Useful Statistics: How to Not get Tricked by Numbers

If you were wondering where I was during the past two or three months, wonder no more!

My new title “Useful Statistics: How to Not get Tricked by Numbers” is now available on Both Amazon Kindle and CreateSpace!

This is an introductory book to statistics especially targeting crap that we see on social media. my favorite caption is “From tumblrite feminists to presidential candidates no one is safe from criticism!”.

Createspace link: https://www.createspace.com/5988397

Amazon Kindle link: http://www.amazon.com/dp/B01AILMCBK

 

Cover ready

Advertisements

The Foot Soldiers of Illogic: The Failure to Understand Statistics (2)

The following picture appeared on my Facebook feed some time ago, and is a perfect example of how when it comes to statistics, many go astray.

Average fail

I originally saw this post on the Facebook page “The LAD Bible”, posted on June of 2014, and now it has almost 350K likes.

Reading the comment section is amusing for someone who understands this mathematical average is complete nonsense. A common theme of the comments are “Math bitch!”, “Science rules!”. Not that I expect much intelligence in the comment section of FB, but still, pretty disheartening.

One of the top comments is “This should be the type of shit they put on exam papers, not bullshit algebra.” It seems to me that our dear commenter is incapable of understanding “this type of shit”, thus calls Algebra “bullshit” and sadly at the end fails in both statistics and mathematics exams. Another commenter suggests “Math.. the silent killer of swag 🙂 ” Indeed, I suspect he means his own “swag”, if he had any.

Simply put, the conclusion is absolute nonsense. “You guys had 3.5 pints each day” is completely false. Instead, “You guys had 3.5 pints on average* each day” is correct. Then we clearly see what went wrong: The average of a set of data does not say anything about its distribution, and therefore is irrelevant here. Without knowing the distribution of drinks in each day between each person calculating its mean is misleading and nonsensical. The group of boys could have drunk on the first day and the last day, and not anything in between. Also, some might have had more drinks than others. In which case the sarcasm of the last sentence simply dies: It very well could have been wild.

____________________________

* Average = Mean

The Foot Soldiers of Illogic: The Failure to Understand Statistics (1.1)

This is simply a follow up on my previous post regarding the failure of some of the so called progressive and SJWs regarding statistics and logical reasoning. Our good twitter user @shezumi has the following tweet for us:

@shezumi_statistics_fail

Again, just like the post I have linked, this is a failure to understand statistics and logic. Let me make it very simple:

(a) “Most white animals are Not rabbits”
(b) “Most rabbits are white”

=> (a) being true does not necessarily prove (b) false.

In the same way, “Most rapists [in US] are white” does not prove “Most Mexicans are rapists” false. You need another set of data for that. The simplest of those is the ratio of say [in this case] Mexican rapists divided by total number of Mexican immigrants.

//

Again this is not to take the side of Trump, but one wonders if progressive cannot do simple logic, then who is worse?

The Foot Soldiers of Illogic: The Failure to Understand Statistics (1)

Anger is a double edged sword, it could derive one to pursue different subjects passionately, but also blind that person to ignore or simply not understand the facts of the matter. Issues of racism in US have much anger and passion surrounding them, and thus many angry fools on all sides making nonsensical claims.

Read the case of our good blogger, Ziggs Unscripted [1]:

Kill the myth that Mexicans and Blacks cause the most rape in the USA

I’ve been pretty mad the last couple of days.  From Donald Trump saying that all Mexicans are rapists and criminals (except that few “good ones”) to that white terrorist saying that blacks are rapists that are taking over this country.

I’ve done some digging and apparently white men are more likely to rape and cause sexual assault in this country.  Not Blacks, Mexicans, Asians, etc. but White men.  And its hardly a contest about 65% of rapists in 2012 were white. Yea just let that sink in.

Here are some Sources: https://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2012/crime-in-the-u.s.-2012/tables/43tabledatadecoverviewpdf

http://www.sarsonline.org/resources-stats/reports-laws-statics

https://www.rainn.org/get-information/statistics/sexual-assault-offenders

http://ovc.ncjrs.gov/sartkit/about/about-sa-so.html

One may not dig, when one does not have the understanding of a high school student to interpret and use the findings to get a meaningful conclusion.

First, one must define the ideas being confirmed or debunked very clearly. We should ask, “What” is actually being debunked? The author of that post wishes to debunk “Most rapes are caused by non-whites”. This is clear from the title, however, at the start of the article we clearly see another claim, namely that of Mr. dead-fox-on-empty-head, Donald Trump and other racists: “All (or most) non-white people are rapists”.

It must be shown to the author of the above post that debunking the first claim does not debunk the latter. In other words, one can prove without contradiction that: “Most rapists are white” is true, while “Most non-whites are rapists” is also true. The truth of the second statement can only be determined when we compare the number of non-white rapists in proportion to their respective population.

This ratio will mathematically translate to the total number of rapists of each ethnicity divided by the total population of that ethnicity, all definitions kept the same.

The author of that post must stop being mad about things that bother him/her, and start thinking more clearly. Maybe take some time to study a bit more about how to analyse statistical facts.

//

Note: These numbers, even corrected, mean nothing on their own. Too many variables (from social to criminal justice system) are involved, and to my knowledge no study links race and violent sexual behavior directly.

_____________________________

[1] http://ziger123.tumblr.com/post/121936726589/kill-the-myth-that-mexicans-and-blacks-cause-the

“Statistics this, Statistics that, Therefore Sexism…”

There are many groups that are very bothersome in this world, religious fanatics, dishonest politicians and intellectual impostures (read the latter as relativists) to name a few. These people do harm, whether it be by means of terrorism, or outright dishonesty, intellectual or otherwise. But there is one group that cannot be seen very easily, and yet exists and persists in its harmful endeavour. But which group is this? Well, not a specific one. And sadly this is the whole issue. This is a group of torturers, not that of living beings, but numbers.

Logically speaking, we must be very careful when we derive conclusions from our premises. Especially, we must care to see if the conclusion of the argument follows the premises. Most of the time it is not very difficult to see bad arguments that do not follow the rule, but when it comes to numbers, the issue becomes much harder to detect.

The following video can demonstrate how facts could be abused, and how one can potentially tackle the issue:

And if you think that this has changed from some decades ago… Well, not so much:

Bad arguments are bad, coming from left or right makes no difference. The right statistics must be used to demonstrate the relevant conclusions. And even before that, the correct method must be chosen to conduct the best argument.

______________________________________________________

Note: I did not present the post to bash Feminism, but to point out something that many are guilty of. However, I have observed that many who call themselves Feminist fall into this category of irrelevant fact propagation.