Censoring (statistics)

Example of censored data

Censoring in statistics occurs when the value of a measurement or observation is only partially known. Censoring is a form of missing data problem that arises in various fields, including clinical trials, survival analysis, and econometrics.

Censoring can occur in different forms, such as right censoring, left censoring, and interval censoring. Understanding and handling censored data is crucial for accurate statistical analysis and inference.

Types of Censoring

Right Censoring

Right censoring occurs when the observation is only known to be above a certain value. This is common in survival analysis, where the event of interest (such as death or failure) has not occurred by the end of the study period. For example, if a study ends before a patient dies, the survival time is right-censored.

Left Censoring

Left censoring happens when the observation is only known to be below a certain value. This can occur in studies where measurements below a detection limit are recorded as being less than that limit.

Interval Censoring

Interval censoring occurs when the observation is only known to lie within a certain interval. This can happen in longitudinal studies where the exact time of an event is not known, but it is known to have occurred between two observation times.

Handling Censored Data

Statistical methods have been developed to handle censored data, ensuring that analyses remain valid and unbiased. Some common methods include:

• Kaplan-Meier estimator: A non-parametric statistic used to estimate the survival function from lifetime data.
• Cox proportional hazards model: A regression model commonly used in the analysis of survival data.
• Tobit model: A statistical model designed to estimate linear relationships between variables when there is censoring.

Applications

Censoring is prevalent in many fields:

• In clinical trials, censoring is often encountered when patients drop out of the study or the study ends before the event occurs.
• In econometrics, censoring can occur in income data where incomes below a certain threshold are not reported.
• In environmental science, measurements below detection limits are often left-censored.

Related pages

• Survival analysis
• Missing data
• Kaplan-Meier estimator
• Cox proportional hazards model