Sample

D

David

I have a list of 2730 numbers in a worksheet, used Sampling in the Tool menu
area, my question is that it picks some of the numbers twice and puts that as
a part of the sample. I need a output of 250, I get the 250, but with some
duplications. Any help you can provide would be greatly appreciated.
 
G

Gary's Student

I ran sample on a list of 700 numbers. The numbers were sequential, however.
I also got multiple hits. It appears that this tool samples with
replacement. That is each sample is drawn from the population independently
from previous draws.

This means that not only is it possible for a number to come up more than
once, it is statistically inevitable.

One approach would be to select, say, 270 or more samples. Then use

Data > Filter > Advanced Filter...
and select unique records only. Keep the first 250.


Good Luck
 
D

Dave Peterson

Another option...

insert a helper column
put =rand() in the top row
and drag down the column.

Sort your data by that column and copy the first 250 in the sorted list.

You could use another column and put
1 (in the top row)
2 (in the second row)

And drag down that column.

If you need to resort to get it back to the original order, sort by that other
helper column.
 
D

Dave Peterson

ps.

put 1 in the first row
put 2 in the second row

Select both cells and drag down the column.

(I left out that part about selecting both cells.)
 
Top