Why School Science Experiments Should Be Replaced By Engineering

Disclaimer: This post is a bunch of stuff I just thought up and is mostly about education. I did ok as a student, but know almost nothing about teaching. So the bullshit meter on this post is higher than usual. And I'm really just writing this as a way to put off doing taxes (as a side note, if Republicans wanted to destroy Democrats they'd move elections to mid April).

The build up to this post is the general concept of evaluating information, especially who to trust (school, scientists, religious leaders, parents, no one decide for self...). I'm going to try to skip past my giant set of rants on this topic, especially since I've touched on it in several posts. But the starting off point is that most of these sources often end up saying just trust me and then can't believe when people put their trust in an alternate source (school and science that includes you).

While school is not a perfect source of information, it is generally better than any of the others and is at least a good starting point. So I'd say that two things schools should aim for are showing why school should mostly (but not absolutely) be trusted and more generally teaching how to evaluate sources of information.

I think that science education is a logical starting point for teaching how to evaluate sources of information and demonstrating the validity of information from school. Especially since it is generally the only subject that has demonstrations of any kind (name another subject in which what is being taught is actually tested in any way, except maybe self consistency in mathematics but that's rather abstract). This means that science experiments are far more important as a piece of education as a whole than they might appear at first.

Problem is, I'd claim, that school science experiments are flawed. Let's say students are doing the time an object falling experiment. The hypothesis is something like "gravity accelerates objects near Earth's surface at approximately 9.81m/s^2" right? Wrong! Everyone knows that's correct and if the experiment proves it wrong the student doesn't get to publish an article in Nature, she/he gets to repeat the experiment again till the "hypothesis" is proved true. That's about as far from science as you can get. The real hypothesis is something along the lines of "what teachers tell me is generally correct and provides information about how the world works", and increasingly has the add on of "science/school is fun".

Well, if that's the hypothesis then doing the object drop experiment is still bad science. Why? Well, let's take another example. Someone on the street says the way to add two numbers is to just add 5 to the first number. Now to prove it try doing 6+5, ok how about 7+5. See the rule works, now trust me. The natural reaction should be, let me come up with a set of numbers to add to test this rule and it would quickly be proven false.

So doing a proscribed experiment doesn't go very far in answering the true hypothesis. The students need to design their own experiments. But the answer to any experiment most students would do is already well known and they'd still be expected to get the correct answer. It is a step in the right direction, but I'd claim that far better would be to ask a student to design something using the knowledge they've been given. When a student is able to use their knowledge to create something that works that is a powerful moment. It shows the knowledge they have works even when tested in the way the student chooses. It means that the student isn't just repeating whatever they do till they get a right answer. And I'd claim it is more fun (although that may just be my opinion).

Now there is a huge issue with building something. It very often won't work. If the project stops there then science is disproved, school is bullshit and it is time to start reading websites about how the moon landing was faked. However, if the student has to figure out why it didn't work and then redesign based on that info then the student is testing whatever science comes up rather than whatever was being taught that week. This becomes a much better test of the reliability of information taught in school. Additionally, many people don't consider science sufficiently "magical", and I think that finding unexpected issues and their causes can help demonstrate how mysterious and "magical" science can be.

Not to mention that if the design is sufficiently interesting the hypothesis can be adjusted to "what teachers tell me is generally correct and provides USEFUL information about how the world works".

In addition to increasing trust in school, now the student knows that some sources provide information that can be used in interactions with the world. Which means that other information sources can be tested for the same ability. Thus, giving the student a tool for evaluating sources of information in general.

Alright, I'll admit the argument doesn't hold together perfectly (not to mention being built on complete ignorance about teaching), but there it is. I guess I should also admit that it isn't surprising that I think the key to making people smarter (aka thinking more like me) is to get them to do more of what I do.