Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
The word research is composed of two syllables, re and search. The dictionary
defines the former as a prefix meaning again, anew or over again and the latter as a
verb meaning to examine closely and carefully, to test and try, or to probe. Together
they form a noun describing a careful, systematic, patient study and investigation in
some field of knowledge, undertaken to establish facts or principles.
A research need not lead to ideal solution but it may give rise to new problems
which may require further research. In other words research is not an end to a
problem since every research gives birth to a new question. It is carried on both for
discovering new facts and verification of old ones.
FEATURES OF RESEARCH
1. To find out the truth which is hidden and which has not been discovered so
far.
2. Aims at advancing systematic knowledge and formulating basic theories about
the forces influencing the relation between groups as well as those acting on
personality development and is adjustment with individuals.
3. Try to improve tools of analysis or to test these against the complex human
behavior and institutions.
4. To understand social life and thereby to gain a greater measure of control
over social behavior.
5. To provide an educational program in the accumulated knowledge of group
dynamics, in skills of research, in techniques of training leaders and in social action.
Process of Research:
for study is the most important step in the entire research process. After selecting
the problem, the researcher has to formulate the problem.
Before formulating the research it is desirable that researcher examines all available
literature, both conceptual and empirical. The conceptual literature is one which
deals with concepts and theories. Empirical literature is that which contains studies
made earlier and so it consists of many facts and figures observed in the earlier
studies.
In any research the problem under study deals with relation between variables. The
variables whose change has affected the other variable, is called independent
variable. Therefore there is a cause and effect relation between the variables. The
research problem must be formulated in such a manner that it highlights the nature,
extent and implications of relation existing between the variables. It is only through
this process of establishing the effective relation between variables that meaningful
conclusions are derived from the study.
SETTING UP OF HYPOTHESIS: -
A research design is a plan that specifies the sources and types of information
relevant to the research problem. It is a strategy which approach will be used for
gathering and analyzing the data. It includes the time and cost budgets since most
studies are done under these two constraints. A research design provides a rational
approach to research enabling one to decide in advance what to do, how to do, in
investigating the subjects.
DETERMINING THE SAMPLE DESIGN:
A sample design is a definite plan determined before any data are actually collected
for obtaining a sample from a given universe. Sample design refers to the technique
or the procedure which the researcher would adopt in selecting some sampling units
from the universe for drawing inferences about the universe. If the proper procedure
is followed to select the sample, definitely the sample will give all dependable
information.
COLLECTING OF DATA: -
There are several ways of collecting the appropriate data. Some of the methods of
collecting primary data are (1) Observation method. (2) Direct personal interview
method. (3) Telephone interview method. (4) Questionnaire method. (5) Schedule
method. A choice of one of these methods.
The researcher has to see that the project is executed in a systematic manner and in
time. He should make necessary preparations for successful conduct of the project.
TESTING OF HYPOTHESIS: -
Depending upon the nature of data and conclusions to be arrived one or two of
these tests can be applied. Testing of hypothesis will results in either accepting or
rejecting the hypothesis. Testing of hypothesis may prove or disprove a theory and a
theory facilitates formulating of a further hypothesis. Testing of hypothesis will result
in contribution to existing theory or the generation of a new theory.
Types of research
1.SetUpaHypothesis:
H0: The medicine is not effective in curing cancer (i.e., there is no significant
difference between the given medicine and other medicines in curing cancer
disease.)
After setting up the hypothesis, the researcher has to set up a suitable level of
significance. The level of significance is the probability with which we may reject a
nullhypothesis when it is true. For example, if level of significance is 5%, it means
that in the longrun, the researcher is rejecting true null hypothesis 5 times out of
every 100 times. Level of significance is denoted by α (alpha).
α = Probability of rejecting H0 when it is true. Generally, the level of significance is
fixed at 1% or 5%.
3. Decide a test criterion: The third step in testing of hypothesis is to select an
appropriate test criterion. Commonly used tests are z-test, t-test, X2 – test, F-test,
etc.
4. Calculation of test statistic: The next step is to calculate the value of the test
statistic using appropriate formula. Thegeneral fromfor computing the value of test
statistic is:-
Value of Test statistic = Difference Standard Error
5. Making Decision: Finally, we may draw conclusions and take decisions. The
decision may be either to accept or reject the null hypothesis.
If the calculated value is more than the table value, we reject the null hypothesis
and accept the alternative hypothesis. If the calculated value is less than the table
value, we accept the null hypothesis.
A research design is a “Blue Print” for collection, measurement and analysis of data.
It outlines how the research will be carried out. It is like glue which sticks together
the entire process of research. It provides answers to various questions like - What
techniques will be used to gather data. What kind of sampling will be used? How
time and cost constraints be dealt with? Etc.
It should be flexible
It must be adequate
Three traditional categories of research design:
Exploratory
Descriptive
Causal
Exploratory research:
Exploratory research is a type of research conducted for a problem that has not
been clearly defined. Exploratory research often relies on secondary research such
as reviewing available literature and sometimes on qualitative approaches such as
informal discussions with consumers, employees, management or competitors, and
more formal approaches through in-depth interviews, focus groups, projective
methods, case studies or pilot studies. Results of exploratory research can provide
significant insight into a given situation.
Descriptive research:
Simple Random Sampling (SRS): Everyone in the entire target population has
an equal chanceof being selected. This is similar to the national lottery. If the
“population” is everyone who has bought a lottery ticket, then each person has an
equal chance of winning the lottery (assuming they all have one ticket each).
Random samples require a way of naming or numbering the target population and
then using some type of raffle method to choose those to make up the sample.
Random samples are the best method of selecting your sample from the population
of interest.
The advantages are that your sample should represent the target population and
eliminate sampling bias, but the disadvantage is that it is very difficult to achieve
(i.e. time, effort and money).
Stratified Sampling: The researcher identifies the different types of people that
make up the target population and works out the proportions needed for the
sample to be representative.
.
Ex: university e.g. Engineering 10%, Social Sciences 15%, English 20%, Sciences
25%, Languages 10%, Law 5%, Medicine 15% The sample must then contain all
these groups in the same proportion as in the target population (university
students).
Example: you wanted to sample 150 children from a school of 1,500; you would take
every 10th name.
Multistage sampling: ampling plans where the sampling is carried out in stages
using smaller and smaller sampling units at each stage 002EMultistage sampling can
be a complex form of cluster sampling because it is a type of sampling which
involves dividing the population into groups (or clusters). Then, one or more clusters
are chosen at random and everyone within the chosen cluster is sampled.
Advantages:
Cost and speed that the survey can be done in
Convenience of finding the survey sample
Normally more accurate than cluster sampling for the same size sample
Disadvantages:
Not as accurate as Simple Random Sample [ambiguous] if the sample is the same
size
More testing is difficult to do
Judgmental/Purposive sampling
Quota Sampling
Convenience sampling
Convenience sampling: the samples are selected because they are accessible to
the researcher. Subjects are chosen simply because they are easy to recruit. This
technique is considered easiest, cheapest and least time consuming.
The primary and secondary data have been collected to cover every aspect of the
study. The primary data are related to behavior and response of employees, dealers
and customers. The secondary data shows the sales of the company product wise.
These data used in combination as per need of the study. These data having
different merits and demerits and have serves our purpose of the research study.
These are explained below:
Primary Data
gather because no one has compiled and published the information in a forum
accessible to the public. Companies generally take the time and allocate the
resources required to gather primary data only when a question, issue or problem
presents itself that is sufficiently important or unique that it warrants the
expenditure necessary to gather the primary data. Primary data are original in
nature and directly related to the issue or problem and current data. Primary data
are the data which the researcher collects through various methods like interviews,
surveys, questionnaires etc. The primary data have own advantages and
disadvantages:
The primary data are original and relevant to the topic of the research study
so the degree of accuracy is very high.
Primary data is that it can be collected from a number of ways like interviews,
telephone surveys, focus groups etc. It can be also collected across the
national borders through emails and posts. It can include a large population
and wide geographical coverage.
Moreover, primary data is current and it can better give a realistic view to the
researcher about the topic under consideration.
Reliability of primary data is very high because these are collected by the
concerned and reliable party.
A lot of time and efforts are required for data collection. By the time the data
collected, analyzed and report is ready the problem of the research becomes
very serious or out dated. So the purpose of the research may be defeated.
It has design problems like how to design the surveys. The questions must be
simple to understand and respond. Some respondents do not give timely
responses.
Sometimes, the respondents may give fake, socially acceptable and sweet
answers and try to cover up the realities. With more people, time and efforts
involvement the cost of the data collection goes high. The importance of the
research may go down.
In some primary data collection methods there is no control over the data
collection. Incomplete questionnaire always give a negative impact on
research.
Secondary Data:
Secondary data are the data collected by a party not related to the research
study but collected these data for some other purpose and at different time in
the past. If the researcher uses these data then these become secondary
data for the current users. These may be available in written, typed or in
electronic forms. A variety of secondary information sources is available to the
researcher gathering data on an industry, potential product applications and
the market place. Secondary data is also used to gain initial insight into the
research problem. Secondary data is classified in terms of its source – either
Secondly, it provides a way to access the work of the best scholars all over
the world.
Thirdly, secondary data gives a frame of mind to the researcher that in which
direction he/she should go for the specific research.
Fourthly secondary data save time, efforts and money and add to the value of
the research study.
The data collected by the third party may not be a reliable party so the
reliability and accuracy of data go down.
Data collected in one location may not be suitable for the other one due
variable environmental factor.
With the passage of time the data becomes obsolete and very old
Secondary data collected can distort the results of the research. For using
secondary data a special care is required to amend or modify for use.
Secondary data can also raise issues of authenticity and copyright.
(a) Questionnaire
For our study purpose a set of questions has been prepared to collect information
relating to the topic of the study. In this study a structured questionnaire has been
used with different types of questions such as closed ended and open ended. Special
case has been taken to select the scales for the questions for collection of responses
very effectively.
Telephone and other devices can be used for collecting data verbally and written on
fax from respondents located away from the researcher and having these facilities
plus the researcher having their contact numbers. Use of interviewers encourages
sample persons to respond, leading to higher response rates. Interviewers can
increase comprehension of questions by answering respondents' questions. It is
fairly cost efficient, depending on local call charge structure. It is good for large
national or international respondents and gives wider coverage. It cannot be used
for non-audio information (graphics, demonstrations, taste/smell samples) this
instrument is not suitable for the respondents where the telephone facility is not
available.
For collection of data from the respondents who are located at a long distance and
do not have any communication facility. They can be contacted through mailed
questionnaire. Only thing is required that the researcher should have the postal
addresses of the respondents. The questionnaire may be handed to the respondents
or mailed to them, but in all cases they are returned to the researcher via mail. The
cost involved is very less but no clarification can be given to the respondents if
required. Respondents can answer at their own convenience. The respondents
cannot be biased by the researchers and the detail information can be collected for
the research purpose. Only one disadvantage this instrument gives is that the
response rate is very less due to lack of interest in the topic of respondents and low
literacy rate.
Interview
In this method the interviewer personally meets the informants and asks
necessary questions to them regarding the subject of enquiry. Usually a set of
questions or a questionnaire is carried by him and questions are also asked
according to that. The interviewer efficiently collects the data from the informants
by cross examining them. The interviewer must be very efficient and tactful to
get the accurate and relevant data from the informants. Interviews like personal
21 | P a g e : PALIVELA PG COLLEGE (ICET CODE : SSSK)
RESEARCH METHODOLOGY FOR MANAGEMENT
The collected data is very reliable since the interviewer tactfully collects
the data by cross examining the responders.
Disadvantages:
It is a time-consuming process.
Measurement
Measurement is a systematic way of assigning numbers or names to objects and
their attributes. It is easy to assign numbers in respect of properties of some
objects, properties like weight, height.
Rules of Measurement
Scales
Scales are devised for measuring variable in social science research. Scaling is the
procedure for determining the quantitative measure of abstract concepts like
leadership style, brand image of product etc
Scale Classification
Level of Measurement
Nominal scales
Ordinal scales
Interval scales
Ratio scale
Nominal Scale: Numbers or letters assigned to objects which serve as labels for
identification or classification. They are Scales “in name only”.
Interval Scale
Interval scales not only indicate order; they also measure order or distance in units
of equal intervals. When an interval scale is used to measure psychological attribute,
the researcher can comment on the magnitude of differences or compare the
average differences on attributes that are measured, but cannot determine the
actual strength of the attitude towards an object
Ratio Scale
A ratio scale has equal interval properties of an interval scale but has two additional
features:
The numerals of the ratio scale have the qualities of real numbers and can be
added, subtracted, multiplied, Divided and expressed in ratio relationships.
Lickert Scale
Respondents indicate their attitude by checking how strongly they agree or disagree
with carefully constructed statements that range from the very positive to the very
negative towards the attitudinal object. Individuals generally choose from five
alternatives: strongly agree, agree, uncertain, disagree and strongly disagree
Semantic Differential
Constant Sum Scale: A measure of attitudes in which the respondents are asked
to divide a constant sum to indicate the relative importance of the attributes
Data continues to be in raw form, unless and until they are processed and analyzed.
Processing is a statistical method by which the collected data is so organized the
further analysis and interpretation of data become easy. It is an intermediary stage
between the collection of data and their analysis and interpretation.
Processing stages
There are four important stages in the processing of data. They are;
Editing
Coding
Classification
Tabulation
Editing: As soon as the researcher receives the data, he should screen it for
accuracy. Editing is the process of examining the data collected through various
methods to detect errors and omissions and correct them for further analysis.
Though editing, it is ensured that the collected data are accurate, consistent with
other facts gathered, uniformly entered and well arranged so that further analysis is
made easier
While editing care has to be taken to see that the data are as accurate and complete
as possible.
1. The editor should familiarize with the copy of instructions given to the
interviewers.
2. The original entry, if found incorrect, should not be destroyed or erased.
On the other hand, it should be crossed out in such a manner that it is still
eligible.
4. All completed schedules must bear signature of the editor an d the date.
Coding
Steps in coding
Prepare a coding manual with the detail of variable names, codes and
instructions.
If the coding manual has already been prepared before the collection of the
data, make the required additions for the open ended and partially coded
questions
Coding rules
Classification
Objectives of classification
Types of classification
In this type of classification, the data that are collected from different places
are placed in different classes.
Simple Classification
Manifold Classification
Here the data are classified to some characteristics that can be measured.
Data are classified on the basis of quantitative characteristics such as age,
height; weight etc. quantitative variables are grouped in to two
Discrete variable: If the variables can take only exact value, it is called
discrete variable.
Continuous variable: The variables that can take any numerical value
within a specified range are called continuous variable.
Tabulation
Objectives of Tabulation
Types of Tables
Simple Table
Here the data are presented only for one variable or characteristic. Any
frequency distribution of a single variable is simple table
Complex table
In this type of table, data of only one characteristic will be shown. It means
that when one type of information is secured about different groups or
individuals, it can be displayed with the help of one- way table
Three-way table
Manifold tables
The title of the table is placed above the table. If there are more than one table in a
research, each should bear a number for easy reference.
It is also termed as “box head”. There may be sub- captions under the main caption.
This is the main body of information needed for the research work.
In research, the data collected may be of complex nature. Diagrams and graphs is
one of the methods which simplifies the complexity of quantitative data and make
them easily intelligible. They present dry and uninteresting statistical facts in the
shape of attracting and appealing pictures. They have a lasting effect on the human
mind than the conventional numbers.
Types of Graphs
Line Graphs
For eg; following data show birth rate per thousands of six countries over a
period.
India 30
Germany 16
UK 20
China 40
Newzeland 30
Sweeden 12
BAR CHARTS
The bar graph is a common type of graph which consists of parallel bars or
rectangles with lengths that are equal to the quantities that occur in a given data
set. The bars can be presented vertically or horizontally to show the contrast and
record information. Bar graphs are used for plotting discontinuous (discrete) data.
Discrete data contains discrete values and are not continuous.
Country Birth
Rate
India 30
Germany 16
UK 20
China 40
Newzeland 30
Sweeden 12
Histogram
Frequency Polygon
Frequency Curves
A frequency distribution can be cumulated in two ways, less than cumulative series
and more than cumulative series. Smoothed frequency curves drawn for these two
cumulative series are called cumulative frequency curves or ogives.
Less than ogive curve: In less than ogive curve the upper limit per limit of each
class interval is taken on x- axis in increasing order. For each such upper limit on x-
axis, the cumulative frequency of all the class intervals from the first class interval to
last class interval are taken on the y-axis.
More than ogive curve: In more than ogive curve the lower limit of each class
interval is taken on x- axis in increasing order. For each such lower limit on x- axis
the cumulative frequency of all the class interval from that class interval to the last
class interval are taken on y-axis.
A pie graph is a circle divided into sections which each display the size of a relative
piece of information. Each section of the graph comes together to form a whole. In a
pie graph, the length of each sector is proportional to the percentage it represents.
Pie graphs work particularly well when each slice of the pie represents 25 to 50
percent of the given data.
Country Birth
Rate
India 30
Germany 16
UK 20
China 40
Newzeland 30
Sweeden 12
Pictograms
Kerala
20
Karnataka
28
Tamil Nadu
39
Andhra Pradesh
42
NON-PARAMETRIC TEST :
Non-parametric tests cover techniques that do not rely on data belonging to any
particular distribution. These include 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 non-parametric statistical models,
inference and statistical tests. Following are the nonparametric tests
KOLMOGOROV-SMIRNOV TEST:
The Kolmogorov-Smirnov test tests whether two arbitrary distributions are the same.
It can be used to compare two empirical data distributions, or to compare one
empirical data distribution to any reference distribution. It’s based on comparing two
cumulative distribution functions (CDFs).
Two Kolmogorov-Smirnov test plots (right column) with histograms of the data being
tested (left column). On the top row, the empirical CDF (green) matches the test
CDF (blue) closely, and the largest difference (dotted vertical red line, near 0.5) is
very small. On the bottom, the empirical CDF is quite different from the test CDF,
and the largest difference is much larger.
Now suppose we want to compare two CDFs, F 1 and F 2 . They might be empirical
CDFs (to compare two different datasets and see whether they’re significantly
different) or one might be a reference CDF (to see whether a particular distribution
is an appropriate choice for a dataset). The Kolmogorov-Smirnov test computes the
statistic Dn:
This compares the two CDFs and looks at the point of maximum discrepancy; see
Figure 5.1 for an example. We can theoretically show that if F 1 is the empirical
distribution of x and F 2 is the true distribution x was drawn from, then limn →∞ Dn
= 0. Similarly, if the two distributions have no overlap at all, the maximum
difference will be 1 (when one CDF is 1 and the other is 0). Therefore, we can test
distribution equality by comparing the statistic Dn to 0 (if Dn is significantly larger
than 0 and close to 1, then we might conclude that the distributions are not equal)
RUNS TEST
The runs test (also called Wald–Wolfowitz test after Abraham Wald and Jacob
Wolfowitz) is a non-parametric statistical test that checks a randomness hypothesis
for a two-valued data sequence. More precisely, it can be used to test the
hypothesis that the elements of the sequence are mutually independent.
The t-test is the standard test for testing that the difference between population
means for two non-paired samples are equal. If the populations are non-normal,
particularly for small samples, then the t-test may not be valid. The rank sum test is
an alternative that can be applied when distributional assumptions are suspect.
The observation in the random model in which the run test is carried out has a
constant variance, and the observations are also probabilistically independent. The
run in a run test is defined as the consecutive sequence of ones and twos. This test
checks whether or not the number of runs are the appropriate number of runs for a
randomly generated series. The observations from the two independent samples are
ranked in increasing order, and each value is coded as a 1 or 2, and the total
number of runs is summed up and used as the test statistics. Small values do not
support suggest different populations and large values suggest identical populations
(the arrangements of the values should be random). Wald Wolfowitz run test is
commonly used.
TEST : Does the X group differ from the Y group in regards to the diet treatment
implemented on both groups?
If the run test is being tested for randomness, then it is assumed that the data
should enter in the dataset as an ordered sample, increasing in magnitude. This
means that for carrying-out the run test for randomness, there should not be any
groupings or other pre-processing.
If the run test is carried out in SPSS, then it is assumed that the variables that are
being tested in the run test should be of numeric type. This means that if the test
variables are of the string type, then the variables must be coded as numbers in
order to make those variables of the numeric type.
Formula of the mean and the variance of the observed number of the runs:
E ( R ) = H + 2 Ha Hb / H
V ( R ) = 2 Ha Hb (2 Ha Hb – H ) / H2 ( H – 1 )
The researcher should note that in the run test for the random type of model, if the
value of the observations is larger than twenty, then the distribution of the observed
number of runs would approximately follow normal distribution. The value of the
standard normal variate of the observed number of runs in the run test is given by
the following:
This follows the normal distribution that has the mean as zero and the variance as
1. This is also called the standard normal distribution that the Z variate must follow.
The sign test is one of the simplest nonparametric tests. It is for use with 2 repeated
(or
correlated) measures and measurement is assumed to be at least ordinal. For each
subject, subtract the 2nd score from the 1st, and write down the sign of the
difference. (That is write “ -” if the difference score is negative, and “+” if it is
positive.) The usual null hypothesis for this test is that there is no difference
between the two treatments. If this is so, then the number of + signs (or - signs, for
that matter) should have a binomial distribution1 with p = .5, and N = the number
of subjects. In other words, the sign test is just a binomial test with + and - in place
of Head and Tail (or Success and Failure).
Applications:
For example, you could use multiple regression to understand whether exam
performance can be predicted based on revision time, test anxiety, lecture
attendance and gender. Alternately, you could use multiple regression to understand
Multiple regressions also allow you to determine the overall fit (variance explained)
of the model and the relative contribution of each of the predictors to the total
variance explained. For example, you might want to know how much of the variation
in exam performance can be explained by revision time, test anxiety, lecture
attendance and gender "as a whole", but also the "relative contribution" of each
independent variable in explaining the variance.
Example:
Assumptions:
At the center of the multiple linear regressions analysis is the task of fitting a single
line through a scatter plot. More specifically the multiple linear regressions fits a line
through a multi-dimensional space of data points. The simplest form has one
dependent and two independent variables. The dependent variable may also be
referred to as the outcome variable or regress and. The independent variables may
also be referred to as the predictor variables or repressors.
There are 3 major uses for multiple linear regression analysis. First, it might be
used to identify the strength of the effect that the independent variables have on a
dependent variable.
Second, it can be used to forecast effects or impacts of changes. That is, multiple
linear regression analysis helps us to understand how much will the dependent
variable change when we change the independent variables. For instance, a
multiple linear regression can tell you how much GPA is expected to increase (or
decrease) for every one point increase (or decrease) in IQ.
Third, multiple linear regression analysis predicts trends and future values. The
multiple linear regression analysis can be used to get point estimates. An example
question may be “what will the price of gold be 6 month from now?”
Assumptions:
Multivariate normality: Independent variables are normal for each level of the
grouping variable
Homogeneity of variance/covariance (homoscedasticity): Variances among
group variables are the same across levels of predictors. Can be tested with
Box's M statistic. It has been suggested, however, that linear discriminante
analysis be used when covariance are equal, and that quadratic discriminante
analysis may be used when covariance are not equal.
Test the data to assure that the assumptions of discriminante analysis are
met. If some assumptions are not met, determine whether discriminante
analysis is robust for those assumptions.