Searching \ for 'Fun with Statistics - Which is better?' in subject line. ()
Help us get a faster server
FAQ page: techref.massmind.org/techref/index.htm?key=fun+with+statistics
Search entire site for: 'Fun with Statistics - Which is better?'.

Truncated match.
'Fun with Statistics - Which is better?'
2003\04\17@140948 by

Lawrence Lile wrote:
{Quote hidden}

All will depend on the kind of test.
If the group of 10 tests are exactly the same, then it is not 20 tests of
10, it was simply 200 tests, and the result is 113 x 132.

Statistics are very tricky.  Suppose the lotto number 35 was not drawn a
single time during the last 52 Saturdays, what are the chances of the ball
#35 to be drawn next Saturday?  Exactly the same as any other ball... since
the past results does not interfere in the next draw, everytime happens to
be a fresh brand new opportunity to all the numbers.  Except of course if
the ball #35 is heavier, lighter, bigger, smaller, etc, what is not the
case, since they always replace the whole set of balls every Saturday for
new ones never used, exactly to avoid this kind of crazy thoughts.

If your result is basically 113 x 132, there is no tricky stat wizard that
will convince you otherwise, but in your place, I would love to see the
explanation why the Unit#2 made 10 in the first and third, and only 2 in
the second group of tests... that would be an interesting explanation...

Wagner.

--
OK here is the lowdown.  These were 200 tests of how accurately I could
juggle two sharp knives and a bottle of nitroglycerine.  Each miss
represents one trip to the hospital.   ;-)

Actually (seriously) the tests are of two barcode scanners.  Each group of
ten was ten consecutive scans of the same product.  In a successful test
the scanner read the UPC code in one pass.  Some packages have different
color ink, wrinkly bags, shiny surfaces, others are flat black ink on
cardboard boxes.   So yes, the sets of ten are *different* but the idea is
to get an overall mix of products that might represent a shopping basket
full.

Of course, the consequences of a miss are not catastrophic, just annoying.
And there is another annoying thing, some products will not scan at all
with either scanner.

I guess my statistics question is this:  Given that the standard deviation
of the data is over 30%, is a difference of  9% between one scanner and
the other *signifigant*?   Or is this fact even relevant?

-- Lawrence Lile

Wagner Lipnharski <wagnerUSTR.NET>
Sent by: pic microcontroller discussion list <PICLISTMITVMA.MIT.EDU>
04/17/2003 01:07 PM
Please respond to pic microcontroller discussion list

To:     PICLISTMITVMA.MIT.EDU
cc:
Subject:        Re: Fun with Statistics - Which is better?

Lawrence Lile wrote:
{Quote hidden}

All will depend on the kind of test.
If the group of 10 tests are exactly the same, then it is not 20 tests of
10, it was simply 200 tests, and the result is 113 x 132.

Statistics are very tricky.  Suppose the lotto number 35 was not drawn a
single time during the last 52 Saturdays, what are the chances of the ball
#35 to be drawn next Saturday?  Exactly the same as any other ball...
since
the past results does not interfere in the next draw, everytime happens to
be a fresh brand new opportunity to all the numbers.  Except of course if
the ball #35 is heavier, lighter, bigger, smaller, etc, what is not the
case, since they always replace the whole set of balls every Saturday for
new ones never used, exactly to avoid this kind of crazy thoughts.

If your result is basically 113 x 132, there is no tricky stat wizard that
will convince you otherwise, but in your place, I would love to see the
explanation why the Unit#2 made 10 in the first and third, and only 2 in
the second group of tests... that would be an interesting explanation...

Wagner.

--

--
{Quote hidden}

No, on the surface they both like like crap.  Throw em both back.

"But ma'm, you didn't get electrocuted the other nine times."

*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
Olin sez,

>No, on the surface they both like like crap.  Throw em both back.

>"But ma'm, you didn't get electrocuted the other nine times."

That's pretty funny, Olin!   Hopefully the mfr. can incrementally improve
these until they don't look like ^\$%#&\$%^.

-- Lawrence Lile
Senior Project Engineer
Toastmaster, Inc.
Division of Salton, Inc.
573-446-5661 voice
573-446-5676 fax

Olin Lathrop <olin_piclistEMBEDINC.COM>
Sent by: pic microcontroller discussion list <PICLISTMITVMA.MIT.EDU>
04/17/2003 02:25 PM
Please respond to pic microcontroller discussion list

To:     PICLISTMITVMA.MIT.EDU
cc:
Subject:        Re: Fun with Statistics - Which is better?

(978) 742-9014, http://www.embedinc.com

--

--
Hi,

to answer correctly, the shape of the distribution should be also checked.
I do not guess they are Gaussian, so some deductions should be taken very
carefully. I suggest other methods, especially q.c. ones such as X-bar and
Pareto charts. Nonetheless, a Mann-Whitney test would not harm, too. At
the time I`m overloaded but I will maybe look at your data...

Imre

+-----------------------------------------------------------------------+
| The information transmitted is intended only for the person or entity |
| to which it is addressed and may contain confidential and/or          |
| privileged material.  Any review, retransmission, dissemination or    |
| other use of, or taking of any action in reliance upon, this          |
| information by persons or entities other than the intended recipient  |
| sender and delete the material from any computer.                     |
+-----------------------------------------------------------------------+

On Thu, 17 Apr 2003, Lawrence Lile wrote:

{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics
In another life I spent a lot of time with optical recognition equipment.
Reject rate is not usually the best evaluation criteria.
Substitution rate is the most dangerous, it gives you bad data.
If the media is out of spec and yet read by the device that is a fault,
substitutions are not far behind!

Another way of saying it is that if you are not sure, it is a reject.

Banking check scanners and postal sorting machinery are a couple of examples
where you need to monitor your reject rate carefully.
Lower rejects usually indicate trouble!

{Original Message removed}
>
>Another way of saying it is that if you are not sure, it is a reject.

Absolutely.  In the case of UPC/EAN, you have a check digit, and well

>Banking check scanners

BTDT, wasn't fun.
E13B was designed by psychotic monkeys on bad drugs.  CMC-7 is far better.

>and postal sorting machinery are a couple of examples
>where you need to monitor your reject rate carefully.
>Lower rejects usually indicate trouble!

Only if that means substitutions are happening.
Subs are the worst sort of error in either area.  People get pretty upset
when you put their money in the wrong account, or charge \$350 for a carrot. :(

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

More... (looser matching)
- Last day of these posts
- In 2003 , 2004 only
- Today
- New search...