Are_different

Are different

Purpose

Experiments are usually done to compare the average response or the proportion of responses among two or more groups. For example, you may wish to compare the mean weight gain of two different feeds given to livestock in order to decide which is better. In these cases, there is no particular value of interest for each feed-type; rather the difference in performance is of interest.

• Paired t-test In many applications it is natural to make two measurements of essencially the same kind. For example, measure a FOM over two reconstructions obtained with two different reconstruction algorithms applied to the same data set. This test checks if we can reject the hipothesys that the two algorithms performs equally. Notice that we do not probe that the two algorithm perfors equally. If you are interested in this test execute:

$ are_different-i Doc_File_Name -min_sig [min significance=0.995]

• Small-sample Comparations... this method is similar to the above one but does not assume that the data is paired. the input is just the mean and the standard deviation of both populations. For this test execute:

$ are_different [N1 avg1+-dev1] [N2 avg2+-dev2] -min_sig [min significance=0.95]

Usage

$ are_different ...

Parameters

• `` Docfile with the data
• -min_sig [min significance0.95] = Minimum value that the significance must have in order to say two populations are different

$ are_different ...

Parameters

• ``
• ```N1 is the number of samples in population 1,avg1` is the average value in population 1, and`dev1` is the standard deviation value in population 1. Notice that there cannot be any space between the average and the standard deviation
• -min_sig [min significance0.95] = Minimum value that the significance must have in order to say two populations are different

Examples and notes

Compare two paired populations:

$ are_different  -i kk1.doc

The result is:

I would NOT say they are the different
Unless you relax your Significance level to 0.873007 

wherekk1.doc is a file that looks like:

; dat/dat   18-Jul-94  AT 14:29:48
1  2   31.4   28.1
2  2   37.0   37.1
3  2   44.0   40.6
4  2   28.8   27.3
5  2   59.9   58.4
6  2   37.6   38.9 

and the data are the 3rd and fourth columns

Compare two non paired populations:

$ are_different 13 6.8+-1.8 15 5.3+-1.6 -min_sig 0.95 

The result is

I would say they are differet
Up to a significance level of 0.972469 

--Main.AlfredoSolano - 31 Jan 2007