Does changing the method of testing change the student grade?
Students usually announce that they are better on multiple choice tests or written tests. I believe that if a student studied sufficiently, it does not matter which question type is used.
I compared the grade for the written half of the exam against the grade for the multiple choice half.Scantrons were employed for MC. Written parts were graded without knowledge of the student or their grade on the MC portion.
Averages on this exam for the MC questions were 17.6 incorrect and for written were 16.8 incorrect. Average grade was 66. A t test revealed that the MC and written grades were not significantly different for the 46 students (2 classes). A linear regression R squared value was .6. Students did perform better on the mutliple choice part of the exam, but not significantly.
46 students took an exam with 50% multiple choice and 50% written. A t test showed no significant different between the grades from the 2 question types. Linear regression analysis showed an R squared value of .6. A student getting -8 on the MC, was likely to get close to a -8 on the written. This suggests to me that test preparation is more important than the type of question asked, but written portions will encourage additional development of writing skills.