Jarmo

IMHO, your problem is that you do not have a statistically valid set. If you plot a
histogram of the R data, you will see that the data is almost random, so that a mean is
meaningless (pun intended!). The mean is 6.58 but the standard deviation 4.10, which is
getting precious near to the mean and the histogram has a fractile skew of 0.62. The
0.95 fractile confidence level is only 2.32. I would expect your SIR results would
produce something approaching a Gaussian curve. The fact that it does not is proof that
either a) your sample is too small for the expected range or b) there are uncontrolled
random factors somewhere along the line. I would opt for b), even though the sample is
not humengous.

If you wish to go really technical, you can use the kurtosis (no, not a disease, unless
it is in the head of statisticians!). This determines the closeness to normality and
the peakiness. A very peaky normal curve will have a kurtosis value approaching the
number of samples. Your data has a platykurtic value of -0.54, which indicates a flat
distribution with little relation to normality. If you wish to use it, I think that
some of the popular spreadsheet software has a KURT function but you will certainly
find it in all the mathematical softwares.

As I said previously, far better would be to calculate the best curve fit for each
sample of R v. t and then average the factors to plot an average curve. However, this
also presupposes that the data are valid, with a reasonably small SD.

Hope this helps

Brian

Kiiski Jarmo wrote:

> Than you all that reply to my question.
>
> Here is an example:
> 12 patterns were energized with 15 VDC and the current of every individual
> were measured with following results (in nAmps):
> 7.90; 1.71; 6.80; 1.23; 3.00; 2.63; 2.90; 1.04; 7.60; 1.54; 3.90; 1.87.
> Equivalent resistance been (in GOhms):
> 1.90; 8.77; 2.21; 12.20; 5.00; 5.70; 5.17; 14.42, 1.97; 9.74; 3.85; 8.02.
>
> Arithmetic mean:
> Arithmetic mean of resistances is 6.58 GOhm.
> Arithmetic mean of currents is 3.51 nA and equivalent resistance is 4.27
> GOhm.
>
> Geometric mean
> Geometric mean of resistances is 5.36 GOhm
> Geometric mean of currents is 2.80 nA and equivalent resistance is 5.36
> GOhm.
>
> Three different results and only one of them is right.
>
> Scope of this example was only to present different methods for calculating
> the average resistance of a sample in SIR-testing -nothing more ore less.
>
> B.R.
>
> Jarmo Kiiski
>
> ##############################################################
> TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
> ##############################################################
> To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
> the body:
> To subscribe:   SUBSCRIBE TECHNET <your full name>
> To unsubscribe:   SIGNOFF TECHNET
> ##############################################################
> Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
> information.
> If you need assistance - contact Keach Sasamori at [log in to unmask] or
> 847-509-9700 ext.5315
> ##############################################################

##############################################################
TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
the body:
To subscribe:   SUBSCRIBE TECHNET <your full name>
To unsubscribe:   SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Keach Sasamori at [log in to unmask] or
847-509-9700 ext.5315
##############################################################