Professional Documents
Culture Documents
5th Sem Business Research Methods
5th Sem Business Research Methods
5th Sem Business Research Methods
RESEARCH METHODOLOGY
CAPSULE NOTES Benefits of Business Research
(Module 1)
1. Information for economic planning – Business
Business Research research collects information on the economic
• Business research is social science research & social structure of the nation.
2. It helps to solve operational & planning
• Business research is a type of research used to
problems.
identify investment opportunities or start of a
3. It helps in business decision making.
business or succeeding improvement.
4. It helps in forecasting demand, production etc.
• According to Mc Daniel & Gates, “Business 5. It helps to formulate policies at government
research is the planning, collection & analysis of levels.
data relevant to business decision making &
communication of the results of this analysis to
Management”.
SOCIAL RESEARCH
Features of Business Research • Research in social science is called social
1. It is a systematic & objective process of research
gathering, recording & analyzing data for • Scientific investigation conducted in the field of
decision making. social and behavioural sciences
2. Business research has a clear objective. • Discovering some facts concealed in a social
3. Business research is multi disciplinary. phenomena or some law governing it.
4. It covers all the regions in which the business • Application of scientific method for
operates, locally & internationally. understanding, studying and analysing a social
5. It helps to judge local problems of the life or phenomena
environment. • To modify, correct and verify the existing the
6. It helps to determine the cost or profitability of knowledge as a system
the business.
7. It is flexible in nature
8. Business research is most often focused on Limitations of Social Research (Difficulties in
demand. conducting Social Research)
• Social data is based on human behaviour which
Objectives of Business Research: - is highly complex and confusing
• It is very difficult to establish cause and effect
1. To promote business in an economically sound
relationship
manner.
• Social phenomena is dynamic (changing)
2. To ensure future success.
• Subject to prejudices and bias
3. Gather relevant information for smooth
conduct of business.
• Social events are non repetitive – difficult to
4. It helps the manager to identify & solve verify
problems. • Problem of collecting information – human
5. To save time & cost. beings are reluctant to disclose information
6. It facilitate managerial decision making.
TYPES OF RESEARCH
Importance of Business Research
Basic Research (Pure/Fundamental/ Theoretical
1. To Investors: - Investors would investigate the
Research)
past performance, business planning & the
• Original type of research
business owners’ expertise etc before making
• Pure research is undertaken for the sake of
investment.
2. To Companies: - Business research helps the knowledge without any intention to apply it into
companies to ensure production of goods & practice.
services according to the demand of the • Undertaken out of intellectual curiosity
consumers.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
2
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
3
Secondary Data Analysis between factors that explain the present status or the
development over a period of time.
Analysis of secondary data helps to define the business
research problem. Secondary sources may be both Examples of a case study are social anthropological
internal & external in character. study of a rural or tribal community, a causative study
of a successful co operative society, a study of labour
a) Internal Secondary Data: - It refers to the participation in Mgt in a particular enterprise etc.
information that already exists within the
company or unit. E.g. record of sales, budget, The case study method describes a case in
advertisement expense, previous market terms of its peculiarities. It gives us an insight into the
research studies etc. typical or extreme cases whose unique features are not
b) External Source of Data: - It refers to reflected by the usual statistical method. A case study
information which is collected by a source helps to secure a wealth of information about the unit
external to business or unit. It may be public or of study, which may provide clues and ideas for further
private document. E.g. Govt. publications, research.
journals, magazines etc.
Pilot Study
Merits of Secondary Data Analysis
A pilot study is a small scale preliminary study
1. Secondary data provide an insight to the total conducted before the main research in order to check
situation. the feasibility or to improve the design of the research.
2. Secondary data provide complementary It is a mini study before the main study to eliminate all
information. the bugs. It helps to know the nature and different
3. Secondary data are economical in nature. aspects of the problem. The pilot study enables the
4. Secondary data help to save time & cost. researcher to gain some systematic knowledge of the
5. Secondary data are most reliable in some universe and its population under study.
cases.
Advantages of the pilot study
Limitations of Secondary Data
1. It helps in the selection of respondents.
1. Since these are collected for another purpose, 2. Helps to expose the inadequacies of the
these may not be suitable for the study of the questionnaire.
researcher. 3. It provides necessary training & motivations for
2. Since these are collected in past, they may not the investigators to carry out the planned job.
be the current information. 4. It provides an opportunity to the researcher to
3. The analyst is not participating in the planning build up self confidence.
& execution of the data collection. 5. It helps to estimate time & resources for
completing the work successfully.
Precautions to be taken while using secondary data
Focus Group Interviews
1. Reliability: - It should be collected from the
reliable sources. It is the method of conducting an interview by a trained
2. Suitability: -The data should be suitable to the moderator in a non structured and natural manner
objective, scope & nature of the study. with a small group of respondents.
3. Adequacy: - The level of accuracy should be
adequate for the study. A focus group is a form of qualitative research in which
a group of people are asked about their perceptions,
Case Study opinions, beliefs and attitudes towards a product, a
service, a concept, an advertisement, an idea or a
A case study is an in depth comprehensive study of a package etc. questions are asked in an interactive
person, an episode, a process, a situation, a group setting where participants are free to talk with
programme, a community, an institution or any other the group members.
social unit. It is one of the most popular types of
research methods. Its purpose may be to understand
the life cycle of the unit under study or the interaction 4. DESCRIPTIVE RESEARCH
• It is the simplest type of research.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
4
• Descriptive research is a fact finding • Some of these steps may be covered quickly and
investigation. some others may require longer time
• It focus on a particular aspect of the problem
• It is designed to gather descriptive information Steps in Research Process
and provides information for formulating more 1. Tentative selection of the topic or Research
sophisticated studies. problem
• It aims at identifying various characteristics of a 2. Initial survey of literature
problem, community or an institution. 3. Finalisation of research problem
• It helps in predicting social life and helpful for 4. Formulation of the problem
planning social oriented programmes. 5. Extensive Literature survey
5. Causal/Predictive/experimental Research 6. Formulation of hypotheses
• Causal research deals with the why questions. 7. Preparation of research design
• It is used to obtain evidence of cause & effect 8. Preparation of sample design
relationships. 9. Collection of data & Execution of the project
• It assesses the effect of a particular variable on a 10. Analysis of data
phenomenon by keeping the other variable 11. Testing hypotheses
constant. 12. Generalisation and Interpretation
• The objective of causal research is to test 13. Preparation of research report/thesis.
hypotheses about cause and effect relationship.
6. Diagnostic Research 1. Tentative selection of the problem
• Research conducted to identify the causes of a • The term problem means a question to be
problem and to find possible solutions answered or an issue to be examined
• It is directed towards what is happening, why it is • First of all the researcher has to select a problem
happening, and what can be done about it. • In case of Govt orgns, it is suggested by the
• Eg. Research conducted to identify the causes of administrators or policy makers.
increasing number of HIV patients and to find • In case of research scholars (students) it is to be
solutions. selected in consultation with the guide.
• Eg. Research conducted to find the causes of • If the study is for a sponsoring agency, they will
huge loss incurred by KSRTC and find solutions. suggest problem
7. Case study • The researcher has to discuss the problem with
• It is a method of explaining or analysing the life his friends, colleagues, experts etc
of a social unit (such as a person, family, social • The main sources of the topic are literatures,
group, a community, an institution etc) seminars, field visits, internships, experiences,
• An in-depth or comprehensive analysis of a social medias etc.
unit. 2. Initial survey of Literature
• Studying everything about something. • After selecting the problem, the researcher is
• It is carried on for understanding a complex issue required to go through the relevant literature to
when it is not discoverable by usual methods. bring out a frame of reference to the work
• Quantitative and qualitative methods are applied • He should review
a) The conceptual literature – Literature explaining
Research Process the concept and theories
• Research is a systematic process which involves b) The empirical literature – Studies made earlier in
number of steps. the topic or related topics.
• There are some variations in the sequence of 3. Finalisation of research topic
these steps suggested by different writers, but • Researcher has to finalise the topic and has to
there is much similarity among their suggesions represent it in analytical and specific terms
• The process is not truly linear, moving from one • Problem/topic should not be too wide or too
step to another in an order narrow
• The order of some steps may be changed, some • It should be suitable for the academic
steps may be avoided, and some steps may background – for which the research is
carried on simultaneously. undertaken
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
5
• Availability of time, money and other resources • Sampling design refers to the technique or the
must also be considered. procedure, the researcher would adopt in
4. Formulation of a problem selecting items as sample.
• Selected problem should be translated or • The researcher can use probability or non
transformed into a scientific research question probability sampling
• It is concerned with specifying what actually the • It is not needed for census survey
research problem and why it is studied. 9. Collection of data/Execution of the project
• It involves • The required data is to be collected from
a) The development of title different sources
b) Define the objectives of the study • Primary data and secondary data are to be
c) Work out the conceptual model etc. collected
5. Extensive Literature Review • Different methods such as observation,
• In order to gain the back ground knowledge of interview, experimentation, survey etc can be
the research topic and to find out the research used
gaps, the researcher has to review books and • Various tools such as questionnaires, inventory
other literatures. schedules etc can be used
• All kind of literatures such as books, articles, • Questionnaires and interview schedules must be
source materials, encyclopaedia, latest formulated carefully and systematically
publications, selected research theses, • Proper training should be given to interviewers
bibliographies of research reports, online 10. Analysis of Data
information, reports of agencies etc. • In order to analyse the data collected, it should
Purpose of Review be edited, classified and tabulated.
• To gain preliminary orientation and back ground • Statistics and quantitative techniques can be
knowledge about the problem used to analyse the data
• To gain up-to-date knowledge • Ratios, percentages, graphs, tables, coefficients
• To know the work already done on the topic or etc can be used to analyse data
related topic • Statistical software packages are also available to
• To identify research gaps in the field analyse the data
• To avoid duplication of work 11. Testing of hypothesis
• To get an idea about the research work • It is the process under which a statistical
6. Formulation of Hypothesis hypothesis is tested and then accepted or
• A hypothesis is a tentative assumption or rejected.
proposition formulated for testing • The test conducted to accept or reject a
• It is a tentative generalisation, the validity of hypothesis is known as Test of hypothesis.
which remains tested • The commonly used statistical tests are Z test, t
• Some type of research does not need hypothesis. test, X2 test and F test
Eg.Mere fact finding investigation • Test help the researcher to conclude the validity
• It is made to draw out and test its logical and of his hypothesis
empirical consequences 12. Generalisation and interpretation
7. Preparation of Research Design • After testing hypotheses, the researcher will be
• It is a comprehensive plan of the series of able to arrive at generalisation.
operations that a researcher intends to carry out • The researcher has to prepare his findings and
to accomplish the research objectives. conclusions
• It specifies the objectives of the study, the • It should be justified with the objectives set
methodology and the techniques to be adopted earlier.
for achieving the goal. • It should be based on proper evidences under
• It prescribes boundaries for the research work study.
and gives a direction to the study 13. Preparation of the Report/Thesis
8. Preparation of Sample design • It is the final stage of the research process
• It is a definite plan for obtaining a sample from a • The researcher has to prepare a research report
given population. which indicates what he has done and how it is
proved.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
6
• It is the formal statement of the research process 3. Lack of Technical knowledge: Lack of Knowledge
and its results in computer, internet, software packages,
• It narrates the problem studied, methods used statistics etc for analysis etc.
for study and the findings and conclusions. 4. Lack of co-operation from respondents:
• The purpose of the report is to communicate the Respondents (including business firms) are
research work to others generally reluctant to provide data because they
are of the opinion that the researcher can misuse
Criteria of a good research problem problem the data.
5. Lack of Code of Conduct for researchers:
Every problem selected for research must satisfy the Different methods and system are followed by
following criteria:-
various disciplines, departments and research
1. It should be original: - The purpose of centres.
research is to fill the gaps in existing 6. Insufficient Interactions: between researchers
knowledge to discover new facts and not to and business, Government agencies etc.
repeat already known facts. Therefore, a 7. Inadequate Assistance: From Govt and other
preliminary survey of the existing literature in agencies.
the proposed area of research should be 8. Difficulty in Social Science Research: Complex
carried out to find out the possibility of making and dynamic social data, problems of collecting
original contribution. information.
2. It should be neither very general nor very 9. Timely availability of published data: from
specific: - If the problem is very general, it is various Government and other agencies
usually too vague to be tested. On the other 10. Huge misleading data on internet: There is no
hand, if the problem is very specific, it is way to check the reliability of data available on
usually too narrow to be important or internet
consequential. Some kind of compromise must 11. Lack of research centres, research guides and
be made between generality and specificity. sponsors: Sufficient number of research centres,
3. It should be solvable: - Generally, a problem
research guides and sponsoring agencies are not
may be unsolvable due to two reasons – (i) it
available in many subjects.
may concern some supernatural or amorphous
12. Improper Library management: Libraries are not
phenomena. For e.g., problems such as “how
managed systematically and hence the
does the mind work?’, “is it possible to change
human nature?” etc. (ii) it cannot be researchers are not able to get the old copies of
operationally defined. books, journals etc.
4. It should be feasible: - The feasibility of 13. High cost of publishing: Publishing research
carrying out research on the selected problem papers in national and international journals is
should be checked against the following highly expensive and discouraging factor.
considerations: -
I. Study design.
II. Access to organization and Theory Building
respondents. • The word theory is derived from the Greek word
III. Sample or universe to be studied.
“theoria” which means contemplation or
IV. Source of data.
speculation.
V. Funds required and availability etc.
• According to Jack Gibbs, theory is a set of
logically interrelated statements in the form of
empirical assertions about properties of infinite
Hurdles(Problems) faced by Researchers class of events or things.
• A good theory is a statement of relations among
1. Time and Money: Researchers have to spent concepts within a set of assumptions and
their time and spend money for the collection, constraints.
analysis and interpretation of data.
2. Lack of scientific training: Lack of knowledge
/training in the methodology of research.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
7
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
8
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
9
2. It helps in systematic planning & execution of For example, in an opinion poll, possible sampling
the study. frames include an electoral register and
3. It helps the researcher to document the a telephone directory.
research activities. MEASUREMENT AND SCALING
4. It ensures proper time schedule for the
implementation of the project. Measurement is a systematic way of assigning numbers
5. It boosts the confidence of the researcher for or names to objects and their attributes.
completing the research work.
6. It provides a sense of success at every stage of Definition – According to Kenneth D Bailey,
work. “measurement is the process of determining the value
or level, either qualitative or quantitative, of a
What is the Validity of a Study? particular attribute for a particular unit of analysis.
Sampling Frame Ratio Scale: - This is known as the ideal scale because
of the characteristics it possesses. A ratio scale
The most straightforward type of sampling frame possesses the characteristics of nominal, ordinal &
interval scales. Examples of variables which are ratio
is a list of elements of the population (preferably
scaled include weights, lengths & times. In ratio scale
the entire population) with appropriate contact there is always a meaningful absolute zero
information.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
11
7 6 5 4 3 2 1
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
12
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
13
• Sample design – plan for selecting sample wider sample. An intensive study covers a few
samples.
• Data collection – which type of data, sources, 5. A survey covers a definite geographical area
methods, tools. like a city, district etc.
Census Method 4. If we want data about each & every item of the
population, this method cannot be used.
The census method involves a complete enumeration
of all units of the population or universe. Census EXPERIMENT
method enables to yield correct factual data. This type
of survey involves great deal of time, money and Experimentation is a research process used to study
energy. When the universe is small, it is better to adopt the causal relationships between variables. It aims at
census method for collection of information. studying the effect of an independent variable on a
dependent variable, by keeping the other independent
Merits of Census Method variable constant through some type of control.
1. The data are collected from each & every item Experimentation requires special efforts. It is often
of the population. extremely difficult to design & it is also a time
2. The results are more accurate & reliable. consuming process. But, experiment is the only
3. Intensive study is possible. method which can show the effect of an independent
4. Collected data can be used for various surveys variable on dependent variable. The fundamental
& analysis. weakness of any non experimental (Survey,
observation etc) study is its inability to specify cause &
Demerits effect. It can show only correlations between variables,
but correlation alone never prove causation.
1. It requires more time, efforts & money.
2. Large number of enumerators is required. Planning & Conducting Experiments
3. If the universe is infinite, this method is not
possible. It is easy to conceive ideas, but difficult to translate the
ideas into a workable, credible, meaningful set of
Sample Survey experimental operations. This to a great extent
depends upon the researcher’s knowledge, imagination
Sample is that part of universe or population which we
& intelligence. However, the general procedure in
select for the purpose of our study. The selected
experimentation can be summarized as follows: -
respondents constitute sample, the selection process is
called sampling & the survey so conducted is called 1. Determine the hypothesis to be tested & the
sample survey. independent and dependent variables
involved in it.
Sampling may be defined as the selection of part of an
2. Operationalize the variables by identifying
aggregate or totality on the basis of which a judgment
their measurable dimensions.
or inference about the aggregate or totality is made.
3. Select the type of experimental plan. The
Merits types of experimental design based on types
of control may be classified into – (1) one
1. It helps to collect more information. group plan, using the same group as
2. It facilitates drawing generalizations about experimental & control group and measuring it
large population on the basis of study of their before & after experimental treatment; (2)
representative sample. matched group plan, consisting of two
3. Various methods can be used to collect data. identical groups, one to be used as control
4. It is simple to administer. group & another as experimental group, with
5. Coding, analysis & interpretation of data is (a) Post test only measurement or (b) pre test
relatively simple. – post test measurements.
6. In certain cases, it is the only method, we can 4. Choose the setting. The setting may be field or
use. laboratory.
5. Make the experimental conditions as nearly
Demerits same as the expected real life conditions.
6. Make a record of pre experimental conditions.
1. The success of this method depends upon the
7. Introduce appropriate methods for controlling
willingness& co operation of the respondents.
extraneous variables that are not manipulated
2. Sampling survey is subject to sampling errors.
in the experiment.
3. It is expensive in some cases.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
15
8. Apply the experimental treatment & record Abelson defines simulation as “the exercise of
observations and measurements using a flexible limitation of process & outcomes for the
appropriate measurement devices. If possible, purpose of clarifying or explaining the underlying
repeat the tests several times in order to mechanism involved”.
insure the accuracy of results.
9. Analyse the results, using appropriate Simulation is a theoretical model of the elements,
statistical devices. relations & processes which symbolize some referent
10. Interpret the results, giving consideration to all system.
possible extraneous conditions.
Simulation Vs Experimentation
Advantages of Experimental Method
A real life experiment may be too costly or may not be
1. The variables can be more effectively possible because of practical difficulties or of the
controlled in this method. complexities of the phenomenon. For Eg the research
2. The element of human errors is reduced to the for an optimal approach to the problem of inventory
minimum. issue & replenishment through experimentation would
3. In this method, better conditions for disrupt operations for a certain period of time & also
conducting experiments may be created. would be very costly. Such difficulties do not arise in
4. Experimentation yields generally exact simulation, because it experiments with models & can
measurement & can be repeated for verifying artificially manipulate variables.
results.
Types of Simulation
Disadvantages
Computer Simulation
1. It is difficult to establish comparable control & Man Simulation
experimental groups. Man Computer Simulation
2. The scope for experimentation with human
OBSERVATION
beings is extremely limited.
3. It is often difficult to design, tends to be Observation means viewing or seeing. Observation
expensive and time consuming. may be defined as a systematic viewing of a specific
4. It is artificial to some extent & may lack phenomenon in its proper setting for the specific
realism. purpose of gathering data for a particular study.
Observation as a method includes both seeing and
Laboratory Experiment: - A laboratory experiment is
hearing. It is accompanied by perceiving as well.
an investigation conducted in a setting created
specifically for the purpose. The researcher Characteristics
manipulates an independent variable & studies its
effect on a dependent variable, keeping other variables 1. It is both a physical & a mental activity.
constant. 2. It is selective. A researcher does not observe
anything and everything, but selects the range
Field Experiment: - This is an experiment conducted in of things to be observed on the basis of the
a real life situation in which the experimenter nature, scope & objectives of his study.
manipulates an independent variable in order to test a 3. Observation is purposive & not casual. It is
hypothesis. Compared with laboratory experiment, a made for the specific purpose of noting things
field experiment incorporates less control. As it is relevant to the study.
difficult to foresee & control extraneous factors in a 4. It captures the natural social context in which
field experiment, its result is subject to large persons’ behavior occurs.
uncontrolled variation. 5. Observation should be exact & be based on
standardized tools of research such as
SIMULATION
observation schedule, social metric scales etc.
Simulation is one of the forms of observational
Types of Observation
methods. It is a process of conducting experiments on
a symbolic model representing a phenomenon. Participant Observation
Non participant Observation
Direct Observation
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
16
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
17
contains the names of all items of the universe. Eg provides a known non zero chance of selection for each
Telephone Directory, Admission Roll etc. population element.
Identify the Sampling Unit: -This is the element or set Features of Probability Sampling
of elements considered for the selection of a sample
like house, households, individuals, blocks, wards etc. 1. In probability sampling, every item of
population has a chance of being selected.
State Sampling Method: - A number of methods are 2. Such chance is known as probability.
used for selecting samples. It includes probability 3. Probability sampling yields a representative
sampling methods and non probability sampling sample, and hence the findings of the sample
methods. Researcher should select one method which survey are generalisable to the population.
is most suitable for his study. 4. Through randomization, the danger of
unknown sampling bias can be minimized.
Determine the Sample Size: - It refers to the number of
items to be selected from the universe. Sample size Methods
should not be large or small. It should be optimum.
That means the sample size elected is one which fulfils Simple Random Sampling: - This sampling technique
the requirement of consistency, competence, flexibility gives each element an equal and independent chance
and representativeness. of being selected. For e.g. in a population of 300, each
element theoretically has 1/300th chance of being
Spell out the Sampling Plan: - In this stage, the selected.
researcher should state the operational procedures for
selection of the sample. Advantages
Select the Sample: - In this stage, the needed samples 1. All elements in the population have an equal
are to be selected. chance of being selected.
2. Of all the probability sampling techniques,
Sampling Techniques or Methods simple random sampling is the easiest to apply.
3. It is the simplest type of probability sampling to
Sampling techniques or methods may be classified into understand.
two generic types: - 4. The amount of sampling error associated with
any sample drawn can easily be computed.
Probability or Random Sampling
Non Probability or Non Random Sampling Disadvantages: -
Probability sampling is of following types: - 1. It is often impractical because of non
availability of population list or enumerating
1. Simple Random Sampling
the population.
2. Stratified Random Sampling
2. This technique does not ensure proportionate
3. Systematic Random Sampling
representation to various groups constituting
4. Cluster Sampling
the population.
5. Area Sampling
3. The size of the sample required to ensure its
6. Multi stage Sampling
representativeness is usually larger under this
method.
4. A simple random design may be expensive in
Non Probability Sampling may be classified into: - time & money.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
18
divided into managers & non managers & each of those 1. This method ignores all elements between two
two groups may be sub divided salary grade wise Kth items selected. Further, except the first
strata. item, other selected items are not chosen at
random.
Advantages 2. As each element does not have an equal
chance of being selected, the resulting sample
1. It enhances the representativeness of the
is not a random one.
sample by giving proper representation to all
3. This method may sometimes give a biased
sub groups in the population.
sample. If by chance, several Kth items chosen
2. It gives higher statistical efficiency than that
represent a particular group, that group would
given by simple random sampling for a given
be over represented in the sample.
sample size.
3. It is easy to carry out this sample method. CLUSTER SAMPLING
Disadvantages Where the population elements are scattered over a
wider area & a list of population elements is not readily
1. A prior knowledge of the composition of the
available, the use of simple or stratified random
population & the distribution of the population
sampling method would be too expensive & time
characteristics are required to adopt this
consuming. In such cases, cluster sampling is usually
method.
adopted.
2. This method is very expensive in time & money
3. The identification of the strata might lead to Cluster sampling means random selection of sampling
classification errors. units consisting of population elements. Each such
sampling unit is a cluster of population elements. Then
Systematic Sampling or Fixed Interval Method
from each selected sampling unit, a sample of
This method is an alternative to random sampling. It population element is drawn by either simple random
consists of taking every Kth item in the population selection or stratified random selection.
after a random start with an item from 1 to K. for e.g.
Difference Between Cluster Sampling & Stratified
suppose it is desired to select a sample of 20 students
Sampling
from a list of 300 students. Divide the population total
300 by the sample size 20, the quotient is 15. Select a Cluster Sampling Stratified Sampling
number at random between 1 & 15, using lottery
method or so. Suppose the selected number is 9. Then, 1 Sampling unit is a 1 Population element
the student numbered 9, 24 (9+15), 39 (24+15), 54, 69, cluster itself is the sampling
84…are selected as sample. unit
2 Clustering is done on 2 Stratification is done
As the interval between the sample units is fixed, this the basis of on the basis of
method is also known as fixed interval method. geographical areas or variables under
administrative study(Educational
Advantages
divisions status etc)
1. It is much simpler than random sampling. It is 3 The resulting sample 3 It yields a higher
easy to use. may give a lower degree of
2. It is easy to instruct this method to degree of representative
investigators. representativeness. sample.
3. It is easier to check whether every Kth item has
been included in the sample.
4. Sample is spread evenly over the population.
5. It is statistically more efficient than a simple Multi Stage Sampling
random sample when population elements are
ordered chronologically, by size, class etc. then Under this method, sample is prepared by stages. The
systematic sampling gives a better population is divided into a number of large sampling
representative sample. units, each of which in turn is divided into smaller
units, and so on. A random sample is taken of the large
Disadvantages units at the first stage & from those selected a further
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
19
random sample i.e. the second stage is collected of the 3. The findings cannot be generalized.
smaller units. Suppose we want to take a sample of
5000 households from Kerala. At the first stage, the Positive or Judgment Sampling
state may be divided into a number of districts & a few
This method means deliberate selection of sample
districts selected at random. At the second stage, each
units that conform to some pre determined criteria.
district may be divided into a number of villages and a
This involves selection of cases which we judge as the
sample of villages may be taken at random. At the third
most appropriate ones for the given study. It is based
stage, 5000 households may be selected from the
on the judgment of the researcher or some experts.
villages selected at the second stage.
Advantages
Area Sampling
1. It is less costly & more convenient.
This is an important form of cluster sampling. In larger
2. It guarantees inclusion of relevant elements in
field surveys, clusters consisting of specific
the sample.
geographical areas like districts, taluks, villages or
blocks in a city are randomly drawn. As the Disadvantages
geographical areas are selected as sampling units in
such cases, their sampling is called area sampling. It is 1. This does not ensure the representativeness of
not a separate method of sampling, but forms part of the sample.
cluster sampling. 2. This is less efficient for generalizing when
compared with random sampling.
NON PROBABILITY SAMPLING METHODS 3. This sampling does not satisfy the underlying
assumption of randomness.
Non probability sampling does not adopt the theory of
probability and it does not give a representative Quota Sampling
sample of the population. Following are important non
probability sampling methods: - This is a form of convenient sampling involving
selection of quota groups of accessible sampling units
Convenience or Accidental Sampling by traits such as sex, age, social class etc.
It means selecting sample units in a just ‘hit and miss’ In this method, the interviewer, instead of receiving a
fashion. Eg. Interviewing people whom we happen to list of names & addresses of persons who are to be
meet. This sampling also means selecting whatever interviewed, receives the quota or the number of
sampling units are conveniently available. Though interviews which he has to conduct of persons who
convenience sampling has no status, it may be used for satisfy certain conditions about sex, age, income etc.
simple purpose such as testing ideas or rough Thus he may be asked to interview 20 businessmen
impression about a subject of interest. over 50 years of age in a particular area. The
interviewer can then select at his own discretion, any
Advantages
20 businessmen of that area who are over 50 years of
1. Convenience sampling is the cheapest & age.
simplest method.
Quota sampling is a method of stratified sampling in
2. It does not require a list population.
which selection within the strata is nonrandom.
3. It does not require any statistical expertise.
Merits
1. It is less costly.
2. Administration is easy.
3. It is most suited in a situation where the field
work has to be done quickly.
Disadvantages 4. It may be the only practical method of
sampling a population for which no suitable
1. It is highly biased because of the researcher’s frame is available.
subjectivity & so it does not yield a
representative sample. Demerits
2. It is the least reliable sampling method.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
20
1. It does not meet the basic requirement of heterogeneous in nature, large number of
randomness. cases may be required.
2. It may not provide a representative sample. 2. Problems related with collection of data: - The
3. Chance for personal bias is very high. resources available including time & money are
to be considered before determining the size
Snowball Sampling of the sample.
3. Type of sampling method: - If stratified
This method of sampling is often used to reach
sampling method is used, small sample will be
research population which is difficult to find out. In this
enough.
case, we initially contact a few potential respondents
4. Degree of accuracy required: - If highest degree
and then ask them whether they know of anybody with
of accuracy is required, size of the sample
the same characteristics that we are looking for in our
should be large.
research. Having decided on the characteristics of the
people to be included in the sample, we find one and .
ask that one person if they know of others like
themselves and so on. Questionnaire & Schedule
1. It does not use the probability statistical This method of data collection is more popular in case
methods. of big surveys. In this method, a well prepared
2. It difficult to apply when the population is questionnaire is sent to the respondents through post
large. with a request to return it to the investigator.
3. It does not ensure the representativeness.
According to Bogardus, a questionnaire is a list of
Sampling & Non sampling Errors questions sent to a number of persons for them to
answer. It secures standardized results that can be
Sample survey does imply the study of a small portion tabulated & treated statistically.
of the population & as such there would naturally be a
certain amount of inaccuracy in the information Designing of a Questionnaire
collected. This natural or standard inaccuracy is known
as sampling error. The steps required to design & administer a
questionnaire include: -
Non sampling errors are those other than sampling
errors which occur due to the fault in planning or 1. Defining the objectives of the survey.
execution of the survey. Non sampling errors include 2. Deciding on the information required.
biases & mistakes. 3. Determining the sampling group.
4. Choose the method of reaching the target
respondents.
5. Writing the questionnaire.
6. Administering the questionnaire.
7. Piloting the questionnaire.
8. Drafting the final questionnaire.
has been already done on the same or similar problem Contents of Questionnaire
in the past.
Introductory Part
Determining the sampling group: - The factors like age, Social Background & Factual Information.
education, income etc of the prospective respondents Main Part of the Questionnaire
should be taken into account while designing the
questionnaire. Introductory Part
Choose the method of reaching the target The details regarding name of the survey, address of
respondents: - There are various methods for reaching the surveyor, serial number etc are to be included in
the target respondents like personal interview, mailed this part.
questionnaire, telephone interview etc. before
Social Background & Factual Information.
designing the questionnaire, researcher is required to
determine which method he intends to use. The details regarding age, marital status, education,
religion, political preference, size of the family,
Writing the questionnaire: - If the researcher has been
occupation, socio economic status etc are to be
loyally executed previous steps, then the most of the
included in this section.
questions will be on clear subjects.
Main part
Administering the questionnaire: - In order to ensure
the accuracy of the data to be collected, questionnaire The information to be collected is given in this part of
should be administered properly. the questionnaire.
Piloting the questionnaire: - It will help to find out the Questions to be included in the questionnaire
shortcomings of the questionnaire.
Closed end Questions
Essentials of a good questionnaire
These questions are also called fixed alternative
1. Size of the Questionnaire: - A questionnaire question. For these questions, the respondents are
should be short & simple. Vocabulary used in given a limited number of alternative responses. From
the questionnaire should also be simple. A these, the respondents have to select one which is
lengthy questionnaire will result in poor most closely matches his opinion or attitude.
response.
2. Appearance: - A questionnaire should impress Merits of closed end questions
the respondent to respond. Hence, the
appearance of paper, printing, lay out etc are 1. It gives the respondent with an easy method of
important. indicating his answer. The respondent need not
3. Clarity: - In case of questionnaire, a respondent think about how to express his answer.
has to read and understand on his own to give 2. The respondent has to depend less on memory
the right answer. Therefore, great care should in answering a question because the option
be taken while framing the same. answers prompts the respondent.
4. Specific & Explicit: - In order to get the desired 3. Responses can be easily classified so analysis
answer, the questions should be made explicit, can be straight forward.
I .e. the questions should be specific & pinpoint
Demerits
the different alternatives.
5. Sequence of questions: - A proper sequence of 1. They do not allow the respondent the
questions avoids misinterpretations of opportunity to give a different response.
questions & improves the rate of response. The 2. They suggest answers that respondents may
first few questions should create interest & not have considered before.
influence the attitude of respondent to
complete the questionnaire.
6. Make questions self explanatory: - The
questions should be self explanatory as far as
possible, so that the desired information can
be gathered & tabulation becomes easy.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
22
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
23
2. The administration of this method is also a areas. In this case, a single person is
problem. interviewed at a time.
3. There is a chance for bias when the field 2. Group Interview: - This interview is conducted
workers influence the respondents. for gathering routine or customary information
by interviewing two or more persons
simultaneously. This type of interview is
conducted to identify people’s reaction on a
common subject.
Differences between Schedule & Questionnaire 3. Formal Interview: - In this, interview is
conducted formally for collecting information
1. A schedule is used by the interviewer for
by presenting a set of well defined questions.
eliciting information from respondents & is
4. Informal Interview: - It is a flexible interview in
filled by the interviewer himself whereas, the
which the interviewer has full freedom to make
questionnaire is sent to the respondent
adequate alterations in the questions to suit
through post & the respondent fills it up &
the specific needs of the interviewer.
returns it to the sender.
5. Focused Interview: - It is meant to focus
2. Collecting data through schedule is a costly
attention on the given experience of the
affaire while collection of data through
respondent & its effects.
questionnaire is economical.
3. In case of schedule, the response is high Benefits of Interview
whereas in case of questionnaire, non
response is high as many people do not 1. More information can be collected.
respond. 2. In depth information can be collected.
4. The identity of respondent is known in 3. Information will be more reliable.
schedule while the identity of respondent may 4. It is a flexible method.
not be known in the case of questionnaire. 5. Personal information can be collected easily.
5. In the case of schedule, even from illiterates, 6. Supplementary information can also be
the data can be collected easily but in case of collected.
questionnaire, data can be collected only when Limitations
respondents are literate & co operative.
1. It is an expensive method.
Interview Method 2. There is the possibility of the personal bias of
Interview may be regarded as a systematic method by the interviewer.
which a person enters more or less imaginatively into 3. It is a time consuming process.
the life of a comparative stranger. 4. Trained staff is required for this.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
25
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
26
3. The captions & stubs should be arranged in a Tools for Statistical Analysis
systematic way.
4. The unit of measurement should be clearly Common Descriptive Techniques
defined. The most common descriptive statistics used in
research consist of percentages and frequency
5. Explanatory foot note if any should be given as
tables
footnotes.
(a)Percentages
6. If the data in a series of tables have been
Percentages are a popular method of displaying
obtained from different sources, indicate the
distribution. Percentages are the most powerful
specific sources in a place just below the table. in making comparisons. In percentages, we
7. Abbreviations should be avoided whenever simplify the data by reducing all numbers in a
possible & ditto marks should not be used in a range of
table. 10 to 100.
8. The table should be made as logical, clear, (b) Frequency Tables
accurate & simple as possible. One of the most common ways to describe a single
Data Entry variable is with a frequency distribution.
Data entry converts information by secondary or Frequency distribution can be depicted in two
primary methods to a medium for viewing & ways, as table or as a graph. If the frequency
manipulation. The researcher can store the data in a distribution is depicted in the form of a table, we
medium. The entire data can be entered in computer call it frequency table.
for statistical packages. (c) Contingency Tables
A Contingency table shows the relationship
Analysis & Interpretation of Data between two variables in tabular form. The term
Contingency table was first used by the statistician
Karl Pearson in 1904. Contingency tables are
Analysis of data means studying the tabulated material
especially used in Chi- square test.
in order to determine inherent facts or meanings. It is
the process of computation of certain indices or
measure or co efficient along with searching for
Graphs and Diagrams
pattern of relationships that exist among data groups.
Analysis can be classified into two – Descriptive
Analysis & Inferential Analysis. Types of Graphs
Descriptive Analysis: - It is the study of distribution of The following graphs are commonly used to
variables. This type of analysis describes the nature of represent data
the object or phenomenon under study. Descriptive 1. Charts or line graphs
analysis is of three types: - 2. Bar charts
1. Uni dimensional Analysis 3. Circle charts or pie diagram
2. Bi variate Analysis 4. Pictograms
3. Multi Variate Analysis
Uni Dimensional Analysis: - If the study is related with 1. Line Graphs
one variable only, it is called uni dimensional analysis. A line graph displays information in a series of
data points that each represents an individual
Bivariate Analysis: - If the study is related with two measurement or piece of data. The series of points
variables, it is called bivariate analysis.(Correlation are then connected by a line to show a visual trend
Analysis, regression analysis etc) in data over a period of time.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
27
consists of parallel bars or rectangles with lengths A pictogram, also called a pictogram or pictograph,
that are equal to the quantities that occur in a is an ideogram that conveys its meaning
given data set. through its pictorial resemblance to a physical
Histogram object. Pictographs are often used in writing and
A histogram is a graph of frequency distributions. graphic systems in which the characters are to a
It is a set of vertical bars whose are considerable extent pictorial in appearance.
proportional to the frequencies. While
constructing histogram, the variable is always
taken on the
x- axis and the frequencies on y-axis. Tools and Statistical Methods For Analysis
Frequency Polygon The tools and technique of statistics can be
The frequency polygon is a graph of frequency studied under two divisions of statistics.
distribution. Here we draw histogram of the (A)Descriptive Statistics
data and then join by straight line and mid points In descriptive statistics we develop certain indices
of upper horizontal sides of these bars. Join both and measures of raw data. They are;
ends of the frequency polygon with the x- Axis. 1. Measures of Central Tendency
Frequency Curves 2. Measures of Dispersion
A continuous frequency distribution can be 3. Measures of skeweness and kurtosis
represented by a smoothed curve known as 4. Measures of correlation
Frequency curves 5. Regression analysis
Ogive or Cumulative Frequency Curve 6. Index numbers
A frequency distribution can be cumulated in two 7. Time series analysis
ways, less than cumulative series and more 8. Coefficient of association
than cumulative series. Smoothed frequency 1.Measures of Central Tendency.
curves drawn for these two cumulative series are The central tendency of a distribution is an
called cumulative frequency curves or ogives. estimate of the "center" of a distribution of
Less than ogive curve: In less than ogive curve values. There are different types of estimates of
the upper limit per limit of each class interval central tendency such as mean, median, mode,
is taken on x- axis in increasing order. For each geometric mean, and harmonic mean.
such upper limit on x-axis, the cumulative Arithmetic Mean
frequency of all the class intervals from the first The Mean or average is probably the most
class interval to last class interval are taken on commonly used method of describing central
the y-axis. tendency. To compute the mean all you do is add
More than ogive curve: In more than ogive up all the values and divide by the number of
curve the lower limit of each class interval is value.
taken on x- axis in increasing order. For each such
lower limit on x- axis the cumulative Arithmetic mean =
frequency of all the class interval from that class
interval to the last class interval are taken on Median
y-axis. The Median is the score found at the exact middle
of the set of values. One way to compute the
3. Circle Charts or Pie Diagram median is to list all scores in numerical order, and
A pie graph is a circle divided into sections which then locate the score in the center of the sample.
each display the size of a relative piece of Mode
information. Each section of the graph comes Mode is the value of the item of a series which
together to form a whole. In a pie graph, the occurs most frequently. In the case of individual
length of each sector is proportional to the series, the value which occurs more number of
percentage it represents. times is mode.
4.Pictograms
2. Index Numbers
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
28
Index numbers are designed to measure the 5. Data Consistency: - If data collected are
magnitude of economic changes over time. inconsistent, it will not be useful to provide
Index numbers are the indicators of the various accurate results.
trends in an economy. For example, Price index 6. Use of Statistical Tools: - If the researcher
numbers indicate the position of prices whether use inappropriate statistical tools,
they are rising or falling and at what rate. inadequate sample size & faulty
calculations, it will produce fake results.
Types of Analysis
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
29
A correlation is a single number that Factor analysis tries to identify the main
describes the degree of relationship factor or the basic factors underlying the
between two variables. If the measure of subjects' response. It is also applied for
Correlation is zero, it indicates no reducing a large set of variables into fewer
correlation. On the other hand, if it is one, factors of manageable size.
it indicates perfect correlation. d) Cluster Analysis: Cluster analysis is a
b)Regression Analysis: It is the study of technique used for classifying objects into
functional relationship existing between groups. This can be used to sort data into
two variables. It helps to predict the homogeneous groups based on their
unknown values of one variable from characteristics. The resultant clusters are
known values of another variable. It is also named by the researcher on the basis of
called causal analysis. their characteristics.
3. Multivariate Analysis
- Testing of Hypothesis
Chi-square Test
- Estimation
• Used to compare between expected and
observed data. Statistical Hypothesis
• A statistical method assessing the goodness of • Hypothesis is an assumption made about the
fit or independence between a set of observed population from which the sample is drawn.
values and those expected theoretically.
• The assumption may be about the form of the
It involves: population or about the parameters of the
population.
• Test of goodness of fit
• Statistical hypothesis can be defined as a
• Test of independence tentative conclusion logically drawn about the
• Test of homogeneity population for testing.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
31
• It states that there is no true difference 1. Set up a Null hypothesis (H0) and Alternative
between sample and population. hypothesis (H1)
Alternative Hypothesis (H1). 2. Decide the Test criterion ( such as Z test. t-test,
F test etc)
• Any hypothesis that contradicts a null
hypothesis is known as an alternate 3. Specify suitable level of significance (Level of
hypothesis. significance may be taken as 5% if it is not
given)
• It states that there is difference between
sample and population. 4. Calculate the value of test statistics using
appropriate formula
Level of Significance
5. Obtain table value of test statistics
• The confidence with which the null hypothesis corresponding to the level of significance and
is accepted or rejected depends on significance degree of freedom
level.
6. Make decision about accepting or rejecting null
• The probability with which we may reject a null
hypothesis.
hypothesis when it is true is called level of
significance. The decision depends on whether the
computed value of test statistics falls in the
• Level of significance is denoted by ᾳ (Alpha) Acceptance region or Rejection region
• Eg. If the LS is 0.05(5%), the possibility of Rejection region is known as Critical region
rejecting a true hypothesis is 5/100.
Acceptance/Rejection Rule
Degree of Freedom
• When the computed value of test statistics is
• Degree Degree of freedom is defined as the numerically less than table value, it falls in the
number of favourable observations that
acceptance region & the Null hypothesis is
supports a null hypothesis. accepted.
• It is obtained by subtracting the number of • When the computed value of test statistics is
constraints from total number of observations numerically more than table value, it falls in
the rejection region & the Null hypothesis is
rejected.
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
32
1. Technical Report
2. Popular Report
3. Interim Report
4. Summary Report
5. Research Abstract
6. Research Article
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
33
academic community. It is a formal long report sponsoring agency to take action without waiting for
covering all the aspects of the research process: a the full report.
description of the problem studied, the objectives of
the study, methods & techniques used, a detailed The interim report contains a narration of what has
account of sampling, field & other research been done so far & what was its outcome. It presents a
procedures, sources of data, tools of data collection, summary of the findings of that part of analysis which
methods of data processing & analysis, detailed has been completed.
findings & conclusions& suggestions. It is so Summary Report
comprehensive & complete that the study can be
replicated by others. A summary report is prepared for the use of the lay
audience i.e. the general public. The preparation of this
The technical report is essentially technical in nature & type of report is desirable for any study whose findings
prepared in technical language. It follows a specified are of general interest. It is written in non technical,
pattern & consists of several prefatory sections with simple language with a liberal use of pictorial charts. It
appropriate headings & paragraphs. just contains brief reference to the objective of the
Popular Report study, its major findings & their implication. It is a short
report of two or three pages. Its size is so limited as to
This type of report is designed for an audience of be suitable for publication in daily newspapers
executives/administrators & other technical users. The
requirement of this audience is different. The reader is Research Abstract
less concerned with methodological details but more This is a short of summary of the technical report. It is
interested in studying quickly the major findings & usually prepared by a doctoral student on the eve of
conclusions. He is interested in applying the findings to submitting his thesis. Its copies are sent by the
decisions. university along with the letters of request to the
examiners invited to evaluate the thesis. It contains a
The organization of this report is very important. The brief presentation of the statement of the problem, the
presentation can be more forceful & persuasive objectives of the study, methods and techniques used
without any distortion of fact. It should be clear, brief and an overview of the report .A brief summary of the
result of the study may also be added. This abstract is
and straight forward. Complicated statistical
primarily meant for enabling the examiner-invitees to
techniques & tables need not be used. Instead, decide whether the study belongs to the area of their
pictorial devices may be extensively used. specialization and interest.
The style may be more journalistic but be precise & it Research Article
should encourage rapid reading & quick
comprehension. It is designed for publication in a professional journal. A
professional journal may have its own special format
Interim Report for reporting research.
When there is a long time lag between data collection Format of Research Report
& the presentation of the result in the case of a
sponsored project, the study may lose its significance & A technical report has a number of clearly defined
usefulness & the sponsor may also lose interest in it. sections. While the headings of the sections & their
One of the most effective ways to avoid such order may vary from one situation to another, the
eventualities is to present an interim report. following is suggested as a broad sequence of contents
of a research report.
This short report may contain either the first results of
the analysis or the final outcome of the analysis of A Prefatory Items
some aspects completely analysed. It facilitates the
1. Title Page
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
34
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
35
It may be divided into several chapters depending upon • Author’s name (last name first), Title of book,
the number of objectives of the study. additional information if any, City of
Publication, Publishing company, date of
Summary, Conclusions & Recommendations publising
This chapter should be a self contained summary of the Format - Magazine and news paper articles
whole report, containing a summary of essential
background information, findings & recommendations. • Name of Author (last name first), Title of
article, Title of the periodical, volume, date,
C. Terminal Items inclusive pages
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery
36
• Encyclopaedia – Format
• Websites
APA Format
Format - Books
Encyclopaedia – Format
Websites
Sarvy Joseph
Capsule notes compiled by Sarvy Joseph, Assistant Professor. Dr. Ambedkar Govt.college, Balussery