LOGO
LOGO
Construction of
Construction of
Questionnaire
Questionnaire
Construct
Construction o
ion o
f
f
Questionnaire
Questionnaire
Lecture # 6 Lecture # 6Contents
Contents
What is
What is questionnaire?questionnaire?
1.
1.
Framework for
Framework for questionsquestions
2.
2.
Types of questionnaires Types of questionnaires3.
3.
Guidelines Guidelines4.
4.
LOGO
LOGO
What is
What is
questionnaire?
questionnaire?
What is
What is
What is
questionnaire?
questionnaire?
A questionnaire is a research
A questionnaire is a research
instrument consisting of a series of
instrument consisting of a series of
questions and other prompts for the
questions and other prompts for the
purpose of gathering information from
purpose of gathering information from
respondents.
respondents.
Although they are often designed for
Although they are often designed for
statistical analysis of the responses,
statistical analysis of the responses,
this is not always the case.
this is not always the case.
Remember, good questionnaires take
Remember, good questionnaires take
time
time
to
to
develop.
develop.
They
They
are
are
not
not
just
just
questions on a page.
Questionnaires are used in sample
Questionnaires are used in sample
surveys or censuses to elicit reports
surveys or censuses to elicit reports
of facts, attitudes, and other
of facts, attitudes, and other
subjective states.
subjective states.
The developments are beginning to
The developments are beginning to
transform survey questionnaire
transform survey questionnaire
construction from an art to a
construction from an art to a
science.
What is not questionnaire?
What is not questionnaire?
A questionnaire is not just a list of
A questionnaire is not just a list of
questions.
questions.
It is tempting to begin questionnaire design
It is tempting to begin questionnaire design
by drafting possible qns on the survey
by drafting possible qns on the survey
topics.
topics.
That is obviously important, but not the right
That is obviously important, but not the right
place to start.
place to start.
We must first think about the following
We must first think about the following
topics.
topics.
Objectives of questionnaire
Objectives of questionnaire
construction?
construction?
To maximize the proportion of
To maximize the proportion of
subjects answering our
subjects answering our
questionnaire - that is,
questionnaire - that is,
the response
the response
rate.
rate.
To obtain accurate relevant
To obtain accurate relevant
information for our survey
History
History
In 1086 William the Conqueror
In 1086 William the Conqueror
surveyed the wealth and
surveyed the wealth and
landholdings of England using a
landholdings of England using a
standard set of inquiries and
standard set of inquiries and
compiled the results in the
compiled the results in the
“Domesday Book.”
“Domesday Book.”
The questionnaire was invented by
The questionnaire was invented by
Sir Francis Galton: 1874
Advantages of Written
Advantages of Written
Questionnaires
Questionnaires
Questionnaires are very cost effectiveQuestionnaires are very cost effective
when compared to face-to-face interviews.
when compared to face-to-face interviews.
This is especially true for studies involving
This is especially true for studies involving
large sample sizes and large geographic
large sample sizes and large geographic
areas. Written questionnaires become
areas. Written questionnaires become
even more cost effective as the number of
even more cost effective as the number of
research questions increases.
research questions increases.
Questionnaires are easy to analyze. DataQuestionnaires are easy to analyze. Data entry and tabulation for nearly all surveys
entry and tabulation for nearly all surveys
can be easily done with many computer
can be easily done with many computer
software packages.
Questionnaires are familiar to mostQuestionnaires are familiar to most
people. Nearly everyone has had some
people. Nearly everyone has had some
experience completing questionnaires and
experience completing questionnaires and
they generally do not make people
they generally do not make people
apprehensive.
apprehensive.
Questionnaires reduce bias. There isQuestionnaires reduce bias. There is uniform question presentation and no
uniform question presentation and no
middle-man bias. The researcher's own
middle-man bias. The researcher's own
opinions will not influence the respondent
opinions will not influence the respondent
to answer questions in a certain manner.
to answer questions in a certain manner.
There are no verbal or visual clues to
There are no verbal or visual clues to
influence the respondent.
Questionnaires are less intrusive
Questionnaires are less intrusive
than telephone or face-to-face
than telephone or face-to-face
surveys. When a respondent
surveys. When a respondent
receives a questionnaire in the mail,
receives a questionnaire in the mail,
he is free to complete the
he is free to complete the
questionnaire on his own time-table.
questionnaire on his own time-table.
Unlike other research methods, the
Unlike other research methods, the
respondent is not interrupted by the
respondent is not interrupted by the
research instrument.
Disadvantages Of Written
Disadvantages Of Written
Questionnaires
Questionnaires
Possibility of low response rates
Possibility of low response rates
:
:
Low response is the curse of
Low response is the curse of
statistical analysis. It can
statistical analysis. It can
dramatically lower our confidence in
dramatically lower our confidence in
the results. Response rates vary
the results. Response rates vary
widely from one questionnaire to
widely from one questionnaire to
another (10% - 90%), however,
another (10% - 90%), however,
well-designed studies consistently
well-designed studies consistently
produce high response rates.
Inability to probe responses.Inability to probe responses. Questionnaires are structured
Questionnaires are structured
instruments. They allow little flexibility to
instruments. They allow little flexibility to
the respondent with respect to response
the respondent with respect to response
format. In essence, they often lose the
format. In essence, they often lose the
"flavor of the response" (i.e., respondents
"flavor of the response" (i.e., respondents
often want to qualify their answers).
often want to qualify their answers).
By allowing frequent space for comments, the rBy allowing frequent space for comments, the researcher canesearcher can
partially overcome this disadvantage. Comments are among
partially overcome this disadvantage. Comments are among
the most helpful of all the information on
the most helpful of all the information on the questionnaire,the questionnaire,
and they usually provide insightful information that would
and they usually provide insightful information that would
have otherwise been lost.
Nearly ninety percent of all communicationNearly ninety percent of all communication
is visual. Gestures and other visual cues
is visual. Gestures and other visual cues
are not available with written
are not available with written
questionnaires. The
questionnaires. The lack of personallack of personal
contact
contact will have different effectswill have different effects
depending on the type of information being
depending on the type of information being
requested. A questionnaire requesting
requested. A questionnaire requesting
factual information will probably not be
factual information will probably not be
affected by the lack of personal contact. A
affected by the lack of personal contact. A
questionnaire probing sensitive issues or
questionnaire probing sensitive issues or
attitudes may be severely affected.
When returned questionnaires arrive in theWhen returned questionnaires arrive in the
mail, it's natural to assume that the
mail, it's natural to assume that the
respondent is the same person you sent
respondent is the same person you sent
the questionnaire to. This may not
the questionnaire to. This may not actuallyactually
be the case.
be the case. Many times businessMany times business
questionnaires get handed to other
questionnaires get handed to other
employees for completion.
employees for completion. HousewivesHousewives
sometimes respond for their
sometimes respond for their husbands.husbands.
Kids respond as a prank. For a
Kids respond as a prank. For a variety of variety of
reasons, the respondent may not be who
reasons, the respondent may not be who
you think it is. It is a confo
you think it is. It is a confounding errorunding error
inherent in questionnaires.
Finally, questionnaires are simply
Finally, questionnaires are simply
not suited for some people
not suited for some people
. For
. For
example, a written survey to a
example, a written survey to a
group of poorly educated people
group of poorly educated people
might not work because of reading
might not work because of reading
skill problems. More frequently,
skill problems. More frequently,
people are turned off by written
people are turned off by written
questionnaires because of misuse.
LOGO
LOGO
Framework for
Framework for
construction of
construction of
questionnaire
questionnaire
Framework for
Framework for
construction of
construction of
questionnaire
questionnaire
Clarify goals
Clarify goals
Conversations with several of the
Conversations with several of the
stakeholders
stakeholders
Focus your goals
Focus your goals
Invite people to take survey
Invite people to take survey
Analyze responses
Analyze responses
Translating data into meaningful
Translating data into meaningful
information
information
Present Results
Present Results
Celebrate your success
Celebrate your success
Steps for Developing
Steps for Developing
Questionnaires
Questionnaires
Develop the PurposeDevelop the Purpose
Create the Conceptual
Create the Conceptual
Framework
Framework
Write Questions
Write Questions
Proper construction of the
Proper construction of the
questionnaire is essential to its
questionnaire is essential to its
success
success
Powell, Ronald R.
Framework: Questionnaire
Framework: Questionnaire
Construction
Construction
Specify What Specify What Information Information Will be Sought Will be Sought Identify Identify Correct Correct Respondent Respondent Determine Determine Method of Method of Administration Administration Question StructureQuestion Structure Question WordingQuestion Wording Question SequencingQuestion Sequencing
PRETEST PRETEST and and Revise Revise
Steps to design a questionnaire:
Steps to design a questionnaire:
1.
1. Write out the primary and secondary aimsWrite out the primary and secondary aims
of your study.
of your study.
2.
2. Write out concepts/information to beWrite out concepts/information to be
collected that relates to these aims.
collected that relates to these aims.
3.
3. Review the current literature to identifyReview the current literature to identify
already validated questionnaires that
already validated questionnaires that
measure your specific area of
measure your specific area of interest.interest.
4.
4. Compose a draft of your questionnaire.Compose a draft of your questionnaire.
5.
5. Revise the draft.Revise the draft.
6.
Questionnaire Design - General
Questionnaire Design - General
Considerations
Considerations
Most problems with questionnaire analysisMost problems with questionnaire analysis can be traced back to the design phase of
can be traced back to the design phase of
the project.
the project. Well-defined goals are theWell-defined goals are the
best way to assure a good questionnaire
best way to assure a good questionnaire
design.
design.
When the goals of a study can beWhen the goals of a study can be expressed in a few clear
expressed in a few clear and conciseand concise
sentences, the design of the questionnaire
sentences, the design of the questionnaire
becomes considerably easier.
becomes considerably easier.
The questionnaire is developed to directlyThe questionnaire is developed to directly address the goals of the study.
One of the best ways to clarify
One of the best ways to clarify
your
your
study goals is to
study goals is to
decide how you
decide how you
intend to use the information.
intend to use the information.
Do
Do
this before you begin designing the
this before you begin designing the
study. This sounds obvious, but
study. This sounds obvious, but
many researchers neglect this task.
many researchers neglect this task.
Why do research if the results will
Why do research if the results will
not be used?
Be sure to commit the study goals to
Be sure to commit the study goals to
writing. Whenever you are unsure of
writing. Whenever you are unsure of
a question, refer to the study goals
a question, refer to the study goals
and a solution will
and a solution will
become clear.
become clear.
Ask only questions that directly
Ask only questions that directly
address the study goals.
address the study goals.
Avoid the
Avoid the
temptation to ask questions because
temptation to ask questions because
it would be "interesting to know".
As a general rule, with only a fewAs a general rule, with only a few
exceptions,
exceptions, long questionnaires get lesslong questionnaires get less
response than short questionnaires.
response than short questionnaires. KeepKeep
your questionnaire short. In fact, the
your questionnaire short. In fact, the
shorter the better.
shorter the better.
Response rate is the single most importantResponse rate is the single most important
indicator of how much confidence you can
indicator of how much confidence you can
place in the results. A low response rate can
place in the results. A low response rate can
be devastating to a study.
be devastating to a study.
One of the most effective methods of One of the most effective methods of
maximizing response is to shorten the
maximizing response is to shorten the
questionnaire.
If your survey is over a few pages,If your survey is over a few pages, try totry to eliminate questions.
eliminate questions.
Many people have difficulty knowing whichMany people have difficulty knowing which questions could be eliminated.
questions could be eliminated.
For the elimination round, read eachFor the elimination round, read each
question and ask, "How am I going to use
question and ask, "How am I going to use
this information?" If the information will
this information?" If the information will
be used in a decision-making process, then
be used in a decision-making process, then
keep the question... it's important. If not,
keep the question... it's important. If not,
throw it out.
One important way to assure a successfulOne important way to assure a successful survey is to
survey is to include other experts andinclude other experts and
relevant decision-makers
relevant decision-makers in thein the
questionnaire design process.
questionnaire design process.
Their suggestions will improve theTheir suggestions will improve the
questionnaire and they will subsequently
questionnaire and they will subsequently
have more confidence in the results.
Formulate a plan for doing the statisticalFormulate a plan for doing the statistical analysis during the design stage
analysis during the design stage of theof the
project. Know how every question will be
project. Know how every question will be
analyzed and be prepared to handle
analyzed and be prepared to handle
missing data. If you cannot specify how
missing data. If you cannot specify how
you intend to analyze a question or use
you intend to analyze a question or use
the information, do not use it in the
the information, do not use it in the
survey.
Give your questionnaire a title Give your questionnaire a title that is short andthat is short and
meaningful to the respondent.
meaningful to the respondent. A questionnaireA questionnaire
with a title is generally perceived to be more
with a title is generally perceived to be more
credible than one without
credible than one without
Include clear and concise instructions on howInclude clear and concise instructions on how
to complete the questionnaire
to complete the questionnaire. These must be. These must be
very easy to understand, so use short
very easy to understand, so use short
sentences and basic vocabulary. Be sure to
sentences and basic vocabulary. Be sure to
print the return address on the questionnaire
print the return address on the questionnaire
itself (since questionnaires often get separated
itself (since questionnaires often get separated
from the reply envelopes).
Begin with a few
Begin with a few
non-threatening and
non-threatening and
interesting items.
interesting items.
If the first items are
If the first items are
too threatening or "boring", there is
too threatening or "boring", there is
little chance that the person will
little chance that the person will
complete the
complete the
questionnair
questionnair
e.
e.
People generally look at the first few
People generally look at the first few
questions before deciding whether or
questions before deciding whether or
not to complete the
not to complete the
questionnai
questionnai
re. Make
re. Make
them want to
them want to
continue by putting
continue by putting
interesting questions first.
Use simple and direct Use simple and direct language.language. The questionsThe questions
must be clearly understood by the respondent.
must be clearly understood by the respondent.
The wording of a question should be simple andThe wording of a question should be simple and
to the point.
to the point.
Do not use uncommon words or long sentences.Do not use uncommon words or long sentences.
Make items as brief as possible. This will
Make items as brief as possible. This will reducereduce
misunderstan
misunderstandings and make dings and make the questionnairethe questionnaire
appear easier to complete.
appear easier to complete.
One way to eliminate One way to eliminate misundersmisunderstandings is totandings is to
emphasize crucial words in each item by using
emphasize crucial words in each item by using
bold, italics or
Leave adequate space for respondents toLeave adequate space for respondents to make comments.
make comments.
One criticism of questionnaires is theirOne criticism of questionnaires is their inability to retain the "flavor" of a
inability to retain the "flavor" of a
response.
response.
Leaving space for comments will provideLeaving space for comments will provide valuable information not captured by the
valuable information not captured by the
response categories.
response categories.
Leaving white space also makes theLeaving white space also makes the questionnaire look easier and this
questionnaire look easier and this
increases response.
Place the most important items in the firstPlace the most important items in the first half of the questionnaire.
half of the questionnaire.
Respondents often send back partiallyRespondents often send back partially completed questionnaires.
completed questionnaires.
By putting the most important items nearBy putting the most important items near the beginning, the partially completed
the beginning, the partially completed
questionnaires will still contain important
questionnaires will still contain important
information.
Use professional production methodsUse professional production methods forfor the questionnaire--either desktop
the questionnaire--either desktop
publishing or typesetting and keylining.
publishing or typesetting and keylining.
Be creative.Be creative.
Try different colored inks and paper. TheTry different colored inks and paper. The object is to make your questionnaire stand
object is to make your questionnaire stand
out from all the others the respondent
out from all the others the respondent
receives.
The final test of a questionnaire is toThe final test of a questionnaire is to try it ontry it on
representatives of the target audience
representatives of the target audience..
If there are problems with the questionnaire, theyIf there are problems with the questionnaire, they
almost always show up here. If possible, be present
almost always show up here. If possible, be present
while a respondent is
while a respondent is completing the questionnairecompleting the questionnaire
and tell her that it
and tell her that it is okay to ask you foris okay to ask you for
clarification of any item.
clarification of any item.
The questions she asks are indicative of problems inThe questions she asks are indicative of problems in
the questionnaire (i.e., the questions on the
the questionnaire (i.e., the questions on the
questionnaire must be without any
questionnaire must be without any ambiguityambiguity
because there will be no chance to clarify a
because there will be no chance to clarify a
question when the survey is
LOGO
LOGO
Types of questions
Types of questions
Types of questions
Types of questions
Classification # 1
Classification # 1
1.
1. Factual questions:Factual questions: age, gender…age, gender…
2.
2. Opinion and attitude questions:Opinion and attitude questions: intended tointended to
determine person’s ideas, inclinations, prejudices,
determine person’s ideas, inclinations, prejudices,
… with attitude scales, and indexes
… with attitude scales, and indexes
3.
3. Information questions:Information questions: designed to measuredesigned to measure
respondent’s knowledge
respondent’s knowledge
4.
4. Self-perception questionsSelf-perception questions: Similar to attitude: Similar to attitude
questions but are restricted to one’s opinions
questions but are restricted to one’s opinions
about himself or herself
about himself or herself
5.
5. Standards of action questionsStandards of action questions: used to determine: used to determine
how respondents would act in certain
how respondents would act in certain
circumstances
circumstances
6.
6. Questions about actual past or present behaviourQuestions about actual past or present behaviour::
7.
Classification # 2
Classification # 2
Four types of response scales for
Four types of response scales for
closed-ended questions are
closed-ended questions are
distinguished:
distinguished:
Dichotomous,Dichotomous, where the respondent has two optionswhere the respondent has two options
Nominal-polytomousNominal-polytomous, where the respondent has , where the respondent has moremore
than two unordered options
than two unordered options
Ordinal-polytomous,Ordinal-polytomous, where the respondent has morewhere the respondent has more
than two ordered options
than two ordered options
(bounded)Continuous(bounded)Continuous, where the respondent is, where the respondent is
presented with a continuous scale
1. Contingency questions
1. Contingency questions - A question that is- A question that is
answered only if the respondent gives a
answered only if the respondent gives a
particular response to a previous
particular response to a previous question.question.
This avoids asking questions of people that
This avoids asking questions of people that
do not apply to them
do not apply to them
2. Matrix questions
2. Matrix questions - Identical response- Identical response
categories are assigned to multiple
categories are assigned to multiple
questions. The questions are placed one
questions. The questions are placed one
under the other, forming a matrix with
under the other, forming a matrix with
response categories along the top and a
response categories along the top and a
list of questions down the side. This
list of questions down the side. This is anis an
efficient use of page space and
efficient use of page space and
respondents’ time.
3. Closed ended
3. Closed ended questionsquestions - Respondents’ answers are- Respondents’ answers are
limited to a fixed set of responses.
limited to a fixed set of responses.
Yes/no questions Yes/no questions - The respondent answers with a “yes” or - The respondent answers with a “yes” or aa
“no”.
“no”.
Multiple choiceMultiple choice - The respondent has several option from which- The respondent has several option from which
to choose.
to choose.
Scaled questionsScaled questions - Responses are graded on a continuum- Responses are graded on a continuum
(example : rate the appearance of the product on
(example : rate the appearance of the product on a scale from 1a scale from 1
to 10, with 10 being
to 10, with 10 being the most preferred appearance). Examplesthe most preferred appearance). Examples
of types of scales include the
of types of scales include the Likert scale, semantic differentialLikert scale, semantic differential
scale, and rank-order scale (See scale for a complete list
scale, and rank-order scale (See scale for a complete list of of
scaling techniques.).
Open ended questionsOpen ended questions -- No options or predefinedNo options or predefined
categories are suggested. Examples of types of open
categories are suggested. Examples of types of open
ended questions include:
ended questions include:
Completely unstructuredCompletely unstructured - For example, “What is your opinion of - For example, “What is your opinion of
questionnaires?”
questionnaires?”
Word associationWord association - Words are presented and the respondent mentions- Words are presented and the respondent mentions
the first word that comes to mind.
the first word that comes to mind.
Sentence completionSentence completion - Respondents complete an incomplete- Respondents complete an incomplete
sentence. For example, “The most important consideration in my
sentence. For example, “The most important consideration in my
decision to buy a new house is . . .”
decision to buy a new house is . . .”
Story completionStory completion - Respondents complete an incomplete story.- Respondents complete an incomplete story.
Picture completionPicture completion - Respondents fill in an empty conversation balloon.- Respondents fill in an empty conversation balloon.
Thematic apperception testThematic apperception test - Respondents explain a picture or make- Respondents explain a picture or make
up a story about what they think is happening in the picture
Question Structure
Question Structure
Open-end questionsOpen-end questions
If you were to purchase a new automobile during 2001, what brand of automobile would
If you were to purchase a new automobile during 2001, what brand of automobile would
y
yoou u mmoosst t lliikkeelly y ppuurrcchhaassee? ? _____________________ _ bbrraanndd
Closed-ended with ordered categories Closed-ended with ordered categories
If you were to purchase a new automobile during 2001, how likely is it that you would
If you were to purchase a new automobile during 2001, how likely is it that you would
purchase the following brands of
purchase the following brands of automobiles?automobiles?
V Veerry y LLiikkeellyy SSoommeewwhhaat t LLiikkeellyy SSoommeewwhhaat t UUnnlliikkeellyy VVeerry y UUnnlliikkeellyy T Tooyyootta a CCaammrry y 1 1 2 2 3 3 44 H Hoonndda a AAccccoorrdd 1 1 2 2 3 3 44 N Niissssaan n MMaaxxiimma a 1 1 2 2 3 3 44
Closed-ended with unordered categories Closed-ended with unordered categories
If you were to purchase a new automobile during 2001, which one of the following
If you were to purchase a new automobile during 2001, which one of the following
automobiles would you most likely purchase?
automobiles would you most likely purchase?
Toyota Camry
Toyota Camry tt Honda AccordHonda Accord tt
Nissan Maxima
Nissan Maxima tt
Partially closed-ended Partially closed-ended
If you were to purchase a new automobile during 2001, which one of the following would
If you were to purchase a new automobile during 2001, which one of the following would
you most likely purchase?
you most likely purchase?
Toyota Camry
Toyota Camry tt Honda AccordHonda Accord tt
Nissan Maxima
Nissan Maxima tt Other (pleaseOther (please
specify)______________
Closed Vs. Open-ended Questions
Closed Vs. Open-ended Questions
A
Addvvaannttaaggeess DDiissaaddvvaannttaaggeess
Closed-ended
Closed-ended ••Easy and quick to answerEasy and quick to answer
•
•Answers across resps easy toAnswers across resps easy to
compare
compare
•
•Answers easier to analyze onAnswers easier to analyze on
computer
computer
•
•Response choices makeResponse choices make
question clearer
question clearer
•
•Easy to replicate studyEasy to replicate study
•
•Can put ideas in resp's headCan put ideas in resp's head
•
•Resps w/ no opinion answer anywayResps w/ no opinion answer anyway
•
•Resps can Resps can feel constrained/frustratedfeel constrained/frustrated
•
•Many choices can be confusingMany choices can be confusing
•
•Can't tell if resp. misinterpreted theCan't tell if resp. misinterpreted the
question
question
•
•Fine distinctions may be lostFine distinctions may be lost
•
•Clerical mistakes easy to makeClerical mistakes easy to make
•
•Force respondents into simpleForce respondents into simple
responses
responses
Open-ended
Open-ended ••Permit unlimited number of Permit unlimited number of
answers
answers
•
•Resps can qualify and clarifyResps can qualify and clarify
responses
responses
•
•Can find the unanticipatedCan find the unanticipated
•
•Reveal resps thinking processesReveal resps thinking processes
•
•Resps give answers w/ diff. level of Resps give answers w/ diff. level of
detail
detail
•
•Answers can be irrelevantAnswers can be irrelevant
•
•Inarticulate or forgetful resps are atInarticulate or forgetful resps are at
disadvantage
disadvantage
•
•Coding responses is subjective andCoding responses is subjective and
tedious
tedious
•
•Requires more resp. time and effortRequires more resp. time and effort
•
•Intimidates respondentsIntimidates respondents
•
•When resp omits a response, can't tellWhen resp omits a response, can't tell
if its because
if its because
of belief or just forgetfulness
Closed Vs. Open-ended Questions
Closed Vs. Open-ended Questions
Evaluation Evaluation Criteria Criteria A Addvvaannttaaggeess DDiissaaddvvaannttaaggeess T Tiimmee LLeesss s ttiimme e ttoo
respond & record
respond & record
Longer time to Longer time to develop develop Quick transfer to Quick transfer to computer form computer form
Evaluation Evaluation Criteria Criteria A Addvvaannttaaggeess DDiissaaddvvaannttaaggeess C Coosstt CChheeaappeer r dduue e tto o lloowweer r recording & recording & interpreting time interpreting time
Evaluation Evaluation Criteria Criteria A Addvvaannttaaggeess DDiissaaddvvaannttaaggeess A Accccuurraaccyy LLeesss s iinntteerrvviieewweer r recording error recording error
Forcing answers into
Forcing answers into
categories categories Respondent Respondent convenience convenience More convenient to More convenient to answer answer
Structured Questions:
Structured Questions:
Effect on Respondent Answers
Effect on Respondent Answers
Response order effects
Response order effects
Visual
Visual presentation
presentation of
of response
response alternatives
alternatives
Primacy effects
Primacy effects
initial items establish a cognitive framework to guideinitial items establish a cognitive framework to guide processing
processing
initial items get deeper cognitive processing & lessinitial items get deeper cognitive processing & less
interference than later items interference than later items
satisficing — minimize psychological costs bysatisficing — minimize psychological costs by
choosing first acceptable response
choosing first acceptable response alternativealternative
SOURCE: Krosnick and Alwin, POQ SOURCE: Krosnick and Alwin, POQ
Response order effects
Response order effects
Oral
Oral presentation
presentation of
of response
response alternatives
alternatives
Recency
Recency
effects
effects
externally based processing
externally based processing
each read alternative terminates processing of previous alternative each read alternative terminates processing of previous alternative increased processing of later response alternatives
increased processing of later response alternatives
Structured Questions:
Structured Questions:
Effect on Respondent Answers
Effect on Respondent Answers
Source
Source:: An evalAn evaluatiouation of a cognn of a cognitive thitive theory of reeory of responssponse-ordee-order effects in sur effects in surveyrvey measurement,” J. Krosnick and D. Alwin,
measurement,” J. Krosnick and D. Alwin, Public Opinion Quarterly Public Opinion Quarterly , 1987,, 1987, 51, 1201-1219.
Question Sequence
Question Sequence
and Order Effects
and Order Effects
Question
Question
Do y
Do you think
ou think a comm
a communist country
unist country lik
lik
reporters come in and send back to t
reporters come in and send back to t
Do y
Do you think
ou think the United States shoul
the United States shoul
other countries come in and send ba
other countries come in and send ba
Q
Qu
ue
es
sttiio
on
n s
se
eq
qu
ue
en
nc
ce
e 2
2
Y
Ye
es
s
Do you think the United States should let communist newspaper reporters Do you think the United States should let communist newspaper reporters from other countries come in and send back to their papers the news as they from other countries come in and send back to their papers the news as they see it?
see it? Do you th
Do you think a commink a communist countunist country like Russia shoury like Russia should let Amld let American erican newnewspaper spaper reporters come in and send back to their pape
reporters come in and send back to their pape rs the news as they see it?rs the news as they see it?
36% 36% 66% 66% Once Committed Once Committed Hard to be Hard to be Unfair Unfair 37% 37% Difference ! Difference ! Source
Questionna
Questionna
ire
ire
Format
Format
Matrix questions are often useful.
Matrix questions are often useful.
They use space efficientlyThey use space efficiently
Respondents can answer quicklyRespondents can answer quickly
Respondents can compare across questions for Respondents can compare across questions for consistency.consistency.
However, matrix questions have some
However, matrix questions have some
problems.
problems.
You might be tempted to put questions into a matrix whenYou might be tempted to put questions into a matrix when
they belong in some other format.
they belong in some other format.
Respondents may pattern their Respondents may pattern their answers inappropriately.answers inappropriately.
Order effects may Order effects may be quite pronounced, with strong potentialbe quite pronounced, with strong potential
for abuse of the survey.
Some special comments on order
Some special comments on order
effects.
effects.
Early answers tend to Early answers tend to influence later answers.influence later answers.
Some respondents (such as less Some respondents (such as less educated respondents) areeducated respondents) are
more subject to order effects than other
more subject to order effects than other respondents.respondents.
RandomiziRandomizing questions tends not to help - it ng questions tends not to help - it just makes thejust makes the
survey more confusing.
survey more confusing.
Fun or interesting questions in the early part of aFun or interesting questions in the early part of a
questionnai
questionnaire can re can maintain interest, but general,maintain interest, but general,
demographic sorts of questions can establish rapport.
demographic sorts of questions can establish rapport.
•
• You might want to adjust the order of questions according tYou might want to adjust the order of questions according too
the type of survey –
Other Types
Other Types
Leading Questions:
Leading Questions:
A leading
A leading
question is one that forces or implies
question is one that forces or implies
a certain type of answer. E.g.
a certain type of answer. E.g.
SuperbSuperb ExcellentExcellent GreatGreat GoodGood Fair Fair
Embarrassing Questions:
Embarrassing Questions:
Embarrassing
Embarrassing
questions dealing with personal or
questions dealing with personal or
private matters should be avoided. Your
private matters should be avoided. Your
data is only as good as the
data is only as good as the trust and
trust and
care that your respondents give you. If
care that your respondents give you. If
you make them feel uncomfortable, you
you make them feel uncomfortable, you
will lose their trust. Do not ask
will lose their trust. Do not ask
embarrassing questions.
Hypothetical Questions:
Hypothetical Questions:
Hypothetical are based, at best, on
Hypothetical are based, at best, on
conjecture and, at worst, on fantasy.
conjecture and, at worst, on fantasy.
I simple question such as:
I simple question such as:
If you were governor, what would you do to combatIf you were governor, what would you do to combat terrorism?
LOGO
LOGO
Most frequent errors
Most frequent errors
in questionnaires
in questionnaires
Most frequent errors
Most frequent errors
in questionnaires
in questionnaires
Issues to Consider
Issues to Consider
To Neutral or Not to Neutral.To Neutral or Not to Neutral.
Think about how a respondent enters a
Think about how a respondent enters a
response
response
scale and how many decisions the person
scale and how many decisions the person
must
must
make.
make.
S Sttrroonngglly y DDiissaaggrreeee DDiissaaggrreeee NNeeiitthheer r AAggrreee e nnoor r Disagree Disagree A Aggrreeee SSttrroonngglly y AAggrreeee 1 2 3 4 5 1 2 3 4 5Pre-coding can save time with data entry.
Pre-coding can save time with data entry.
Has this presentation increased your
Has this presentation increased your understanding of questionnaireunderstanding of questionnaire
development?
development?
1
1 YesYes 22 NoNo
There is more than one way to
There is more than one way to
ask a question. Use a
ask a question. Use a
variety of question formats.
variety of question formats.
Pre-code
What do you think?
What do you think?
Were you satisfied with the quality of
Were you satisfied with the quality of
our
our
services and programs?
services and programs?
A. Yes
A. Yes
B. No
B. No
Double
Double
Barrel:
Barrel:
Ask
Ask
only
only
one
one
question
question
at
at
a
a
time
most frequent errors in
most frequent errors in
questionnaires
questionnaires
questions which don't quite mean whatquestions which don't quite mean what the researcher intended
the researcher intended
questions which don't probe to find outquestions which don't probe to find out what the respondent really meant
what the respondent really meant
long and complicated questions
long and complicated questions
questions which repeat what has alreadyquestions which repeat what has already been asked
been asked
questions which don't allow thequestions which don't allow the
respondent to answer in a way which is
respondent to answer in a way which is
relevant
relevant
questions which are inappropriate to thequestions which are inappropriate to the research method
bad routing which leaves the interviewerbad routing which leaves the interviewer wondering which question to ask next, or
wondering which question to ask next, or
worse, routing to the wrong one
worse, routing to the wrong one
too many questions
too many questions
poorly laid out questions which are badlypoorly laid out questions which are badly grouped in the questionnaire
grouped in the questionnaire
questions which have been missed outquestions which have been missed out completely
completely
pages of the questionnaire which arepages of the questionnaire which are missing or out of order.
Research showing that small
Research showing that small
changes in question wording or
changes in question wording or
order can substantially affect
order can substantially affect
responses has reinforced the
responses has reinforced the
assumption that questions must be
assumption that questions must be
asked exactly as worded, and in the
asked exactly as worded, and in the
same order, to
same order, to
produce comparable
produce comparable
data.
LOGO
LOGO
Guidelines
Guidelines
Guidelines
Guidelines
Quality aims in survey research
Quality aims in survey research
Goal is to collect information that is:
Goal is to collect information that is:
Valid:Valid: measures the quantity or concept thatmeasures the quantity or concept that
is supposed to be measured
is supposed to be measured
Reliable:Reliable: measures the quantity or concept inmeasures the quantity or concept in
a consistent or reproducible manner
a consistent or reproducible manner
Unbiased:Unbiased: measures the quantity or conceptmeasures the quantity or concept
in a way that does
in a way that does not systematically under-not systematically
under-or overestimate the true value
or overestimate the true value
Discriminating:Discriminating: can distinguish adequatelycan distinguish adequately
between respondents for whom the underlying
between respondents for whom the underlying
level of the quantity or concept is
Qualities of a Good Question
Qualities of a Good Question
Evokes the truth.Evokes the truth. Questions must be non-Questions must be
non-threatening. When a respondent is
threatening. When a respondent is
concerned about the consequences of
concerned about the consequences of
answering a question in a particular
answering a question in a particular
manner, there is a good
manner, there is a good possibility that thepossibility that the
answer will not be truthful. Anonymous
answer will not be truthful. Anonymous
questionnaires that contain no identifying
questionnaires that contain no identifying
information are more likely to produce
information are more likely to produce
honest responses than those identifying
honest responses than those identifying
the respondent. If your questionnaire does
the respondent. If your questionnaire does
contain sensitive items, be sure to clearly
contain sensitive items, be sure to clearly
state your policy on confidentiality.
Asks for an answer on only one dimensionAsks for an answer on only one dimension..
The purpose of a survey is to find out
The purpose of a survey is to find out
information. A question that asks for a
information. A question that asks for a
response on more than one dimension will
response on more than one dimension will
not provide the information you are
not provide the information you are
seeking. For example, a
seeking. For example, a researcherresearcher
investigating a new food snack asks "Do
investigating a new food snack asks "Do
you like the texture and flavor of the
you like the texture and flavor of the
snack?" If a
snack?" If a respondent answers "no",respondent answers "no",
then the researcher will not know if the
then the researcher will not know if the
respondent dislikes the texture or the
respondent dislikes the texture or the
flavor, or both
Can accommodate all possible answers.Can accommodate all possible answers.
Multiple choice items are the most popular
Multiple choice items are the most popular
type of survey questions because they are
type of survey questions because they are
generally the easiest for a respondent to
generally the easiest for a respondent to
answer and the easiest to analyze. Asking
answer and the easiest to analyze. Asking
a question that does not accommodate all
a question that does not accommodate all
possible responses can confuse and
possible responses can confuse and
frustrate the respondent. For example,
frustrate the respondent. For example,
consider the question:
consider the question:
What brand of computer do you own? What brand of computer do you own? __ __
A. IBM PC A. IBM PC
Has mutually exclusive options.Has mutually exclusive options. A goodA good
question leaves no ambiguity in the mind of
question leaves no ambiguity in the mind of
the respondent. There should be only one
the respondent. There should be only one
correct or appropriate choice for the
correct or appropriate choice for the
respondent to make. An obvious example is:
respondent to make. An obvious example is:
Where did you grow up? Where did you grow up? __ __
A. country A. country
B. farmB. farm
C. city C. city
A person who grew up on a farm in theA person who grew up on a farm in the
country would not know whether to select
country would not know whether to select
choice A or B.
Produces variability of responses.Produces variability of responses. When aWhen a
question produces no variability in
question produces no variability in
responses, we are left with considerable
responses, we are left with considerable
uncertainty about why we asked the
uncertainty about why we asked the
question and what we learned from
question and what we learned from thethe
information. If a question does not
information. If a question does not
produce variability in responses, it will not
produce variability in responses, it will not
be possible to perform any statistical
be possible to perform any statistical
analyses on the item. For example:
analyses on the item. For example:
What do you think about this report? What do you think about this report? __ __
A. It's the worst report I've read A. It's the worst report I've read
B. It's somewhere between the worst and best B. It's somewhere between the worst and best
C. It's the best report I've read C. It's the best report I've read
Order of questions:
Order of questions:
Go from general to Go from general to particular.particular.
Go from easy to difficult.Go from easy to difficult.
Go from factual to abstract.Go from factual to abstract.
Start with closed format Start with closed format questions.questions.
Start with questions relevant to the Start with questions relevant to the main subject.main subject.
Do not start Do not start with demographic and personalwith demographic and personal
questions.
Follows comfortably from the previousFollows comfortably from the previous question.
question. Writing a questionnaire is similarWriting a questionnaire is similar
to writing anything else. Transitions
to writing anything else. Transitions
between questions should be smooth.
between questions should be smooth.
Grouping questions that are similar will
Grouping questions that are similar will
make the questionnaire easier to
make the questionnaire easier to
complete, and the respondent will feel
complete, and the respondent will feel
more comfortable. Questionnaires that
more comfortable. Questionnaires that
jump from one unrelated topic to another
jump from one unrelated topic to another
feel disjointed and are not likely to
feel disjointed and are not likely to
produce high response rates.
Does not presuppose a certain state of Does not presuppose a certain state of
affairs.
affairs. Among the most subtle mistakes inAmong the most subtle mistakes in
questionnaire design are questions that
questionnaire design are questions that
make an unwarranted assumption. An
make an unwarranted assumption. An
example of this type of mistake is:
example of this type of mistake is:
Are you Are you satisfied with your satisfied with your current auto current auto insurance? insurance? (Yes or (Yes or No)No)
This question will present a problem forThis question will present a problem for
someone who does not currently have auto
someone who does not currently have auto
insurance. Better:
insurance. Better:
Are you satisfied with your Are you satisfied with your current auto insurance? current auto insurance?
___
___ YesYes __ __ NoNo
___
___
Don't have auto
Don't have auto
insurance
insurance
Does not imply a desired answer.Does not imply a desired answer. TheThe
wording of a question is extremely
wording of a question is extremely
important. We are striving for objectivity in
important. We are striving for objectivity in
our surveys and, therefore, must be careful
our surveys and, therefore, must be careful
not to lead the respondent into giving the
not to lead the respondent into giving the
answer we would like to receive. Leading
answer we would like to receive. Leading
questions are usually easily spotted
questions are usually easily spotted
because they use negative phraseology. As
because they use negative phraseology. As
examples:
examples:
Wouldn't you like to receive our free brochure? Wouldn't you like to receive our free brochure?
Does not use emotionally loaded orDoes not use emotionally loaded or vaguely defined words.
vaguely defined words. This is one of theThis is one of the
areas overlooked by both beginners and
areas overlooked by both beginners and
experienced researchers. Quantifying
experienced researchers. Quantifying
adjectives (e.g., most, least, majority) are
adjectives (e.g., most, least, majority) are
frequently used in questions. It is
frequently used in questions. It is
important to understand that these
important to understand that these
adjectives mean different things to
adjectives mean different things to
different people.
Does not use unfamiliar words orDoes not use unfamiliar words or
abbreviations.
abbreviations. Remember who your audienceRemember who your audience
is and write your questionnaire for them. Do
is and write your questionnaire for them. Do
not use uncommon words or compound
not use uncommon words or compound
sentences. Write short sentences.
sentences. Write short sentences.
Abbreviations are okay if you are absolutely
Abbreviations are okay if you are absolutely
certain that every single respondent will
certain that every single respondent will
understand their meanings. If there is any
understand their meanings. If there is any
doubt at all, do not use the abbreviation. The
doubt at all, do not use the abbreviation. The
following question might be okay if all the
following question might be okay if all the
respondents are accountants:
respondents are accountants:
Is not dependent on responses to previousIs not dependent on responses to previous
questions.
questions. Branching in written questionnairesBranching in written questionnaires
should be avoided. While branching can be used as
should be avoided. While branching can be used as
an effective probing technique in telephone and
an effective probing technique in telephone and
face-to-face interviews, it should not be used
face-to-face interviews, it should not be used inin
written questionnai
written questionnaires because res because it sometimesit sometimes
confuses respondents. An example of
confuses respondents. An example of branchinbranching is:g is:
1. Do you currently have a life insurance policy ? (Yes or No) If 1. Do you currently have a life insurance policy ? (Yes or No) If no, gono, go to question 3
to question 3
2. How much is your annual life insurance premium ? ______2. How much is your annual life insurance premium ? ______
Alternate Option: Alternate Option:
How much did you spend last year for life insurance ? ______How much did you spend last year for life insurance ? ______
. How much did you spend last year for life insurance ? ______
. How much did you spend last year for life insurance ? ______
. How much did you spend last year for life insurance ? ______
. How much did you spend last year for life insurance ? ______
. How much did you spend last year for life insurance ? ______
Does not ask the respondent to order or ranDoes not ask the respondent to order or rank ak a
series of more than five items.
series of more than five items. This becomesThis becomes
increasingly difficult as the number of items
increasingly difficult as the number of items
increases, and the answers become less reliable.
increases, and the answers become less reliable.
This becomes especially problematic when asking
This becomes especially problematic when asking
respondents to assign a percentage to a series of
respondents to assign a percentage to a series of
items. In order to successfully complete this task,
items. In order to successfully complete this task,
the respondent must mentally continue to r
the respondent must mentally continue to re-
e-adjust his answers until they total one hundred
adjust his answers until they total one hundred
percent. Limiting the number of items to
percent. Limiting the number of items to five willfive will
make it easier for the respondent to answer.
Questionnaire Construction
Questionnaire Construction
Keep it Keep it short, simple, and specific (KISSS)short, simple, and specific (KISSS)
Logical question progressionLogical question progression
Flow more important than grammaticalFlow more important than grammatical
correctness
correctness
Key questions first, sensitive Key questions first, sensitive lastlast
Keep allied questions togetherKeep allied questions together
Be Be reasonable reasonable when when requesting requesting forecastsforecasts
Mix Mix fully-sfully-structured tructured and and open-ended open-ended questionsquestions