, LECTURE I
Introduction:What is survey research about?
2 Produce statistics/draw inferences abouta target audience.
C ↳ Disadvantages
Advantages
->
Relatively low cost ->
Generally limited to scaled or check-list questions
->
Relatively fastdelivery -> No control over response rate
->
Rapid data processing -> Provide information more than
understanding
-> can search large population
->
Increasing public resistance
-> various mode:phone, mail, Internet, f2f... -
Difficult to exploreissues in depth.
C Two inferences
How
->
closely the sample responding mirrors the population -> external validity
-> How well answers measure characteristics internal validity
Errors in surveys
① Common sources of errors
C A worded question
poorly
↳ Deviations from the script by the interviewer
C
Misunderstanding on the part of the interviewee
Cy
Memory problem on the part of the interviewer
C. The way the information is recorded by the interviewer
② Types of errors
↳ Random
Sample: sampling errors
->
-
a Questions:invalidity
↳ Biased
Samplebias:not equal chance for everyone in the population
-
Question bias
->
Designing questions
① General rules ② Common mistakes
↳ Remember C.
your RQ Ambiguous terms
C. Decide Co
exactly what you want to find out Long questions
↳ Consider Cn Double-barreled
your audience questions
↳ C.
Imagine yourself respondent
as a Very general questions
Would you answer the question? C
Leading questions
->
How would you answer the question? Cn Double
negative
->
Identify any vague misleading questions
->
or
1
, ③ What to consider?
↳ Is
your systolic pressure normal atthis moment?
↳ Do respondents have the knowledge to answer the questions?
↳ Do you want just a yes/no answer, have you given more possibilities?
↳ Have you thought whether you should include "don't know option?
↳ Are
relying too
you much on the respondents' memory?
or "I
④ Midpoint don'tknow"
↳
If you include don't know option/middle category, people will use it.
->
If you need people to choose one or the other:do not include "don't know"
->
If itis possible that people lack knowledge required to answer a
particular question:
including don't know can be sensible.
-> some will spend less time thinking about difficult question.
↳ Beware
-
Leaving "don'tknow"/middle
out option may cause irritation.
sometimes middle
->
ground/don't know is a
possibility.
⑤
Answer options and measurement level
(DN-0-1-R
↳y based on what researcher desires
⑥ Being specific
↳ How detailed should the answer be?
↳ Sometimes, is more burdensome.
more precise
⑦ Social desirability
↳ You answer as what society expect you to answer
↳ Make undesirable answer less threatening
Everybody makes mistake
->
->
Anonymous data
->
Desirability varies between groups and culture
⑧ Balanced question
↳ Balanced
pros and cons in a question
↳ Is it A, or is it B?
⑨ Primacy and intimacy
↓ ↓
first ans last ans
cannot leave out individual bias
category category -
↳ Rotate answer -- ↓ bias at the
aggregate level
, ⑩ Open a close question
C
Open-ended question
-> Noless bias due to
options
validity
C increase
->
can be
exhausting for a respondent
->
Explorative questions
↳
Unanticipated responses
Very important to clarify the question
->
->
Transformations after datacollection / lots of differentresponse.
-> sometimes answer categories are
preferred
->
Advantages
In
Respondents answer in their own term
Cn Allow for
new,unexpected responses
C
Exploratory generate fixedanswer categories
-
Disadvantages
->
Cn
Time-consuming
C
Difficultto code
2 Interviewer variation in
recording answers
Pretest
G
Pretesting an interview
Evaluate question
->
↳ It is worded
easy to read as
In
Respondents can understand in a consistent way
->
Behaviors
*
Whether or not the interviewers read the questions as worded
↳ Whether not the
or respondents ask for clarification
C. Whether or not the respondents initially give an adequate answer thatrequires
interviewer
probing
Results are systematic and replicable
->
Cnpretesting a self-administered questionnaires -> in person, w/a group of potential respondents
↳
Debugging a computer-assisted instrument
-> The ease of use
The contigency instructions
->
are working as
advantages
intended - computer's
Cp
Survey instruments length
Introduction:What is survey research about?
2 Produce statistics/draw inferences abouta target audience.
C ↳ Disadvantages
Advantages
->
Relatively low cost ->
Generally limited to scaled or check-list questions
->
Relatively fastdelivery -> No control over response rate
->
Rapid data processing -> Provide information more than
understanding
-> can search large population
->
Increasing public resistance
-> various mode:phone, mail, Internet, f2f... -
Difficult to exploreissues in depth.
C Two inferences
How
->
closely the sample responding mirrors the population -> external validity
-> How well answers measure characteristics internal validity
Errors in surveys
① Common sources of errors
C A worded question
poorly
↳ Deviations from the script by the interviewer
C
Misunderstanding on the part of the interviewee
Cy
Memory problem on the part of the interviewer
C. The way the information is recorded by the interviewer
② Types of errors
↳ Random
Sample: sampling errors
->
-
a Questions:invalidity
↳ Biased
Samplebias:not equal chance for everyone in the population
-
Question bias
->
Designing questions
① General rules ② Common mistakes
↳ Remember C.
your RQ Ambiguous terms
C. Decide Co
exactly what you want to find out Long questions
↳ Consider Cn Double-barreled
your audience questions
↳ C.
Imagine yourself respondent
as a Very general questions
Would you answer the question? C
Leading questions
->
How would you answer the question? Cn Double
negative
->
Identify any vague misleading questions
->
or
1
, ③ What to consider?
↳ Is
your systolic pressure normal atthis moment?
↳ Do respondents have the knowledge to answer the questions?
↳ Do you want just a yes/no answer, have you given more possibilities?
↳ Have you thought whether you should include "don't know option?
↳ Are
relying too
you much on the respondents' memory?
or "I
④ Midpoint don'tknow"
↳
If you include don't know option/middle category, people will use it.
->
If you need people to choose one or the other:do not include "don't know"
->
If itis possible that people lack knowledge required to answer a
particular question:
including don't know can be sensible.
-> some will spend less time thinking about difficult question.
↳ Beware
-
Leaving "don'tknow"/middle
out option may cause irritation.
sometimes middle
->
ground/don't know is a
possibility.
⑤
Answer options and measurement level
(DN-0-1-R
↳y based on what researcher desires
⑥ Being specific
↳ How detailed should the answer be?
↳ Sometimes, is more burdensome.
more precise
⑦ Social desirability
↳ You answer as what society expect you to answer
↳ Make undesirable answer less threatening
Everybody makes mistake
->
->
Anonymous data
->
Desirability varies between groups and culture
⑧ Balanced question
↳ Balanced
pros and cons in a question
↳ Is it A, or is it B?
⑨ Primacy and intimacy
↓ ↓
first ans last ans
cannot leave out individual bias
category category -
↳ Rotate answer -- ↓ bias at the
aggregate level
, ⑩ Open a close question
C
Open-ended question
-> Noless bias due to
options
validity
C increase
->
can be
exhausting for a respondent
->
Explorative questions
↳
Unanticipated responses
Very important to clarify the question
->
->
Transformations after datacollection / lots of differentresponse.
-> sometimes answer categories are
preferred
->
Advantages
In
Respondents answer in their own term
Cn Allow for
new,unexpected responses
C
Exploratory generate fixedanswer categories
-
Disadvantages
->
Cn
Time-consuming
C
Difficultto code
2 Interviewer variation in
recording answers
Pretest
G
Pretesting an interview
Evaluate question
->
↳ It is worded
easy to read as
In
Respondents can understand in a consistent way
->
Behaviors
*
Whether or not the interviewers read the questions as worded
↳ Whether not the
or respondents ask for clarification
C. Whether or not the respondents initially give an adequate answer thatrequires
interviewer
probing
Results are systematic and replicable
->
Cnpretesting a self-administered questionnaires -> in person, w/a group of potential respondents
↳
Debugging a computer-assisted instrument
-> The ease of use
The contigency instructions
->
are working as
advantages
intended - computer's
Cp
Survey instruments length
