Imagine you are enjoying a pleasant evening on the pedestrian mall in downtown Iowa City. As the evening comes to a close, you feel fine and begin to drive home. After driving a few blocks a member of the Iowa City Police Department activates his lights and pulls you over for having a license plate light out. Fast forward now to you sitting at the Iowa City police station. The officer requests a breath test and you consent. After blowing in the machine a small piece of paper which resembles a cash register receipt prints out the result. So, how accurate is the result?
The answer is it depends on who you ask. The prosecution will produce a certification from the Department of Criminal Investigation (DCI) laboratory that professes a “margin of error” for the datamaster as being “.004 or plus or minus 5%, whichever is greater.” Thus, it would appear at first blush that the datamaster is extremely accurate. On closer examination, this “margin of error” would be better described as the “margin of illusion.”
The DCI lab created this margin of error formula without a working definition of margin of error. It essentially is a measure of how precise the datamaster can measure in a laboratory setting with strict controls using a simulator device instead of human breath.
In the scenario where you are sitting in the Iowa City police station being tested, the datamaster is of course not concerned with accuracy of simulators, it matters instead as to how accurate it is in measuring alcohol in human breath. The National Safety Council, for examples, states that when testing for human breath alcohol content, two tests should be taken from the same subject within two to ten minutes of each other. In that test scenario, the breath test should result should be within .02 of each other and the lesser result should be used. Thus that .087 breath test result could easily render a .067 result two minutes later if a duplicate test was administered.
Indeed, if the DCI lab was involved in a study in which human breath and blood where drawn contemporaneously from the same subjects. In many instances the breath registered at least .01 greater than the blood.
So, whats the margin or error for the datamaster when testing for human breath? The one thing that is certain is that it is NOT “.004 or plus or minus 5%, whichever is greater.”