Chapter 7

Data Collection: Questions and Questionnaires

The most common format for tests that measure respondent’s attitudes is a Likert scale (named after its developer, Rensis Likert). A Likert scale presents an item and a range of responses such as strongly disagree, disagree, no opinion, agree, and strongly agree. A respondent simply checks the response that is closest to his or her attitude.

Badly Worded Questions

1. Spanking and paddling are effective methods of child discipline
Strongly disagree            Disagree            No opinion                Agree            Strongly agree

2. Sometimes when a child misbehaves, physical punishment should be used.
Strongly disagree            Disagree            No opinion                Agree            Strongly agree

3. Revocation of privileges is an effective form of discipline that should be used judiciously
Strongly disagree            Disagree            No opinion                Agree            Strongly agree

4. Parents should not spank children
Strongly disagree            Disagree            No opinion                Agree            Strongly agree

5. We should have laws that prohibit physical discipline
Strongly disagree            Disagree            No opinion                Agree            Strongly agree

The problem with Item 1 above is that it asks about two methods of physical discipline, spanking and paddling. Such items are referred to as compound items. It is easy to imagine a person who believes that a quick swat on the behind is acceptable, but that using a paddle is not. How should such a person respond? Thus, one rule is that an item should be about only one thing. Hint: the word and is one tip-off that an item is a compound item. The solution to this problem is to write two separate items, one for spanking and one for paddling.


The problem with Item 2 is the word, Sometimes. The word sometimes creates an ambiguity. Imagine a person who thinks that children should always be physically punished when they misbehave. Such a person would respond “strongly disagree.” (They should always be punished.) Another person who does not approve of physical punishment at all would also respond “strongly disagree.” (They should never be punished.) Thus, this item has two people with very different attitudes who respond the same way. The solution is to remove the word sometimes. A good way to detect ambiguities is to play a game with yourself. For each item, adopt a succession of quite different attitudes and write your response. If a strange result occurs (as happens with the sometimes item), you have an ambiguity. An easier way to detect ambiguities is to get friends to read and discuss items with you.


There are several problems with Item 3. To begin with, it is hard to comprehend. It is a long item that has two uncommon words, revocation and judiciously. Another problem is that, like Item 1, it is a compound item. It asks whether revocation of privileges is an effective form of punishment and it also asks if revocation should be used judiciously. A simple shorter form is, “Loss of privileges is an effective form of discipline.” However, another problem is that the item may not measure attitude toward physical discipline. The item is about discipline that is not physical, which may or may not give an indication of attitude toward physical discipline. Thus, Item 3, even in its improved form, lacks content validity.


Item 4 includes a negative, not, which creates a problem. Readers often miss the not in a sentence. (You may have had this experience yourself.) Besides the possibility of missing an important word, respondents who favor physical discipline have to comprehend a double negative to correctly register their opinion. They are saying, I disagree that parents should not spank their children. Double negatives are difficult to understand. Best to rewrite the item to remove the negative. The rewritten item becomes, “It is acceptable for parents to spank their children.”


The problem with Item 5 is that it evokes attitudes about the judicial system as well as attitudes about physical discipline. It may be that such an item can be a reliable measure of attitude toward physical discipline, but additional testing is required.
The rules we have illustrated in this exercise are summarized in Below For more complete instruction in how to write test items, see Babbie (2004). The best overall advice we know of is that of Ira Bernstein (1988). “The most important thing to keep in mind about constructing items is to keep the wording simple” (p. 377). Indeed, the best published tests in psychology are characterized by short and simple items.

"Rules" for composing questions

Example Home Made Questionnaires Used at SAU

  1. Womens Abuse
  2. First Sexual Experience
  3. Photographs and Parents
  4. Identify Males and Females (Class demonstration, not research)
  5. Parental Discipline
  6. Internet Service Providers
  7. Future Plans Questionnaire
  8. Computer-based Teaching
  9. Autism
  10. Children and Feeding
  11. WWW Search Strategies
  12. Normal Sexual Behavior (Published questionnaire used in first class)
  13. GPA and Musical Background
  14. MPA Telephone Questionnaire
  15. Computer Hassles Scale (Published questionnaire)
  16. Generational Differences in Education

Published and Home Made Questionnaires Used at SAU (Selected List)

  1. Adair: BMI and Sexuality
  2. Burkham: Draw-a-Scientist
  3. Eley: Religiousity, Self-Esteem
  4. Garrard: Machiavellian Personality
  5. Hogue: Corporal Punishment and Child Abuse
  6. Henderson: Casual Sex
  7. Legge: Contraception
  8. McLaughlin: Shyness by Media Type
  9. Payne: Male Attitudes Towards Rape
  10. Riddle: Seat Belt Use
  11. Ranson: Self-Esteem and Cosmetics Use
  12. Wan: Self-Esteem by Year in College

Web Pages

Testing and assessment
From APA, this FAQ gives information about finding and using published psychological tests.
http://www.apa.org/science/faq-findtests.html

Buros Institute of Mental Measurements
Site allows users to search for published tests from their Mental Measurements Yearbook by name or by category.
http://buros.unl.edu/buros/jsp/search.jsp

Questionnaire Design
Long page gives tips on how to write and test custom-made questions and tests.
http://www.cc.gatech.edu/classes/cs6751_97_winter/Topics/quest-design/

Writing Effective Tests
Helps users write better test questions.
http://www.glencoe.com/sec/teachingtoday/educationupclose.phtml/40

 


Back to Main RMPA Page