Lesson 2 Self Test

For each item in this self test, read the item and then click on the circle you wish to select for a correct response. After you have made your selection, feedback will be provided to you. A total score will not be calculated.

1. Which kind of information should we expect to find in a table of specifications for an assessment tool?

a. Names of topics covered b. Reliability of the scores c. Description of the norm group d. Number of items that were identified as biased Feedback

2. In which stage of the test development process is subjectivity likely to influence decisions?

a. In developing test specifications b. In writing items c. In checking for bias and fairness d. All of the above Feedback

3. Which type of information would most likely be obtained from a tryout of items for a constructed-response test?

a. Whether students can write well b. Whether the reading level of the items is appropriate c. Whether there are several acceptable, correct responses to some items d. Whether students prefer constructed-response over multiple-choice items Feedback

4. Which of the following is most likely to negatively affect the “internal consistency” reliability estimates for the scores from an objective test?

a. How much guessing students do b. Who scores the test c. Who gives the test d. How much studying students do Feedback

5. Suppose most boys in a class get a certain test item wrong, but most girls in the class get it right. Which of these is most reasonable to conclude?

a. The item definitely is not biased; girls learned more than boys. b. The item might be biased; it should be examined further. c. The item definitely is biased; boys are at a disadvantage. Feedback

6. Which of these factors is generally most responsible for how much bias might be present in a paper/pencil test?

a. Who administers the test b. Who writes the items c. Who develops the test specifications d. Who obtains the reliability estimates for the scores Feedback

7. Which of these pairs of tests is likely to be most equivalent?

a. Tests 1 and 2 have the same number of items. b. Tests 1 and 2 were reviewed for fairness by the same panel of judges. c. Tests 1 and 2 both contain multiple-choice and constructed-response items. d. Tests 1 and 2 were created using the same table of specifications. Feedback

8. Which set of national norms is probably most useful for score interpretation?

a. One that includes at least 10,000 students per grade. b. One that includes students from every state. c. One that was obtained only five years ago. d. One that includes an equal number of students from each racial/ethnic group. Feedback

9. What kind of information does decision consistency usually convey?

a. How much the test scores might be affected by measurement errors b. How well judges agreed about what should be in the test specifications c. How well reviewers were able to decide about the presence of bias d. How well the test development plan was actually followed Feedback

10. When achievement levels are used for interpreting group scores, which part tells us most about what the students’ performance means?

a. The labels used for the achievement levels (e.g., High, Low) b. The kind of score used to describe the cut points (e.g., percent-correct, percentile rank) c. The verbal descriptions of each achievement level Feedback

11. Which of these best indicates what an achievement level is?

a. A collection of content standards and benchmarks b. A description of a certain range of performance c. A continuum that describes the full range of achievement in a curriculum area Feedback

12. Methods of internal consistency are used to estimate the influence of rater subjectivity in scoring a test.

a. True b. False Feedback

13. If two tests are considered “equivalent,” it is reasonable to conclude that their items are similar in terms of general level of cognitive complexity.

a. True b. False Feedback

14. Norms usually describe how student should score rather than how they actually do score.

a. True b. False Feedback

15. The scores obtained from tests developed by experts often yield reliability coefficients that are greater than 1.

a. True b. False Feedback

End of Lesson 2 Self Test

All contents copyright ©2003.The University of Iowa®. All rights reserved.