Estimating the standard deviation from a histogram/boxplot

Let me know in the comments section below what other videos you would like made and what course or Exam you are studying for! (Ans: Range/6 = (Max value - lowest value)/6)
Use the 68-95-99.7% rule to estimate the standard deviation.
Ans: Range/6 = (Max value - lowest value)/6
Another approximation is Range/4. BUT if question states using 68-95-99.7% rule you would use method as in video.
I believe Range/6 is a better approximation. If you calculate the Range/4 you are essentially finding 25% of data and saying that this number covers 50% data below and above the mean. BUT We know that 68% of the data lies within one standard deviation above and below the mean. So Range/6 is a better approximation.

[@Aussiebuck]

big up you for making this means alot, even 7 years later still helpful and descriptive.

[@DanielDJB]

Thanks, glad it is clear for most people :)

[@Picplosions]

This helped me so much dude, you have a nice teaching style :)

[@DanielDJB]

Michael Geros thank you. Nice to hear it helped

[@edwardnelson9130]

awwwwwwww

[@rainapatel1899]

LOVE YOU DUUUUDEEEEEEEE ROCK ON MY GUYYYYYYYYY

[@chriscothran8744]

Who was yelling in the background? 😅

[@thirteennote3206]

Ur mum

[@balls4924]

Good stuff! It really helped me for my engineering class!

[@FiveMissiles]

thanks bro u the man!

[@triglyceryde4439]

Good Job i learnt sooo much

[@hannananan9427]

Amazing

[@zanerubaie3226]

Thanks, this helped with Further Maths

[@asfandyarimtiaz7322]

Thank you so much. Exactly what I wanted!

[@DanielDJB]

Just wondering what subject you needed this for?

[@asfandyarimtiaz7322]

@@DanielDJB data interpretation section for GRE

[@DanielDJB]

@@asfandyarimtiaz7322 Many thanks. Just wondering as a lot of people from U.S have watched. Hope you went well!

[@isaacd.380]

i have no idea what is going on

[@aoa0687]

me too...

[@eggyolk8067]

If there are outliers in the box plot, do I need to factor them in?

[@MRCMatheus]

Thanks for the video! Helped a lot :)

[@satindollyxx]

the background noise is insane

[@adrianli916]

Best video ever!!!!!!

[@stephhgreig1117]

This makes no sense to me... why can't I understand this?! I hate statistics!! Aggggh!!!

[@erenmateo8186]

Felt that

[@saku-ny2ob]

real asf

[@shriennathsevva852]

whole lotta turbulence

[@guibertthechampion2652]

Thank you.

[@DanielDJB]

Glad it helped.. Just wondering what subject you needed this for?

[@leejohnson197733]

i am trying to sketch a normal distribution graph based on a mean of 21.8k and sd of 1k with a range of 19-24k with a nominal value of 22k any advice? i have worked out the numbers but not sure how to show on a bell graph
i worked out 19-21.8/2 = -2.8 and 24-21.8/2 = +2.2 i need to show 2 graphs and compare, changing the sd to 21.5 and see which one gives better results closer to 22k.

[@kichwachagoogle5138]

mmmm i am still do not understand

[@paperplate9301]

Very helpful thank you

[@joshuavanname1246]

Thank you... subscribed

[@DanielDJB]

Glad it helped.. Just wondering what subject you needed this for?

[@helena-zd6sw]

if your lower limit was a negative number and the mean was 0 would it be
(60-(-60))/6?

[@DanielDJB]

I think you mean smallest value is -60 and average/mean is 0. So because of symmetry the "upper limit" would be +60. So estimate would be 120/6.

[@helena-zd6sw]

@@DanielDJB thank you!!

[@michaelcui]

ty bb

[@kol2456]

wouldn't you divide by 4?

[@DanielDJB]

That is another approximation/estimate but if question states using 68-95-99.7% rule you would use method as in video. I believe Range/6 is a better approximation. If you do Range/4 you are essentially finding 25% of data and saying that this number covers 50% data below and above the mean. BUT We know that 68% of the data lies within one standard deviation above and below the mean. So Range/6 would be better.

[@melakjenkins8696]

lol this is the exact same question in my further book

[@2921mozart]

everyone teaching me math on youtube are mostly foreigners

[@jedholckner4734]

How dare you disrespect the Brosche

[@edwardnelson9130]

Yeah you tell 'em!

[@Mangko]

how is it even possible to define foreigner on the internet lmao

[@kanivakil198]

@ JOVANTZL Yeah, it's either some fucking Indian accent or some accent for the UK or Australia or New Zealand. Like fuck!

[@kanivakil198]

@Robert O' Conner; it's anyone that doesn't speak with an American accent. It means the RUclipsr is using a foreign accent.

[@silverwolftg5498]

Was this in the Further Maths exam? Which exam was that?

[@DanielDJB]

I am pretty sure it is from a Further Exam

[@vampiredutchess1]

You lost me at 1:10 😭

[@DanielDJB]

ruclips.net/video/cgxPcdPbujI/видео.html

[@DanielDJB]

This guy is good at explaining concepts. Basically a lot of normal data follows this curve.. think heights of 14 year old boys or something similar.. they may vary between 140 to 180 cm. And the standard deviation is a number which nreaks up this data into intervals..

[@AdhavR7]

I hate this unit so much I don’t understand anything

[@DanielDJB]

Maybe the easiest way of thinking about standard deviation is a means of how far spread out the data is.
Consider 10 data points
Zero standard deviation
100, 100, 100, 100, 100, 100, 100, 100, 100, 100
The mean is 100
The standard deviation is zero (no variation about the mean - there is zero range)
small standard deviation
98, 98, 98, 99, 100, 100, 101, 102, 102, 105
The mean is roughly 100 and the spread (range) about the mean is small
Larger standard deviation
10, 20, 60, 80, 100, 100, 120, 130, 140, 140
The mean is roughly 100
But because data is more spread out, the standard deviation will be larger.
The 68, 95, 99.7% rule tells you about how much data is within the intervals.
Consider a selection of 100 people with normal distribution
If the average height is 150 cm and a standard deviation of 10 cm. This means about 68 people (68% of 100) will have heights between 140 and 160 cm.... and so on....

[@lukehannebery9953]

this vid was bad ngl

[@joeybertalli4464]

dont be dog it was best

[@exikroo7000]

come on doggy, dont dog the Brosche like that ...

[@tomgenovese3320]

get off my boy ya dog

[@latin4248]

get off my guy Broche