Re: mean vs average

Okay, now explain standard deviation, please.

Joseph Marton wrote:
> On Wed, 14 May 2008 19:27:58 +0000, Doug Black wrote:
> 
>> mode is... something I
>> always have to look up.
> 
> The mode is simply the value(s) which appear(s) with the most frequency in
> a list.  Then there's the midrange, which is where you take the highest
> value added to the lowest value and then divide by two.  Or there's just
> the range which is the lowest value subtracted from the highest value.
> 
> Yes, I just took my stats final on Monday night.  We'll see how quickly
> this stuff disappears from my brain.

In <VNHWj.12605$Dh4.1936@kovat.provo.novell.com>, Mary Wood wrote:
> Okay, now explain standard deviation, please.

For each data point, find how far it is from the mean value -- square
that value, and find the average of those squares.  That gives a figure
called the variance of the dataset.

The standard deviation is the square root of that figure, and so gives
a measure of how much the data is spread about the mean.  The smaller
the standard deviation, the more clustered around the mean the data is.

You can then use this figure to, for instance, determine which points
lie within two standard deviations of the mean, so that you discard any
freak outlying points.

... at least, that's how I remember it working :)

-- 
Regards,
Ben A L Jemmett.

Mary Wood wrote:

> Okay, now explain standard deviation, please.

There are two ways.  First there's Ben's way.  Then there's my way. 
Take your TI-83 Plus, enter the list of values, then hit stat, calc, 
1-Var stats, the list (e.g. L1), and enter.  You'll now have both the 
sample standard deviation and the population standard deviation.  Square 
each of those and you'll have the respective variances.

-- 
Joe
"Those of you who think you know everything are annoying to those of us 

In <W6JWj.12642$Dh4.6318@kovat.provo.novell.com>, Joseph Marton wrote:
> There are two ways.  First there's Ben's way.  Then there's my way.
> Take your TI-83 Plus, enter the list of values, 

LOL -- yes, although I used a Casio.  One of the earliest colour
models, after my lovely fx-7000GB got squashed by a classmate.  I miss
that calculator...

(The long way was the only way when it came to exams, though.  I scored
148/150 on Statistics I at A-level; where those two went has worried me
ever since.  That was only eight years ago and I've forgotten most of
it already :(  Oh well.)

-- 
Regards,
Ben A L Jemmett.