I am using "essay assessment" as a third data type in my research. There are 80 essays: first draft and final draft.
a. In order to get a general picture of students' level of writing, should I use the final draft score only or the mean of first and final draft score?
b. Total score (Content: 6 points, Language Use: 6 points, Organization: 5 points, Punctuation: 3 points) a student can get is 20.
Should I find the percentage of each part: If a student gets 3 out of 6 for content, should I say this student gets 50 out of 100 for content? how should I show the general picture of student success?
Though you are very much clear, I still have some suggestion. The difference between the two scores (Final - initial) will give you a new score which you can regard as an improvement if it is positive. You can conduct one sample t-test for each of the variable (For instance, content, style, or .....) and total score. You should use test value as zero for all the tests. If the result is positive and significant then it means there is improvement. The scale whether in numbers or percentage would not affect your analysis greatly, you can go for any of them, but I would prefer the raw scores.
Thank you. There is also a second rater. I'll be conducting interrater reliability. For the general picture, should I use only my evaluation of final drafts?
- Badges
- Science topic