DATA PREPARATION AND DESCRIPTION

DATA PREPARATION
Editing
Coding
Data Entry

EDITING
Editing detects errors and omissions, correct them possible and certifies that minimum data
quality standards have been achieved.
After editing data must be
Accurate
Consistent with intent of the question and other information in the survey
Complete
Arranged to simplify coding and tabulation
The researcher should ensure
The data obtained is complete in all respect.
It is accurate in terms of information recorded and responses sought.
Questionnaires are correctly deciphered, especially open ended questions.
The response format is in the form as was instructed.
The data is structured in a manner that entering the information will not be a problem.

FIELD EDITING
When entry gaps are present from interview a callback should be made .
To validate the field results, field supervisor interview some percentage of the respondents.

CENTRAL EDITING
A thorough editing
Use of a single editor produces maximum consistency

CODING
Coding involves assigning numbers or other symbols to answers so that responses can be
grouped into a limited number of classes or categories.

CODING RULES
Appropriate for research problem and purpose
Exhaustive
Mutually exclusive
Single dimension

CODEBOOK
A codebook or coding scheme, contains each variable in the study and specifies the
application of coding rules to the variables.

CODING CLOSED QUESTIONS

Precoding can be used.
RANKING QUESTIONS

SCALED QUESTIONS
Compared to the past (5-10 years)
1. The individual customer today shops more. SA A N D SD
2. The consumer is well informed about market offerings. SA A N D SD
3. The consumer today has more money to spend. SA A N D SD

SAMPLE QUESTIONNAIRE ITEMS

What is the pin code of your residence?-----------
What is your year of birth?----
Gender (1) Male Indicate appropriate no.----
(2) Female
Marital Status: (1) Married Indicate appropriate no.-
(2) Widow (er)
(3) Divorced
(4) Unmarried

CODING OPEN-ENDED QUESTIONS

Open questions are more difficult to code since answers are not prepared in advance.
A systematic method for analysing open questions are called Content analysis.
It uses preselected sampling units to produce frequency counts and other insights into data
patterns.

Question : How can management-employee relations can be improved?

DONT KNOW RESPONSES
May be legitimate response
May be researchers failure to get appropriate information

What is the chemical formula of common salt?

Do you believe companys recruitment policy is sound?
Do you like your present job?

DEALING WITH UNDESIRED DK RESPONSES

Proper designing of the questionnaire
If it is legitimate, put them in a separate category
For large DK responses, distribute them among the other answers in the same ratio that the
other answers occur.

TABULATION/DATA ENTRY
Keyboard Entry
Optical Mark Recognition

DESCRIPTIVE STATISTICAL SUMMARIES

Describing distributions:
Normal Symmetric, Positive or rightly skewed, Negative or left skewed
Kurtosis: Mesokurtic, Leptokurtic, Platykurtic

DESCRIPTIVE STATISTICAL SUMMARIES

Measures of Location: Mean, Median and Mode
Measures of Spread: Standard Deviation, Range, Interquartile range (IQR), Quartile
deviation
Measures of Shape: Skewness and Kurtosis

SKEWNESS
A measure of a distributions deviation from symmetry.
A distribution that has cases stretching toward one tail or other is called skewed.

KURTOSIS
A measure of a distributions peakedness (or flatness).
Leptokurtic: Cluster heavily in the centre
Platykurtic: Scores more evenly distributed, tails flatter than normal distribution
Mesokurtic: Neither too picked nor too flat