math question

Curt Ashendel ashendel at aclcb.purdue.edu
Thu Jan 9 10:13:40 EST 1997


On 8 Jan 1997 21:17:32 GMT, 
Anthony Palombella  <palomb at beagle.Colorado.EDU > wrote:

>
>I have a collection of specific activity data, each with an SE.  For 
>duplicate assays of the same enzyme, how do I get an SD or SE for the
>average of the two specific activities?  In other words, how do I go from
>
>A1 +/- SE1, A2 +/- SE2
>
>to an average activity with its own SE?  

Although you need to understand the relationship of SE to SD to 
sample variance, as well as partial derivatives (trivial in 
this case), chapter 4 of the book "Data Reduction and Error Analysis 
for the Physical Sciences by Phillip R. Bevington (1969, 
McGraw Hill) describes how to propagate sample variation through any 
type of calculation. For the simple sum of two numbers the variance 
(SD squared) of the sum is the sum of the variances of the two numbers.

Or

Variance of (A1+A2) = variance of A1 + variance of A2

You need to convert your SE values to SD values, square them to get 
the variances, sum these squares and then take the square root to get 
the SD of the sum. This may be able to be converted to SE, but I am 
not sure what value of N to use.

Hope this helps.


Curt Ashendel
Purdue University, West Lafayette, IN
ashendel at purdue.edu



More information about the Methods mailing list