Kids should be building rockets and robots, not taking standardized tests

Standardized tests serve a number of roles in public education, very few of which relate to either enhancing or testing a student's ability to get a job.

The main applications relate to checking in on students' progress through the education system relative to their peers. It is also a way to compare the performance of particular groups of students (or schools) relative to others, which is very controversial, but obviously easier than trying to judge the quality of the robots they produce.

By far the greatest advantage of standardized tests over other assessment methods is that they scale beautifully. When you want to ask the question "how are we as a country/state/district doing compared to others", standardized tests provide a reasonably easy-to-interpret, reliable reference point. It's not just about what a test score says about an individual student (which ultimately is very little), but what the test scores indicate in aggregate.

> It sure isn't. Anyone who can read and comprehend text can answer it with no prior knowledge of the subject.

This reminds me of people who think the manpages are a sufficient resource to completely understand the unix cli environment.

I understand what you meant, that's why I qualified my statement by saying your choice of words was unwitting. My point is that, precisely because it was not deliberate, your choice of words reflects a subconscious perspective that teaching is no different than other jobs, where formal qualifications and a professional attitude are reasonable substitutes for an in-depth assessment.

Of course, we could get into a discussion about whether that's a valid assumption for any profession, but that's neither here nor there.

I can only speak from personnal experience: the first time I used a microscope I did break a glass by trying to zoom in first. I had no manual but I wouldn't have read it anyway (nowadays I do read instructions before using anything). When I read this particular article I "played" the game and answered correctly (hooray for me). I agree this kind of test has nothing to do with STEM because it is interchangeable and could be devised for any courses ; my point is that it's "just" a comprehensive reading test and if you fail it it means you fail at comprehensive reading which is one of the most basic skills one pupil should acquire. I agree it has nothing to do with validating specific STEM knowledge but nevertheless the explanations are correct and useful in the STEM field and understanding them is important. Now, if american science courses are only about those kind of tests it's a little bit concerning: Scoring A's in every courses just because you understand the questions versus scoring A because you understand the courses.

Not: I am not american so I don't know how education is organised in the US.

Most of those ideas are nice, but not very practical. Some thoughts:

1) Real, actual, real world problems can't be solved in a class room. They are solved in 'the real world', i.e. business and government. Business and government won't let kids have a go at such things because it's too expensive, risky and messy. Companies continually complain about young employees' lack of (VET/STEM)skills, yet when it comes down to it they're really not prepared to run the risk that comes with having young kids (not) solving their actual problems. Now, schools could 'simulate' real problems, and do try to do that, but it's enormously expensive and difficult and not very practical. Also, in our services-based economy most problems are solved through repetitive procedures. This is why there's such a focus on 'transversal' skills these days. Learning how to do look up information, read texts etc are important skills to have. They're the foundation underpinning those transversal skills. The idea is to not let kids learn about just one problem, but about a basic skill set that can be applied to any problem. Combine this focus with the hesitancy of companies, and you can see why these dreaded standardized tests come enter the equation...

2) Many schools already do this or try to do this. The problem is that there are some essential skills kids need to learn, which cannot be removed from the curriculum. Teaching those skills takes up a lot of time. This is why there is little room for such 'free play' in a class room. Creativity is wonderful and essential, but very hard to consistently implement in schools. Especially with the complaints that todays students lack the basics: math&english.

3) Schools already track progress. It's the one thing schools are actually pretty good at, and it has led to a complete focus on metrics that has had the unfortunate side effect of making the test results more important than actual learning outcomes and skills. This is also why I'm pretty cynical about web-based learning: a metrics based system breeds metrics based testing, schooling and studying. IT is a nice luxury, but not (yet?) the solution to our problems. 4) I like this idea. Especially involving the parents. 5) This is really the same idea as option 2. The same issues apply there. I agree that kids should do projects and have so leeway in their choices, but you underestimate the real world problems teachers encounter in a class room. Most students are in fact NOT responsible, smart or capable enough to choose a (relevant, teachable, educational) project and tackle it. Add social problems, broken families, shitty reading comprehension and problem kids to the equation and you can see why this doesn't work for most students. It's a great idea for excellent students though - add it to the curriculum for the better students who do well on tests anyway and it would be fantastic. In fact, don't do it on Fridays. Let them do projects based on what the rest of the class is doing. When everybody else is learning about the planets and the moon, give those 'smart' kids an advanced project about planetary movements and let them present the results to the class.