Scribd
Upload
63 views31 pages

Lecture 29

This document discusses the process of data analysis, which involves transforming raw data into useful information through various stages. The key stages discussed are editing, coding, and error checking of data. Editing involves checking data for completeness and consistency before coding. Coding assigns numeric or other symbols to data values for computer processing. Error checking verifies that all codes are legitimate and without mistakes during coding and data entry. The overall goal of data analysis is to take recorded measures from a study and present them in a usable format for making decisions.

Uploaded by

mankhokhar
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
63 views31 pages

Lecture 29

This document discusses the process of data analysis, which involves transforming raw data into useful information through various stages. The key stages discussed are editing, coding, and error checking of data. Editing involves checking data for completeness and consistency before coding. Coding assigns numeric or other symbols to data values for computer processing. Error checking verifies that all codes are legitimate and without mistakes during coding and data entry. The overall goal of data analysis is to take recorded measures from a study and present them in a usable format for making decisions.

Uploaded by

mankhokhar
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 31

RESEARCH

METHODS

Lecture 29
DATA ANALYSIS
Data Analysis
• Data processing and analysis is part of
research design – decisions already made.
• During analysis several interrelated
procedures are performed to summarize
and rearrange data.
• Goal of most research is to provide
information.
• Information: a body of facts that are in a
format suitable for decision making.
• Data: simply recorded measures of certain
phenomenon.
Stages of Data Analysis
EDITING

CODING
ERROR
CHECKING
AND
DATA ENTRY VERIFICATION

DATA ANALYSIS
Transformation of raw
data into information
• It is a lengthy process. Starts with:
• Editing, coding, and data storage.
• For storage and processing, many
advantages to use computer.
• For computer storage follow the
coding procedure.
Errors
• Possibility of mistakes by
fieldworkers e.g.
• - Recording improbable answer
• - Interviewing ineligible respondent
• - Contradictory answers
• Editing should take care of such
mistakes prior to coding
Editing
• Process of checking and adjusting the data
for omissions, legibility, and consistency.
• Purpose is to ensure the completeness,
consistency, and readability of data to be
transferred storage.
• Editor may have to reconstruct some data
(weekly income converted into monthly
income)
• Bring to light the hidden values without
bias.
Editing is not coding.
• Coding is assignment of numbers of
symbols to previously edited data.
Editing
FIELD EDITING

IN-HOUSE
EDITING
Field Editing
• Fieldworkers and then supervisors to do
the field edits.
• Do the editing the same day.
• Catch technical omissions (blank page)
• Check legibility of hand writing.
• Clarify inconsistent responses.
• Supervisor can question thee interviewers.
• Callback if needed.
• Do the interviewers need additional
training.
In-House Editing
• Rigorously investigates the results
of data collection.
• For mailed questionnaire in-house
editing is the only option.
Editing for consistency
• Adjust the inconsistent and contradictory
responses (wrong person interviewed).
Editor to make sure that the sampling unit
is consistent with the objectives.
• Check for logical consistency of responses
Answer to one question is consistent with
answer to another related question.
• Are the questions applicable to respondent.
• Editor should adjust for consistency
Editing for completeness
• Respondent may have answered only the
2nd portion of the 2 part question e.g. Does
your organization have more than one
Internet Web site? Yes ___ No ___
• In the following question he answers 3 web
sites.
• Editor to complete the missing answer.
Item non-response
• Non-response a technical term for an
unanswered question on an otherwise complete
questionnaire.
• Specific decision rules:
• - do nothing (missing value)
• - If necessary  use plug value – *Plug in an
average or neutral value. *Mean value of all who
have answered that question. * Mean value of
responses of this respondent. * Alternate the
response categories (yes, no, yes, no)
• Decide whether or not the entire questionnaire
is usable.
“I Don’t Know”

Legitimate don’t know


Reluctant don’t know
Confused don’t know
Tabulation of “don’t know”
• Make a decision about “don’t know”
• Make a separate response category
but eliminate the “don’t know”
answers from the percentage base.
• Can we distribute the “don’t know”
among other categories
(proportionately). Not recommended.
Editing questions
answered out of order
• Usually happens in interviewing
• An answer to an open ended question
• A respondent may have provided answer to
a subsequent question in his answer to an
earlier question.
• Interviewer may have avoided to asking the
subsequent question. Skipped the question
• The editor may move the out-of-order
answer to the section related to the skipped
question.
Coding
• Coding involves assigning numbers or
other symbols – responses can be
grouped into classes or categories
• Helps in efficient analysis of data.
• Keep the data in raw form so far it is
possible. Computer can help in grouping.
• Codes allow data processing in a
computer.
Organize data into:
• A field: collection of characters
(single No., letter, symbol)
representing single type of data.
• A record: collection of related
fields.
• A file: collection of related records.
Coding procedure
• It is a set of rules stating that certain
numbers are assigned to a variable
attributes e.g. code male as 1 and female as
2. Missing information also needs a code.
• Coding procedure should be recorded.
• A codebook is a document describing the
coding procedure and the location of data
for variables in a format that computers can
use.
Pre-coding
• Researchers think of coding procedures
before they collect data.
• Survey researcher pre-codes a
questionnaire
• Pre-coding means placing the code for each
of the categories on the questionnaire.
• Otherwise coding done after editing.
• Transfer the coded data into a format that
computers can read.
Code Construction
• For closed ended questions, the
number of categories requiring codes
is determined during questionnaire
design.
• Conventionally code 8 and 9 are given
to “don’t know” (DK) and “no answer”
(NA)
Two basic rules for code
construction
• Coding categories should be
exhaustive i.e. coding category
provided for every possible
response.
• Coding categories should be
mutually exclusive and
independent.
Coding open-ended questions
• Pre-coding easier for closed-ended
questions.
• Framework for classifying responses to
open-ended question not possible before
data collection.
• Give thought after editing.
• Reduce large number of responses to a few
general categories of answers – assign
numerical scores
• Depends upon the researcher’s judgment.
Codebook
• A book identifying each variable in a
study and its position in the data
matrix. The book is used to identify a
variable’s description, code name, and
field.
Codebook
• Q/V No. Field/ col. No. Code values
• -- 1-5 Study number
• - 6 City
• 1 = Lahore
• 2 = Rawalpindi
• 3 = Karachi
• 7 -9 Interview No.
• Sex 10 1 = Male
• 2 = Female
• Age 11-12 Actual
• Education 13 1 = Non
literate
2 = Literate
Production Coding
• Transferring the data from the
questionnaire after data collection.
• Depending upon the nature of data
collection form, codes may be
written directly on the instrument,
or on a special coding sheet.
Data Entries
• The process of transforming data from the
research project to computers.
• Optical scanning systems
– Marked-sensed questionnaires
Recoding
• Recoding is the process of
using a computer to convert
original codes used for raw
data to codes that are more
suitable for analysis.
• Var1 = 8 - Var1
Cleaning data
• Error checking and verification –
make sure that all codes are
legitimate.
• No errors in coding process, as
well as entering data into a
computer.
RESEARCH
METHODS

Lecture 29

You might also like