Questionnaires and Surveys: A quick and dirty overview
Purposes: To rapidly determine how a relatively large group of people respond to
predetermined issues and focused questions. Unlike interviews which allow the
researcher to probe a relatively small number of respondents in depth, surveys are less
flexible and lack the ability to ask the follow up question. Surveys are often a second
step after pilot interviews or open-ended surveys have been completed.
Content: may be on issues of fact or as a way to assess attitudes or opinions.
Surveys about factual issues are easier to construct.
Opinions are the expressions of deeper, guiding attitudes.
Attitudes are convictions, fears, inclinations.
Attitudes are expressed when “triggered” by some object or event. (e.g., response
to the Katrina fiasco).
It’s relatively easy to assess opinions by using a simple questions and a rating
Attitudes require sets of 5-20 related questions with which respondent has to
agree or disagree.
Survey Formats and getting people to respond: People need to see it as worthwhile –
altruistic or serving some need of their own.
Surveys may be
2. distributed electronically via email or forum,
3. given out in classes, or
4. conducted verbally (an interview) either in person or by telephone..
Writing clear, concise questions that relate to the issues of interest is the most difficult
part of using survey methods.
open ended. No fixed answer.
Open ended questions are very difficult to evaluate and should be used sparingly once the
pilot stage has passed and categories of answers have been identified.
Close-ended: Have a fixed set of answers to respond to. May be
factual or attitudinal.
These are quick and easy to evaluate. However, providing answer categories can
introduce bias by giving people responses they may not have thought of, or, by not
including a person’s choice the person can become negatively influenced toward the
Close-Ended Questions: Response Formats
1. Fixed choice
e.g., three ways to structure responses to closed ended questions. Which do you prefer?
How many time per week Circle the best Check the best Circle the
do you answer answer number of the
meet each science class on
2 or 3 ___ 2 or 3 1. 2 or 3
4 ___ 4 2. 4
5 ___ 5 3. 5
other ___ other 4. other
2. Rating scales
These are used to measure intensity of response on some set of ordered
Strongly disagree disagree agree
strongly agree no opinion
I like ice cream __ __ __
Joe is doing a good job 1 2 3
Too little About right
The amount of reading is I--------------------------------------I--------------------
3. Ranking schemes
Respondent must prioritize a set of options. Important that directions be
clear to eliminate multiple scores of any number.
e.g. Please tell what is most important to you in a teacher. Rank the most
important item “1,” the next most important item “2,” and so on. Do not
use a ranking more than once. Use only whole numbers.
_____ Being prepared
_____ Being fair
_____ Knowing how to motivate students
_____ Knowing subject matter
_____ Good classroom management
Criteria for writing close-ended questions.
Response categories must be:
clearly defined in terms the respondent is likely to understand
1. Not exhaustive
Sciences courses taken Chemistry _____
2. Not mutually exclusive
Number of different preps per day _____ 1-3
3. Not clearly understood
How many science classrooms are in your school? ___ none
4. Violates all three:
What kind of science furnishings are in science classrooms at your
_____ lab benches
_____ storage cabinets
Sequencing the questions:
Use the inverted funnel model.
Start with narrow questions, followed by broader ones. Go from facts to generalizations,
from specifics to abstract ideas.
The first questions:
1 Are the ones most often responded to favorably. They also often shape the
respondents attitude to the survey and subsequent responses.
2. Should put the responder at ease because they are easy to answer and don’t
touch on sensitive issues.
3. Are usually closed ended
1. What grade were you in when you first encountered science in school?
2. What were the science lessons like?
a. Did you use a text?
b. Did you have a lab?
c. Did you do investigations?
d. Did you write papers?
3. In general, how effective do you think your early science experiences were?
4. In your opinion, what would a very good first science experience be like?
For every question or group of questions with similar response types, you need to write a
clear set of instructions.
They can be simple. “Circle the one choice that best reflects your opinion”
Or, they can be complex. See the rank ordering example above.
If your questionnaire is long, you may need to introduce different sections of the
questionnaire with separate instructions.
Telling the Purpose
If you are going to need parental permission or are going to mail the survey, you need a
cover letter telling the purpose of the survey. Otherwise, the survey needs a little
introduction at the beginning that helps to make the likelihood of people responding
higher. These guidelines are similar to information to be given before conducting
1. Identify who is doing the survey
2. Tell why the survey is being done. There may be a conflict between research
needs and ethics here.
3. Tell why it is important that they respond to the survey (how they benefit)
4. Confidentiality and how it will be maintained (if appropriate).
1. Question placement on the list may bias the responses.
2. The wording may bias a person’s answers.
a. If they don’t understand the vocabulary
b. If words open to too much interpretation are used, like “liberal” or “republican”
3. Response set bias. Having all of a set of opinion questions be rated in the same direction.
This is especially a problem in matrix formatted questions.
4. Leading questions.
a. The responder can tell the way they are supposed to answer/
E.g., Don’t you think the best teachers ……?
e.g., You wouldn’t say this text was too bad, would you?
e.g., Most people say that…., what do you think?
b. Questions that use emotionally loaded words can bias the way people respond.
e.g., issues like starvation,
Food should be sent to starving people in Bosnia SA A
D SD N
5. Threatening questions: These touch on anxiety arousing topics like sex, drugs,
religion, gambling. People tend to underreport their activities. Can increase by:
Using a long explanatory introduction.
instead of closed questions
6. Double-barreled questions: These are two in one.
e.g. Would you say most science is inquiry based and fun?
If both elements are important, then sort out the questions and ask them separately.
Or, you could have people rate them.
Which of the two elements of a science program is most important?
1. that it be inquiry based
2. That it be fun
David Hopkins, A Teacher’s Guide to Classroom Research, Philadelphia: Open
University Press, 1985