Any BITOGers know how to calculate Standard Deviation (statistics)?

[GON]

Sorry I had to post this thread here- I was unable to post to the general off topic forum.

I am taking some courses on-line at Northwestern University. I have always struggled with statistics. Seeing if a BITOGER might be able to look at the standard deviation chart I filled out. 76,359.7 seems awful high. And funny, the deviation subject is 2008 Crown Victoria crashing at 5mph.

 

[GON]

Found an online standard deviation calculator and the calculator shows = 92.129 as the standard deviation

 

[JeffKeryk]

Luckily I never had to code std dev. Programming languages like Transact SQL have built in functions. Now Linear Regression is another story...

 

[Gradient]

Standard Deviation Calculator.

 

[JayhawkRoy]

Correct first line to 29172.64, then re-do. An online calculator says you should get 103.0033980022. Calculator below shows the math so you can follow and find your errors.

I used https://www.calculator.net/standard-deviation-calculator.html?numberinputs=236,409,420,462,507&ctype=s&x=68&y=6

 

[GON]

Jayhawk R,

Thanks for finding my (huge) error in line one. It works now!!

Very grateful!!!!

 

[JayhawkRoy]

You're very welcome. Best of luck as you do your Northwestern studies.

 

[2strokeNorthstar]

I did 20 years ago in college. Never used it again. It was interesting to find out most news and polling is BS.

 

[JeffKeryk]

I did 20 years ago in college. Never used it again. It was interesting to find out most news and polling is BS.

I used it many times in business analytics.

 

[MNgopher]

I did 20 years ago in college. Never used it again. It was interesting to find out most news and polling is BS.

Umm... It is? Polling, when done correctly and understanding the limitations of the data is a useful tool. Most people zero in on exactly what they want to hear, and nothing else.

 

[2strokeNorthstar]

Umm... It is? Polling, when done correctly and understanding the limitations of the data is a useful tool. Most people zero in on exactly what they want to hear, and nothing else.

Polling is almost never done correctly was the point.

 

[bbhero]

Exactly ^^^^^^

Saw a great number of stupidly performed polls in my time...

They are like computer models in forecasting the weather.

Can be helpful when performed correctly and worthless when not done that way.

 

[PWMDMD]

Lol…I’m taking a business stats course for my MBA and I’m doing my homework as I post this. I make a table of mean, sum of squares, etc, and calculate variance and standard deviation.

So, the values of (X-u)^2 have to be plugged into the variance formula, THEN take the square root of that number to get S.D.


The sum of squares was 42428 then divide by 4 = 10609.7 and take sqrt = 103

 

[GON]

Lol…I’m taking a business stats course for my MBA and I’m doing my homework as I post this. I make a table of mean, sum of squares, etc, calculate variance and standard deviation.

PWMDMD,

Your a doctor- doing statics as part of an MBA is as easy as a hungry ten year old eating a nice pancake.

BTW- meant to call you out when you said you buy toothpaste just like every other "joe" at the store. We all know you have seven thousand tubes of tooth paste samples in your home.

Finally, I buy Aim toothpaste at Wal-Mart. Bottom shelf hard to find, but only .99 for the big tube!

 

[Number_35]

@GON, one of the better bits of advice I ever got w.r.t. using a calculator is to do some simple mental math first - in this case, you're squaring -170.8.

So do 100 x 100 in your head. Easy, it's 10,000. You know your answer has to be greater than that.

Now do 200 x 200. That's 40,000. Your answer has to be less than that.

So you know the answer will be between 10,000 and 40,000. (Your 292,172.64 isn't in that range, and that should be a red flag.)

You can easily get closer though. 170.8^2 shouldn't be that far off from 150 x 200, which is 30,000.

Closer yet ... 17^2 = 289, so 170 x 170 will be 28,900.

But you're going a little over 170, so figure on the answer being a bit over 29,000.

Now crunch it on your calculator - 29,172.64. That extra digit in there threw you out almost 10x.

Anyway, false key closures happen, so do some quick mental math first to get a ballpark estimate.

Best of luck in the course!

 

[GON]

Number 35,

Thanks so much for taking the time to write/post some awesome tips/techniques. I am grateful.

 

[chemman]

Arggggg... Haven't had to think about statistics for a good long time and glad that I always had a calculator or computer program (SigmaPlot or Excel) to do the nitty gritty calculations along with confidence limits, t-tests, etc.

 

[Number_35]

Number 35,

Thanks so much for taking the time to write/post some awesome tips/techniques. I am grateful.

GON, you contribute so much excellent stuff to the forum that so many of us appreciate, so it's a pleasure to be able to give a little bit back.

 

[GON]

Arggggg... Haven't had to think about statistics for a good long time and glad that I always had a calculator or computer program (SigmaPlot or Excel) to do the nitty gritty calculations along with confidence limits, t-tests, etc.

Cm,

The only class I failed in college was statistics. Every one in our class failed. Our professor was super nice, but we were his first class. We all ended up passing, but we all failed his class.

I am.sure statistics is learnable for a knucklehead by me, but will take a true effort and I am not so motivated to learn..but I have decided to learn as I have four more weeks of statistics and want to not fail the subject twice....

 

[GON]

GON, you contribute so much excellent stuff to the forum that so many of us appreciate, so it's a pleasure to be able to give a little bit back.

Thanks n35, but I am sure you have me confused with another member of BITOG. But thanks so much, it is a pleasure to be around so many good people that bring so much knowledge to one place. I am a student here.