Display a Thompson-Howarth Plot of Duplicate Measurements
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.
thplot1(x1, x2, xname = "", ifzero = 0.01, xlow = NA, xhih = NA, yhih = NA, rsd = 5, ptile = 95, main = "", ...)
x1 |
a column vector from a matrix or data frame, |
x2 |
another column vector from a matrix or data frame, |
xname |
by default the character string for |
ifzero |
as the Thompson-Howarth plot is log-scaled values of zero cannot be displayed, therefore the parameter |
xlow |
if is desired to produce plots with consistent scaling this may be achieved by defining |
xhih |
enter an appropriate value of |
yhih |
enter an appropriate value of |
rsd |
to assist in QA/QC inspection a target precision may be defined as a RSD%, a default of |
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 |
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 |
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.
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)
Robert G. Garrett
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.
## 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)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.