As an engineer, this book makes me happy. A great discussion of how to break *any* problem down into quantifiable metrics, how to figure out which of those metrics is valuable, and how to measure them. The book is fairly actionable, there is a complementary website with lots of handy excel tools, and there are plenty of examples to help you along. The only downside is that this is largely a stats book in disguise, so some parts are fairly dry and a the difficulty level jumps around a little bit. If you make important decisions, especially in business, this book is for you.
Some great quotes:
Anything can be measured. If a thing can be observed in any way at all, it lends itself to some type of measurement method. No matter how “fuzzy” the measurement is, it’s still a measurement if it tells you more
than you knew before. And those very things most likely to be seen as immeasurable are, virtually always, solved by relatively simple measurement methods.
Measurement: a quantitatively expressed reduction of uncertainty based on one or more observations.
So a measurement doesn’t have to eliminate uncertainty after all. A mere _reduction_ in uncertainty counts as a measurement and possibly can be worth much more than the cost of the measurement.
A problem well stated is a problem half solved.
—Charles Kettering (1876–1958)
The clarification chain is just a short series of connections that should bring us from thinking of something as an intangible to thinking of it as a tangible. First, we recognize that if X is something that we care about, then X, by definition, must be detectable in some way. How could we care about things like “quality,” “risk,” “security,” or “public image” if these things were totally undetectable, in any way, directly or indirectly? If we have reason to care about some unknown quantity, it is because we think it corresponds to desirable or undesirable results in some way. Second, if this thing is detectable, then it must be detectable in some amount. If you can observe a thing at all, you can observe more of it or less of it. Once we accept that much, the final step is perhaps the easiest. If we can observe it in some amount, then it must be measurable.
Rule of five: There is a 93.75% chance that the median of a population is between the smallest and largest values in any random sample of five from that population.
An important lesson comes from the origin of the word experiment. “Ex- periment” comes from the Latin ex-, meaning “of/from,” and periri, mean- ing “try/attempt.” It means, in other words, to get something by trying. The statistician David Moore, the 1998 president of the American Statistical Association, goes so far as to say: “If you don’t know what to measure, measure anyway. You’ll learn what to measure.”
Four useful measurement assumptions:
1. Your problem is not as unique as you think.
2. You have more data than you think.
3. You need less stated that you think.
4. And adequate amount of new data is more accessible than you think.
Don’t assume that the only way to reduce your uncertainty is to use an impractically sophisticated method. Are you trying to get published in a peer-reviewed journal, or are you just trying to reduce your uncertainty about a real-life business decision? Think of measurement as iterative. Start measuring it. You can always adjust the method based on initial findings.
In business cases, most of the variables have an "information value" at or near zero. But usually at least some variables have an information value that is so high that some deliberate measurement is easily justified.
While there are certainly variables that do not justify measurement, a persistent misconception is that unless a measurement meets an arbitrary standard (e.g., adequate for publication in an academic journal or meets generally accepted accounting standards), it has no value. This is a slight oversimplification, but what really makes a measurement of high value is a lot of uncertainty combined with a high cost of being wrong. Whether it meets some other standard is irrelevant.
When people say “You can prove anything with statistics,” they probably don’t really mean “statistics,” they just mean broadly the use of numbers (especially, for some reason, percentages). And they really don’t mean “anything” or “prove.” What they really mean is that “numbers can be used to confuse people, especially the gullible ones lacking basic skills with numbers.” With this, I completely agree but it is an entirely different claim.
The fact is that the preference for ignorance over even marginal reductions in ignorance is never the moral high ground. If decisions are made under a self-imposed state of higher uncertainty, policy makers (or even businesses like, say, airplane manufacturers) are betting on our lives with a higher chance of erroneous allocation of limited resources. In measurement, as in many other human endeavors, ignorance is not only wasteful but can be dangerous.
If we can’t identify a decision that could be affected by a proposed measurement and how it could change those decisions, then the measurement simply has no value.
The lack of having an exact number is not the same as knowing nothing.
The McNamara Fallacy: The first step is to measure whatever can be easily measured. This is OK as far as it goes. The second step is to disregard that which can’t easily be measured or to give it an arbitrary quantitative value. This is artificial and misleading. The third step is to presume that what can’t be measured easily isn’t important. This is blindness. The fourth step is to say that what can’t easily be measured really doesn’t exist. This is suicide.
First, we know that the early part of any measurement usually is the high-value part. Don’t attempt a massive study to measure something if you have a lot of uncertainty about it now. Measure a little bit, remove some uncertainty, and evaluate what you have learned. Were you surprised? Is further measurement still necessary? Did what you learned in the beginning of the measurement give you some ideas about how to change the method? Iterative measurement gives you the most flexibility and the best bang for the buck.
This point might be disconcerting to some who would like more certainty in their world, but everything we know from “experience” is just a sample. We didn’t actually experience everything; we experienced some things and we extrapolated from there. That is all we get—fleeting glimpses of a mostly unobserved world from which we draw conclusions about all the stuff we didn’t see. Yet people seem to feel confident in the conclusions they draw from limited samples. The reason they feel this way is because experience tells them sampling often works. (Of course, that experience, too, is based on a sample.)
Anything you need to quantify can be measured in some way that is superior to not measuring it at all.