Monday, 19 June 2017

Developing Better Tests and Quizzes

One of the biggest challenges during my practicum was writing good quiz and test questions. I didn't know if the questions that I made would actually test the students understanding of the content or if what I was writing down was just going to have the students rely on solely on how much they remember from class. In my opinion my very first test I made, a safety test for Science 8, highlights the problems I was having with making up good questions for tests. About 3/4 of the test were multiple choice and true/false questions with the last part being short answers. When making this test I found that while multiple choice questions were easy to make there were hard to make interesting or challenging. Most of the multiple choice questions required students to remember information that they learned previously. The short answer questions are slightly better but again relied on the students to recall materials that they remembered or make simple observations of a picture. If I was to write the test now I would reduce the number of multiple choice questions or even remove them entirely and include more short answer questions. I would change the short answer questions to ask students to explain their answers as I found that students wrote one word answers for the questions the first time around. In hindsight I would unknowingly implement these changes in future tests and quizzes of mine.

When developing my first math test I asked my SA how he designed his tests. He designed his test differently from all the other tests I've seen and taken as a student. He designed his test by splitting up his questions into 4 sections, C-, C+, B, and A questions. If students are able to get all the questions right for either sections then they would earn that grade. For example if students were able to get all the A level questions right then they would get an A for that test. Percentages was a subjective value that he would determine on a later date based on his observations on students. He had some flexibility when marking the test. For example if a student only got some questions right for the C- and B level questions he would give them a C+ as they demonstrated they had a solid understanding of the materials. The philosophy he used when designing questions for these types of tests was that C- questions should demonstrate that the students have a basic understanding of the material, enough for them to past the unit. C+ and B questions should be questions that are more difficult but something that they've encountered before if they had done the handout or textbook questions. A level questions were questions that they probably didn't see much of during the course but would instead require them to have a deep understanding of the content learned in the unit in order to get the questions right. For my SA test should accurately determine the students understanding of the content and that A should only be given if the student has an strong understanding of it.

I adopted this philosophy when I made the test for my statistics unit test. The first section is simple and straight forward asking students to analyze and calculate data, as well as explain why certain numbers are given based on the data set. The C+ questions followed the same idea with students doing more complicated calculations and having to go into more detail when explaining how certain calculations are done and how each part of the equations work. The B level questions bump the difficulty higher, such as asking students to calculate the standard deviation of a small data set and applying their understanding of the content to a contextual question. The A level questions are much more difficult and are questions that students wouldn't normally encounter. For example students are asked to develop their own data table based on certain conditions that are set in the questions.

I found making test based on my SA's philosophy to be more interesting and gives the students a better opportunity to demonstrate their understanding. However this method works the best when students are given the chance of doing a retest or some sort of opportunity to improve their mark. My SA gave students the opportunity to do a second unit test with the ability of starting only doing questions based on sections that they were not able to complete. For example if students got a C+ on their first test they would only have to do the B and A level questions. I only had students do one unit test, however I did allow students to do a retest based on the above method in order for students to understand their growth.

While I enjoyed writing my tests using the above method I followed a more traditional method on making test for chemistry, focused on giving students a number to show students if they got an answer right or not. For example my unit test for the Mole Unit assigned numbers to each questions. However I followed the same method on writing questions for the unit test for chemistry as I did when I wrote questions for the unit test for math. Most of my questions asked students to think through the information given to come up with the answers and some asked them to describe the process they would go through to make a solution with a specific concentration or how to convert between different units.

No comments:

Post a Comment