The uncertainty or "error" in data that is due to the procedures used by statisticians or inexact responses provided by subjects or others who record information. Statistical uncertainty is usually thought of as including uncertainty due to sampling [rather than measuring an entire population] and measurement [such as the effect of the wording of a questionnaire.] It can also include the effect of how a population is defined, what questions are asked or when the data is collected.
Statistical weights
Weights applied to samples or observations to infer conclusions about a broader population.
Structured protocol
A set of procedures, questions or observations that occurs in a specified order according to a defined method.
Surveillance
Ongoing systematic collection, analysis and dissemination of health data to those who need to have them.
Surveillance, Active
Surveillance that uses a protocol that does not depend on the decisions of observers--for example conducting a survey using interviews.
Surveillance, Passive
Surveillance that relies on observers [such as physicians] to report the presentation of a condition.
Survival rate
The number of people with a condition [for example lung cancer] who are alive at a later time.
Transparent indicators
Indicators with a clear description of the methods or procedures used to collect, analyze and present them.
Treatment protocol
The course of treatment provided in response to a diagnosis.
Validity of data
The extent to which data reflect the concepts measured. Validity depends on how the data are collected compared with the ideas that analysts have about their meaning.
