Chapter B (Survey Instrument, Probing, Processing, Coding)

Question 1

Questions that require respondents to
provide an answer in their own words since a
complete list of answer categories are not
provided

Answer 1

Open Ended

Question 2

Questions that provide a complete set of
alternative answers from which respondents
select their choice

Answer 2

Close Ended

Question 3

Questions that have the option for the
respondents to provide answers other than
the categories provided

Answer 3

Semi-open

Question 4

Uses a general-to-specific question flow

Answer 4

Funnel approach

Question 5

Uses a general-to-specific question flow

Answer 5

Work approach

Question 6

▧ Aims to determine any errors in the
questionnaires or areas that need
improvements such as the ordering of
questions, wording, and the natural flow
from one section to another
▧ There are no rules regarding the
respondents who will be part of the pretest
and the sample size

Answer 6

Pretest

Question 7

▧ Respondents are informed about a practice
run and are asked to give comments and
suggestions to the questionnaire

Answer 7

Participating Pretest

Question 8

▧ Respondents are not informed that the
questionnaire is only a draft and the
interviewer plays it straight

Answer 8

Undeclared Pretest

Question 9

The most commonly used measure of reliability
▧ If alpha is high (at least 0.80, closer to 1.00),
then this suggests that all of the items are
reliable and the entire questionnaire is internally consistent
▧ If alpha is low (less than 0.60), then at least one of the items is unreliable and must be identified via the item analysis procedure

Answer 9

Cronbach’s Alpha

Question 10

▧ It is not the same as reliability
▧ The ability of the questionnaire to measure
what it intends to measure (the latent
variable, e.g. happiness, is measured using all
items in the questionnaire)
▧ Types of validity include face validity,
content validity, criterion-related validity, and
construct validity – test1: Confirmatory
Factor Analysis and Rasch Analysis)

Answer 10

Validity

Question 11

▧ A small trial run of the entire survey process
▧ It is undertaken after pretest
▧ It is completed prior the final survey goes
into the field, to alert the enumerator to
possible difficulties that were not
anticipated

Answer 11

Pilot Survey

Question 12

To clarify answers in a non-directive way
▧ To motivate the respondent
▧ To make the respondent focus on a
particular question

Answer 12

Probing

Question 13

The rambler R how to deal?

Answer 13

The silent treatment. Stop the “uh-huh” and
keep quiet.
The rude approach. Wait until R stops for a
breath, then quickly ask the next Q. If R still
doesn’t answer the Q, wait for another breath,
say “I see,” then repeat Q.
Change focus approach. Drop all pencils on the
floor. Everything stops cold while picking up all
the pencils.

Question 14

▧ R is deaf or has severe speech problems or
other disabilities that make completion of an
interview extremely difficult for them

Answer 14

Thanks, R, for his/her time. Mark this interview
as “no eligible respondent.”

Question 15

The process of inspecting the raw data from
a questionnaire and correcting for any error
to ensure its accuracy and reliability (data
screening)

Answer 15

Data Editing

Question 16

▧ The process of converting answers to survey
questions into numerical form (codes) to
help compile the survey statistics

Answer 16

Data Coding

Question 17

▧ Correspond to a list of choices or possible
responses
▧ Successive integers, beginning with 0 or 1,
are assigned to the items in the list of
possible responses

Answer 17

Listing Codes

Question 18

▧ Represent ranges or class intervals
▧ These are frequently used for numerical
responses like income and age in which the
respondent may not be able to provide an
exact answer

Answer 18

Bracket Codes

Question 19

▧ Represent grouping of responses that vary in
degree or intensity along a continuum
▧ Commonly used for coding responses to
attitudinal questions (Likert scaling)

Answer 19

Scale Codes

Question 20

▧ Multi-digit codes are divided into two sets –
(1) leading digits represent a major
classification, and (2) trailing digits represent
a specific category within the major
classification
▧ This coding scheme can be extended to more
than two levels

Answer 20

Series Code

Question 21

▧ For each possible response, create a
dichotomous variable with a value of 1 to
indicate that the response was given and a
value of 0 if not

Answer 21

Methods of Coding Multiple Response
Multiple Dichotomy Method

Question 22

Create new categories that are combinations
of the possible responses

Answer 22

Methods of Coding Multiple Response
Coding Combination Method

Question 23

▧ It should be coded NO or given a code
number directly following the other
categories
▧ Code: Yes ‘1’; No ‘0’ and No Opinion/No
Comment a code ‘2’

Answer 23

No Opinion/No Comment

Question 24

▧ It happens if the attribute is not present; the question becomes irrelevant ▧ Code: NA; ‘7', ‘77‘, ‘777’, or ‘7777’

Answer 24

Not Applicable

Question 25

▧ It occurs when we ask knowledge questions ▧ Code: ‘8’, ’88’, ‘888’, or ‘8888’

Answer 25

Don't Know

Question 26

▧ It may be due to an offensive question or altogether sensitive topics ▧ Code: '9', or '99', ‘999’, or '9999'

Answer 26

Refuse To Answer