"The Art of Electronics" and the companion student manual.
This is NOT an electronics for dummies book; no illustrations with charming stick figures running along a wire carrying "buckets of charge" or the like (I don't have my copy handy, so of course the probability of exactly that illustration appearing on page 1 approaches unity) but investing the time to understand the book and do the exercises will pay off.
Be advised that the third edition is anticipated Real Soon Now, undoubtedly appearing the day after Mr Someone gets their copy of the second edition. So, tell Someone to hurry up and order!
I've heard Physics Prof's who are trying to teach their students a bit of electronics complain that AoE is not the best book for a first introduction. (Sorry I don't have any better recommendation.) To OP, if you go this route I would definitely recommend that they get the student lab manual that goes with AoE. If they work through that they will learn something! George H.
What's effective will depend in some fair measure upon past education and aptitudes. What's relevant depends on what interests their are... again, quite personal. But truly, the most effective and relevant education will likely come from trained friends and closer associates who can engage in a two-way dialog of learning. Hopefully, that is available. I didn't have access to either formal training or friends and I know, first hand, it's a veritable pain in the butt.
The Art of Electronics, 2nd edition, should be used together with its Student Manual -- for self-education uses. In some areas for someone doing self-study, there simply isn't enough worked out in the main volume. It covers a great deal, but in most cases expects there is a teacher and classroom setting to expand the material and engage it. Or it certainly seems to require it. However, much of that is saved by the Student Manual. For example, the student manual goes into a detailed, step-by-step organized, design of a common emitter BJT amplifier. Something left out of the book, itself.
A good math background is almost a must for parts of AofE. Some of the concepts are quickly glossed over in the book, such as Euler's and it's application with complex numbers used in electronics. For someone with a sufficient math background, that's all that is needed. It's already pretty well understand, in detail, and all that is needed is the hand-waving found in AofE to make sense of it. But for someone with none of that in hand, it's just random noise and will simply pass right over their heads.
This is still just a hobby for me. But as a teenager, someone gave me some manuals from the military that were designed to teach electronics to soldiers and were written, I think, in either the 1940's or 1950's. They were pretty good, as I recall. VERY THICK! Lots of them, too. A few feet along a shelf. I think I remember seeing something similar in a variety of PDF chapters, available on the web. Might try looking for them. (I apologize for not taking the time to search, right now.)
MIT has open courseware on the web. It's quite likely electronics is part of that. See:
I haven't read this one (and it is scanned in and HUGE), but it was supposedly a good, early book (100 years ago, or so):
A smaller scan of a 1943 version is at:
Early books often were targeted at self-education, which is why I suggest looking in that direction, too.
Best of luck in all this.
Also, if possible, connect up with a local community college class. The teachers are often there because they _want_ to teach this material. (They darned well aren't there to work on pet projects via grant programs.) It's usually "dumbed down" because that's where a lot of their clientele is at -- community colleges are complaining that their students' first year is mostly spent in remedial education, in my state. But I've seen some very good teachers doing a wonderful job getting across somewhat complex things (matrix methods for solving linear component circuits) to students generally ill-prepared for them and doing it successfully.
One thing I'll add: Textbooks from the UK often seemed to be aimed a bit more at practicing engineers than those in the U.S. are. This means that while you miss out on some of the deeper theory you find in, e.g., Sedra & Smith, you do get more design-oriented lessons than what The Art of Electronics has. (The U.S. model still seems to be, "we teach you theory in school, you learn practice from your employer" whereas the UK model is a more mixed approach?)
That's for sure... community colleges aren't that many notches up on the payscale from, e.g., social workers helping the indigent. They also have the benefit that the administration doesn't require PhD "qualifications" to teach Circuit Theory 101.
You've probably seen the news today about students at the University of California protesting tuition hikes (e.g.,
I'm quite convinced that if you have a pretty focused subject area such as PC or embedded software development, digital or analog circuit design, etc., one could come up with a school that costs far less and requies less time to complete than a traditional college, while simultaneously producing more-qualified (on average) students. The trick would be to get such a place accredited by, e.g., ABET so that the students could actually get jobs when they graduated. (Grumble, grumble...)