Blog Entry #5

The readings this week covered assessment. The bulk of the readings were concerned with proper testing methods. I was most interested this week in the book's ideas on constructing a test, the notion that testing needs to be more than a letter grade to aim for, and a few of the other methods of assessing learning.

Constructing a test in essence is as simple as putting down a few questions that cover the material and don't take too long to answer. While this may technically be true it is a lot more in depth. To adequately write a test you need to look at what you want the students to learn as well as what the material will allow you to do. I honestly thought this would be the easy part. Just pick a few main concepts, mix a few different question types and done. There are even some test making websites out there that will format the exam for you! This chapter made me realize that I needed to be a little more selective than that. It describes what the different types of question can accomplish(pg. 88 Teaching Tips). I would likely use multiple choice and short answer questions for lower level classes to build the foundation of knowledge and allow some creative explanation while avoiding a grading nightmare involving hundreds of essays for the typical 100-200 level Biology class. In the higher level courses, I would use a mixture of multiple choice/short answer and essay questions. I still want to see what they are learning for base information, but these are the classes where students really need to show that they can apply the knowledge. Many classes in science have math that needs to be tested as well. I would likely put this in the short answer section for problem solving and grade as much on recognizing what to use as whether or not they can do the math. For more information on how to construct a test check Tools for Teaching from UC Berkeley. As an undergrad, I always liked multiple choice tests, but looking back on it, that was not because I really wanted something to help me progress in learning. Rather it was so that I didn't need to be bothered by studying the material. I would prefer my students learn the material rather than like the test. This leads into my next topic of testing being more than assigning grades.

While it may be the commonly held belief that the point of tests is to get a good grade, this is not always true. It can be a valuable tool for teaching students. Occasionally tests will be simply a means to evaluate knowledge, but they can be much more in the college setting. I think that using the book's idea of making tests cumulative is going to be very helpful in my field. In science we usually start out with a decent amount of fact or method learning. While a teacher may say that tests are not cumulative there is almost no way around it. To effectively learn in science a student needs to learn concepts first then build on them. The really important ideas will never leave, but they will be changed and applied in different ways. I would look into building a test that flows from what they already know to what they learned to what those two mean together. When I was an undergrad, I always dreaded the cumulative tests that brought back random ideas that were never talked about after they were mentioned once. If a teacher explained that certain things were important enough to bring back up on several tests, I would feel like it must be important enough to know. In this way I learned things that were not reflected in my grades. Grades are still important, but the point of college is not to get all "A"s, but to learn things that will prepare you for your career. An "A" isn't worth much if you can't remember how you got it.

Some of the other methods of testing can be very effective in learning and evaluating at the same time. Graphic representations of concepts can be very useful in science. It is a way to show that, not only do they know the parts they should know, but that they know how those parts interact with each other. I would include this as a test question or as an alternative to writing an essay where applicable. Rather than writing about the Krebs Cycle a student could diagram it and show where parts come in to make the process work. If it is a complex diagram I could make the outline on the test and allow them to fill in areas. This may make people less nervous because they would know how much they need to write, but it would also stifle creativity and make a good idea a simple fill in the blank question. I can also see this replacing a short paper. It can show that they know what they are talking about and force students to look at broader implications of what they are learning about. A good example of how to create a concept map can be found here. I would actually like to use something like this as a final project for a class in which there is a central theme. A map is made linking all of the concepts learned with the main theme. The concepts are then linked to each other reinforcing how the knowledge is not irrelevant to each other. Ideally I would hope that the teacher learns from these too and can shape their future courses based off of their students diagrams.

While I am still very likely to make several tests that include many different types of questions this chapter, more than most so far, has changed my mind on how to do things. It is difficult to see that tests are used for learning when you need to take them. In that position you focus on what doing poorly will do to you rather than for you. I really liked the concept mapping idea and will likely at least try to incorporate that into a future class.