Status Updates From How to Lie with Statistics

Statistically Significant

3. Don’t include your sample size/keep running your experiment until you get the result you want

4. Standard deviations might mean differences are not statistically significant

Quick and easy read so far.

Nothing earth shattering for today’s readers I don’t think... these tactics are well known and understood by most these days I believe.

Quick recap:

1. How you sample your demographic will give different answers - who is included/excluded

2. Average - mean/median/mode? Use whatever is convenient.

Public pressure and hasty journalism often launch a treatment that is unproved, particularly when the demand is great and the statistical background hazy. So it was with the cold vaccines that were popular some years back and the antihistamines more recently. A good deal of these unsuccessful "cures" sprang from the unreliable nature of the ailment and from a defect of logic.

Just read the first chapter. About random samples and how the way the data is collected skews the results.

For example, polling at night could exclude people who go to movies, or work evenings etc

I am really loving the book!

If I didn't knew this already, it would have been an absolute fun to read.

o que dizer desse livro que nem terminei e já considero pacas?

«Существуют три вида лжи: ложь, наглая ложь и статистика. — Бенджамин Дизраэли»

More often than not, false impression hide the real facts. The figure/number was there, people just choose to show what favours the readers and in the end would benefit them the most.

The chapter I've just read reminds me of this line from Designated Survivor: "I believe that nothing in this world is more valuable than data, applied with responsibility and interpreted honestly"

What stopping us from doing that?

am into data science these past months and to learn it without solid fundamental concepts of statistics is definitely useless; i will not going anywhere as statistics is data science building blocks (well, statistics by definition is actually data science itself. people just prefer techy-er and fancier name to make it less intimidating, no?). almost halfway of this book and me taking notes✍🏻

the word "average" is INDEED having a very loose meaning tho

para escuela

today

before. do not know when

The fact is that, despite its mathematical base, statistics is as much an art as it is a science. A great many manipulations and even distortions are possible within the bounds of propriety. Often the statistician must choose among methods, a subjective process, and find the one that he will sue to represent the facts.

Misinforming people by the use of statistical material might be called statistical manipulation; in a word (though not every good one), statisticulation.

To avoid falling for the post hoc fallacy and this wind up believing many things that are not so, you need to put any statement of relationship through a sharp inspection. The correlation, that convincingly precise figure that seems to prove that something is because of something, can actually be any of several types.

... the probable error of the Stanford-Binet IQ has been found to be three percent. This has nothing to do with how good the test is basically, only with how consistently it measures whatever it measures.

How accurately your sample can be taken to represent the whole field is a measure that can be represented in figures: the probable error and the standard error.

Knowing nothing about a subject is frequently healthier than knowing what is not so, and a little learning may be a dangerous thing.

How results that are not indicative of anything can be produced by pure chance- given a small enough number of cases- is something you can test for yourself at small cost.

The basic example is the kind called "random." It is selected by pure chance from the "universe," a word by which the statistician means the whole of which the sample is a part. Every tenth name is pulled from a file of index cards.

Like many a more sophisticated statistic it as guilty of semiattachment: It assumed that newspaper space given to crime reporting is a measure of crime rate.

Very basic statistics, not as interesting as I thought it’d be

More people are killed in the evening than in the morning simply because more people are on the high ways then to be killed.

The guilt does not always lie with the medical profession alone. Public pressure and hasty journalism often launch a treatment that is unproved, particularly when the demand is great and the statistical background hazy.

Picked this up again after watching Moneyball :)
Very interesting, light read, particularly for those starting with either statistics, data science, data analysis or even UX research. Easy to read and opens your eyes to subtleties you maybe wouldn't see otherwise.
Wish I read it in college before my Statistics II classes kicked in!

1954-ais išleista, bet tokia gera, labai smagu ir linksma skaityti. :D

10 pages read in 12 min, via @leioapp (http://leio.co).