64.4k views
4 votes
When examining statistical data from a situation where the same group of students were given two equivalent forms, Form B and Form D, of a standardized test, you would expect evidence for high test-retest reliability would result in a

a) High correlation between the scores on Form B and Form D
b) Low correlation between the scores on Form B and Form D
c) Negative correlation between the scores on Form B and Form D
d) No correlation between the scores on Form B and Form D

User BenFenner
by
8.1k points

1 Answer

2 votes

Final answer:

High test-retest reliability in a standardized testing scenario would result in a high correlation between the scores on two equivalent forms of the test, Form B and Form D.

Step-by-step explanation:

Examining statistical data from a situation where the same group of students were given two equivalent forms of a standardized test, high test-retest reliability would result in a high correlation between the scores on Form B and Form D. High correlation indicates that if a student did well on Form B, it is likely they also did well on Form D, and vice versa, if the tests are reliable measures of the same construct. Test-retest reliability is used to assess the consistency of a test across time.

In the context of the scenario described in the question, concluding that there is a significant linear relationship between two sets of scores because the correlation coefficient is significantly different from zero supports the suggestion of high reliability between test forms. If the correlation coefficient was low, negative, or nonexistent, it would suggest poor reliability, as scores would vary greatly between test administrations.

User Quantumplate
by
8.4k points