean greater than 10?
 Is the mean less than 10?
If you choose an alpha level of 5%, and the f statistic is in the top 2.5% or
bottom 2.5% of the probability distribution, then there is a significant
difference in the means. That situation will also result in a p-value of less than
0.05. A small p-value gives you a reason to reject the null hypothesis.

Two tailed F test

An f test tells you if two population variances are equal. A two tailed f test is
the standard type of f test which will tell you if the variances are equal or not
equal. The two tailed version of test will test if one variance is greater than, or
less than, the other variance. This is in comparison to the one tailed f test,
which is used when you only want to test if one variance is greater than the
other or that one variance is less than the other (but not both).

Difference Between T-test and F-test

Hypothesis testing starts with setting up the premises, which is followed by
selecting a significance level. Next, we have to choose the test statistic, i.e. t-
test or f-test. While t-test is used to compare two related samples, f-test is
used to test the equality of two populations.

47
 The hypothesis is a simple proposition that can be proved or disproved
through various scientific techniques and establishes the relationship
between independent and some dependent variable. It is capable of being
tested and verified to ascertain its validity, by an unbiased examination.
Testing of a hypothesis attempts to make clear, whether or not the
supposition is valid.
For a researcher, it is imperative to choose the right test for his/her
hypothesis as the entire decision of validating or refusing the null hypothesis
is based on it. Take a read of the given article to understand the difference
between t-test and f-test.

Comparison Chart
BASIS FOR
COMPARISO T-TEST F-TEST
N

Meaning T-test is a F-test is

univariate statistical test,
hypothesis test, that
that is applied determines the
when standard equality of the
deviation is not variances of
known and the the two normal
sample size is populations.
small.

Test statistic T-statistic F-statistic

follows Student follows
t-distribution, Snedecor f-
under null distribution,
hypothesis. under null
hypothesis.

Application Comparing the Comparing two

means of two population
populations. variances.
Definition of T-test
A t-test is a form of the statistical hypothesis test, based on Student’s t-
statistic and t-distribution to find out the p-value (probability) which can be
used to accept or reject the null hypothesis.
T-test analyses if the means of two data sets are greatly different from each
other, i.e. whether the population mean is equal to or different from the
standard mean. It can also be used to ascertain whether the regression line
has a slope different from zero. The test relies on a number of assumptions,
which are:
 The population is infinite and normal.
 Population variance is unknown and estimated from the sample.
 The mean is known.
 Sample observations are random and independent.
 The sample size is small.
 H0 may be one sided or two sided.

48
 Mean and standard deviation of the two sample are used to make comparison
between them, such that:
where,

x̄1 = Mean of the first dataset

x̄2 = Mean of the second dataset
S1 = Standard deviation of the first dataset
S2 = Standard deviation of the second dataset
n1 = Size of first dataset
n2 = Size of second dataset
Definition of F-test
F-test is described as a type of hypothesis test, that is based on Snedecor f-
distribution, under the null hypothesis. The test is performed when it is not
known whether the two populations have the same variance.
F-test can also be used to check if the data conforms to a regression model,
which is acquired through least square analysis. When there is multiple linear
regression analysis, it examines the overall validity of the model or
determines whether any of the independent variables is having a linear
relationship with the dependent variable. A number of predictions can be
made through, the comparison of the two datasets. The expression of the f-
test value is in the ratio of variances of the two observations, which is shown
as under:
Where, σ2 = variance

The assumptions on which f-test relies are:

 The population is normally distributed.
 Samples have been drawn randomly.
 Observations are independent.
 H0 may be one sided or two sided.

What is a Non Parametric Test?

A non parametric test (sometimes called a distribution free test) does not
assume anything about the underlying distribution (for example, that the data
comes from a normal distribution). That’s compared to parametric test, which
makes assumptions about a population’s parameters (for example,
the mean or standard deviation); When the word “non parametric” is used in
stats, it doesn’t quite mean that you know nothing about the population. It
usually means that you know the population data does not have a normal
distribution.

Nonparametric statistics refer to a statistical method in which the data is not

required to fit a normal distribution. Nonparametric statistics uses data that is
often ordinal, meaning it does not rely on numbers, but rather a ranking or
order of sorts.

49
 For example, one assumption for the one way ANOVA is that the data comes
from a normal distribution. If your data isn’t normally distributed, you can’t
run an ANOVA, but you can run the nonparametric alternative–the Kruskal-
Wallis test.

If at all possible, you should us parametric tests, as they tend to be more

accurate. Parametric tests have greater statistical power, which means they
are likely to find a true significant effect. Use nonparametric tests only if you
have to (i.e. you know that assumptions like normality are being violated).
Nonparametric tests can perform well with non-normal continuous data if you
have a sufficiently large sample size (generally 15-20 items in each group).

When to use it

Non parametric tests are used when your data isn’t normal. Therefore the key
is to figure out if you have normally distributed data. For example, you could
look at the distribution of your data. If your data is approximately normal,
then you can use parametric statistical tests.
Q. If you don’t have a graph, how do you figure out if your data is normally
distributed?
A. Check the skewness and Kurtosis of the distribution using software like
Excel (See: Skewness in Excel 2013 andKurtosis in Excel 2013).
A normal distribution has no skew. Basically, it’s a centered and symmetrical
in shape. Kurtosis refers to how much of the data is in the tails and the center.
The skewness and kurtosis for a normal distribution is about 1.

Negative kurtosis (left) and positive kurtosis (right)

If your distribution is not normal (in other words, the skewness and kurtosis
deviate a lot from 1.0), you should use a non parametric test like chi-square
test. Otherwise you run the risk that your results will be meaningless.
Data Types
Does your data allow for a parametric test, or do you have to use a non
parametric test like chi-square? The rule of thumb is:
 For nominal scales or ordinal scales, use non parametric statistics.
 For interval scales or ratio scales use parametric statistics.

A skewed distribution is one reason to run a nonparametric test.

50
 Other reasons to run nonparametric tests:

 One or more assumptions of a parametric test have been violated.

 Your sample size is too small to run a parametric test.
 Your data has outliers that cannot be removed.
 You want to test for the median rather than the mean (you might want to do
this if you have a very skewed distribution).

Types of Nonparametric Tests

When the word “parametric” is used in stats, it usually means tests

like ANOVA or a t test. Those tests both assume that the population data has
a normal distribution. Non parametric do not assume that the data is
normally distributed. The only non parametric test you are likely to come
across in elementary stats is the chi-square test. However, there are several
others. For example: the Kruskal Willis test is the non parametric alternative
to the One way ANOVA and the Mann Whitney is the non parametric
alternative to the two sample t test.
The main nonparamteric tests are:
 1-sample sign test. Use this test to estimate the median of a population
and compare it to a reference value or target value.
 1-sample Wilcoxon signed rank test. With this test, you also estimate the
population median and compare it to a reference/target value. However, the
test assumes your data comes from a symmetric distribution (like the Cauchy
distribution or uniform distribution).
 Friedman test. This test is used to test for differences between groups
with ordinal dependent variables. It can also be used for continuous data if
the one-way ANOVA with repeated measures is inappropriate (i.e. some
assumption has been violated).
 Goodman Kruska’s Gamma: a test of association for ranked variables.
 Kruskal-Wallis test. Use this test instead of a one-way ANOVA to find out if
two or more medians are different. Ranks of the data points are used for the
calculations, rather than the data points themselves.
 The Mann-Kendall Trend Test looks for trends in time-series data.
 Mann-Whitney test. Use this test to compare differences between two
independent groups when dependent variables are either ordinal or
continuous.
 Mood’s Median test. Use this test instead of the sign test when you have
two independent samples.
 Spearman Rank Correlation.Use when you want to find a correlation
between two sets of data.

How to select a test?

Identify the scope and conclude on technical feasibility
As the first step, it is important to identify the complete scope requirements
and various development technologies involved in the architecture of the
application under test. The requirements need to be clear, precise, and
thoroughly reviewed before the tool selection process comes into the picture.
Once the requirements are well documented the appropriate tools should be
shortlisted which can best meet the needs with lesser cost and
implementation effort – but more about those issues later.

51
Building the business case
Organizations must first decide on their desired Return on Investment or do a
cost-benefit analysis before selecting the right tool based on the scope of the
project. If you want to know how to go about doing that – this post of
ours may help! As per the results, if the analysis of the outcome is that
implementing the tool will considerably improve productivity of test execution
or considerably reduce testing effort then it should be pursued further. Some
of the automation tools can help create automated tests without the need to
code – this impacts the type, and hence the cost of resources to be deployed.
The cost of developing tests by hiring specialists versus the cost of investing
in the automation tool needs to be compared and analyzed.
Open source vs licensed tools
There is little doubt that licensed tools have more inbuilt features, but they
always come with a higher cost. For smaller and mid-size projects, open
source tools along with minimal customization can serve the purpose rather
than investing in licensed tools. Some of the open source tools have limited
features, for example, they can execute only specific tests or have support for
specific languages or operating systems, whereas some of the licensed tools
have multiple features and functionality. The pros and cons of choosing the
right tool need to be weighed. In the case of licensed tools, the post
production or after sales support from the vendor needs to taken into
consideration as well.
Proof of concept
Before finally drawing a conclusion, sometimes it may be advisable for
organizations to do a small proof of concept project with the automation tool
that is selected. Setting up the correct success criteria is an important aspect
of a proof of concept for objective decision making. Licensed tools often have
trial versions which can be downloaded and used for the proof of concept.
Doing a proof of concept also gives an idea if test automation is indeed
required or manual testing can suffice. Based on the outcome of the proof of
concept organizations can narrow down the automation tool that best suits
their needs which could be generating test results in a specific format, simple
to implement, better test coverage, etc. In case the proof of concept is not
successful or does not produce desired results, it can help to save time, effort
and costs for the organization before the full-fledged project is executed.

Tools with enhanced features

As stated earlier, it is a must to choose the automation tool which fits the
organization or project requirements. However, many automation tools have
additional nice-to-have features which can benefit the organization in the long
run. For a large organization working on several projects, it is good to have
best of the breed automation software. Additional features like cross-platform
and multi-language support, mobile device support, ability to connect to
multiple test data sources, ability to generate detailed reports, integration
with automated build software, and version control tools are few of the add-
ons which an organization needs to consider.

Selecting the right automation tool is not a simple task and needs to be given
considerable thought. This is because a wrong choice or decision may
severely impact the project, the testing phase, or the end business goals.
What has been your experience of choosing a test automation tool – what
factors did you take into consideration?
52
 Sign test
The sign test is a statistical method to test for consistent differences between
pairs of observations, such as the weight of subjects before and after
treatment. Given pairs of observations (such as weight pre- and post-
treatment) for each subject, the sign test determines if one member of the
pair (such as pre-treatment) tends to be greater than (or less than) the other
member of the pair (such as post-treatment).

What is the Sign Test?

The sign test is a statistical test to compare the sizes of two groups. It is
a non-parametric or “distribution free” test, which means the test doesn’t
assume the data comes from a particular distribution, like the normal
distribution. The sign test is an alternative to a one sample t test or a paired t
test. It can also be used for ordered (ranked) categorical data.
The null hypothesis for the sign test is that the difference between medians is
zero.

ow to Calculate a Paired/Matched Sample Sign Test

Assumptions for the test (your data should meet these requirements before
running the test) are:
1. The data should be from two samples.
2. The two dependent samples should be paired or matched. For example,
depression scores from before a medical procedure and after.
To set up the test, put your two sets of sample data into a table (I used Excel).
This set of data represents test scores at the end of Spring and the beginning
of the Fall semesters. The hypothesis is that the summer break means a
significant drop in test scores.
H0: No difference in median of the signed differences.
H1: Median of the signed differences is less than zero.
Step 1: Subtract set 2 from set 1 and put the result in the third column.

Step 2: Add a fourth column indicating the sign of the number in column 3.
Step 3: Count the number of positives and negatives.
 4 positives.
 12 negatives.
53
 12 negatives seems like a lot, but we can’t say for sure that it’s significant
(i.e. that it didn’t happen by chance) until we run the sign test.
Step 3: Add up the number of items in your sample and subtract any you had
a difference of zero for (in column 3). The sample size in this question was 17,
with one zero, so n = 16.
Step 4: Find the p-value using a binomial distribution table or use a binomial
calculator. I used the calculator, putting in:
 .5 for the probability. The null hypothesis is that there are an equal number of
signs (i.e. 50/50). Therefore, the test is a simple binomial experiment with a .
5 chance of the sign being negative and .5 of it being positive (assuming the
null hypothesis is true).
 16 for the number of trials.
 4 for the number of successes. “Successes” here is the smaller of either
the positive or negative signs from Step 2.
The p-value is 0.038, which is smaller than the alpha level of 0.05. We can
reject the null hypothesis and say there is a significant difference.

Introduction to Correlation and Regression Analysis

In this section we will first discuss correlation analysis, which is used to
quantify the association between two continuous variables (e.g., between an
independent and a dependent variable or between two independent
variables). Regression analysis is a related technique to assess the
relationship between an outcome variable and one or more risk factors or
confounding variables. The outcome variable is also called
the response or dependent variable and the risk factors and confounders
are called the predictors, or explanatory or independent variables. In
regression analysis, the dependent variable is denoted "y" and the
independent variables are denoted by "x".

Correlation Analysis
In correlation analysis, we estimate a sample correlation coefficient, more
specifically the Pearson Product Moment correlation coefficient. The
sample correlation coefficient, denoted r,
ranges between -1 and +1 and quantifies the direction and strength of
the linear association between the two variables. The correlation between
two variables can be positive (i.e., higher levels of one variable are associated
with higher levels of the other) or negative (i.e., higher levels of one variable
are associated with lower levels of the other).
The sign of the correlation coefficient indicates the direction of the
association. The magnitude of the correlation coefficient indicates the
strength of the association.
For example, a correlation of r = 0.9 suggests a strong, positive association
between two variables, whereas a correlation of r = -0.2 suggest a weak,
negative association. A correlation close to zero suggests no linear
association between two continuous variables

Correlation coefficient
The degree of association is measured by a correlation coefficient, denoted by r. It is 
sometimes called Pearson's correlation coefficient after its originator and is a measure of linear 
54
association. If a curved line is needed to express the relationship, other and more complicated 
measures of the correlation must be used.

The correlation coefficient is measured on a scale that varies from + 1 through 0 to ­ 1. 
Complete correlation between two variables is expressed by either + 1 or ­1. When one variable
increases as the other increases the correlation is positive; when one decreases as the other 
increases it is negative. Complete absence of correlation is represented by 0. Figure 11.1 gives 
some graphical representations of correlation.

Figure 11.1 Correlation illustrated.

Looking at data: scatter diagrams
When an investigator has collected two series of observations and wishes to see whether there 
is a relationship between them, he or she should first construct a scatter diagram. The vertical 
scale represents one set of measurements and the horizontal scale the other. If one set of 
observations consists of experimental results and the other consists of a time scale or observed
classification of some kind, it is usual to put the experimental results on the vertical axis. These 
represent what is called the "dependent variable". The "independent variable", such as time or 
height or some other observed classification, is measured along the horizontal axis, or baseline.

The words "independent" and "dependent" could puzzle the beginner because it is sometimes 
not clear what is dependent on what. This confusion is a triumph of common sense over 
misleading terminology, because often each variable is dependent on some third variable, 
which may or may not be mentioned. It is reasonable, for instance, to think of the height of 
children as dependent on age rather than the converse but consider a positive correlation 
between mean tar yield and nicotine yield of certain brands of cigarette.' The nicotine liberated 
is unlikely to have its origin in the tar: both vary in parallel with some other factor or factors in 
the composition of the cigarettes. The yield of the one does not seem to be "dependent" on the 
other in the sense that, on average, the height of a child depends on his age. In such cases it 
often does not matter which scale is put on which axis of the scatter diagram. However, if the 
intention is to make inferences about one variable from the other, the observations from which 
the inferences are to be made are usually put on the baseline. As a further example, a plot of 
monthly deaths from heart disease against monthly sales of ice cream would show a negative 
association. However, it is hardly likely that eating ice cream protects from heart disease! It is 
simply that the mortality rate from heart disease is inversely related ­ and ice cream 
consumption positively related ­ to a third factor, namely environmental temperature.

Calculation of the correlation coefficient
A paediatric registrar has measured the pulmonary anatomical dead space (in ml) and height 
(in cm) of 15 children. The data are given in table 11.1 and the scatter diagram shown in figure 
11.2 Each dot represents one child, and it is placed at the point corresponding to the 
measurement of the height (horizontal axis) and the dead space (vertical axis). The registrar 
now inspects the pattern to see whether it seems likely that the area covered by the dots 
centres on a straight line or whether a curved line is needed. In this case the paediatrician 

55
decides that a straight line can adequately describe the general trend of the dots. His next step 
will therefore be to calculate the correlation coefficient.

When making the scatter diagram (figure 11.2 ) to show the heights and pulmonary anatomical 
dead spaces in the 15 children, the paediatrician set out figures as in columns (1), (2), and (3) 
of table 11.1 . It is helpful to arrange the observations in serial order of the independent variable
when one of the two variables is clearly identifiable as independent. The corresponding figures 
for the dependent variable can then be examined in relation to the increasing series for the 
independent variable. In this way we get the same picture, but in numerical form, as appears in 
the scatter diagram.

Figure 11.2 Scatter diagram of relation in 15 children between height and pulmonary anatomical
dead space.

The calculation of the correlation coefficient is as follows, with x representing the values of the 
independent variable (in this case height) and y representing the values of the dependent 
variable (in this case anatomical dead space). The formula to be used is:

which can be shown to be equal to:

56
Calculator procedure
Find the mean and standard deviation of x, as described in

Find the mean and standard deviation of y:

Subtract 1 from n and multiply by SD(x) and SD(y), (n ­ 1)SD(x)SD(y)

This gives us the denominator of the formula. (Remember to exit from "Stat" mode.)

For the numerator multiply each value of x by the corresponding value of y, add these values 
together and store them.

110 x 44 = Min
116 x 31 = M+
etc.

This stores in memory. Subtract

MR ­ 15 x 144.6 x 66.93 (5426.6)

Finally divide the numerator by the denominator.

r = 5426.6/6412.0609 = 0.846.

The correlation coefficient of 0.846 indicates a strong positive correlation between size of 
pulmonary anatomical dead space and height of child. But in interpreting correlation it is 
important to remember that correlation is not causation. There may or may not be a causative 
connection between the two correlated variables. Moreover, if there is a connection it may be 
indirect.

A part of the variation in one of the variables (as measured by its variance) can be thought of as
being due to its relationship with the other variable and another part as due to undetermined 
(often "random") causes. The part due to the dependence of one variable on the other is 
measured by Rho . For these data Rho= 0.716 so we can say that 72% of the variation 
between children in size of the anatomical dead space is accounted for by the height of the 
child. If we wish to label the strength of the association, for absolute values of r, 0­0.19 is 
regarded as very weak, 0.2­0.39 as weak, 0.40­0.59 as moderate, 0.6­0.79 as strong and 0.8­1 
as very strong correlation, but these are rather arbitrary limits, and the context of the results 
should be considered.

Significance test
To test whether the association is merely apparent, and might have arisen by chance use 
the ttest in the following calculation:
57
 The t Appendix Table B.pdf
is entered at n ­ 2 degrees of freedom.

For example, the correlation coefficient for these data was 0.846.

The number of pairs of observations was 15. Applying equation 11.1, we have:

Entering table B at 15 ­ 2 = 13 degrees of freedom we find that at t = 5.72, P<0.001 so the 
correlation coefficient may be regarded as highly significant. Thus (as could be seen 
immediately from the scatter plot) we have a very strong correlation between dead space and 
height which is most unlikely to have arisen by chance.

The assumptions governing this test are:

1. That both variables are plausibly Normally distributed.

2. That there is a linear relationship between them.
3. The null hypothesis is that there is no association between them.
The test should not be used for comparing two methods of measuring the same quantity, such 
as two methods of measuring peak expiratory flow rate. Its use in this way appears to be a 
common mistake, with a significant result being interpreted as meaning that one method is 
equivalent to the other. The reasons have been extensively discussed(2) but it is worth recalling
that a significant result tells us little about the strength of a relationship. From the formula it 
should be clear that with even with a very weak relationship (say r = 0.1) we would get a 
significant result with a large enough sample (say n over 1000).

Spearman rank correlation
A plot of the data may reveal outlying points well away from the main body of the data, which 
could unduly influence the calculation of the correlation coefficient. Alternatively the variables 
may be quantitative discrete such as a mole count, or ordered categorical such as a pain score.
A non­parametric procedure, due to Spearman, is to replace the observations by their ranks in 
the calculation of the correlation coefficient.

This results in a simple formula for Spearman's rank correlation, Rho.

where d is the difference in the ranks of the two variables for a given individual. Thus we can 
derive table 11.2 from the data in table 11.1 .

58
From this we get that

In this case the value is very close to that of the Pearson correlation coefficient. For n> 10, the 
Spearman rank correlation coefficient can be tested for significance using the t test given earlier

Unit – V

Multivariate Data Analysis:

Multivariate Data Analysis refers to any statistical technique used to analyze
data that arises from more than one variable. This essentially models reality
where each situation, product, or decision involves more than a single
variable.

Multivariate analysis can be complicated by the desire to include physics-

based analysis to calculate the effects of variables for a hierarchical "system-
of-systems". Often, studies that wish to use multivariate analysis are stalled
by the dimensionality of the problem. These concerns are often eased

59
through the use of surrogate models, highly accurate approximations of the
physics-based code. Since surrogate models take the form of an equation,
they can be evaluated very quickly. This becomes an enabler for large-scale
MVA studies: while a Monte Carlo simulation across the design space is
difficult with physics-based codes, it becomes trivial when evaluating
surrogate models, which often take the form of response-surface equations.

Factor analysis is a technique that is used to reduce a large number of

variables into fewer numbers of factors. This technique extracts maximum
common variance from all variables and puts them into a common score. As
an index of all variables, we can use this score for further analysis. Factor
analysis is part of general linear model (GLM) and this method also assumes
several assumptions: there is linear relationship, there is no multi collinearity,
it includes relevant variables into analysis, and there is true correlation
between variables and factors. Several methods are available, but principle
component analysis is used most commonly.

Types of factoring:
There are different types of methods used to extract the factor from the data
set:
1. Principal component analysis: This is the most common method used
by researchers. PCA starts extracting the maximum variance and puts them
into the first factor. After that, it removes that variance explained by the first
factors and then starts extracting maximum variance for the second factor.
This process goes to the last factor.
2. Common factor analysis: The second most preferred method by
researchers, it extracts the common variance and puts them into factors.
This method does not include the unique variance of all variables. This
method is used in SEM.
3. Image factoring: This method is based on correlation matrix. OLS
Regression method is used to predict the factor in image factoring.
4. Maximum likelihood method: This method also works on correlation
metric but it uses maximum likelihood method to factor.
5. Other methods of factor analysis: Alfa factoring outweighs least
squares. Weight square is another regression based method which is used for
factoring.

Cluster analysis is a statistical tool used to classify objects into groups, such
that the objects belonging to one group are much more similar to each other
and rather different from objects belonging to other groups.

What is Cluster Analysis? • Cluster: a collection of data objects – Similar to

one another within the same cluster – Dissimilar to the objects in other
clusters • Cluster analysis – Grouping a set of data objects into clusters •
Clustering is unsupervised classification: no predefined classes • Typical
applications – As a stand-alone tool to get insight into data distribution – As a
preprocessing step for other algorithms

Examples of Clustering Applications • Marketing: Help marketers

discover distinct groups in their customer bases, and then use this knowledge
to develop targeted marketing programs • Land use: Identification of areas of
similar land use in an earth observation database • Insurance: Identifying
60
groups of motor insurance policy holders with a high average claim cost •
City-planning: Identifying groups of houses according to their house type,
value, and geographical location • Earth-quake studies: Observed earth
quake epicenters should be clustered along continent faults

What Is Good Clustering? • A good clustering method will produce high

quality clusters with – high intra-class similarity – low inter-class similarity •
The quality of a clustering result depends on both the similarity measure used
by the method and its implementation. • The quality of a clustering method is
also measured by its ability to discover some or all of the hidden patterns.

Measure the Quality of Clustering • Dissimilarity/Similarity metric:

Similarity is expressed in terms of a distance function, which is typically
metric: d(i, j) • There is a separate “quality” function that measures the
“goodness” of a cluster. • The definitions of distance functions are usually
very different for interval-scaled, boolean, categorical, and ordinal variables. •
Weights should be associated with different variables based on applications
and data semantics. • It is hard to define “similar enough” or “good enough”
– the answer is typically highly subjective.

Discriminant analysis is a statistical method that is used by researchers to

help them understand the relationship between a "dependent variable" and
one or more "independent variables." A dependent variable is the variable
that a researcher is trying to explain or predict from the values of the
independent variables. Discriminant analysis is similar to regression analysis
and analysis of variance (ANOVA). The principal difference between
discriminant analysis and the other two methods is with regard to the nature
of the dependent variable.

Discriminant analysis requires the researcher to have measures of the

dependent variable and all of the independent variables for a large number of
cases. In regression analysis and ANOVA, the dependent variable must be a
"continuous variable." A numeric variable indicates the degree to which a
subject possesses some characteristic, so that the higher the value of the
variable, the greater the level of the characteristic. A good example of a
continuous variable is a person's income.

In discriminant analysis, the dependent variable must be a "categorical

variable." The values of a categorical variable serve only to name groups and
do not necessarily indicate the degree to which some characteristic is
present. An example of a categorical variable is a measure indicating to which
one of several different market segments a customer belongs; another
example is a measure indicating whether or not a particular employee is a
"high potential" worker. The categories must be mutually exclusive; that is, a
subject can belong to one and only one of the groups indicated by the
categorical variable. While a categorical variable must have at least two
values (as in the "high potential" case), it may have numerous values (as in
the case of the market segmentation measure). As
the mathematical methods used in discriminant analysis are complex, they
are described here only in general terms. We will do this by providing an

61
 example of a simple case in which the dependent variable has only two
categories.

The steps involved in conducting discriminant analysis are as follows:

• The problem is formulated before conducting.
• The discriminant function coefficients are estimated.
• The next step is the determination of the significance of these discriminant
functions.
• One must interpret the results obtained.
• The last and the most important step is to assess the validity.

What is SPSS and what is it used for?

SPSS is the acronym of Statistical Package for the Social Science. SPSS is
one of the most popular statistical packages which can perform highly
complex data manipulation and analysis with simple instructions. It is
designed for both interactive and non-interactive (batch) uses.

1. INTRODUCTION SPSS is a software package used for conducting statistical

analysis, manipulating data, and generating table and graphs that summarize
data. SPSS performs statistical analysis range from basic descriptive
statistics, such as average and prevalence, to advanced inferential statistical,
such as regression model, analysis of variance (ANOVA), factor analysis etc.
SPSS also contains several tools for manipulating data, including functions for
recording data, macros programming on visual basic editor, merging data,
and aggregating complex data sets.

2. FEATURES AND BENEFITS For Small And Medium Enterprises (SME)

Resources & best practices. Techniques for cleaning data. Access Data in
Relational Databases. For Large Scale Enterprises (LSE) Real time processing
and online mapping. Online Database connectivity different networks. Data
import & export data from different medium like SAS, Statistica. Compliances
and Validations.

3. What are the options? Many commercial products available SAS, Statistica,
Minitab, and others Widely available (part of MS Office Suite) Not a statistical
software – spreadsheet Finance, math, and statistics applications SPSS Robust
software for sophisticated statistical applications Upgrade with add-ins

4. WHY SPSS? Strengths Very robust statistical software Many complex

statistical tests available Good “stats coach” help with interpreting results
Easily and quickly displays data tables Can be expanded Using the syntax
feature Purchasing add-ins Limitations Can be expensive Not intuitive to use
Typically requires additional training to maximize features (at a cost)
Graphing feature not as simple as Excel

5. RESEARCH & DEVELOPMENT SPSS given the technique of specialize in

design, development, and implementation of corporate analytical and data
mining solution integrated with the existing data acquisition and storage of
systems. Its continuously given the solution in Pharmaceutical and Medical
research. Quality and Manufacturing. Compliance and ValidationEmpower
62
 Researcher with Tools. Regulatory Safety Testing. Clinical Trails. Data Mining in
Drug Discovery.

6. QUALITY AND MANUFACTURING SPSS can provide a single, standardized

quality control and improvement solution that can be easily validated that
cover the whole range of manufacturing statistics, including simple
descriptive statistics and graphics, QC charting, capability analyses for
normal and Non-normal distributions, design of experiment (DOE) much more.
Online and offline SPC (Statistical Process Control) Process Capability
Analysis. Measurement System Analysis. Design of Experiment Process
Validation.

7. COMPLIANCE AND VALIDATION SPSS offers a pre-validation core platform

with fully secure user administration and configuration control, audit trails on
system configuration changes, data manipulation, analysis results and
integrated data modules. Together these features can radically change the
way of your organization works and remove the need for extensive manual
documentation of analytical processes. Validation of Software Analysis of
Audit Trails. Computer Systems Validation

8. EMPOWER RESEARCHER WITH TOOLS Even small improvement in

research and development process directly affect an organization’s bottom
line and many pharmaceuticals research organizations have experienced the
positive impact of deploying SPSS. Offering a choice of desktop (stand-alone),
networks, and server based analytical platforms. Statistician has provided
analytical tools that are easy to use, relevant and integrated with their data
sources. Comprehensive Statistical Test & Graphical Tools. Efficient use of
limited Statistician & Training Resources. Integrated with Existing System &
Data Sources. Rapid return on investment.

9. REGULATORY SAFETY TESTING From simple bioassay and dose-response

experiments to long-term survival and carcinogenicity studies, SPSS has the
tools to handle your data. Where the standard study designed are used. SPSS
tools provides the most efficient mechanism for automating and validating
routine analysis report. Power and Sample Size Calculation. Predictive Data
Mining. Forecasting and Report Generation. Analysis of Complex Drug’s Assay.

INTERPRETATION is the act of explaining, reframing, or otherwise showing

your own understanding of something. A person who translates one language
into another is called an interpreter because they are explaining what a
person is saying to someone who doesn't understand. Interpretation requires
you to first understand the piece of music, text, language, or idea, and then
give your explanation of it. A computer may produce masses of data, but it
will require your interpretation of the data for people to understand it.

Definitions of interpretation:
The act of interpreting something as expressed in an artistic performance
An interpretation of something is an opinion about what it means.

Techniques of interpretation

63
Consecutive interpretation
Consecutive interpretation is a technique in which the interpreter serves as
an intermediary between the speakers. The interpreter listens to the speech
and the speaker pauses and allows the interpreter to repeat their statements
in another language. The interpreter begins interpreting after the speaker has
paused. Consecutive Interpretation is commonly used for depositions, court
hearings, business meetings and negotiations, medical appointments, tours,
informal meetings and other social events.

Simultaneous interpretation
Simultaneous interpretation is a technique in which the interpreter renders
the interpretation simultaneously while the speaker is still speaking.
Simultaneous interpretation is used in large conferences, conventions, and
seminars where there is only one speaker addressing an audience. The
speaker speaks into a microphone and the interpreter wears a headset which
enables them to listen to the speaker and almost simultaneously render the
message into a microphone in the target language. Those people needing
interpretation wear an earpiece which allows them to hear the interpreter.
This technique is often used in media interpreting and larger conferences. A
simultaneous interpreter must have an absolute command of the target and
the source languages, the ability to interpret instantaneously and accurately,
and the knowledge of specialized terminology. TransCultures’ interpreters
possess all of these skills thanks to our rigorous screening and qualifying
process. Many of our conference interpreters are members of prestigious
interpreter organizations such as ATA, AIIC and TAALS.

Whispered interpretation / Chuchotage

Whispered interpretation or Chuchotage is a technique where the interpreter
provides interpretation simultaneously to a small audience, usually less than
four people. In this setting, the interpreter sits or stands close to the audience
and whispers the interpretation into their ears. It is used mainly in bilateral
meetings and small settings. Headsets are not needed since it is a small
group and no extra time is needed to interpret because the interpretation is
done while the speaker is speaking.

Sight Translation
Sight translation is a technique that requires the interpreter to render the
content of a written document in the source language orally into the target
language. This technique of interpretation is often used during legal and
medical interpretation.

Liaison Interpretation The interpreter translates a few of the speaker’s

sentences at a time; his counterpart replies and the interpreter translates
these sentences back to the speaker. Liaison interpreters work both ways, so
conversations take twice as long. Suitable for: business presentations, press
conferences, guided tours, negotiations, interviews or executing deeds and
contracts.

What Is Report Writing?

Report Writing is a well-organized writing process that needs enough skill,
research, and details.

64
This report writing process is usually time-consuming and often requires a
detailed research for which the report writer may have to visit places, meet
people and find a solid conclusion at the end of the day to write a good
report.
Use of report writing is to explain a matter or issue and inform higher
authorities to help them take a right decision or action in regards to the
matter or issue.
There is a little room for creativity in the report writing structure as the main
focus is to enlighten the readers about a matter or issue and make the whole
thing self-explanatory for easy understanding.
Report Writing is the primary tool of media personnel through which they pass
on specific information about an incident or topic.
Apart from the media usage, report writing is required in various sectors like
corporate, government, politics etc.
To offer a better understanding, learning about the essence of this form of
writing – ‘The Report’ is very important.

What is a report?
A report is a methodical, well planned document which outlines and evaluates
a subject or problem, and which may include:
• The record of an order of events
• Explanation of the implication of these events or facts
• Evaluation of the facts or results of research presented
• Discussion of the consequences of a conclusion or course of action
• Conclusions
• References

65
Reports must always be:
• Correct
• Crisp
• Clear
• Well-Structured

Report Writing Format

Every organization has a predefined report writing format and hence a writer
can’t remain stuck to a specific format for different reports.
While writing a formal report, the freelance writer must keep in mind that the
target readers need don’t have to do unnecessary research to take a decision
or action after reading the report.

Types of Report Writing

Everything must be in details. There are many types of report writing for
organizations that are used for various purposes. To format your report to
serve varied needs, have a look at the primary report writing types
mentioned below.

Formal Report Writing

Formal Report Writing is pretty complex and time-consuming. Usually, it
demands an immense research, explanation, references, links, lists and many
other things to make the primary point clear enough for the readers.

This type of report writing is usually preferred for an important incident, issue
or matter by big organizations. Formal report writing is generally long and
expensive.
Formal Report writing has an internationally accepted pattern that includes
various components that are mentioned below.

Cover – The cover of a report is something that drives a reader’s attention

first. The report’s cover leaves a huge impression on the reader’s mind and
he/she can get an idea regarding the report’s topic or quality through the
cover design.
There is a saying – ‘’don’t judge a book by its cover’’ but it is also true – what
seems well, sells well and hence the report writer must take care of the cover
of the report with has a major role to depict the report brief.

Title Of The Report – This component includes the report’s title and the
name of the writer. Apart from these things, the title can also have a date and
the name of the organization for which the report has been prepared.
The cover also has these things but putting them in details in the title section
is mandatory.

Table Of Contents – This section includes headings and subheadings of the

primary text written. This is a very important portion of report writing. It helps
your readers to reach desired sections in your report in a hassle-free manner.

Summary – Here you basically provide the synopsis of the whole report’s
primary text and you can also call it an informative summary.

66
Many times, it is referred as ‘executive summary’. You can also use
descriptive summary which is a simple table of contents. The format is always
decided by the organization.

Introduction – This is the most important section of the main text. The main
text always includes three components – introduction, discussion, and
conclusion.
Here, you introduce the main text of your report in the most intriguing a
detailed manner so that all types of readers can get your point without much
effort.

Discussion – In the discussion section, a report writer discusses the main

story of the report. According to your reader’s convenience, you decide the
order of the report’s results.
You can also do a result to theory comparison here along with the analysis,
evaluation and interpretation of the data included.

Conclusion/Recommendation – You can present the summary of the

discussion section here. Here, you mention your findings and recommend the
elements to your readers as per your overall evaluation.

Appendix – In the appendix portion, you can attach the graphs, lists, survey
and suchlike stuff that are related to your report and helps your readers to
understand the report comprehensively.

Reference/Bibliography – You can cite your report’s references in this

portion.

EXECUTIVE SUMMARY:

Definition:
An executive summary of a business plan is an overview. Its purpose is to
summarize the key points of a document for its readers, saving them
time and preparing them for the upcoming content.

Think of the executive summary as an advance organizer for the reader.

Above all else, it must be clear and concise. But it also has to entice the
reader to read the rest of the business plan.
This is why the executive summary is often called the most important part of
the business plan.

What Information Goes in an Executive Summary?

The information you need to include varies somewhat depending on whether
your business is a startup or an established business.
For a startup business typically one of the main goals of the business plan
is to convince banks, angel investors, or venture capitalists to invest in your
business by providing startup capital in the form of debt or equity financing.
In order to do so you will have to provide a solid case for your business
idea which makes your executive summary all the more important.
A typical executive summary for a startup company includes the following
sections:
67
 The business opportunity - describe the need or the opportunity.
 Taking advantage of the opportunity - explain how will your business will
serve the market.
 The target market - describe the customer base you will be targeting.
 Business model - describe your products or services and and what will make
them appealing to the target market.
 Marketing and sales strategy - briefly outline your plans for marketing
your products/services.
 The competition - describe your competition and your strategy for getting
market share.

What is your competitive advantage, e.g. what will you offer to

customers that your competitors cannot?

 Financial analysis - summarize the financial plan including projections for at

least the next three years.
 Owners/Staff - describe the owners and the key staff members and the
expertise they bring to the venture.
 Implementation plan - outline the schedule for taking your business from
the planning stage to opening your doors.
For established businesses the executive summary typically includes
information about achievements, growth plans, etc. A typical executive
summary outline for an established business includes:
 Mission Statement – Articulates the purpose of your business. In a few
sentences describe what your company does and your core values and
business philosophy.
 Company Information – Give a brief history of your company - describe
your products and/or services, when and where it was formed, who the
owners and key employees are, statistics such as the number of employees,
business locations, etc.
 Business Highlights – describe the evolution of the business - how it has
grown, including year-over-year revenue increases, profitability, increases in
market share, number of customers, etc.
 Financial Summary – if the purpose of updating the business plan is to seek
additional financing for expansion, then give a brief financial summary.
 Future goals – describe your goals for the business. If you are seeking
financing explain how additional funding will be used to expand the business
or otherwise increase profits.

How Do I Write an Executive Summary of a Business Plan?

Start by following the list above and writing one to two sentences about each
topic (depending on whether your business is a startup or an established
business). No more!
The Easy Way of Writing One

Having trouble getting started?

The easiest way of writing the executive summary is to review your business
plan and take a summary sentence or two from each of the business plan
sections you’ve already written. If you compare the list above to the sections
outlined in the Business Plan Outline, you’ll see that this could work very well.

68
 Then finish your business plan’s executive summary with a clinching closing
sentence or two that answers the reader’s question, “Why is this a winning
business?”
For example, an executive summary for a pet-sitting business might
conclude: “The loving on-site professional care that Pet Grandma will provide
is sure to appeal to both cat and dog owners throughout the West Vancouver
area.”
(You may find it useful to read the entire Pet Grandma executive summary
example before you write your own.)
Tips for Writing the Business Plan’s Executive Summary
 Focus on providing a summary. The business plan itself will provide the
details and whether bank managers or investors, the readers of your plan
don’t want to have their time wasted.
 Keep your language strong and positive. Don’t weaken your executive
summary with weak language. Instead of writing, “Dogstar Industries might
be in an excellent position to win government contracts”, write “Dogstar
Industries will be in an excellent position...”
 Keep it short - no more than two pages long. Resist the temptation to
pad your business plan’s executive summary with details (or pleas). The job
of the executive summary is to present the facts and entice your reader to
read the rest of the business plan, not tell him everything.
 Polish your executive summary. Read it aloud. Does it flow or does it
sound choppy? Is it clear and succinct? Once it sounds good to you, have
someone else who knows nothing about your business read it and make
suggestions for improvement.
 Tailor it to your audience. If the purpose of your business plan is to entice
investors, for instance, your executive summary should focus on the
opportunity your business provides investors and why the opportunity is
special. If the purpose of your business plan is to get a small business loan,
focus on highlighting what traditional lenders want to see, such as
management's experience in the industry and the fact that you have
both collateral and strategies in place to minimize the lender's risk.
 Put yourself in your readers’ place... and read your executive summary
again. Does it generate interest or excitement in the reader? If not, why?
Remember, the executive summary will be the first thing your readers read. If
it's poorly written, it will also be the last thing they read, as they set the rest
of your business plan aside unread!

14 Precautions A Researcher Should Keep in Mind While Preparing

the Research Report

A research report forms a channel of communication between the research

findings given by researcher on the one hand and the readers of the report on
the other hand and effective communication of research findings to the
readers is the basic task of the research report.

Hence researcher should keep in mind the following precautions while

preparing the research report:
1. The length of the report should be kept keeping in view the fact that it
should cover the subject in length but it should be short enough to maintain
the interest of readers.

69
2. The report should not be dull. It should be such as to sustain one’s interest.

3. The report should possess the features of simplicity and avoiding

vagueness.
4. Readers of the report are more interested in acquiring a quick knowledge of
the main findings and the report should provide a ready availability of the
findings with the help of charts, graphs and statistical tables.

5. The contents of the report should be well thought out keeping in view the
objective of the research problem.

6. The grammatical mistakes should be avoided and the report should be

prepared strictly in accordance with the techniques of composition of report.
Writing such as the use of quotations, foot notes, proper punctuations and the
use of abbreviations in the foot notes etc.

7. Logical analysis should be the very basis of preparing a research report.

8. The report should also suggest the policy implication relating to the
problem under consideration and should help us in future forecasting.

9. All the technical data should be appended in the report.

10. A list of bibliography of various sources consulted for the research study
should be prepared and be given in the report.

11. A good report should also possess an index of subjects and authors as
well. It should be appended at the end of the report.

12. Report should be attractive in shape, neat and clean whether typed or
printed.

13. Calculated confidence limits must also be mentioned in the report and the
various limitations experienced by the researcher in his study should be
stated in the report.

14. A good report should also mention the objective of the study the nature of
the problem the methods and analysis techniques used. All these things
should be given in the form of introduction in the report.

Thus after describing all the things related with the interpretation and report
writing, we should always keep in view the fact that the process of writing a
report is an art which is learnt by practice and experience.

Tips on effective use of tables and figures in research papers

Several studies, journal guidelines, and discourses on scientific writing1-

10
affirm the critical role that tables, figures, and graphs (or display items)
play in enhancing the quality of manuscripts. Scientific tables and graphs can
be utilized to represent sizeable numerical or statistical data in a time- and
space-effective manner. Readers are often drawn towards tables and figures,
because they perceive it as easy-reading, as compared to reading a verbose
70
 account of the same content. They rightly assume that these display items
will provide them with a larger amount of information in a shorter time
span. At the manuscript screening stage, these display items offer reviewers
and journal editors a quick overview of the study findings, and once the paper
is published, they do the same for readers (some of whom look only at these
display items and not at the rest of the manuscript7). However, tables and
figures only add value to the format of a research report, if they are brief yet
sufficiently informative.
These visual elements help authors present detailed results and complex
relationships, patterns, and trends clearly and concisely;1-8,11,13,14 reduce the
length of the manuscript;3,5,13,15 and enhance readers’ understanding of the
study results.11,12 Therefore, these tools are integral to the format of a
research paper because, if clear and well-organized, they speed up the
comprehension and interpretation of the study’s findings. But while well-
presented tables and figures in research papers can efficiently capture and
present information, poorly crafted tables and figures can confuse readers
and impair the effectiveness of a paper.16 To help authors get the balance
right, this article presents some essential guidelines to the effective use of
tables and figures in research papers.

Planning your paper: When to use tables and figures in scientific

papers

Producing effective tables and figures requires careful planning that begins at
the manuscript writing stage itself. Here’s how to go about it:

 First, check out what your target journal has to say on the issue. Some
journals limit the number of tables and figures and also have specific
guidelines on the design aspects of these display items.
 Next, decide whether to use tables and figures or text to put across key
information.(Refer to Table 1 below for help on making this decision.)
 After you’ve decided to use a display item, choose the display item that best
fits your purpose based on what you wish readers to focus on and what you
want to present.(Refer to Table 1 below for more information.)
 Finally, follow the best-practice guidelines outlined in section 3 and review the
examples presented in section 4 of this paper to ensure that your tables and
figures are well-designed.

Table 1: How to choose between tables, figures, and text to present data

71
Best practices for presentation of tables and figures in scientific
papers

General guidelines:
1. Ensure that display items are self-explanatory: Some readers (and certainly
reviewers and journal editors) turn their attention to the tables and figures
before they read the entire text, so these display items should be self-
contained.6,8,9,12,16,20
2. Refer, but don’t repeat: Use the text to draw the reader’s attention to the
significance and key points of the table/figure, but don’t repeat
details.5,12,16,19,21,22 So for example, you could highlight your main finding (e.g.,
“We found that the treatment was effective in only 24% of the cases, as
shown in Figure 1”), but don’t repeat exact values (e.g., “As Table 2 shows,
32% of the subjects chose Option 1, 12% chose Option 2, 10% chose Option
3, and 46% chose Option 4”). This defeats the very purpose (efficiency and
clarity) of having a table or figure.
3. Be consistent: Ensure consistency between values or details in a table
(e.g., abbreviations, group names, treatment names) and those in the text.
4. Give clear, informative titles:16,17 Table and figure titles should not be
vague9,19 but should concisely describe the purpose or contents of the
table/figure and should ideally draw the reader’s attention to what you want
72
him/her to notice17 (e.g., Advantages and disadvantages of using sleep
therapy with patients suffering from schizophrenia). Also ensure that column
heads, axis labels, figure labels, etc., are clearly and appropriately labelled.
5. Adhere to journal guidelines: Check what your target journal has to say
about issues like the number of tables and figures, the style of numbering,
titles, image resolution, file formats, etc., and follow these instructions
carefully.18,23

Guidelines for tables:

1. Combine repetitive tables: Tables and figures that present repetitive
information will impair communication rather than enhance it.8,12 Examine the
titles of all your tables and figures and check if they talk about the same or
similar things.8 If they do, rethink the presentation and combine or delete the
tables/graphs.

Do you have any questions related to scientific writing?Ask Your

Question Now

2. Divide the data:7 When presenting large amounts of information, divide the
data into clear and appropriate categories and present them in columns titled
accurately and descriptively.
3. Watch the extent of data in your tables: If the data you have to present is
extensive and would make the tables too cluttered or long, consider making
the tables a part of the Appendix or supplemental material.8
4. De-clutter your table: Ensure that there is sufficient spacing between
columns and rows7 and that the layout does not make the table look too
messy or crowded.

Guidelines for figures:

1. Ensure image clarity: Make sure that all the parts of the figure are
clear:18 Use standard font; check that labels are legible against the figure
background; and ensure that images are sharp.24
2. Use legends to explain the key message:25 Figure legends are pivotal to the
effectiveness of a figure. Use them to draw attention to the central message
as well as to explain abbreviations and symbols.
3. Label all important parts: Label the key sections and parts of schematic
diagrams and photographs, and all axes, curves, and data sets in graphs and
data plots.6,7,16,18
4. Give specifics: Include scale bars in images and maps; specify units
wherever quantities are listed; include legends in maps and schematics; and
specify latitudes and longitudes on maps.

Examples of well-prepared tables and figures

This section presents one example each of a well-prepared table and a well-
designed figure.
The table below is taken from a dietary study on chick-rearing macaroni
penguins26 and is an example of an effective table for the following reasons:

73
1. The title clearly describes what the table is about.
2. The column heads are descriptive and clearly indicate the nature of the
data presented.
3. The data is divided into categories for clarity.
4. It is self-contained and can be understood quite well even without
reference to the entire paper.
5. Superscript letters and notes are used to offer additional, clarifying
information.
6. Sufficient spacing is present between columns and rows; the layout is
clean; and the font is legible.

Examples of an effective figure (graph)

The figure below from a paper on the efficacy of oyster reefs as natural
breakwaters27, scores on several counts:

74
 1. The informative title that immediately tells the reader what to expect
in the graph.
2. The axes are labeled clearly.
3. The key clearly identifies what each element in the graph stands for.
4. A figure legend at the bottom draws the reader’s attention to the
graph's key
points.
5. A note at the bottom acknowledges the source.
6. The graph is 2-dimensional, with no clutter.
Conclusion
Figures and tables, or display items, are powerful communication tools—they
give your manuscript a professional feel, attract and sustain the interest of
readers, and efficiently present large amounts of complex
information.7 Moreover, as most journals editors and reviewers will glance at
these display items before they begin a full reading of your paper, their
importance cannot be overemphasized.

75