The assumption that we’re training scientists in K-12 (the target of NAEP) misses the mark. Scientists can get all of the lab practical training necessary in college. That’s the time for apprenticeship. I’m a scientist with a B.S. in chemistry from Caltech and a PhD from Columbia University. I was the chair of the 3,500-member Northeastern Section of the American Chemical Society and a university professor. I know about this topic.

K-12 schools must have physical campuses but can use virtual tools in them. Some students can, at a sufficiently advanced age, go entirely virtual, of course. But it’s not everyone’s cup of tea.

It will be long, if ever, before computers can substitute for good teachers. However, teachers must have more support than ever in their quest to provide good learning. That support must include computer technology but not the bad examples that too many have experienced.

Science should be learned properly by K-12 students, and it is not so today. It was seven years ago that the National Research Council published “America’s Lab Report” castigating our science classes for their mismanagement of science lab experiences. Their statements go against the usual virtual lab too because they require that students take real data from the real (they call it “material”) world.

I go a step further and suggest that students at lower sophistication levels should take their own data point by point. That approach argues against using technology such as probeware where students do not take data but rather have it presented to them.

Dissections aren’t really science labs. A science lab has a prediction (or even an hypothesis) that will be supported or refuted by data collected. All of the hue and cry over dissections is misplaced. Operating a microscope is not a crucial life skill for most students. And so it goes. Understand what a lab is for so that you don’t waste all of that valuable time and money.

On kitchen labs, I agree that many are like magic tricks. Even lab kits, which are way too expensive, have a sort of cookbook aspect to them. They have to in order to ensure success by students who are often alone and left to their own devices. However, this need not be so. I’ve seen kitchen lab plans by teachers who managed to avoid this pitfall and were very excellent investigation experiences.

Some people, most notably MIT with their iLabs, have created remote robotic labs to aid in learning science and engineering. IMO, this idea mostly helps the latter. The range of science experiments possible in this mode is quite limited, and the experience comes out much like probeware. There has to be a better way.

1. There’s not enough time, money, space, or safety to do the lab investigations that would provide a complete science learning experience particularly in grades 6-12. Really good classroom labs and field labs are great but have their limits.

2. Animated simulations have improved in the last 30 years, but they’re still not substitute for the real thing and even can detract from learning science if substituted for classroom labs. Use them for visualization or for model comparison with real data (but only for more sophisticated students).

3. Technology can help us solve many of our education problems but not if implemented blindly or without understanding the technology, the subject, and the pedagogy involved. You cannot simply bend, fold, and stretch business software into great educational software. Also, amateur programmers build amateur software that should not be allowed in classrooms.

You can get a good science education in K-12 and even in college if you’re not a science major with a combination of online hands-on labs and so-called kitchen labs. Science majors must get in the lab in college. To me, a kitchen lab is one that uses ordinary materials that are relatively inexpensive and safe.

What’s an online hands-on lab? It uses prerecorded real experiments, lots of them, to comprise a lab along with software that enables tight interaction with the videos to collect data from them point by point, much as in a classroom lab.

Furthermore, students should have the opportunity to perform extended investigations into the real world once or twice a semester in secondary schools. These would not be virtual — at least until the technology has advanced considerably. Sometimes these are called projects, but beware the projects that are just engineering build-its and are not science investigations. Engineering is great but is not science.

It can be the answer in some cases for some subject areas but certainly not all.