Indexes, Scales and Typologies

1
Indexes, Scales and Typologies
2
Content validity

• Achieved by including all the dimensions of a
  concept
• Most non-demographic variables require more than
  one measure

3
Composite Methods

• Used pre-dominantly in quantitative studies
• Indexes and Scales allow a summary of several
  measure of the same concept

4
Scales

• Reflect degree of the variable
• Give more information about the concept
• Especially good for measuring attitudes

5
Sample Likert-Scale Questions

• Rate each item on the scale shown to indicate
  your level of agreement
• I believe in the Bill of Rights.
• strongly agree agree uncertain
  disagree strongly disagree
• I think that everyone should vote.
• strongly agree agree uncertain
  disagree strongly disagree
• Most politicians cannot be trusted.
• strongly agree agree uncertain
  disagree strongly disagree

6
Scales

• More difficult to code than indexes
• Coding considerations
• desirable range
• weighting

7
Sample Scale of Job-Related Depression

• I feel down-hearted and blue
• often sometimes rarely
  never
• I get tired for no reason
• often sometimes rarely
  never
• I find myself restless and cant keep still
• often sometimes rarely
  never
• I am more irritable than usual
• often sometimes rarely
  never

8
Sample Scale of Political Involvement

• I think that everyone should vote.
• strongly agree agree uncertain
  disagree strongly disagree
• Most politicians cannot be trusted.
• strongly agree agree uncertain
  disagree strongly disagree
• Joining a political party should be mandatory.
• strongly agree agree uncertain
  disagree strongly disagree
• All votes should be counted.
• strongly agree agree uncertain
  disagree strongly disagree

9
Different Types of Scales

• Bogardus Social Distance Scale
• Thurstone Scales
• Likert Scales
• Semantic Differential
• Guttman Scaling

10
Indexes

• Usually use a number of questions about the same
  thing
• Answers can be scored and added up to produce a
  total score for the concept

11
Items Measuring School Violence and Delinquency

• Question wording
• 1. Have you been suspended? 
• 2. Have you hit another student?
• 3. Have you hit a teacher?
• 4. Have you threatened another student?
• 5. Have you threatened a teacher?
• 6. Have you stolen something from someone?
• 7. Have you carried a weapon?
• 8. Have you used drugs or alcohol?
• 9. Have you sold drugs?
• 10. Have you been caught fighting with another
  student?
• 11. Have you been involved in gang fights?
• 12. Have you been caught fighting with a
  teacher?
• 13. Have you been suspended for fighting?

Robber, M.L. (2000). Scary Schools Specifying
General Strain Theory.
12
Developing a Survey Instrument

• Use clear language and explicit instructions
• Types and number of questions will depend on the
  purpose and the type of survey planned
• Types of items
• Open-ended items
• Multiple-choice items
• Yes / No items
• be mindful of lack of information here

13
Guidelines for Writing Survey Questions

• Items should be clear
• Avoid Double-Barreled Questions
• Respondents Should be Competent
• Questions Should be Relevant
• Short Items are Best
• Avoid Negative Items
• Avoid Biased Items and Terms

14
Sample Multiple-Choice Questions

• How frequently do you take a sick day from work?
• a) never
• b) once or twice a year
• c) 3 to 5 times a year
• d) 6 to 12 times a year
• e) at least once a month

• Identify the issue that you believe is most
  critical to this countrys future.
• a) the economy
• b) education
• c) integrity in government
• d) national defense
• e) some other issue

15
Typologies

• Examine two variables at once
• Example
• Political orientations of newspapers in terms of
  domestic and foreign policy coverage
• Difficult to analyze and are rarely seen