Pivot tables and Standard Deviation
 

I have columns of data that decrease in value that I collate using a pivot
table.

Table
Object Run Time Mass
100 0 36.1791
100 1 36.1709
100 2 36.1617
100 3 36.1527
100 4 36.1449
100 5 36.1367
100 6 36.1292
100 7 36.1209
100 8 36.1138
101 0 36.2029
101 1 36.1931
101 2 36.1837
101 3 36.1756
101 4 36.1682
101 5 36.1604
101 6 36.153
101 7 36.1455
101 8 36.1373
102 0 34.5683
102 1 34.5598
102 2 34.5515
102 3 34.5439
102 4 34.5366
102 5 34.5285
102 6 34.5215
102 7 34.5142
102 8 34.5066

After setting up the pivot table with 'Mass' as the target, I double click
the 'Count of Mass' and modify the table to my needs. ie Average, StdDev and
so forth.
Using the "PivotTable Field" -options I selected Difference from (previous)
for in this particular case "Run Time".

Problem.
If I use the StdDev function I get negative numbers showing up in the table.
Since standard deviation is by definition the root of a square, I find the
result disturbing not to mention a little disconcerting for my research since
I have been depending on Excel for other number crunching statistics.

What is going on?

Tom

Pivot tables and Standard Deviation
 

Hi tom,

You are seeing the difference of StdDev(at RunTime t) from StdDev(at RunTime
t-1), so yes you will get negative numbers.

Ed Ferrero
http://edferrero.m6.net/


I have columns of data that decrease in value that I collate using a pivot
table.

Table
Object Run Time Mass
100 0 36.1791
100 1 36.1709
100 2 36.1617
100 3 36.1527
100 4 36.1449
100 5 36.1367
100 6 36.1292
100 7 36.1209
100 8 36.1138
101 0 36.2029
101 1 36.1931
101 2 36.1837
101 3 36.1756
101 4 36.1682
101 5 36.1604
101 6 36.153
101 7 36.1455
101 8 36.1373
102 0 34.5683
102 1 34.5598
102 2 34.5515
102 3 34.5439
102 4 34.5366
102 5 34.5285
102 6 34.5215
102 7 34.5142
102 8 34.5066

After setting up the pivot table with 'Mass' as the target, I double click
the 'Count of Mass' and modify the table to my needs. ie Average, StdDev
and
so forth.
Using the "PivotTable Field" -options I selected Difference from
(previous)
for in this particular case "Run Time".

Problem.
If I use the StdDev function I get negative numbers showing up in the
table.
Since standard deviation is by definition the root of a square, I find the
result disturbing not to mention a little disconcerting for my research
since
I have been depending on Excel for other number crunching statistics.

What is going on?

Tom