Respond to this
Godish et al. (2014) describe within their illustration three different target studies. The first study has values that are scattered about and don’t come close to the center of the target. The second study has a tight bundle of values all within the center of the target. The third study has a bundle of values that are tightly held together but are no way near the center of the target. Godish et al. (2014) explain the meaning behind this illustration as the first study having low accuracy and poor precision. The second study having results that are both precise and accurate, and the final study having excellent precision, but the results deviate from the true value due to study bias.
Bias occurs as a result of such things as calibration errors and any instrumentation issues that may due to hardware or systematic software problems (Godish, Davis, & Flu, 2014). This bias due to calibration or instrumentation problems could cause immense problems within an air monitoring program. Calibration, checking, and record keeping (QC/QA) is key to any air quality program. If there were no calibration and record keeping how could someone look back at the data and determine when a deviation may have occurred and determine the breadth of a situation. As a company that has a waste water facility on site, I can tell you all that our paperwork is crucial for us to prove to the county and state that our waste water stays within a specific pH throughout everyday, and that we verify and record our pH probes daily and calibrate and record our probes weekly. This is required because we also record the number of gallons of waste water discharged and from all of these recordings we can discern the exact time a pH probe failed and the amount of water that may or may not have accidentally been discharged into the local sewer system.
I feel this is what Godish, Davis, and Fu are trying to explain in there representation of this graphical illustration.
Recent Comments