- #1
Ackbach
Gold Member
MHB
- 4,155
- 92
Having taught physics for several years now, both algebra-based and calculus-based, and seeing what works and what doesn't in labs, I thought I would post some of my findings here in one place for your edification.
1. I am decidedly NOT a fan of what I call cookie-cutter, follow-the-recipe, paint-by-numbers approach to labs. This is not experimental science. In real experimental science, you don't even usually have the hypothesis given to you! Essentially nothing is given to you. However, in the exigencies of time, I usually have given my students the hypothesis.
2. The students need to arrive at the place where they are designing the experiment, running it, collecting data, analyzing data, writing up their own lab reports from scratch, and coming to a reasonable conclusion from their data. How to get them to the point where they can do this?
3. The most powerful method I have found for doing this is to model the entire process for them. That is, take a beginning sort of hypothesis, design an experiment to test that hypothesis (especially including techniques for controlling for undesirable variables), run the experiment, collect the data, analyze the data, and then write up the lab report all in front of the students. This, in conjunction with giving them general guidelines for writing up their lab reports, has resulted in an astounding improvement in the quality of the lab reports. They get the logic of how it all fits together! Lab reports are a very logical chain of reasoning, but if students aren't told why things are put where they are, they're not going to understand. But even telling them and showing them isn't enough, of course: you have to make them do it.
4. So I have attached my lab report guidelines as well as a sample lab report that I do for mechanics, for your edification.
https://www.physicsforums.com/attachments/4499._xfImport
https://www.physicsforums.com/attachments/4500._xfImport
5. The students need to write up their lab reports from scratch, using correct English spelling and usage. I generally require MLA format, but you can do what you like with that.
6. This procedure is much closer to real experimental science than most labs are. However, there is still one component missing: statistics. Science is inherently statistical and probabalistic. However, the typical lab does not leave time for doing enough runs to get a statistically significant sample size. I found this out when I was teaching statistics, and I mentioned off-hand that statistics was the back-bone of science. And one of my students said, "Really?" And I thought, "Oh, no!" Something failed to transfer here! This was a junior in high school, and she had had multiple, real science classes, and yet had no clue that statistics had anything to do with real science!
7. So a colleague of mine had a brilliant idea: have a "running lab" that is quick to set up and take down, such that students can get 5 to 10 data points every time they do this running lab. By the end of the term, they should have a statistically significant sample size. For a recent course in electricity and magnetism, I had the students measure resistor values to see if they were in spec, and whether the average value was equal to the nominal value. I have attached my instructions for analyzing the data for your edification, as well as a sample spreadsheet.
https://www.physicsforums.com/attachments/4501._xfImport
https://www.physicsforums.com/attachments/4502._xfImport
I have not gotten any feedback from these instructions, but I am hoping that the students get the idea that if they run an experiment once, they haven't shown a thing. You need lots of runs! Repeatability is a hallmark of science.
So these are some of my practices for improving laboratory instruction in physics classes. I get that chemistry classes have an important safety component to them, and you've have to modify these ideas a bit to make them fit with a chemistry class. But the basic idea should still carry over.
Questions, comments, feedback, and other charitable donations would be very much appreciated on this thread -->>
http://mathhelpboards.com/commentary-threads-53/commentary-quot-thoughts-physics-labs-quot-15732.html#post73616.
1. I am decidedly NOT a fan of what I call cookie-cutter, follow-the-recipe, paint-by-numbers approach to labs. This is not experimental science. In real experimental science, you don't even usually have the hypothesis given to you! Essentially nothing is given to you. However, in the exigencies of time, I usually have given my students the hypothesis.
2. The students need to arrive at the place where they are designing the experiment, running it, collecting data, analyzing data, writing up their own lab reports from scratch, and coming to a reasonable conclusion from their data. How to get them to the point where they can do this?
3. The most powerful method I have found for doing this is to model the entire process for them. That is, take a beginning sort of hypothesis, design an experiment to test that hypothesis (especially including techniques for controlling for undesirable variables), run the experiment, collect the data, analyze the data, and then write up the lab report all in front of the students. This, in conjunction with giving them general guidelines for writing up their lab reports, has resulted in an astounding improvement in the quality of the lab reports. They get the logic of how it all fits together! Lab reports are a very logical chain of reasoning, but if students aren't told why things are put where they are, they're not going to understand. But even telling them and showing them isn't enough, of course: you have to make them do it.
4. So I have attached my lab report guidelines as well as a sample lab report that I do for mechanics, for your edification.
https://www.physicsforums.com/attachments/4499._xfImport
https://www.physicsforums.com/attachments/4500._xfImport
5. The students need to write up their lab reports from scratch, using correct English spelling and usage. I generally require MLA format, but you can do what you like with that.
6. This procedure is much closer to real experimental science than most labs are. However, there is still one component missing: statistics. Science is inherently statistical and probabalistic. However, the typical lab does not leave time for doing enough runs to get a statistically significant sample size. I found this out when I was teaching statistics, and I mentioned off-hand that statistics was the back-bone of science. And one of my students said, "Really?" And I thought, "Oh, no!" Something failed to transfer here! This was a junior in high school, and she had had multiple, real science classes, and yet had no clue that statistics had anything to do with real science!
7. So a colleague of mine had a brilliant idea: have a "running lab" that is quick to set up and take down, such that students can get 5 to 10 data points every time they do this running lab. By the end of the term, they should have a statistically significant sample size. For a recent course in electricity and magnetism, I had the students measure resistor values to see if they were in spec, and whether the average value was equal to the nominal value. I have attached my instructions for analyzing the data for your edification, as well as a sample spreadsheet.
https://www.physicsforums.com/attachments/4501._xfImport
https://www.physicsforums.com/attachments/4502._xfImport
I have not gotten any feedback from these instructions, but I am hoping that the students get the idea that if they run an experiment once, they haven't shown a thing. You need lots of runs! Repeatability is a hallmark of science.
So these are some of my practices for improving laboratory instruction in physics classes. I get that chemistry classes have an important safety component to them, and you've have to modify these ideas a bit to make them fit with a chemistry class. But the basic idea should still carry over.
Questions, comments, feedback, and other charitable donations would be very much appreciated on this thread -->>
http://mathhelpboards.com/commentary-threads-53/commentary-quot-thoughts-physics-labs-quot-15732.html#post73616.