44 of 45
Lesson 4:
Outline
|
Notes
|
Glossary
|
Presentation
|
Activities
|
Directed Questions
|
Assessment
Directed Questions for Developing Standardized Tests
Multiple Choice
Attention: ONLINE RESPONDING IS DISABLED
Javascripting is either turned off or is not recognized by your browser.
1.
What is a construct and how can we test for it?
example:
A construct is a theorized phenomenon that cannot be directly observed or measured. To measure it, it must be operationally defined into observable and measurable behaviors.
2.
When determining the purpose of your exam, what are some questions you should ask yourself regarding the exam?
example:
Who is my intended audience? What level of difficulty should the items be to achieve my purpose? What do I want to learn from the exam? Etc...
3.
What is domain sampling and why is it necessary?
example:
Domain sampling is the process of selecting a representative sample set of items from a test´s content domain. This is necessary whenever the domain is too large to be tested completely in a reasonable period of time.
4.
What information should be included in test specifications?
example:
How a test´s items should be constructed, from what content areas, and in what proportions
5.
What is an experimental section and why do tests frequently include an experimental section?
example:
An experimental section of a test is a part of the test that contains new items being field tested for future versions of the test. These are frequently included because it is an efficient way to field test new items.
6.
What are difficulty level, discrimination index, and differential item functioning and how are they used to determine the quality of an item?
example:
Difficulty level, discrimination index, and differential item functioning are all characteristics of an item.
Difficulty level measures how difficult it is for an examinee to correctly respond to an item. The discrimination index refers to how well the item discriminates between low and high ability students. Differential item functioning is used to detect item bias.
7.
What is equating and why is it necessary?
example:
Equating is the process by which raw scores from different tests or different versions of the same test are translated to a new scale.
Equating is necessary because it is common for one version of a test to be slightly easier or more difficult than another version.
8.
What is vertical equating?
example:
Vertical equating allows you to compare performance on a single test across grades or age ranges.
9.
What is horizontal equating?
example:
Horizontal equating allows you to compare performance over time of students in the same grade.
Principles of Measurement
