Hello,
I have administered a post-test for evaluating reading comprehension of an experimental group of EST learners.
The student have been asked to give (1) the meaning, (2) the equivalent in French, and finally (3) the Arabic equivalent of some technical terms chosen from the text they have been offered.
So, how should I put weights to these three response components so that the whole mark reflects their real comprehension of the text?
Thank you in advance.
Hello Amer,
The weights are entirely arbitrary, but should depend on the intended purpose/use of the test scores. For example, if the primary purpose was that of feedback to instructor(s), then the individual scores would likely be more informative than a combined score. If the primary purpose was accurate prediction of a some subsequent performance, then weighting the three scores based on the results of a regression model might be the better plan. If there is expert consensus as to what is meant by "real comprehension," for the target group of learners, then that might help guide the determination of how (if at all) the scores should be combined.
Sorry if that's a bit open-ended, but it's the use of the scores that's key here.
Good luck with your work!
- Badges
- Science topic
- Similar topics
- Organizational Psychology
- Training