Sept 29 Discussion Questions
In chapter 7, McKeachie talks about peer assessment as an alternative method of assessing learning. Students would use a computer-based peer reviewing system to grade other students' work online using a grading rubric developed by the professor. Should a student's grade be based partially on the grading of another student?
I like McKeachie's suggestion for alternative assessment using application of knowledge to field situations, especially for upper division courses, how might this work in an intro or lower division course?
Would you rather have a bunch of small assignments or a couple of really big ones. For example, this class has several smaller assignments, versus a course which has only one midterm, one papaer, and a final? Which do you prefer?
Our author mentions several ways to access student learning that differ from the traditional methods of testing. If we were to use such methods, such as the student portfolio, would it be important to stick to some set "number based" grading system, or could grade assignment be more "progress" related? Which seems the most appropriate?
Do you think that the first test in a course should be weighted less, or should every test be weighted the same?
The book brings up the topic of group testing (having the students take an individual test, and then a group test, and basing their grades on a combination of the two tests), to evaluate students learning. Is this an effective way to assess how well the students are understanding the material, and is it ever used?
Are essay tests effectivive at measuring the learning of introduction to psyc students?
McKeachie describes grading essay questions in a 9 step process. This seems very long winded and for a large intro class would take too much time. Are there any suggestions for quicker more efficient ways to grade essays?
How often should an instructor give tests? Does the university have specific requirements? If you give too many tests, there will not be sufficient time for lectures, and not enough tests and you run the risk of students forgetting past information. Is there a magic number in the academic world?
It seems like so many tests are multiple choice on scantrons. Do teachers prefer this because of ease of grading? Although the McKeachie likes essay questions, and I, too, feel it is a better way to see if students really are "getting it" is it really feasible in a class of 150 or 300?
I like the idea of giving tests early in the term to get students used to your style and used to studying, but is it really advisable to give early tests less weight than later exams?
McKeachie suggest using a grid in test
development, has anyone used this method? How time
consuming is it?
Chapter 8 - When creating multiple choice questions, is
it best to choose all possible answers for the value they
contibute to a discrimination test, or is it best to
create a test bank and allow the computer to do this
randomly? It seems like allowing the computer to do it
randomly allows for more possible mistakes!
McKeachie suggests that one should
"avoid evaluation devices that increase anxiety and
competition." If students are taking learning
seriously, can some anxiety or competition be helpful?
(Ch 8): If you do as McKeachie suggests and weight
earlier tests very little, will that have an effect on
the the way students study and, therefore, the accuracy
of the feedback the exam will give you?
When teaching a large introductory type class, do you think that it's more effective to include various types of questions (multiple choice, essay, short answer, etc) in each test, or should you have a test that is all multiple choice, then have a dedicated essay, etc?
Are there ways other than peer review of assignments to stimulate students' self-assessment of their work?
Do you think it is a good idea to use the peer assessment process? A big issue of whether or not thepeer graded fairly would come up and everyone who is dissatisfied with their grade would eventually come to the teacher.