Same as me. This is one of those little tasks which we just struggle on with any old how - because the risk seems so trivial. And given the weight of a bike, and the fact that it's reasonably easy to manhandle, it probably is a trivial risk to your back. Certainly not in the league of moving fridges up stairs or stacking sacks of potatoes.
So I'd say giving instruction on safe lifting is the wrong way to skin this cat, because it's likely that people will keep ignoring it. If you want to help people, then have a look to see if there's anything you can provide which they'll use because they perceive it to be better/easier. With a bit of lateral thinking, you could come up with all sorts of things, and the idea of using a mirror or camera, as already mentioned, is a good start. Here's another low-fi example of something else you could try: a length of cord hanging from a rafter, with a means of quickly attaching it to the bike - e.g. under the saddle or at the seat tube/seat stay/top tube junction. The attachment could be done via a strap with velcro or 'side buckles' or a carabiner to hook back up to a loop in the cord. This would allow the bike to be lifted by grabbing the frame near the bottom bracket and at a chainstay, and lifting it up to you whilst the rest of the bike leans away from you. This means you could keep your hands close to your body whilst you're lifting thereby minimising the leverage the bike's weight is exerting on your lower back. Obviously, this would be a terrible way to hold the bike if you actually wanted to work on it, but it would be fine for taking a look for a number on the bottom of the frame, and would avoid the problems mentioned above, e.g. damaging mudguards or brake master cylinders.
In reality, you'll probably still get some people turning the bike over or laying it down on its side, but I'd say this approach is your best chance of influencing how people do it. If it was a workplace, finding little helpers like this would earn you brownie points, because safe lifting instruction should be the last resort when it comes to manual handling risk assessments: avoidance or reduction of lifting should be your first port of call.