Making Classroom Assessments Reliable and Valid by Marzano Robert J.;

Author:Marzano, Robert J.;
Language: eng
Format: epub
Publisher: Solution Tree
Published: 2017-08-15T00:00:00+00:00

As described in table I.1 in the introduction, the column Initial Administration represents the scores of ten students on a test. The column Second Administration (A) represents an administration of the test soon after the first administration under the condition that students forgot about the first administration. As described, the pattern of similarities between the pairs of scores for students indicates a high reliability, but we now have a correlation between scores on these two administrations that quantifies the degree of relationship. That correlation is 0.96 and represents the reliability coefficient under the traditional definition of reliability. As described in the introduction, the pattern of dissimilarities between the pairs of scores in Second Administration (B) and Initial Administration indicates low reliability. We now have the correlation between these two administrations (0.32) that verifies the low reliability. The reliability coefficient obtained by correlating two administrations of the same test to the same students seems to provide a clear and unequivocal index of reliability.

Again, the idea is straightforward. Correlations of multiple administrations of the same test to the same students is an intuitively appealing way to compute reliability. Unfortunately, this approach to computing reliability is difficult, if not impossible, to implement in practice. This changed with work by Louis Guttman (1945).

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.