evaluation of a new machine against one in current use

[moonya]

hie guys i need a bit of help here. i am currently evaluating a new laboratory machine against one that is currently in use. i took 20 different samples and measured 30 different analytes in the sample using both machines. so basically i have 20 readings of each analyte for both machines. how can i analyse this data in such a way that i can show whether the instruments are in agreement or not and present the data to a group of people whose understanding of statistics is limited.

[moleculardude]

Hello there! and welcome to the forum!!

I would use Excel an easy and quick way to analyze statistics.

input all 20 values from both machines into cells in the rows.

each machine should have one row.

then you can do this:
in a cell put

=(click on function average - and select the whole row of your values )

do this for each row to get average of each

then do in a new cell underneath each row:


=(click on function SD standard deviation - and select the whole row of your values )
this gives you error

then do a T-Test in a cell:
= (click on T-test function Fn and do a t-test between the two rows.)

if its less than 0.05 they are similar and not differently significant

Cheers!