Linearity is the ability of the test method to provide results proportional to the concentration of microorganisms present in a sample across a given range. Using a combination of the compendial organisms and routine isolates, cultures are prepared and diluted across the useable range of the methods being compared. For plate count, this would equate to approximately 300 organisms per plate at the top of the range, and a single organism at the bottom. At least five replicate determinations at each concentration across the range for each organism is used for both methods. The easiest way to then compare the data is graphically, by plotting the results obtained against the dilutions used taking the dilution providing the highest result as 100%. Alternatively the data can be statistically manipulated using the correlation coefficient r2 to measure linearity. A value for r2 of 0.9 or better should be expected.
Was this article helpful?