ABSTRACT
In the last chapter we discussed calibration and detection limits. Often in environmental data analysis values are reported simply as being “below detection limit” along with the stated detection limit (e.g., USEPA, 1992c, p. 25; Porter et al., 1988). A sample of data contains censored observations if some of the observations are reported only as being below or above some censoring level. Although this results in some loss of information, we can still use data that contain nondetects for graphical and statistical analyses. Statistical methods for dealing with censored data have a long history in the field of survival analysis and life testing (Bain and Engelhardt, 1991; Kalbfleisch and Prentice, 1980; Lee, 1980; Miller, 1981b; Nelson, 1982; Parmar and Machin, 1995). In this chapter, we will discuss how to create graphs, estimate distribution parameters, perform goodness-of-fit tests, compare distributions, and fit linear regression models using censored data.
