FIELD STUDIES

QUESTIONNAIRES

690A- Advanced Methods in HCI

Prof. Narges Mahyar

Slides from Prof. Joanna McGrenere and Dr. Leila Aflatoony

1
Includes slides from Prof. Karon MacLean and Jessica Dawson
TODAY
• Questionnaires [30 min]

• In class activity [20 min]

• Redesign a questionnaire

• Discussion of readings [20min]

2
LEARNING GOALS
• explain when and why questionnaires may be
appropriate evaluation technique choice; discuss their
pros/cons
• list different styles of questions (open, closed, likert, etc.)
and give examples of what they are appropriate for;
• give examples of data different kinds of questions can
collect
• discuss important considerations for designing and
administering a questionnaire

3
QUESTIONNAIRES
WHEN & WHY?
• evaluating to understand: good for reaching lots of people
early on
• evaluation of prototypes: typically used in combination with
other methods (but not always)

• also called ‘surveys’

• Survey is a complete methodological approach: a process
for gathering data that could involve a wide variety of data
collection methods, including a questionnaire (list of
questions).

4
QUESTIONNAIRES
WHAT CAN YOU DO WITH THEM?
• closed or open questions
• evidence of wide general opinion

pros/cons:
• can reach a wide subject group (e.g. mail or email)
• does not require presence of evaluator
• many results can be quantified
• can have low response rate and/or low quality response

5
WHAT KINDS OF DATA CAN YOU
COLLECT?
questionnaires can gather both:
• subjective AND objective data
• qualitative AND quantitative data

6
STYLES OF QUESTIONS:
OPEN-ENDED
• asks for opinions
• good for general subjective information
• but difficult to analyze rigorously

for example, “Can you suggest any improvements to the

interface?”

7
STYLES OF QUESTIONS:
CLOSED
• restricts responses by supplying the choices for answers
• can be easily analyzed …
• but can still be hard to interpret, if questions / responses not
well designed!
• options should be very specific

Do you use computers at work:

O often O sometimes O rarely

In your typical work day, do you use computers:

O over 4 hrs a day
O between 2 and 4 hrs daily
O between 1 and 2 hrs daily
O less than 1 hr a day

8
STYLES OF QUESTIONS:
COMBINING OPEN-ENDED & CLOSED
QUESTIONS
• gets specific response, but allows room for user’s opinion

It is easy to recover from mistakes:

disagree agree
1 2 3 4 5

comment:…
…the undo facility is great!...

9
STYLES OF QUESTIONS (CLOSED):
SCALAR --- LIKERT SCALE
• measure opinions, attitudes, and beliefs
• ask user to judge a specific statement on a numeric scale
• scale usually corresponds to agreement or disagreement with
a statement
• odd or even numbered (what’s the difference?)

Characters on the computer screen are hard to read:

strongly agree strongly disagree
1 2 3 4 5

10
STYLES OF QUESTIONS (CLOSED):
SCALAR --- SEMANTIC
DIFFERENTIAL SCALE
• similar to likert scales - also measure opinions, attitudes, beliefs
• but explore a range of bipolar attitudes about a particular item
• each pair of attitudes is represented as a pair of adjectives
à generally easier cognitively to answer than likert

Moodle is:

poorly 1 2 3 4 5 well designed

clear 1 2 3 4 5 confusing
attractive 1 2 3 4 5 ugly

11
STYLES OF QUESTIONS (CLOSED):
RANKED
• respondent places an ordering on items in a list
• useful to indicate a user’s preferences
• forced choice

Rank the usefulness of these methods of issuing a command

(1 most useful, 2 next most useful..., 0 if not used)
__2__ command line c
__1__ menu selection
__3__ control key accelerator

12
STYLES OF QUESTIONS (CLOSED):
MULTI-CHOICE
• respondent offered a choice of explicit responses

How do you most often get help with the system? (tick one)
O on-line manual
O paper manual
O ask a colleague

Which types of software have you used? (tick all that apply)
O word processor
O data base
O spreadsheet
O compiler

13
DESIGNING A QUESTIONNAIRE
establish the purpose of the questionnaire:
• what information is sought?
• how would you analyze the results?
• what would you do with your analysis?

determine the audience you want to reach

• typical when using questionnaire for understanding: random
sample of between 50 and 1000 users of the product

test everything before sending it out:

• test the wording
• test the timing
• test the validity
• test the analysis

14
DESIGNING GOOD QUESTIONS
unlike interviews, hard to ask a follow-up questions
² extra important to get questions right

a few general guidelines:

• be specific and clear about how users should answer
• keep questions short and easy to follow
• avoid ‘double-’ and ‘triple-barreled’ questions
• e.g., how often have you used the system and what do
you like about it?
• avoid ambiguity and too much room for interpretation
• avoid biasing responses as much as possible

15
COMMUNITYCRIT: INVITING THE PUBLIC TO
IMPROVE AND EVALUATE URBAN DESIGN IDEAS
THROUGH MICRO-ACTIVITIES

Narges Mahyar, Michael R. James, Michelle M. Ng, Reginald A. Wu, Steven

P. Dow, ACM Human Factors in Computing Systems (CHI 2018).

16
1) PICK AN IDEA 2) DO OR SKIP ACTIVITIES

17
3) SUBMIT A NEW IDEA 4) VIEW CONTRIBUTIONS

18
VALIDITY
are your questions getting at what you want?
can increase validity by. . .
• piloting (see how people answer)
• triangulation (target hypotheses with multiple questions)
• use previously validated questionnaires (studied extensively to
confirm they gather what they intend to gather)

19
TRADEOFFS
questionnaires are limited by length and complexity
• can’t always ask about everything you want to

try to focus questions on what you really want to learn

• a few focused questions more useful than many general ones.
• if the answer is obvious, you probably don’t need to ask it!

but be careful of focusing too much on what you expect to the

exclusion of other explanations

20
ADMINISTERING QUESTIONNAIRES
in-person • requires time to administer, but highest
administration completion rate

“take home” • often subjects don’t complete / return the

(conventional) questionnaire

• permits subjects to answer on their own time

• responses may tend to be more free-form
email • attachments may be a problem
• response rates depend on trust in source

web-based • standardize formats and responses

forms • Java/Javascript to ensure correct / complete
• payment or incentives
general issues • anonymity
• self-selection

21
IN-CLASS DISCUSSION:
HOW SHOULD THEY DIFFER GIVEN CONTEXT?

to successfully deploy a questionnaire: account for the

context and the nature of the questions you’re asking.

what are the important considerations?

E.g., in a hospital setting where…

1. You want to ask patients who just tried a new insulin
pump what they thought of it
Versus…
2. You want to widely distributed a survey about general
health concerns to people between ages of 20 - 55

22
BE CONSIDERATE OF YOUR
RESPONDENTS
AND THE CONTEXT YOU ACCESS THEM IN
• questionnaire length (short is good)
• think in terms of reasonable completion times
• do not ask questions whose answers you will not use!
• privacy invasions/anonymity
• be careful how / what you ask
• motivation
• why should the respondent bother?
• usually need to offer something in return
• ability
• limitations like literacy and disability can come into play

23
ACTIVITY [20 MIN]
questionnaire critique and redesign
• Work in groups of 2-3

24
DISCUSSION ON QUESTIONNAIRE
READINGS [20 MIN]
Get into group of 3-4 answering the following
questions:
• What surprised you? or
• What you disagreed with?
• Others?

25
ON DECK…
Next class (Thursday) …
1. Readings (as posted) and researcher journals
2. First Interim Project milestone
² due on Monday Feb 11

26
EXTRA SLIDES

27
SUMMARY:
QUESTIONNAIRES
1. establish purpose
2. determine audience
3. variety of administration methods
(for different audiences)
4. design questions:
• many kinds, depend on what you want to learn
• most important distinction: open/closed (like
structured/unstructured interview questions)
5. be considerate of your respondents
6. motivate your respondents (without biasing them).

28
QUESTIONNAIRE EXAMPLE
**Citation of Research Paper that uses the Example Survey:**

Rock Leung, Charlotte Tang, Shathel Haddad, Joanna Mcgrenere,

Peter Graf, and Vilia Ingriany. 2012. How Older Adults Learn to Use
Mobile Devices: Survey and Field Investigations. ACM Trans.
Access. Comput. 4, 3, Article 11 (December 2012), 33 pages.
DOI=http://dx.doi.org/10.1145/2399193.2399195

29