Chapter 3 THE ILLUSION OF INTANGIBLES: WHY IMMEASURABLES AREN'T

 



3 reasons why people think something cant be measured. They are rather misunderstandings.

  1. Concept - People misunderstand the concept of measurement itself. Once, you get it, you will say a lot more things are measurable than you thought before.
  2. Object - The things being measured is not very well defined. Sloppy and Ambiguous language is used.
  3. Method - Many people are simply not familiar with different methods to measure things.

3 reasons why people think something shouldnt be measured.

  1. Economic Objection - Too expensive to measure
  2. General Objection to meaningless of Statistics - You can prove anything with Stats. Circle Jerking BS.
  3. Ethical Objection - Dont wanna ask the hard questions needed to measure it.

The Concept of Measurement

Only an inexact man would call things exact. Measurement is not perfect nor certain.

Defn of Measurement - Quantitaively expressed, reduction of uncertainity, based on one or more observations.

Measurement is mere reduction of uncertainity, not elimination. It is a probabilistic exercise.

Since nothing is really certain about the world, you must also then speak the language of reality

"the average yield of the corn farms using this new seed increased between 10% to 18% (95% Confidence Interval)

This is how you speak if you want to represent reality in your quantitative language.

Information Theory is a field developed by some fellow called Claude Shannon. This guy proposed a mathematical definition for Information, which sounds like amount of uncertainity reduction in a signal. (Everything is a signal from the perspective of Infromation).

But here is the interesting part, he described it mathematically in terms of "entropy" removed by the signal.

I know right? How is this not TENET like? Reversing the entropy? Removing the entropy? Fuck.

HTMA is TENET. I knew there was a reason and a place for TENET in my life.

Stanley Smith Stevens gave use the widely known scales of measurement. Interesting this guy was also a psychologist. Far too many psychologists are poking their noses into Statistics, Math and Measurements, which is very very curious.

Nominal Scales - Fetus is Male or Female?, Yes or No?. Its just a list of possibilities. You get no knowledge about the order or size of things.

Oridinal Scales - Ratings, Simply tells you that one value is greater than other but not by how much. 4 stars restaurant is not two times better than 2 star restaurant.

Ratio Scales - Most widely understood, Dollars, Kilometers, liters. Tell us the order and the difference, unlike ordinal. Can be added, substracted, multiplied, divided.

Interval Scales - Almost like Ratio Scales, except Zero is not exactly empty. Zero also has some value. Celsius. (Temp in Kelvin is Ratio Scale)

Point of giving out all this info is that people forget that Nominal and Ordinal are also Measurements.

Measuring %Yes and %No survey is a nominal measurement.

Bayesian Measurement : A Pragmatic Concept for Decisions

Where does uncertainity lives is the debate between Bayesian and Frequentist.

For Bayesian, Uncertainity is the feature of the observer.

For Frequentist, Uncertainity is an objective aspect of reality.

Bayesian Interpretation means one thing, in the light of new information prior probabilities can be readjusted.

Frequentist's believe that probability is something that does not live inside you but is a part of the external environment/system.

Bayesian spend time calculating prior states of probability and the change between states of probability. Frequentists dont.

Measurement is not just employment of numbers, it has to reduce some uncertainty for you. Otherwise, its useless.

THE OBJECT OF MEASUREMENT

There is no greater impediment ti the advancement of knowledge than the ambiguity of words.

Thomas Reid

Clean the Question - What do you mean exactly by Productivity, mentorship, efficiency, Security?

We label with high syllable words, but they dont really mean anything useful. If you dig deeper, you will realize what the manager meant with IT security was that he wanted to reduce the intrusion and virus attacks.

So ask them again "What do you mean by " " and why do you care?

Dont go to measure Bullshit parameters that dont matter to you.

CLARIFICATION CHAIN

  1. If it matters at all, it is detectable/observable
  2. If it is detectable, it can be detected as an amount (or range of possible amounts)
  3. If it can be detected as a range of possible amounts, it can be measured.

ALIEN SCIENTIST THOUGHT EXPERIMENT

TEST and CONTROL is a pretty robust and common method to experiment about something and understand cause and effect.

So you can be confidently using it in your personal life as well.

Lets say you are an alien scientist who can replicate organizations, so you replicate two organization, to one of them you give more "Employee Empowerment" and to the other one give a little less "how would you see the difference?

Remember Scott and his brother, the american astronaut.

THE METHODS OF MEASUREMENT

You dont always need direct access to the object of measurements, Make use of indirect deductions and inferences..

Dont lament about not having enough data.

Philosopher, economist and a legal expert Gottfried Achenwall coined the word Statistics. Its amazing renowned guys in early statistics can from crazy backgrounds like Philosophy, Psychology etc.

Statistics fucks will students in School and College and does more harm to their understanding than good. One thing it fails to teach students is how even with small samples can tell you something that improves the odds of making a better bet in real decisions.

RULE OF FIVE - The Median Guesser

There is a 93.75% chance that the median of a population is between the smallest and the largest values in any random sample of five from that population

Did you know that you could use this sample of only five to estimate the median of the entire population?

Well, not exactly median but a small range of values that will contain the median with 93.75% confidence between its highest and lowest values.

Why does this work though?

Median is point where 50% of the data is above it and 50% is below it, so it is like a coin toss.

The only way your highest lowest range from 5 values wont contain the median is if all the values end up above median or all values end up below median.

Thats like getting all heads 5 times in a row (or tails)

0.5 x 0.5 x 0.5 x 0.5 x 0.5 = 1/32 = 3.125

Similarly getting 5 times tails is also 3.125

So the chances of not getting all heads or all tails is 100 - (3.125 + 3.125) = 93.75%

This book is flipping your thinking, to think of statistics as not a perfect way with no room for error but rather as a good enough way to work with least amount of resources.

The Urn of Mystery ; Single Sample majority rule

Lets say you have an urn filled with red marbles and green marbles. The portion could be anywhere between 0 to 100% but if Red is 45% then, Green would be 55% (remaining)

So here is the crazy part, if you are allowed to pick one marble at random from this urn and see its color, then there is 75% chance that this color is the colour of the majority.

So if you bet as every right call gets you $10 and every wrong call you pay 20$, the your AVG WIN PER BET would be 10x75% - 20X25% = 7.5 - 5 = 2.5$ win per bet,

So if you play this same bet 100 times, you have strong chance of making some dough

The Math for why this works is in Downloads.

What we think of Small Samples Vs. the Math for it

Stop complaining about small sample size. In order to appreciate the power of these small sample sizes, you need to remember how little we knew before the samples.

There are a lot more binary things with 50% probability in the world than you think - Coin Flips, Male Female, Yes No.

Human beings are not good on placing intuitions about random sampling. We are terrible at this. So its better to calculate them rather than make premature intutions about them.

Economic Objections to Measurements.

The author somewhere thinks that economic argument has some value so he places a key step in the process, which is calculation of the economic value of information.

Author has Excel Macro designed to find Economic value of the information. GET THAT!!!!

Usually we few things matter in business but they usually matter a lot.

What makes the measurement valuable?

  1. Preexisting level of uncertainity
  2. Importance of the decision.

If you are betting a lot of money on the outcome of a certain variable and there is a lot of uncertainity around it, then even marginal reduction in your uncertainity has a computable monetary value.

When some say that measurement is too expensive or too difficult, ask them compared to what?

Expert Judgement Vs Measurements, Measurements wins. A psychologists Paul Meehl tested this out. (Again, a psychologist)

Book Reference - Expert Political Judgement: How good is it? How can we know?

Objection to Usefulness of Measurement

You can prove anything with Statistics.

Meehl oftene noted that sometimes the objection to using stats boils down to nothing more than an irrational fear of numbers causing some to believe math somehow detracts from understanding or appreciation.

Ethical Objections to Measurement

Measurements can be perceived as dehumanizing because it compells to ask hard questions.

How many premature deaths? What is a premature death? Old people death vs Young kids disease?

Toothless Argument

You have Far more data than you think

Work with indirect observations.

Does insurance companies ever argue that they are unable to compute your insurance premium stymied by the lack of data on your death.

Book Recommendation : PULSE : THE NEW SCIENCE OF HARNESSING INTERNET BUZZ TO TRACK THREATS AND OPPORTUNITIES.

I need to read this book.

You need far less data than you think

When you know nothing, almost anything will tell you something.

Directly Stealing this -

Effectiveness of Online Teaching - They sampled randomly with some sampling plan a bunch of videos, 2 mins of it and watched that to assess.

Useful New Observations are more accessible than you think

When it comes to methods to gather new data, try working with the assumption that the first approach you think of is the "hard way" to measure.

Directly Stealing this

Satisfaction Survey for Opera Vs. Counting Standing Ovation -

You are not trying to get published in a peer reviewed journal. You are just reducing the uncertianity about the real life business decision?

Experimenting is a habit.

Philosophical Interludes

Bayesian Vs Frequentists

Ronald A. Fisher - Frequentist Biologist Vs. Harold Jeffreys - Bayesian Geophysicist and astronomer

Comments

Post a Comment

Popular posts from this blog

Chapter 6 Quantifying Risk through Modeling

Image
  Quantifying Risk through Modeling It is better to be approximately right than to be precisely wrong. HOW NOT TO QUANTIFY RISK Scoring method with all that colourful heatmap is utter bullshit. There is no evidence that it works and in fact there is evidence that it works in a bad way because it increases people's confidence in their risk management without actually managing risk. More on this perhaps in Hubbard's other book called "The Failure of Risk Management", which I will have to read someday. Insurance Premiums are good example of how rightly they quantify risk to a certain dollar value. Based on the level of risk you want to mitigate, your insurance premiums will be charged. They dont score risk in these instances as "Low" "Medium" or "High" LETS TALK ABOUT THE REAL SHIT - MONTE CARLO When you have a precise value of something it is easy to add, substract, multiply and divide them, but what do you do when you dont have a ...
Read more

Chapter 1 The Challenge of Intangibles

Image
  PEOPLE WHO GAVE US UNITS OF MEASUREMENT The book starts with a quote from Lord Kelvin, this is the guy who gave us the unit of measurement for Heat. I remember his weird experiment to measure heat from my school days. Its interesting that this guy is mentioned at the start of this book. In a way this quote is not just point to him, but everyone in the past who introduced any units of measurement. The whole concept of coming up with a unit of measurement requires a sophisticated understanding, so all these guys may be a potential mentor. Anyway the quote goes like... "When you can measure what you are speaking about, an express it in numbers, you know something about it; but when you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely in your thoughts advanced to the state of science." Basically Einstein said, unless you can teach it, you have not really learnt it.. and Lord Kelv...
Read more