Lecture 3 - Test Development 1: part 1 - Conceptualization (DN)

Question 1

Why are psychological tests designed or revised?

Answer 1

learning objective

Question 2

What is the difference between measurement scales & scaling?

Answer 2

learning objective

Question 3

What types of scaling methods are available to test developers

Answer 3

learning objective

Question 4

How are test items written?

Answer 4

learning objective

Question 5

What is test conceptualization?

Answer 5

conceiving the need for a test

7:55

Question 6

1. What would urge one to develop a new test?
   and
2. Why develop a new test if one already exists?

Answer 6

1. Normally a question in ones mind (e.g., there should be an instrument to measure some psychological construct).
2. Even if a measure of a particular psychological construct does exist, the test may suffer from poor psychometric properties (e.g., low reliability, suspect content validity, poor construct validity)

Question 7

What questions relate to Assumption 1:

Psychological traits and states exist

Answer 7

1. What is the test designed to measure? 2. Is there a need for this test?
2. What content will the test cover?

Question 8

What questions relate to Assumption 2:

Psychological traits and states can be quantified and measured

Answer 8

4. What is the ideal format of the test?

5. What types of responses will be required of test takers?

Question 9

What questions relate to Assumption 3:

Test-related behaviour predicts non-test-related behaviour

Answer 9

6. Should more than one form of the test be developed?

7. How will meaning be attributed to scores on this test?

Question 10

What questions relate to Assumption 4:

Tests have strengths and weaknesses

Answer 10

8. Who benefits from administration of this test?

9. Is there any potential for harm as a result of the administration of the test?

Question 11

What is pilot work?

Answer 11

the preliminary work surrounding the creation of a prototype of a test

Question 12

What does a test developer do during Pilot Work

Answer 12

attempts to determine how to best measure a targeted construct using a number of strategies
17:40

Question 13

What are four preliminary questions one might ask when developing a measure of behaviour-specific emotions?

Answer 13

Q: What is the test designed to measure
A: The emotional experience that is associated with different types of human behaviours

Q: Is there a need for this test?
A: Yes, if (1) measures of the specific emotions typically experienced during different types of specific behaviours are not available and (2) psychologists want to study these emotions

Q: What content will the test cover?
A: The specific emotions that are judged by laypeople (i.e., not emotion scientists) as associated with specific behaviours

Q: What is the ideal format of the test?
A: Likert-type scales are easily interpretable and yield richer information than binary response scales (i.e., yes/no; true/false). Thus, they may be appropriate for our instrument too.

Question 14

What two strategies will we use to determine how we should best

(1) conceptualise and
(2) measure specific emotions associated with specific human behaviours?
38: 10

Answer 14

Literature reviews
E.g., research on basic emotions, particularly positive and negative affect
and moral emotions and

Experimentation
E.g., Asking ~ 830 participants to rate the extent to which 100 different emotions are associated with 2 specific behaviours:
- Not getting enough sleep (lab report focus)
or
- Performing calculations for a statistics assignment (lecture focus)
38:10