tsk: Trimmed Spearman-Karber Method

In tsk: Trimmed Spearman-Karber Method

Description

Calculates the ED50 (effective dose) from a dose-response curve using the Trimmed Spearman-Karber method.

Usage

tsk(...)
## Default S3 method:
tsk(x, n, r, control = 0, trim = 0, 
    conf.level = 0.95, use.log.doses = TRUE,...)
## S3 method for class 'data.frame'
tsk(input, control = 0, trim = 0, 
    conf.level = 0.95, use.log.doses = TRUE,...)
## S3 method for class 'tskresult'
print(x,...)

Arguments

x	a non-empty numeric vector of doses.
n	a non-empty numeric vector of number of subjects at each dose, of the same length as x, or a number if the number of subjects are the same at each dose.
r	a non-empty numeric vector of the number of responding subjects at each dose.
input	Data frame containing the data to be analyzed. Must contain the three components x (doses), n (number of subjects at each dose), and r (number of subjects that respond).
control	fraction of subjects responding in a control experiment
trim	a number between 0 and .5. 0 gives the untrimmed Spearman-Karber method.
conf.level	confidence level to use for confidence intervals.
use.log.doses	if TRUE then the log of the doses in x will be taken and used for calculation.
...	further arguments to be passed to or from methods.

Value

A list of type "tskresult" containing the following components:

use.log.doses	same as above.
trim	same as above.
LD50	The estimated effective dose for 50 percent of the population.
gsd	The geometric standard deviation of the LD50 estimate.
conf.level	same as above.
conf.int	A confidence interval for the measurement, at the level of conf.level.

If use.log.doses = FALSE, instead of LD50 and gsd, the list will contain:

mu	The estimated LD50, given a different name to prevent confusion with calculations on a log scale
sd	The standard deviation of the mu estimate.

Note

This code expects the data to have an increasing slope once smoothed. However, some studies may result in data with a decreasing trend. For instance, one user had data measuring the survival of freshwater fish under increasing water salinity: at low salinity all fish survive, while at high salinity all fish die. If using data with a decreasing trend, simply use the opposite response, e.g., swap the count that survive for the count that do not, or replace r with n-r.

Note

This code was validated using the data given in Hamilton et al. (1977) Note that the confidence intervals given in Hamilton et al. (1977) are incorrect, as per Hamilton et al. (1978). This code replicates the results of the DOS program provided by the US EPA to perform the Spearman-Karber method, available at http://www.epa.gov/eerd/stat2.htm.

Author(s)

Brenton R. Stone, [e-mail]

References

Hamilton, M. A.; Russo, R. C.; Thurston, R. V. Trimmed Spearman-Karber Method for Estimating Median Lethal Concentrations in Toxicity Bioassays. Enviro. Sci. Tech. 1977, 11 (7), 714-719. http://dx.doi.org/10.1021/es60130a004

Ibid, 1978, 12 (4), 417. http://dx.doi.org/10.1021/es60140a017

See Also

hamilton.

Examples

tsk( c(1, 10, 100, 1000), 20, c(0, 3, 17, 20) )
dr <- data.frame( x=c(1, 10, 100, 1000), n=c(20, 20, 19, 20), 
                  r=c(0, 3, 17, 18) )
tsk( dr, 0, 0.1, 0.99 )

data(hamilton)
tsk(hamilton$dr1a)

tsk documentation built on May 2, 2019, 5:16 p.m.