If you're looking for a book to help you with PMP certification concepts and the related mathematics, this book is likely to do you much more harm than good. It is one thing to have study material that contains the occasional errata, but when the errors become so numerous and compounded that you end up spending most of your time trying to work out what in the book is correct and what is not, you have firmly crossed the line between study guide and doorstop.
I don't even own a copy of this book. My girlfriend was trying to use it to prepare for her PMP exam, occasionally asking for help with bits of the math. After struggling with the EMV decision tree section, she gave it to me look through. This section looks for all the world like it was written by someone with minimal experience with math and risk management who was given a 10-minute crash course on it, and then six months later, sick with the flu, was told to write up a study guide. Really, it is that bad.
There is no way to understand what a mess this is without giving some specifics. I feel duty-bound to do this here. It is easy to say a book is awful; I want to carefully show why.
The section purports to teach a single simple form of a decision tree problem, where there are two possible decisions, each with an up-front cost and two distinct outcomes that have their own impact and probability. The object is to evaluate which is the decision with least risk, that is, with the highest expected value outcome.
The sample problem invites us to consider a company's make-or-buy decision:
If "make", initial cost is $100,000, with an 80% chance of "success" (impact $0) and 20% chance of "failure" (impact $40,000).
If "buy", initial cost is $75,000, with a 40% chance of "success" (impact $80,000) and 60% chance of "failure" (impact $70,000).
The make and buy costs are given without indicating that they are costs. No demerits for this, because it is obvious... except it becomes clear eventually that to the author it is anything but obvious, as I will show.
There is no indication whether an amount given for an impact is positive or negative. In the example problems that follow the sample problem, the problem solutions infer that all failure outcomes should be interpreted as negative amounts, but in the sample problem, the entire issue is hopelessly muddled. The book's brief description of how a decision tree works doesn't give any clues. (A failure outcome in real life can certainly be a positive amount, usually just an amount less than the initial outlay for the decision leading up to it.) The fact that the whole thing is left ambiguous is, in my opinion, because the author does not quite know what they are doing, and is essentially punting on it to avoid opening up a can of worms that might mean missing the print deadline.
Don't believe me? Read on.
In Step 1 of the solution, the book says the expected monetary value (EMV) of the buy/success outcome is $75,000 + (40/100) * $80,000 = $107,000. The math here is correct, but the answer is wrong, because the $75,000 is a cost, not a positive impact. It should be -$75,000 + (40/100) * $80,000 = -$43,000. But that hardly seems like a positive outcome. Perhaps the impact value of $80,000 has a different meaning? What could that be, though? Is it the net outcome (i.e. with the cost built in)? Obviously not; otherwise, why add in the cost in the first place?
Here is what I think might have happened:
a) The author picked the numbers for the sample problem randomly
b) Then, the author plugged them into the correct formula, got a negative number, and thought "how can that be right?" (in fact, a "successful" outcome might only be relatively so in real life, therefore sometimes it can be negative). What to do? Perhaps go back and figure out where they went wrong?
c) Nah! Just change that -$75,000 to $75,000 and rock on!
The author probably had some very minimal understanding of EMV and was probably just trying to muddle through using equations from some other book.
In Step 2 of the solution, the book says that the EMV of the buy/failure outcome is, and here I need to quote:
"$75,000 + $70,000 = (-) $145,000 (recall that because this outcome has a negative impact on the project, the EMV is shown with a minus sign.)"
This really gives the game away; whoever wrote this has no idea what they're doing and can't be bothered to figure it out. For starters, the 60% probability is nowhere to be seen. But mainly, the author's mind is just insisting on a result that makes sense to them, and is making stuff up to justify it. I searched the book in vain for the part I was supposed to "recall" which told me that an EMV with a negative outcome must be shown with a minus sign. Here is what I think happened:
a) The author was in a hurry and forgot about the 60%
b) Then, the answer came out to $145,000, which the author was awake enough to realize couldn't be right, because it's larger than the "success" outcome! What to do? Perhaps go back and figure out where they went wrong?
c) Nah! Put a minus sign in front of that, dude! Problem solved.
If an action taken turns out to be a "failure", why does this make $1 suddenly equal -$1? If the action is a "success", why does the $75,000 cost suddenly become a $75,000 windfall? These are just two of many obvious questions that spring immediately to mind.
Imagine a reader at this point whose math is rusty and is just trying to get the hang of this decision tree stuff. Will they ever recover from this?
In Step 3 of the solution, the book adds the results from Step 1 and Step 2 together to get an overall EMV of -$138,000. Well, now that is bad math. That sum is actually -$38,000. OK, just a typo. There are lots in this book.
The real answer, by the way, is -$75,000 + ((40/100) * $80,000) + ((60/100) * -$70,000) = -$85,000. That is assuming that the "failure" impact should be read as negative.
So the exposition of the sample solved problem is just hopelessly wrong, and no amount of pure carelessness or pure incompetence can account for it. It's obviously both. It leaves the aspiring student in such a hopeless position, that I really wish there were such a thing as authorial malpractice for instructional books.
In the exercise problems immediately following the sample solved problem, the author seems to forget about their maverick rule of negating the "failure" EMV value. Instead, the solution method is almost entirely correct, except the author persists in counting the up-front cost of a decision as income and not a loss, so the answers are nonsensical. For example, in Exercise 1, the reader must determine whether to build or upgrade given:
Build $100,000: 65% chance of success, impact $50,000; 35% chance of failure, impact $90,000
Upgrade $50,000: 80% chance of success, impact $60,000; 35% chance of failure, impact $50,000
Now obviously, the $100,000 and $50,000 are up-front costs associated with the build and upgrade decisions, respectively. So just glancing at this problem it is easy to see that the decision to build is a loser, but because the author treats the up-front figures as benefits and not costs, the book gives "build" as the right answer. The exact same thing is done for all ten example problems, which all have the same form.
So in the exercise problems the author has the formula right, but never stopped to think, in writing all ten example problems, what the meaning of the dollar amount associated with the initial choice was. Someone can only make this mistake if they don't understand project management. Not very reassuring, to put it mildly.