1. +
How to Ask Survey
Questions
Workshop on Survey Design
2. +
Learning Objectives
After this class you will be able to:
1. Construct questions that are clear to your
intended audience
2. Construct questions requiring ordered
responses or scales
3. Choose among question types
4. Check that questions will provide data that
meets your goal
3. +
Exercise 1: Survey Analysis
Look over and take the Green Attitudes survey
What are your reactions to the survey?
What did you like about the survey? Dislike?
What do you think about the content and length of the survey?
4. +
Overview of Previous Workshop
Concepts
Checklist of Survey Tasks
Overview of the survey process
Guidelines for Asking Questions
Best practices for question wording
5. +
Review:
Guidelines for Asking Questions
Ask purposeful, concrete questions that use conventional
language
Avoid bias in words and phrasing
Avoid two-edged questions and negative questions
Use shorter questions
Choosing between Open and Closed Questions
6. +
Review: Selecting Open and Closed
Ended Questions
Copy over
Use Open Use Closed
Purpose Respondents’ own words
essential
Want data that are rated or
ranked and can order the ratings
in advance
Respondents Are capable and willing to
provide answers in their
own words
Are unwilling or unable to
express themselves while being
surveyed
Questions Prefer open because
choices are unknown
Using a pre-specified set of
response choices
Analysis Have ability to analyze
variety of comments
Prefer to count the number of
choices
Reporting Provide individual or
grouped verbal
responses
Reports statistical data
7. +
Types of Closed Questions
Categorical: The number next to each response has no
meaning except as a placeholder for that response.
Multiple Choice: Select one or more answers.
Ordinal: Assigns a meaningful number to responses, puts
things in order.
Likert Scale: To assess a person's feelings about something.
Ranking: Rank things in relation to other things.
Numerical: For real numbers, like age, number of months.
8. +
Advantages of Closed-Ended
Questions
Easy to code, enter, and analyze
Easy to present
Quick turnaround
Enhanced reliability
Less researcher bias
High degree of anonymity
9. +
Disadvantages of Closed Ended
Questions
Harder to develop questions and response categories
May force invalid responses
Less depth and substance
Respondents unable to explain, qualify, or clarify answer
During the past month, have you felt depressed?
0 = No
1 = Yes, once in a while
2 = Yes, some of the time
3 = Yes, most of the time
4 = Yes, all of the time
11. +
Categorical Questions, ex. 1-2
Gender
Male
Female
Transgender
Color of eyes
Blue
Black
Brown
Hazel
Ask respondents to tell which
category they fit into
Response choices should be
mutually exclusive
Response choices should be
inclusive and exhaustive
Use groups that make sense in
the survey and will be useful in
reporting results
Can allow respondents to
choose more than one answer
12. +
Categorical Questions, ex. 3
Ask respondents to tell which
category they fit into
Response choices should be
mutually exclusive
Response choices should be
inclusive and exhaustive
Use groups that make sense in
the survey and will be useful in
reporting results
Can allow respondents to
choose more than one answer
4. My child/children attend(s)
Public school
Non-local public school
Charter school
Private school
Does not attend school yet
Graduated
Other (please specify)
__________________
13. +
Exercise 2: Improve Poor Categorical
Questions
1. Which one of the following best describes your primary
expertise?
Landlord-tenant problems
Consumer problems
Traffic cases
Other (specify)
2. Which of the following best describes you?
Professional
Registered nurse
Nurse practitioner
Administrator
Nurse midwife Professional
RN
Practitioner
14. +
Ordinal Questions, ex. 1-2
How satisfied are you with how
long you can check out books?
1. Very dissatisfied
2. Dissatisfied
3. Satisfied
4. Very Satisfied
What is the highest level of
education that you have
completed?
1. Elementary school
2. High school graduate
3. Some college
4. College graduate
Use a meaningful scale
Balance all responses
Use don’t know or neutral
carefully
Use 4- to 7-point rating scales,
with negative end first
Keep uncluttered and easy to
complete
15. +
Ordinal Questions: Likert Scales
Considering your reading habits, during the past year how often
did you read the following newspapers, journals, and
magazines? Circle one number for each periodical.
Periodical Never
(1)
Rarely
(2)
Sometimes
(3)
Frequently
(4)
Always
(5)
New York
Times
1 2 3 4 5
Wall Street
Journal
1 2 3 4 5
Cosmopolitan 1 2 3 4 5
New England
Journal of
Medicine
1 2 3 4 5
Sports
Illustrated
1 2 3 4 5
16. +
Ordinal Questions: Ranking ex. 1
Rank items against each other
Each item is assigned a number
Don’t make list too long
Can be very confusing, use sparingly
Based upon what you have seen, heard,
and experienced, please rank the following
brands according to their reliability. Place
a "1" next to the brand that is most reliable,
a "2" next to the brand that is next most
reliable, and so on. Remember, no two
cars can have the same ranking .
__ Honda
__ Toyota
__ Mazda
__ Ford
17. +
Ordinal Questions: Ranking ex. 2
Rank items against each other
Each item is assigned a number
Don’t make list too long
Can be very confusing, use sparingly
18. +
Exercise 3: Improve Poor Ordinal
Questions
1. How satisfied were you with your hotel stay?
Completely Satisfied, Very Satisfied, Pretty Satisfied,
Satisfied, A Little Satisfied, Slightly Satisfied,
Unsatisfied, Don’t Know
2. Please rate your instructor on the following criteria
19. +
Numerical Questions
Numerical: For real numbers, like age, number of
months.
1. How old are you?
2. In what year were you born?
3. what is your annual income?
20. +
Types of Closed Questions
Categorical: The number next to each response has no
meaning except as a placeholder for that response.
Multiple Choice: Select one or more answers.
Ordinal: Assigns a meaningful number to responses, puts
things in order.
Likert Scale: To assess a person's feelings about something.
Ranking: Rank things in relation to other things.
Numerical: For real numbers, like age, number of months.
21. +
Survey Goals Revisited
What question or questions are you trying to answer?
What information do you need to decide upon an answer?
What actions will you take when you have the survey’s results?
Start or stop doing something?
Make a purchase?
Select content for a class?
Who is your audience?
22. +
Every Question Should Have a
Purpose
Keep your survey as short as possible
Every question should provide data you will use to make a
decision
Don’t automatically include questions you’ve seen before
Don’t just use questions that are about your topic without thinking
about the resulting actions possible from the data
Be sure that every answer will cause you to do something
differently
Don’t ask questions just to satisfy curiosity
23. +
Exercise 4: Survey Audience and
Goals
Look over your group’s survey
Green Attitudes : 5-8, 10, 29
Great Streets : 20, 26-28, 35
Vogue Knitting: 5-6, 9, 12, 28, 29, 47
What were the goals of the survey creator?
Who is the audience of the survey?
Look carefully at each question
Will the data gathered meet the goals of the survey?
Will the answers to any individual question make a choice clearer?
24. +
Kinds of Information
Copy over Description Example
Knowledge What people know; how
well they understand
something
True or False:The most effective
weight loss plan includes
exercise.
Belief What people think is true;
an opinion
Do you think that lower food
prices would increase
consumption?
Attitude How people feel about
something; a preference
Do you favor or oppose a
smoking ban in your county?
Behavior What people do; can be
mental or physical
Have you ever attended a
Library workshop?
Attributes What people are; what
people have
Demographics: What year do
you plan to graduate?
25. +
When to Use Categorical Questions
Can be used to find all types of information,
but answer choices are limited
To ask many questions in a short time period
When there are a finite number of options
Be careful in analysis
Cannot compare one item to another
Can create a frequency distribution
If only one answer allowed, you don’t have data
about the other answers
26. +
When to Use Ordinal Questions:
Ranking
Best used to assess attitudes and
knowledge
To rate things in relation to other
things
When there are a finite number of options
Be careful in analysis
There is not a meaningful interval between choices
Can create a frequency distribution or cumulative frequency
distribution
Cannot average, add, subtract data
27. +
When to Use Ordinal Questions:
Likert-Scale
Best used to assess beliefs and attitudes
Sample scales
Very dissatisfied, somewhat dissatisfied, somewhat satisfied, very
satisfied
Strongly disagree, disagree somewhat, uncertain, agree somewhat,
strongly agree
Be careful in analysis
Be sure your intervals are meaningful and equal
Can average, add, subtract, and create frequency distributions
28. +
When to Use Numerical Questions
Best used for attributes (demographics)
For precise real numbers, not ranges
Can analyze using most methods
Multiply, divide, average, add, subtract, and frequency distributions
If your survey tool doesn’t support this data type, you can
approximate with an open-ended question
You will have to manually analyze data
29. +
Exercise 5: Matching Question
Types to Goals
1. Name one action you might you take with each goal’s results.
2. What type of data do you need to meet each goal?
3. What question type will help you get that data?
4. Write a sample question for each goal.
Goal 1: Determine if patrons know what services we offer at the
Reference Desk.
Goal 2: Determine if people are satisfied with the wait times at
the Circulation Desk.
Goals 3: Determine if the respondents to your survey are
representative of the whole campus.
30. +
Don’t Forget…
Ask enough demographic questions to ensure that your
intended audience is represented
Ask as few questions as possible
Don’t ask too many open-ended questions
Will lower response rate
One is usually enough
Don’t have too many Likert-scale questions in a large block
31. +
Learning Objectives Revisited
After this class you will be able to:
1. Construct questions that are clear to your intended audience
2. Construct questions requiring ordered responses or scales
3. Choose among question types
4. Check that questions will provide data that meets your goal
How many of you participated in the first survey workshop?
The previous workshop focused on the process from conception to delivery. Today’s workshop focuses on the design of the survey’s questions.
Please keep in mind your own experiences with taking and designing surveys.
RV
RV
5 minutes to take
5 minutes to discuss
Did you feel your opinions, attitudes, behavior were captured in these questions?
How was the balance of types of questions, open, closed, rankings?
RV
(open Checklist on LibWorks)
Ask participants to recall some tasks and guidelines. Task to highlight: considering stakeholders and authority to act.
Success depends on preparation.
RV
A concrete question is precise and unambiguous. Use time periods that are related to the importance of the question.
Periods of a year or more can be used for major life events while periods of a month or less should be used for questions that are less important. Asking people to remember relatively unimportant events over long periods of time leads to too much guessing.
A survey is not a conversation. To get accurate information, survey questions rely on standard grammar, punctuation, and spelling. You should use words that maximize understanding for everyone in the survey. Use complete sentences, avoid abbreviations, avoid slang and colloquial expressions, and be careful of jargon and technical expressions.
Negatives and double-negatives can be confusing.
Purpose of loading is to encourage respondents to give a “true” response rather than one that is socially acceptable. Use loaded questions, if necessary, but be cautious as respondents may see through them, get annoyed, answer inaccurately or not at all.
Avoid ‘two-edged’ questions which contain two ideas—look for “and” in question.
Can you think of some examples of not concrete vs concrete, bias vs non-bias, etc.
What do you think of the class and the instructor? (two-edged)
What is your opinion of the library’s Access Services (jargon)
Do you support the govt taking away personal choice by imposing govt mandated universal healthcare? (bias)
RV
Use open ended questions when you want to have a range of responses, use close ended when you want to narrow down a range. Consider the purpose behind posing the question, the capabilities and willingness of the respondents, the level of analysis you will be able to apply to the results, and how you will report out the results. Again, preparation of the questions is very important in order to get the data you are looking for.
RV
Likert Scale : Often used to assess a person's feelings about something.
Multiple Choice : When there are a finite number of options, can allow respondents to choose more than one.
RV
As a respondent, you may feel frustrated with the options you are given.
cs
Rina has outlined the advantages and disadvantages of closed end-ed questions. Understanding the pro’s and con’s of your question type will help you understand the what the data you collect can and cannot tell you. So first I’m going to look more in-depth about these question types and then Sarah will continue with choosing the type. Let’s take a closer look at these three types of closed ended questions.
CS
Categorical Questions—you may also hear them referred to as Nominal QuestionsQuestions for which the responses are assigned a number with no meaning. The number is used as a placeholder for that response.
-Can add an other, free response category
-Includes True/False and Yes/No questions
A categorical variable (sometimes called a nominal variable) is one that has two or more categories, but there is no intrinsic ordering to the categories. For example, gender is a categorical variable having two categories (male and female) and there is no intrinsic ordering to the categories. Hair color is also a categorical variable having a number of categories (blonde, brown, brunette, red, etc.) and again, there is no agreed way to order these from highest to lowest. A purely categorical variable is one that simply allows you to assign categories but you cannot clearly order the variables. If the variable has a clear ordering, then that variable would be an ordinal variable, as described below.
CS
Categorical Questions—you may also hear them referred to as Nominal QuestionsQuestions for which the responses are assigned a number with no meaning. The number is used as a placeholder for that response.
-Can add an other, free response category
-Includes True/False and Yes/No questions
A categorical variable (sometimes called a nominal variable) is one that has two or more categories, but there is no intrinsic ordering to the categories. For example, gender is a categorical variable having two categories (male and female) and there is no intrinsic ordering to the categories. Hair color is also a categorical variable having a number of categories (blonde, brown, brunette, red, etc.) and again, there is no agreed way to order these from highest to lowest. A purely categorical variable is one that simply allows you to assign categories but you cannot clearly order the variables. If the variable has a clear ordering, then that variable would be an ordinal variable, as described below.
CS
Categories are not exhaustive. Survey likely to produce more responses in the other category than in the other three. Improve by adding more categories.
Categories for the responses are not mutually exclusive. The choice professional can include all of the remaining categories, and registered nurse can include several of the other categories. To improve change to Which of the following apply to you? Answer yes or no for each category.
CS
Nominal QuestionsQuestions for which the responses are assigned a number with no meaning. The number is used as a placeholder for that response.
-Can add an other, free response category-Includes True/False and Yes/No questionsAn ordinal variable is similar to a categorical variable. The difference between the two is that there is a clear ordering of the variables. For example, suppose you have a variable, economic status, with three categories (low, medium and high). In addition to being able to classify people into these three categories, you can order the categories as low, medium and high. Now consider a variable like educational experience (with values such as elementary school graduate, high school graduate, some college and college graduate). These also can be ordered as elementary school, high school, some college, and college graduate. Even though we can order these from lowest to highest, the spacing between the values may not be the same across the levels of the variables. Say we assign scores 1, 2, 3 and 4 to these four levels of educational experience and we compare the difference in education between categories one and two with the difference in educational experience between categories two and three, or the difference between categories three and four. The difference between categories one and two (elementary and high school) is probably much bigger than the difference between categories two and three (high school and some college). In this example, we can order the people in level of educational experience but the size of the difference between categories is inconsistent (because the spacing between categories one and two is bigger than categories two and three). If these categories were equally spaced, then the variable would be an interval variable, like a Likert scale.
CS
Examples include frequency and intensity Likert scales
When crafting a Likert scale, try to keep the values of each interval roughly the same. Likert scale questions are similar to other ordinal questions, except that the intervals between the answers are equally spaced. For example, suppose you have a variable such as annual income that is measured in dollars, and we have three people who make $10,000, $15,000 and $20,000. The second person makes $5,000 more than the first person and $5,000 less than the third person, and the size of these intervals is the same. If there were two other people who make $90,000 and $95,000, the size of that interval between these two people is also the same ($5,000).
CS
Looks a lot like a likert scale question, but it can be confusing to the user. Because only one ranking can be selected per category of ranking. Ex. Only one of the responses can be a 5, one a 4 and one a 3…etc.
Difference between answers not a meaningful interval
CS
Looks a lot like a likert scale question, but it can be confusing to the user. Because only one ranking can be selected per category of ranking. Ex. Only one of the responses can be a 5, one a 4 and one a 3…etc.
Difference between answers not a meaningful interval
CS
cgs
Likert Scale : Often used to assess a person's feelings about something.
Multiple Choice : When there are a finite number of options, can allow respondents to choose more than one.
cgs
Likert Scale : Often used to assess a person's feelings about something.
Multiple Choice : When there are a finite number of options, can allow respondents to choose more than one.
SB
As mentioned in previous class, there is a reason or goal for each survey. It could be a large question or concept. For example, What services do graduate students expect from their library? Once you have the survey goal or question, you need to think about the actions that will result from the survey. Will you stop doing something? Start doing something? Purchase something? Include different content in a class?
Work backwards from goals to data to questions
SB
SB
Multiple surveys, 3-5 (political, library), different groups do the work
Indentify some questions on each survey for the group to pay review for the questions in the exercise above
Green Attitudes: 5-8, 10, 29
Vogue Knitting: #5-6, 9, 12, 28, 29, 47
Great Streets: #20, 26-28, 35
SB
SB
Turn into handout!!!
SB
SB
Be sure to use a meaningful scale, not-biased
SB
Not too many in a big block
Odd number provides neutral, even does not
SB
SB
Each goal will require more than one question to be fully answered. That’s ok. Please select one to work on.
SB
Ask class for things they hate about surveys we can avoid?