I am putting this here because I know I won’t remember it but I think it’s moderately useful to know. As with type I censoring, it is really about a study design rather than anything special about the censoring of the data which is right censoring.
Details #
Censored data (see entry) is data where at least some values are not know accurately but are known to be bigger than a given value, less than a given value, or between two values.
Type II censoring happens in a study where a certain number of participants enter a study and are followed through until a predetermined number of the events of interest have happened. So if we were looking at returns to therapy we design the study to recruit until 500 clients had been recruited and to terminate follow-up when 50 have returned to therapy. Assuming that a minority return (or have some other event of interest) then this can be a good way to design a study to have the precision you want in estimating say the median time to return.
As always, censoring matters because if we ignore the censoring we introduce a bias into our estimation of general durations of therapies for that study. However, there are statistical methods that handle censoring and will avoid that bias (given some sensible assumptions mostly that the pattern of duration hasn’t actually changed across the period of the study). See survival analysis for probably the biggest class of such methods.
Try also #
Censored data
Estimation
Frailty analysis
Interval censoring
Precision
Right censoring
Survival analysis
Type I censoring
Chapters #
Not covered in the OMbook.
Online resources #
None forseeable.
Dates #
First created 15.xii.24.
