Study and practice tactics.
Was Jeremy Silman wrong?

Thanks for the comments. But it is rather confusing when many people advise one to study endgames if one wishes to improve.

I study one level ahead in Silman's endgame book and it hasn't done anything bad to me.
It really depends on your style whether tactics or endgames are more important, but you should be as good as you possibly can at both (and certainly study them more than strategy or openings).
For a dry conservative player (Capablanca, Karpov, Kramnik style) endgames would be more important, and for a speculative combinational player (Morphy, Tal, Korchnoi) tactics would be more important. But everyone that I just mentioned was a master of both tactics and endgames, and you will have to be too if you want to be any good at chess.

Study and practice tactics.
Yup. At your level, you shouldn't be studying endgames, just the basic ones, like P + K and Q + K... Study more tactics. ^_^
I'm sorry I disagree and believe your logic is faulty. Endgames by their very nature are not only tactical, but they are tactics stripped down to their bare minimum. So if the mantra is "tactics, tactics, tactics" it could just as well be "endgames, endgames, endgames".

Silman was not wrong at all. He emphasizes that there are essential endgames that the beginner and low intermediate must know COLD in order to play them almost without thought. I'm not going to list them, because he's already done that in his book.
But after that, the player must be able to wend his way through the thicket of the middlegame to even get a chance to play that ending. What good is knowing the Lucena position, for instance, if you get blindsided by a knight fork on move 17? This is where tactics--and playing lots and lots of slow games--will serve you well.
I assume we are talking about the overkill mates, what can and cannot produce checkmate and basic king and pawn endgames?
I am really not sure that anyone really knows how to teach someone with a sub 1200 rating anything. If they take chess seriously enough to study it, how can they be that bad? You just can't see what you need to be able to see. Apparently the fastest way to learn to see involves studying tactics. Until you can see, the only endgame knowledge you need is the endgame knowledge that forms the basis of "positional" play.
I hope Silman weighs in. I really want to know what a chess teaching professional has to say. My own efforts to teach beginers usually devolves into trash talking, personal insults and attempts to "force feed" basic positional concepts into an empty void.

How many points do you miss out on by misplaying the endgame? If most of your losses come from middle game tactics, then that's probably what you should be studying. Studying more endgame certainly won't hurt,but most people want to improve performance as fast as possible.

I read the book fully without regard for his recommendations, and I'd suggest got more out of the book for it. It is a very good endgame book, make good use of it.

If you don't trust an author's advice when he tells you not to use his book, you need serious reasons.

@BorgQueen: I lose some games in the middle game, but I also lose plenty in the endgame. And I do tactics puzzles (not on this site because of the time controls), but feel I also need endgame study.
Look at it this way: If you don't study and gain adequate skills in the basics of openings, tactics/strategies and middle games, how much is the end game really going to matter? Chances are you'd stumble into the end game already at a disadvantage. ;) That's just my own learning process though; to each his own :)

Endings and an understanding of them are a crucial element of learning how to play real chess. Knowing what winning ending positions look like and why they are winning enables players to realize such potential advantages at earlier stages in the game. For example, if you can triple your opponents pawns and isolate them from their counterparts early in the game, you might conceive a plan such as this. " material is even, the main difference in this position is pawn structure, my structure is perfect, my opponents is fragmented, he has isolated and tripled pawns everywhere. Therefore, my plan will be to trade off all pieces, which if carried out correctly should enable me to actualize my advantage and win in a K an P endgame. In other words, a sound understanding of the endgame should allow players to improve their opening and middle game play because they can move towards the endgame while accumulating advantages during the opening and middle game that will decide the outcome of the game in the endgame .
In his "Complete Endgame Course", Jeremy Silman gives some instruction on a few basic endgames, then tells his reader to put the book away until he has reached a higher rating. At the rate I'm improving, I'll never be picking the book up again. Was his advice wrong? How can I improve if I can't study more endgames?