Visual Sample Plan (VSP) Models and Code Verification Page: 4 of 61
This report is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
Visual Sample Plan (VSP) is an easy-to-use visual and graphic software tool being developed to
select the right number and location of environmental samples so that the results of statistical tests
performed to provide input to environmental decisions have the required confidence and performance. It
is a significant help in implementing Steps 6 and 7 of the Data Quality Objectives (DQO) planning
process ("Specify Tolerable Limits on Decision Errors" and "Optimize the Design for Obtaining Data,"
This report documents 1) the models (equations) in the VSP program that are used to compute
the approximate number of samples needed for an environmental sampling study; 2) the activities
conducted by Pacific Northwest National Laboratory (PNNL) to verify that those models are
scientifically sound and correctly programmed and implemented in VSP; and 3) an evaluation of the
nonstatistical user-interface aspect of the VSP program.
PNNL researchers performed this evaluation on Version 0.9H; needed corrections were
incorporated into Version 0.9I.
This report first discusses
" the equations used in VSP to approximate the number of samples (sample size) needed for
statistical tests of hypotheses
" the testing conducted previously to verify the accuracy of VSP calculations of the sample size
needed to detect "hot spots" or of the probability of detecting "hot spots" for a specific number
" the assumptions that underlie each sample-size equation
" the technical basis and scientific source (peer-reviewed scientific papers, books) of each sample-
" the calculations conducted by hand and by an independent computer program, S-PLUS, to verify
the accurateness of the sample-size computations in VSP
" possible alternative methods for determining the number of samples that may be of interest for
future implementation within VSP.
The calculations of sample size conducted by hand as well as by S-PLUS confirmed that Version
0.9H of VSP was correctly and accurately computing sample sizes except for the one-sample test of
proportion when the null hypothesis was stated as "the site is clean." The VSP code was modified to
correct this error for Version 0.9I. Also, the method used in Version 0.9H for computing the two-sample
test of proportions was changed in Version 0.9I to be consistent with the corrected method for the one-
sample test of proportion. The only other significant problem identified was that, in some cases, Version
0.9H displayed an incorrectly drawn Decision Performance Goal Diagram (DPGD) for the MARSSIM
sign test and the Wilcoxon rank sum test. This problem also was corrected in Version 0.9I.
In addition, this report discusses
" a series of steps to test the installation of VSP on various versions of the Windows operating
" verification tests of the import, export, and removal of random sampling locations
Here’s what’s next.
This report can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Report.
Gilbert, Richard O.; Davidson, James R.; Wilson, John E. & Pulsipher, Brent A. Visual Sample Plan (VSP) Models and Code Verification, report, March 6, 2001; Richland, Washington. (https://digital.library.unt.edu/ark:/67531/metadc935513/m1/4/: accessed April 18, 2019), University of North Texas Libraries, Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.