Marketing Research: CH-8 Data Processing and Data Analysis
Uploaded by
shaukat74Marketing Research: CH-8 Data Processing and Data Analysis
Uploaded by
shaukat74MARKETING RESEARCH
CH-8
Data Processing and data analysis
Backdrop
• The data, once collected, are to be processed,
analyzed and interpreted. The efforts put in by
the researcher and his team will go waste in
case the information collected is not properly
processed and analyzed. Data processing is to
do with the specialized activities of editing,
coding, classification and tabulation of the
collected data so that it can be taken for
analysis and interpretation.
Data Processing
• In fact, data processing means reduction of data
to such an extent that one is able to do analysis
and later on the interpretation. Data as collected
are unduly large which have to be boiled down
to a definite size and shape. It means taking out
unwanted data, correcting it to make it more
relevant and meaningful.
COMPONENTS OF DATA PROCESSING
I. EDITING
• Editing is the process of examining the
collected data to detect errors and omissions
with a view to correct. In other words, editing
is the process by which data are prepared for
subsequent coding. Editing involves careful
scrutiny of the completed questionnaires.
Data Processing
• The aim of editing is to assure that the data
are accurate, consistent with other facts as
gathered, uniformly entered as complete as
possible and have been well arranged to
proceed for coding, classification and
tabulation.
STAGES OF EDITING
Field Editing:
• Field editing is one which is carried by the
interviewer himself or herself just after the
interview. Field editing is significant because,
it is not possible for the field investigator or
the interviewer to fill in all the entries at the
time of interview. He or she many use
symbols or brief notes to record the answers.
STAGES OF EDITING
Office Editing
Office editing is also called as central editing.
Once the field investigator has
collected the information from the
respondents, the questionnaires or interview
schedules are sent to the researcher’s office.
Office editing assumes all the more importance
in cases of mail surveys conducted, because
the editing work can be possible only when
questionnaires reach the central office.
POSSIBLE PRECAUTIONS IN EDITING
1. Conversant with Instructions and Directions.
2. Corrections with Original Data.
3. Corrections with Initials.
4. Corrections with Original Data.
5. Corrections with Initials.
6. Dating of Corrections.
PROBLEMS ASSOCIATED WITH
EDITING
1. Fictitious Interviews
2. Inconsistencies
3. Incorrect Answers
4. Incomplete Answers
5. Unclear Responses
6. “Don’t Know” and “No Answers”
CODING
• Coding is the process of assigning numbers or
symbols to the answers to prepare them
for tabulation. Coding is the procedure of
classifying the answers to a question into
meaningful categories. The symbols or
numbers used to indicate these categories are
called codes. Coding is a must to carryout the
subsequent operations of tabulating and
analyzing the data.
CLASSIFICATION
• It is quite obvious that most research studies
result in large volume of raw data which must
be reduced into homogeneous groups if one is
interested is getting meaningful relationships.
This warrants classification of data scientifically.
Classification is the process of arranging things-
either actually or notionally-in groups or classes
according to their resemblances and affinities.
Classification is the process of condensation. It is
the process of arranging the data into group.
Why Tabulation
(1) It simplifies the comparison of data.
(2) It makes the study complete and accurate
in every respect.
(3) It makes comparisons easier together with
saving a lot of time and energy in interpreting
them.
(4) It makes the statistical treatment a
possibility.
CLASSIFICATION OF TABLES
• One way of classification is that of ‘simple’
and ‘manifold’. A ‘simple’ table is normal,
general table which is mainly divided into two
columns and rows. In general, simple tables
are easy to construct and to follow where all
possible ambiguities and duplications are
done away. On the other hand, ‘manifold’
table is one that consists of multiple rows and
columns.
STRUCTURE OF TABLE
1. The Title
2. The Unit
3. The Captions
4. The Stub
5. The Body
6. The Source or Sources
7. Foot Notes
WHAT MAKES GOOD TABLE OR
TABULATION WORK?
1. Cover Maximum Data
2. Maintain Basic Structure
3. Approximate the Values
4. Project Comparable and Intelligible Data
5. Meet the Basic Need of Preparation
6. Colour the Data
7. Numbering the Tables
What is data analysis
• Stated simply, analysis of data is a process to
arrive at a meaning or sense of the data
collected. It involves converting a series of
data into descriptive statements about
variables and inferences and about
relationship among variables. Data analysis
provides an answer or answers to questions
one may ask.
THE PRE-REQUISITES OF ANALYSIS
PROCESS
1. Constructive Imagination
2. Critical Imagination
3. Open Mindedness
TECHNIQUES OF DATA ANALYSIS
• Depending on the variables analytical
methods are used. Analysis of data might
involve a single variable, two variables and
more than two variables.
UNIVARIATIVE ANALYSIS
• MEASURES OF CENTRAL TENDENCY
1. ARITHMETIC MEAN –
It is also called as average or mean. It is a sum of
all items divided by number of items.
UNIVARIATIVE ANALYSIS
2. MEDIAN
Median is the middle value of an ordered or
arranged set of numbers. In other words,
median is the measure of the central item when
all the items in a series are arranged either in
Ascending or descending order of magnitude.
Median is quite a popular in population and
income.
UNIVARIATIVE ANALYSIS
3. MODE
• It is the most frequently occurring value of a
group values. It is the value at the point
around which the items are most heavily
concentrated. It is most least frequently used
measure of central tendency.
UNIVARIATIVE ANALYSIS
MEASURES OF DISPERSION
1. Range:
‘Range’ is the simplest measure of dispersion. It
is equal to the maximum minus minimum
value in the array of data.
2. Inter-Quartile Range:
Inter-quartile range is the difference between
the two extreme quartiles of a series.
UNIVARIATIVE ANALYSIS
MEASURES OF DISPERSION
3.Mean Deviation
Mean Deviation is the arithmetical average of
the deviations of item of a series taken from
its Central Value ignoring the plus and minus
signs. Here, the central value may be any type
of average value. However, preference is given
to Mean, Median and Mode. Again, among
these three average Median is the most
preferable.
UNIVARIATIVE ANALYSIS
MEASURES OF DISPERSION
4. Standard Deviation:
Standard deviation is the square root of the
arithmetic average of the squares of deviation
taken from arithmetic average of a series. SD is
just like the average deviation except that
instead of taking the absolute value of the
numbers to eliminate the negative signs, the
SD takes squares of the deviations of individual
values from arithmetic mean.
BIVARIATIVE ANALYSIS
• The limitation of univarivate analysis is that the
analysis is confined to only one variable.
Practical world for a marketing researcher is not
made up of only one variable but he often faces
situations or problems which are complex involving
two or more variables, in which case analysis based
on one variable will not be relevant. Therefore,
bivariate and multivariate analysis becomes an
appropriate tool than univariative analysis. When the
researcher is interested in the degree of relationship
between two variables, he uses bivariative tools
namely/correlation and regressive analysis.
BIVARIATIVE ANALYSIS
1. CORRELATION
Correlation, in statistics, refers to relationship between
any two variables say height and weight,
rainfall and yield, price and demand, income and
expenditure, wages and price index, production
and employment. The two variables are said to be
correlated if a change is the value of one variable
there arises a change in the value of another variable. On
the other hand, if a change in value
of one variable does not bring any change in the value of
another variable, the two variables are
said to have no relation with each other..
BIVARIATIVE ANALYSIS
2. REGRESSION ANALYSIS
The term ‘regression’ implies opposite of ‘progression’
which means’ going back’ or “stepping down”. It’s
meaning in statistics is quite different. Regression
analysis is a statistical tool for measuring the average
relationship between any two or more closely related
variables in terms of the original units of their data-
positively or negatively. It is that statistical tool which
helps the researcher to estimate or predict the values
of unknown variables from the known values of
another variables.
MULTIVARIATIVEANALYSIS
• Multivariative analysis is the analysis of the
simultaneous relations among three or more
phenomena. While in case of univariative analysis the
focus is on the level or average and distribution or
variance of the phenomena in bivariative analysis the
focus shifts to the degree of relationships like in
correlations or covariate between the phenomena.
However, in case of multivariate analysis, it focus is
shifted from paired relationships to the more complex
simultaneous relationships among the phenomena.
MULTIVARIATIVE ANALYSIS
Independent Methods –
Factor Analysis, Cluster Analysis, Conjoint Analysis.
Dependent Methods –
Correlation and Regression Analysis, Discriminant
Analysis, Analysis of variance.
Thanks
