Data Processing And Analysis
• Raw data collected from the field are not in a
proper form for interpretation.
• The data have to go through processing -
involves a sequence of operations -editing,
coding, classification and tabulation of collected
data
• The objective of data analysis is to prepare raw
data for statistical analysis and presentation.
•
Stages of processing
a) Editing-process by which the questionnaire or
interview schedules are scrutinized with the intention
of eliminating errors and omissions.
-Its purpose is to ensure that data collected are
accurate and unambiguous.
-Questions should be checked for completeness,
accuracy and uniformity.
-i) Completeness involves checking whether all
questions have been answered.
-ii) Accuracy involves checking the data for
consistency. (contradictory information)
-If a Qs has a lot of errors, it is advisable to discard it.
• iii) Uniformity involves checking whether
interviewers have interpreted questions and
instructions uniformly.
• Editing should start from the field.
• If parts of the questionnaire do not apply to a
respondent, it is appropriate to include
provision for a response such as “don’t know”
or “not applicable.”
• A response to each question is necessary to
show that the interviewer has covered every
topic.
Coding
• A process by which response categories are
summarized by certain symbols.
• The purpose is to classify the answers to a
question into meaningful categories to bring out
a pattern.
• The first step is to decide on the categories to
be used and the second is to allocate individual
answers to them.
Classification
• This is the process of arranging data in groups
or classes on the basis of common
characteristics.
• The statistical nature of the data is more likely
to be observable when they are classified into a
relatively small number of groupings.
Tabulation
• This is the process of summarizing raw data
and displaying it in compact form for further
analysis.
• Tabulation basically involves counting of the
number of cases falling into each of several
classes.
• The objective is to organize information and
prepare a summary that highlights its features.
• There are different types of tabulation – these
are;-
a) Tabulation sheet
-This is a simple table of rows and columns.
-Each column used to store answer to questions
(variable) and each questionnaire occupies a row
of the table.
(see text document)
Tally sheet tabulation
• If analysis requires counting proportions, totals
and averages of single variables these can be
done in a tally sheet.
• More often cross tabulations between two
variables is needed.
• From the tabulation sheet above, a tally sheet
could be used to compare educational level and
No. of employees. (see text).
Data checking
• There are three aspects of data checking
• a) Data verification which is an examination of
completeness and consistency.
• When data are processed they are passed from
one document to another.
• The two principal types of transfer are the initial
data entry to a tabulation sheet or a computer
file, and for subsequent transfers of part or all
of the data from one sheet to another.
Each transfer will give rise to a number of
possible errors.
• Some data may be lost
• Some data may be repeated
• The values of some data items may be
changed
• Changes may be made to groups of data.
• b) Data validation –Aimed at ensuring that data
items satisfy the criteria of magnitude and logic.
• Concerned with the quality of the data collected
• One system of checking is to check a sub
sample immediately after survey. This is
designed to check the quality of enumeration.
• An item of data can be checked in three ways; -
that it lies within a permissible range of values,
that its order of magnitude or response
category is plausible.
• c) Close scrutiny required to spot fabricated
data.
• The best simple check of data is the visual
check when entering data.
• Key variables for each interviewer should be
summarized and distribution statistics be
calculated.
• Very high or very low coefficient of variation
should prompt the researcher to explore the
data further.
Distributions
This is the first step in data analysis-concerned with organizing
the raw data into more usable and interpretable format.
Three purposes are served by the initial organization of
data into distributions and graphs; -
-The organization allows for preliminary interpretation of
study results by recording the data into a more understandable
format.
-The organization of raw data into distributions allows
further statistical analysis.
-The organization of the data provides for a preliminary test of
some assumptions regarding the nature of data.
Distributions involve the use of frequencies, relative frequencies,
Data Analysis
• This involves the interpretation of survey data.
• It involves choosing the variables to examine
and what relationships to explore.
• Once totals, averages, ratios and proportions
have been calculated estimates from different
sub populations or periods of time can be can
be compared with each other.
• Data analysis involves the use of statistical
methods.
