DATA PROCESSING

DR PARMAR
• After data collection, the process of data has to be
done , this involves editing, coding, classification &
tabulation of data. The steps involved in the
organization of data are
1. Preliminary screening of the data collected.
2. Editing of data.
3. Coding of data.
4. Classification of data.
5. Tabulation of data.
• Data collected through filled/completed questionnaires
are raw data which cannot be used directly for tabulation,
for detailed analysis & finally for drawing conclusions. In
the absence of processing of data the analysis of data will
be defective, the conclusions drawn may not be reliable &
the recommendations made may not be useful for solving
the marketing problems faced by the sponsoring company
processing of data is an important step in the research
process. Filled questionnaires is a huge body of raw data
or processing such data, editing, coding, classification &
tabulation are required.
 I. Preliminary screening of the data collected.

• It is prior to the editing. The purpose of initial

screening is to verify accuracy & completeness
of data for further processing processing. The
screening of data is usually on the basis of the
following three considerations.
1. Relevance.
2. Reliability.
3. Practical use/utility.
 Editing of data

• It is the process of eliminatory errors in the raw data

collected so that the data may subsequently undergo
tabulation & classification. The purpose of editing is
to eliminate errors or known bias & to prepare data
for subsequent coding, tabulation & analysis. It is the
process of inspecting correcting & modifying the
information submitted on each questionnaire, so that
it is stated correctly adequately & on a basis common
to all similar categories of reply.
 Need of editing of data

1. To check accuracy of data.

2. To remove errors.
3. To improve the quality of data.
4. To evaluate the performance of field staff.
5. To make uniform decisions on incorrect answers.
 Procedures for editing of data.

• In the editing of data, legibility of answers given in

the questionnaire etc are checked. The ultimate
purpose is to see that data selected for analysis is
accurate, complete, uniform, consistent & relevant.
In editing decision is taken as regards answers which
are incorrect & incomplete. Such answers may be
rejected or accepted by changing them in a suitable
manner.
 Types of Data editing.

• 1. Field editing (2) Office editing.

• 1. Field editing: It means editing of data during the
process of field survey itself. This work is normally given
to interviewers & their supervisors when the data
collection work is in progress. The interviewers can do
such editing work, while noting the information
immediately offer the interview. The interviewer has to
see that his completed questionnaires are clear & verify
them immediately . Field editing is also possible at the
level of the supervisor.
• 2. Office editing: It is also called as central
editing. It is possible when all questionnaires are
collected at the head office for tabulation. In
office this work can be given to a team of editors
with detailed instructions as regards the rules &
procedures of editing work. Office editing is
more systematic & uniform in character as
compared to field editing. There are two ways of
editing (a) Manual editing, (b) Computer editing.
 Coding of data
• It is necessary for the purpose of tabulation. Coding is the
procedure of classifying the answers to a question into
meaningful categories. It helps in arranging/grouping of
varied responses into meaningful categories for the
purpose of classification, tabulation & analysis of data. In
the absence of coding it will not be possible to reduce a
large number of heterogeneous responses into
meaningful categories. The symbols used to indicate
these categories are called codes. Eg M may be used for
male, F may be used for female, similarly numbers like
1,2,3,4 can be used as code numbers
• Coding simplifies & speeds up the work of
tabulation of data, A code is a number/symbol
which stands for each type of reply. It means
assigning a number, symbol or other marks to every
answer. It is one useful tool of analysis when a large
number of questionnaires are required to be
handled. Coding is possible at the time of preparing
questionnaires. As an alternative coding is done
after editing. Editing may be done simultaneously
along with editing or it may be done after editing.
 Guidelines for data coding

• Coding of data involves two steps 1st to specify

the different categories/classes into which the
responses are to be classified, 2nd to allocate
individual answers to different categories.
Following points should be considered while
coding.
1. Coding should cover all types of answers
given by the respondents. It should be of
wide range.
• 2. Coding categories should not overlap. The categories
introduced should be all inclusive & mutually exclusive. Every
answer should legitimately fit into only one category & not in
many categories.
• 3. A code sheet containing the code number of all categories
of answers for each question in the questionnaire should be
prepared for quick reference.
• 4. Coding work should be given to a knowledgeable person.
• 5. Appropriate use of numbers, symbols & alphabets should be
made for coding of information.
• 6. More stress should be given to pre-coding over post-coding.
 Classification of data

• Data collected through questionnaire are not directly useful for

tabulation. Classification of data is essential. The aim of
classification is to arrange data systematically for quick & easy
understanding/analysis. Classification is a process of arranging data
in different groups/class according to resemblances & similarities.
• Classification means grouping together the data which are
homogeneous in character for interpretation purpose. It is the
sorting of collected data on the basis of certain well defined
characteristics or norms. Classification is the division of data into
homogeneous groups on the basis of certain norms. It facilitates
comparison & also suggests the salient features of data in a
compact form.
 Objectives of data classification

1. To condense the available data in a compact

manner.
2. To make data ready for further processing &
analysis.
3. To find out important facts & also to facilitate
comparison with data of similar nature.
4. To make data ready for tabulation.
5. To facilitate easy & quick interpretation of data.
 Basic principles of data classification

• Classification should not be ambiguous-various

categories/groups should be clearly defined in classification.
• Should be on the basis of a single classification principle eg
buyers may be classified on the basis of economic or social
status or frequency of purchases or amount spent.
• Should be mutually exclusive-no overlapping of the
categories after classification.
• Should be mutually exhaustive.
• Should be distinct.
• Should be pertinent to MR project.