Reference no: EM132848768

EDFL 581-100 Principles and Mearsurements in Teaching

1. Which method of estimating reliability provides the most useful information for each of the following? Why?

a. Selecting a scholastic aptitude test for predicting future achievement

b. Reassessing students to determine standing relative to a fixed performance standard

c. Determining whether attest measures a homogeneous characteristic

d. Determining the degree to which essay scores obtained from different raters are affected by differences in the stringency of raters

2. What effect would the following most likely have on reliability?

a. Increasing the number of tasks in an assessment

b. Removing ambiguous tasks

c. Changing from a multiple choice test to an essay test covering the same material

3. What types of reliability data are reported of test manuals for a few standardized achievement tests? Are these types of data valuable in deciding whether to choose tests?

4. In reviewing the reliability data in a test manual, a teacher noted the following reliability coefficients:

a. correlation of Form A test scores over a 1-month interval 50.90

b. correlation of Form A with Form B scores over a 1-month interval 50.85

c. correlation of scores based on two halves (odd-even) of Form A 50.95

How would you account for these differences in reliability coefficients (assume the group tested were the same? Which estimate of reliability provides the most useful information? Why?