Surveys and Scales

Question 1

Constructs and operational definitions:

Answer 1

Measurement of an abstract concept (construct) requires procedures for translating it into concrete representations (variables) that can be directly measured according to specified methods (operational definitions). All theoretical constructs have multiple qualities or dimensions. As a consequence, no single operational definition can yield a measure that entirely captures the construct that it represents. Furthermore, any single operational definition produces an inaccurate and imperfect measure of the construct. Yet operational definitions are necessary because they provide the only means for obtaining scientific measures of a construct. The recommended solution for best measuring a theoretical construct is to employ multiple operational definitions.

Question 2

Using theory to generate operational definitions:

Answer 2

The first step in generating an operational definition is to define the relevant construct at an abstract level in terms of other theoretically-related constructs, a definition known as the construct’s nomological net. The information that you use for defining a construct by other inter-related constructs is drawn from one or more theories that involve the relevant construct.

Note that a particular nomological net is shaped by its sources and thus it represents just one way to define the construct. After constructing the nomological net, you then use the related constructs to generate operational definitions. Think carefully about how your selected measures are related to the construct and draw upon your common sense knowledge to determine whether or not the measures have simple linear relationships with the construct.

Refer to Judd, Smith and Kidder (1991) for examples of a nomological net and of nonlinear relationships between variables and their related constructs.

Methods other than a nomological net can also be used to produce multiple operational definitions for a theoretical construct.

Four of the more useful techniques follow:

1. Interviews and theme analysis: Question other people about what they understand is meant by the construct. Identify the most frequent categories that occur in the answers and treat these as the construct’s major components. Finally, generate operational definitions from these components.
2. Research review: Extract and use the operational definitions that have been used by other research into the relevant construct.
3. Direct observation: Directly observe the situations or circumstances in which the construct occurs and identify relevant behaviours that can be used as operational definitions.
4. Expert judgment: Obtain operational definitions from experts who have direct experience with the construct.

Question 3

Constructing self-report measures:

Answer 3

Self-report measures are frequently used in correlational research. In this subsection we review some of the advantages and disadvantages of this technique. In addition, you will learn about some of the practical issues that researchers need to grapple with in developing good self-report measures.

Advantages:

• Direct
• easy, cheap to administer

Disadvantages:

• Distortion of responses, people may lie

Question 4

Type of Self-Report

Answer 4

• Open-ended questions
  
  • (more qualitative, can be inflated)
• Restricted questions
  
  • Constricted, easier to manage data
• Rating Scales
  
  • also restricted, Likert and such. keeps it quantitative
  • may lose richness of information captured

Question 5

Types of Validity

Answer 5

Question 6

Administering surveys and scales:

Answer 6

Researchers also need to determine the best way to administer their scales. This can include via an interview, mail, telephone or the internet. In this subsection, we review each of these main methods available to researchers for administering surveys and scales. The advantages and disadvantages associated with each method will also be addressed.

Methods

* Mail
* Telephone
* In-person
* Internet

Perks

 Expense
 Convenience
 Response rate
 Nonresponse bias: only those most interested in the survey topic will take part so will not be a representative sample
 Who completed it?
 Interviewer bias: leading questions, expectations

Question 7

Mail surveys

Answer 7

Advantages

• Convenient and nonthreatening for participants
• Easy to administer
• No time restrictions

Disadvantages

• Expensive
• Low response rate * and nonresponse bias
• Can’t clarify questions/answers