thplot1: Display a Thompson-Howarth Plot of Duplicate Measurements

In rgr: Applied Geochemistry EDA

Description

Function displays a Thompson-Howarth (1973 & 1978) plot for a set of duplicate measurements to visually inspect them as a part of the QA/QC process. By inputting a target precision the data may be visually checked to determine if they meet that criterion. The user is prompted for the location of the two legend items.

Usage

thplot1(x1, x2, xname = "", ifzero = 0.01, xlow = NA, xhih = NA, 
	yhih = NA, rsd = 5, ptile = 95, main = "", ...)

Arguments

x1	a column vector from a matrix or data frame, x1[1], ..., x1[n].
x2	another column vector from a matrix or data frame, x2[1], ..., x2[n]. x1,x2 must be of identical length, n, where x2 is a duplicate measurement of x1.
xname	by default the character string for x1 is used for the title. An alternate title can be displayed with xlab = "text string", see Examples.
ifzero	as the Thompson-Howarth plot is log-scaled values of zero cannot be displayed, therefore the parameter ifzero has to be specified. A suitable choice is a value one order of magnitude lower than the value of the detection limit. A default value of ifzero = 0.01 units is provided, corresponding to a detection limit of 0.1 units.
xlow	if is desired to produce plots with consistent scaling this may be achieved by defining xlow, xhih and yhih, the ylow value is set equal to ifzero. Enter an appropriate value of xlow to ensure all data are displayed on all plots.
xhih	enter an appropriate value of xhih to ensure all data are displayed on all plots.
yhih	enter an appropriate value of yhih to ensure all data are displayed on all plots.
rsd	to assist in QA/QC inspection a target precision may be defined as a RSD%, a default of rsd = 5 is provided. See comments concerning RSD in Details below.
ptile	defines the confidence interval for a line to be drawn on the plot above which only 100 - ptile% of the points should plot if the defined target RSD is being met. A default of ptile = 95 is provided. The function counts the number of points falling ‘out of limits’ and reports the probability that this number would have fallen ‘out of limits’ by chance alone.
main	a title may be added optionally above the display, see Example.
...	further arguments to be passed to methods. For example, the size of the axis scale annotation can be change by setting cex.axis, the size of the axis titles by seetting cex.lab, and the size of the plot title by setting cex.main. For example, if it is required to make the plot title smaller, add cex.main = 0.9 to reduce the font size by 10%.

Details

This function expects the RSD% as a measure of repeatability (precision), which is more familiar to the current generation of applied geochemists, rather than the precision at the 2 Standard Deviation level. The necessary calculations to conform with the Thompson and Howarth procedure are made internally.

Duplicate pairs containing any NAs are omitted from the calculations.

If the data are as a single concatenated vector from a matrix or data frame as x[1], ..., x[n] followed by x[n+1], ..., x[2n], or alternated as x[1] and x[2] being a pair through to x[2*i+1] and x[2*i+2], for the i in 1:n duplicate pairs use function thplot2.

The user is prompted for the location of the two legend items added to the plot, the number of duplicate pairs, and whether or not the duplicates have met the RSD% criterion. In both instances the user is prompted for the location of left end of the text line, or the top left corner of the text block. The probability that the plotted data have met the RSD criterion is displayed ib the text block; if the probability exceeds 0.9999 it is displayed as 0.9999.

Note

Any less than detection limit values represented by negative values, or zeros or other numeric codes representing blanks in the data, must be removed prior to executing this function, see ltdl.fix.df.

Duplicate pairs x1,x2 containing any NAs are omitted from the calculations.

This script was published by Garrett and Grunsky (2003)

Author(s)

Robert G. Garrett

References

Garrett, R.G. & Grunsky, E.C., 2003. S and R functions to display Thompson-Howarth plots. Computers & Geosciences, 29(2):239-242.

Stanley, C.R., 2003. THPLOT.M: A MATLAB function to implement generalized Thompson-Howarth error analysis using replicate data. Computers & Geosciences, 29(2):225-237.

Thompson, M. and Howarth, R.J, 1973. The rapid estimation and control of precision by duplicate determinations. The Analyst, 98(1164):153-160.

Thompson, M. and Howarth, R.J., 1978. A new approach to the estimation of analytical precision. Journal of Geochemical Exploration, 9(1):23-30.

See Also

thplot2, ltdl.fix.df, remove.na

Examples

## NOTE: the examples below are commented out as thplot1 makes a
## call to the locator function that fails when the examples are run
## during package checking and building
 
## Make the Stanley (2003) test data available
##data(ms.data1)
##attach(ms.data1)

## Display the default plot
##thplot1(MS.1, MS.2, xname = "Magnetic Susceptibility", 
##	main = "Stanley (2003) Test Data")

## Display a Thompson-Howarth plot for a RSD of 7.5% and a draw the limit
## for a confidence interval of 90%
##thplot1(MS.1, MS.2, xname = "Magnetic Susceptibility", rsd = 7.5, 
##	ptile = 90, main = "Stanley (2003) Test Data")
     
## Detach test data
##detach(ms.data1)

Example output

Loading required package: MASS
Loading required package: fastICA

rgr documentation built on May 2, 2019, 6:09 a.m.