Feedback on this I finally understand that all measurements, or collection of data using a tool contain errors. We can only hope to minimize errors but cannot prevent them 100% of the time. One type of error is measurement error. We cannot really measure someone's blood pressure directly. Instead, we rely on a tool and a process to collect this data. That introduces error into the system. Another type of error is based on a yes or no outcomes. Now if an error is made if it's a little bit wrong, it's all the way wrong. Lastly, this measurement error deals with percentages, incidence and rates. We cannot truly know how many people live in the United States at any time. We can use census data and other tools to conduct a best estimate. However, those will always have some degree of measurement error. There is no such thing as a perfect census