Ch 8.1 Test Development Flashcards

1
Q

Test Conceptualization

A

Coming up with an idea for a test

Why is a test needed?

  • Existing measures are not good enough
  • There are no existing measures (e.g. new medical disease)
  • Literature review
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Norm - referenced

A
  • Distinguish between high & low scorers
  • Relative standing
  • EX: WAIS-IV
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Criterion - referenced

A
  • Distinguish between mastery and non - mastery
  • Pass/fail
  • Ex: Driver’s license exam
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Writing Items

A

Need to develop item pool

  • Reservoir from which items will be drawn or discarded for the final version of the test
  • Twice as many as the final version
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Where do items come from?

A
  • Literature reviews
  • Expert opinion
  • Personal experience, knowledge
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Question Interpretation

A

Matching what you think you are asking wo what participants think you are asking

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Considerations for writing questions

A

-Be careful about eliciting a certain type of response

  • Avoid leading questions
  • Double barreled questions
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Rules for writing questions

A
  • Follow the rules of communication
  • Provide sufficient instructions
  • Allow for variability

– Make measurement scale meaningful

– Floor and ceiling effects

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Survey Formats [placeholder]

A

.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Forced Choice

A

Questions require respondents to select from a limited number of answers

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Categorical Responses

A
  • Dichotomous questions
  • Nominal response format
  • Ordinal responses using rank order
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Interval level response

A
  • Likert scale
  • Semantic differential
  • Response brackets
  • Guttman scale or cumulative
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Guttman scale or Cumulative

A
  • Items range sequentially from weaker to stronger expressions of the construct being measured
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Open-ended (free or constructed response)

A
  • Fill in the blank
  • Short answer
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Filter or contingency questions

A
  • Do not ask questions that cannot be answered by all participants
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Scoring and Scaling

A

Scoring

  • One vs multiple raters
  • Partial credit vs correct/incorrect
  • Are instructions or training required?

Unit Weighting

Or Weighting by:

  • item difficulty
  • effort or time spent
  • reliability (importance)
17
Q

Test Tryout [placeholder]

A

.

18
Q

Piloting

A

Pilot - (work, study, research)

  • Preliminary test surrounding the creation of a prototype of the test
  • Should items be included in the final version?
  • Determine the best measurement methodology
19
Q

Test Tryout

A
  • Tryout with people similar to those for whom test was designed
  • 5 - 10 people for each item
  • Should items be kept, revised, or discarded?
  • Use standardized conditions