Assess: User`s Guide
Contents
1. using the arrow keys to move the cursor bar Notice that the first 42 lines are historical information and will be ignored You may pass over afield thus assigning a missing value to that particular field 9999 0 in the data file represents a missing value Notice that by pressing the enter key you will be positioned at the Revise option on the menu bar Select it again to return to the input screen Once all of the data is entered take the following actions Option Action Field Value Save Data Enter Data file Saved2 fil Execute Enter The screen of Figure 3 11 will be displayed Notice that through the Save Data option you may save your data at anytime The format of this file is Input with labels This point is important if you would want to use this file again to add or delete information Therefore on the first screen you would select the Input with labels toggle field of the Data Option if there is a need to use this file as input to ASSESS This example is used to verify your results Analysis of data is similar to the analysis discussed in the previous section Quit Enter Quit Enter Quit Prompt Y 3 13 DECEMBER 1990 Figure 3 11 The Results Screen nt ASSESS 1 0 DECEMBER 1990 ALTERNATIVE DESIGN AN EXAMPLE Alternative Design Results ASSESS 1 0 Figure 3 12 Screen The alternative QA design was developed to assess measurement error variances in absence of FES and2. Two example data files are provided As noted above Example fil is an example of a file generated by ASSESS that can be used as an input file When selecting the Data option in the Data Quality Objectives menu a question appears asking which format to use If the toggle Input with labels is chosen then the file Example fil may be selected If the toggle Standard no labels is chosen then the file Standard fil may be selected An explanation of the formats used in Example fil the reporting of results report sent to the printer and in Standard fil follow Certain lines of historical information will have no information after the descriptive label This is specific to the example at hand and not represen tative of data files in general A 1 DECEMBER 1990 Data File Example Example fil Example fil is a data file created by ASSESS that contains the computed variances as well as any historical information for notational purposes displayed on the screens Quality assessment data is also included This file is also an input data file so if results and data changes are to be saved for later use this file type should be created This file is generated by selecting the Report option of the Results screen menu and then selecting the option File input format The quality assessment data is listed at the end of the file It includes 9999 values which represent missing values see the descri
3. ment screen are used presently for information purposes No calculations are performed using the information from these three screens The informa tion is written to a user specified file along with the results of other calculations 2 4 DECEMBER 1990 Figure 2 2 The Menu Hierarchy ARTO NN ASSESS 1 0 DECEMBER 1990 31 OVERVIEW ASSESS 1 0 Section 3 Using ASSESS in an Assessment Study Example This section will demonstrate how to use the ASSESS software to conduct an assessment study We start with an example data set obtained from an actual Superfund site which was contaminated by lead deposition from a smelter however the arrangement of the data into batches and data from field evaluation samples are fictional and are included for illustrative purposes Through this example data set we will explore the ASSESS utilities and analyze its results Two other data sets for the alternative QA design will also be shown The data set standard fil has been included with the software so that you may repeat the exercise as a tutorial or to test the software For a detailed explanation oft he data refer to the Rationale document l The pilot study was conducted over a representative area to determine spatial variability and extent of the lead contamination in order to develop an efficient sampling network for obtaining representative measurements of contamination over a large area The quality asse
4. bar Toggle through the possible fields which are HS amp FD and RS amp PS Remember to use the space bar to toggle throughout the possible options Select the RS amp FD field and press the lt enter gt key Note the message on the message line and press any key to view the scattergram The following is the plot of the original data untransformed ee ee te et een Figure 3 5 illustrates the need for a transform of the original data The plot shows that the standard deviation of the data multiplied by 2 from routine samples and field duplicates increases with the average concentration of these samples A logarithmic transform of the data will stabilize the variance of the data over measured concentration range After this transformation has occurred the variances may be computed for the purpose of assessing variability throughout the measurement process Press the letter Q to quit the graph and observe the following Trans forms screen DECEMBER 1990 Figure 3 6 Transforms Screen Unary Operation This option allows the user to toggle through five unary operations as well as a Retain option which allows the original data to be retained thereby eliminating any confusion as to the type of data being transformed These six options are also listed on the screen Select the Unary operation and use the space bar to toggle through the possible selections Select the Ln operator to take the
5. from which the program ASSESS has been derived be used in conjunction with this user manual Please forward any questions comments or bug reports to the following address Jeff van Ee Assess USEPA EMSL LV EAD P O Box 93478 Las Vegas NV 89193 3478 1 2 DECEMBER 1990 1 2 EQUIPMENT REQUIREMENTS 1 3 USER PROFILE 1 4 INSTALLING THE SYSTEM 1 4 7 ASSESS Data Files ASSESS 1 0 ASSESS was designed to run under DOS Disk Operating System on an IBM PC XT AT PS2 or compatible Graphics capability is not required but is highly recommended as graphics output is produced Graphics support is provided for the Hercules graphics card laptops with monochrome displays having graphic capabilities the Color Graphics Adapter CGA and the Enhanced Graphics Adapter EGA Support for Video Graphics Array VGA is not available however VGA does emulate EGA and therefore graphics support is provided for VGA indirectly At least 512 kilobytes Ko of random access memory RAM is required but 640 Kb is recommended An arithmetic co processorchip is recommended due to t he computational intensive nature of the program but is not required for use ASSESS may be run from floppy diskette or from a fixed disk The system storage requirement is approximately 420 kilobytes For a hardcopy of results a graphics printer IBM graphics compatible is required Support is provided for plotters which accept HPGL plotting commands To use ASSESS
6. one should have some familiarity with personal computers and DOS Disk Operating System One should also understand basic DOS commands such as DIR directory CD change directory and how to insert and use diskettes For more information on these topics consult a DOS user s manual The manual titled A Rationale for the Assessment of Errors in the Sampling of Soils 1 must be followed For a list of references refer to Appendix B References ASSESS in its executable form is entirely in the public domain In the future it may be downloaded from the U S EPA Bulletin Board System ASSESS exe is the only file required to run ASSESS If a metacode file is to be converted to a file with an HPGL format then Hpplot exe and Hershy bar are required Hershy bar is a character font file and is required to execute Hpplot exe The executable file example data set and optional files are as follows ASSESS 422778 bytes required Optional Standard fil 2576 bytes example data set Alt1 fil 1471 bytes example data set Alt2 fil 1636 bytes example data set Saved1 fil 4744 bytes used for comparison Saved2 fil 3771 bytes used for comparison Barchart met 1440 bytes wad for comparison Scatter met 2880 bytes used for comparison Results fil 4744 bytes used for comparison Smdata fil 1714 bytes example data set Example fil 4744 bytes example data set Hpplot 98417 bytes conversion program Herahy bar 176000 by
7. 9999 000 i 2 9999 000 9999 000 9999 000 475 000 498 000 9999 000 9999 000 HISTORICAL PRECISION ASSESSMENTS VAR 3 9999 000 9999 000 9999 000 423 000 424 000 9999 000 9999 000 sampling 1000 4 389 000 9999 000 430 000 9999 000 9999 000 9999 000 9999 000 source this is some thing new 8 246 000 410 000 9999 000 9999 000 9999 000 9999 000 9999 000 handling transportation preparation 0700 33 400 9999 000 32 100 9999 000 9999 000 9999 000 9999 000 source 7 960 000 780 000 9999 000 9999 000 9999 000 9999 000 9999 000 subsampling 0500 221 000 9999 000 244 000 9999 000 9999 000 9999 000 9999 000 source this is old 9 180 000 208 000 9999 000 9999 000 9999 000 9999 000 9999 000 laboratory analytical 2000 10 60 000 9999 000 72 000 9999 000 9999 000 9999 000 9999 000 source 11 87 000 221 000 9999 000 9999 000 9999 000 9999 000 9999 000 data analysis and interpretation 0500 12 275 000 9999 000 233 000 9999 000 9999 000 9999 000 9999 000 source 13 349 000 400 000 9999 000 9999 000 9999 000 9999 000 9999 000 between batch 0000 14 474 000 9999 000 446 000 9999 000 9999 000 9999 000 9999 000 source 15 478 000 382 000 9999 000 9999 000 9999 000 9999 000 9999 000 TOTAL MEASUREMENT VARIANCE 0600 16 33 500 9999 000 32 700 9999 000 9999 000 9999 000 9999 000 source 17 33 000 33 300 9999 000 9999 000 9999 000 9999 000 9999 000 18 1360 000 9999 000 1340 000 9999 00
8. ELES pairs Formulas in Table 6 in the Rationale Document l are used to calculate the results The required data for the alternative design are RS FD PS BFD Batch Field Duplicates the location of BFD values and FES1 and ELES1 values The FES1 and ELES1 values are not used in the computations and are only used to represent bias Note ASSESS presently does not calculate bias in the data In the following example the data file Alt1 fil is used to illustrate the case where batch field duplicates are all obtained from one sampling location The format of the file is the same as for the data file standard fil Take the following sequence of actions to display the variance estimates produce error bar graphs and scattergrams Option Action Field Value Data Enter Data File Yes Standard with no labels Alt1 fil Alternative Enter Sample Location Yes No If you answer either Yes to both questions or No to both you will be notified by an error message asking you to start over Cursor bar returns to Data option valuate Data Enter xecute Enter The following screen will be displayed PERLE OE E RE SA DECEMBER 1990 Note that the variance component associated with handling cannot be separated from that associated with sample collection and that the variance component associated with subsampling cannot be separated from that associated with analysis This loss of information is a consequence o
9. e aa d d Y 0 05 A ASSESS 1 0 DECEMBER 1990 2 2 2 Types of Screen input Fields ASSESS 1 0 0 The Screen Frame This is the rectangular box which encloses each screen Program inputs and results are displayed in this area Typically the screen frame is divided into smaller single line rectangles Each of these smaller rectangles contains a functionally related group of one or more input parameters or program results e The Message Line This is the rectangle at the bottom of the screen frame This area is used to display program error messages yes no prompts prompts for additional information or instructions for using a program option e The Menu Line This is the line of text just below the screen frame It contains a set of menu option names and a highlighted box cursor bar The cursor bar can be moved along the menu line by using the left and right arrow keys As the cursor bar is moved over a menu option name a short description of the menu option is displayed on the line just below the menu line This line is called the menu description line You may explore the possible choices in a program by moving the cursor bar and reading the descriptive messages which accompany each menu op tion To select a menu option move the cursor bar over the desired menu option name and press enter An alternative and faster way to select a menu option is to press the key which corresponds to
10. the Log Transformed The Resulls CNC NNI CRI I EH Fa FN UM The Intermediate Results Screen wc E The Error Bar Graph The Results SCOR Alternative Design Results Screen wun Alternative Design Error Bar Graph Variance Estimates for the Alternative Design 8 17 ASSESS 1 0 V DECEMBER 1990 11 OVERVIEW ASSESS 1 0 Section 1 Introduction ASSESS is an interactive program designed to assist the user in statistically determining the quality of data from field samples The program permits quality assessment data and historical information historical information is for notational purposes only such as data quality objectives sampling considerations and historical assessments to be saved Measurement error variances are computed Scatter plots of the variance represented as Y2 times the standard deviation versus the average concentration may be produced for Routine Samples RS with either Field Duplicates FD or Preparation Splits PS Transforms may be applied to the entire data set Error bar graphs of variance estimates from quality assessment QA samples may be produced Hardcopies of all plots may be obtained by depressing the P key The plots may also be stored in a file of device independent plotting commands metacode An HPGL pl
11. the first letter in the menu option name The result is the same as using the cursor control keys and pressing enter For example you would choose the Revise option by pressing R from the main menu Parameter Groups Typically a functionally related group of program input parameters fields are enclosed together on the screen by a single line rectangle These groups of parameters are accessed through the menu When a menu option is selected a cursor bar appears at the screen field and a message describing what action to take appears on the message line When such a group contains several fields the cursor control keys are used to move to subsequent fields Exiting from the last field in the group will return the cursor bar to the menu line Several types of input files are provided to allow flexibility in program parameter specification Below is a list of these types and an example of each field type in the first screen of ASSESS Figure 2 1 Alphanumeric Fields These fields may contain character strings of alphabetic or numeric characters Any alphanumeric characters may be entered The Data menu option in the screen of Figure 2 1 requires an alphanumeric value to be entered To specify a data file name select the data option on the menu and type the name of the input data file 2 3 DECEMBER 1990 2 2 3 The Menu Tree 2 2 4 ASSESS Screens and Menus ASSESS 1 0 Numeric Fields Only numeric data may be ent
12. 0 2 9999 000 9999 000 9999 000 475 000 4 8 000 9999 000 9999 000 a ASSESS 1 0 1990 De HW A YN 9 10 11 12 33 33 13 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Figure 4 Explanation of Standard fi data file continued Data Quality Objectives screen gcriptiona Bite 64 Method 14 Analyte 14 Analytical Method 14 Desired Accuracy 4 Desired Precision I 4 Bias I 4 Precision I 4 Sampling Considerations creen Number of Samples I 10 Number of Batches I 10 Sample Collection F 10 2 Analysis F 10 2 gt ha A 7 A 15 A 15 must appear naf ASA Rh Sampling Source Bandling Preparation Source Subsampling Source Laboratory analytical Source Data analysis Source Between batch Source Total measurement variance Source Bampling no contamination Sampling contamination Source Handling Prep no contamination A Me Me Mu De a Cap Dg Bas a a Tn Set ASSESS 1 0 If more than 10 are present then hl a gt a M y Descriptions The descriptions broken down into screen and include the screen label type of field A Alphanumeric 1 Integer F Floating point and length of fie
13. 0 9999 000 9999 000 9999 000 HISTORICAL MEASUREMENT BIAS ASSESSMENTS 19 104 000 128 000 9999 000 9999 000 9999 000 9999 000 9999 000 sample collection no contamination 0000 20 313 000 9999 000 294 000 9999 000 9999 000 9999 000 9999 000 sample collection contamination 0000 21 201 000 161 000 9999 000 9999 000 9999 000 9999 000 9999 000 souroe 22 67 000 9999 000 67 000 9999 000 9999 000 9999 000 9999 000 handling preparation no contamination 0000 23 215 000 199 000 9999 000 9999 000 9999 000 9999 000 9999 000 handling preparation contamination i 0000 2 sources 1 ASSESS 1 0 2 1990 Data File Example Report of Results The report data file is created by ASSESS and contains the computed variances as well as any historical information displayed on the screens Quality assessment data are also included This file is the same as the Example fil discussed in the preceding section with one exception the quality assessment data has blanks where the 9999 values were located As a result this file cannot be used as an input data file This file is generated by selecting the Report option of the Results screen menu and then selecting the option File report format The file may be sent to the printer by selecting the Printer option Only the quality assessment data are shown below as the rest of the file is the same as Example fil shown on the previous pag
14. 91 001 q D Sr lt 5 c 5 Oo c pum gt 5 5 lt gt ES 5 E y SE Bo gt Y gt LLI Per O O gt a 3 otc ec 2 gt cc 3478 93478 Las Vegas NV 89193 P Agency Research and Development ASSESS EPA Guide User NOTICE Version 1 0 of this software is a prototype Additional modifications are planned for the future The information in this document does not represent the views or policy of the Environmental Protection Agency DISCLAIMER ASSESS software and documentation are provided as is without guaran tee or warranty of any kind expressed or implied The Environmental Monitoring Systems Laboratory U S Environmental Protection Agency and Computer Sciences Corporation will not be liable for any damages losses or claims consequent to use of this software or documentation ASSESS 1 0 DECEMBER 1990 1 Introduction 2 System Operation 3 Using ASSESS in an Assessment Study Example ASSESS 1 0 Contents BONIENIS davon iron Rx GER Un A ei ii LISD OF FIQUICS n 1 1 OVERVIEW 1 1 1 2 EQUIPMENT REQUIREMENTS s 1 3 1 3 USER PROFILE nnnm 1 3 1 4 INSTALLING THE SYSTEM ee 1 3 1 4 1 ASSESS Data Files 1 3 1 4 2 Hard Disk Installation 1 4 1 4 3 Using ASSESS on Flo
15. amples are floating point 11 IMPORTANT The 9999 represents an empty field On line 43 of the above example only columns five and sir have values other than 9999 This means that FES1 and FES2 have values of 448 and 505 respectively and that RS FD PS 11581 and ELES2 have no values for that row of data The 9999 must be entered in the column of the sample that has no value for that row If it is not entered then Assess will either read the data incorrectly or will display an error message 44 Second row of quality evaluation data such data may be entered A maximum of 1000 rows of DECEMBER 1990 Heference U S EPA 1990 A Rationale for the Assessment of Errors in the Sampling of Soils Environmental Monitoring Systems Laboratory Las Vegas Nevada EPA 600 4 90 013 ASSESS 1 0 DECEMBER 1990 ASSESS 1 0 RS FD PS FES1 FES2 ELES1 ELES2 BFD Location 2 SFD 2 Sps 2 SWFES 2 SBFES 2 SWLES Appendix C Nomenmclature Routine sample Field duplicate Preparation split Field evaluation sample 1 Field evaluation sample 2 External laboratory evaluation sample 1 External laboratory evaluation sample 2 Batch field duplicate Location of a batch field duplicate sample Field duplicates error estimate Preparation splits error estimate Field evaluation samples error estimate within batches Field evaluation sam
16. d to select this option The Yes No prompt is a means of ensuring that a series of lt Q gt keystrokes will not cause inadvertent termination of the program You may now select the Sampling Considerations option from the menu bar The following screen will be displayed Figure 3 2 Sampling Considerations RR We will now discuss the options accessed through this screen Revise This option is exactly similar to the discussion of the Revise option above Let us change the analysis cost to 500 00 Select the Revise option use ASSESS 1 0 DECEMBER 1990 Figure 3 3 Historical Assessment Screen ASSESS 1 0 the down arrow key to move to the Analysis field and type 500 00 Notice that if you hit an enter key after entering the number 500 00 you will be placed into the next part of the screen regarding the batch data To get back to the Revise option move your cursor bar to the extreme left or right sides of the screen frame using the arrow keys Historical Assessment This option displays the Historical Assessment menu This menu shows a screen with the bias and measurement data of the input data file DO NOT elect this option yet We will discuss it after examining the remaining options on the screen Evaluate Data This option is exactly similar to the option discussed for the main screen Quit This option will return the user to the main menu of the main screen We may now select th
17. ductory information When you press a key to proceed you will see the program main screen and menu as displayed in Figure 2 1 The bottom line on the screen provides the list of available options The first three options move you to an area on the screen or to a new screen where you can input or select program parameters We begin by running through the program using the menu options Throughout this tutorial you will seethe phrase select an option used often You select an option by positioning the cursor baron the option or field and pressing the enter key In the data analysis section of this document the following points are noted 1 The line Insufficient samples exist for this variance for any error variance estimates means that enough samples do not exist to assess the variance 2 if the value of any variance estimate is less than zero a value of zero will be assigned to that variance Data The Data option is used to decide whether a data file is present or data is to be entered on the screen Select the data option and answer Yes to the prompt Do you want to specify a data file use space bar This is done by pressing the enter key on the toggle field Yes You are then asked about the format of the input file Hit the space bar to toggle through the options and select the standard with no labels field in response to the What is the input file format use space bar prompt A
18. e Historical Assessment option After pressing the enter key on the Historical Assessment option the following screen will be displayed We will now discuss the options accessed through this screen Revise Through this option you may alter any of the screen fields as discussed in the previous explanation of Revise Select the Revise option and change the Total Measurement Variance field to 0 09 DECEMBER 1990 Figure 3 4 Quality Assessment Data ASSESS 1 0 Evaluate Data This option is essentially the same as previously discussed Quit This option will return you to the previous menu which is the Sampling Considerations screen e will now select the Evaluate Data menu option to display the Quality ssessment Data screen AA AS ee The screen above shows the contents of your input data file It only shows data for 10 batches within the RS FD and PS columns The Scroll option described below will let you explore the additional batches and the remaining columns not seen on the screen We will now discuss the options accessed through this screen Units This option will allow you to enter the concentration unit Select this option and type mg kg When you hit the enter key after typing in the concentration unit you will be back on the units option Scroll This option will allow you to scroll through the input screen The data file standard fil contains 23 batches but
19. e Note that the 9999 values have been replaced with blanks Figure A 2 Report data file without the historical information DLP QUALITY EVALUATION DATA Transform None Concentrations mg kg FES1 FES2 ASSESS 1 0 uu N e 389 000 246 000 33 400 960 000 221 000 180 000 60 000 87 000 000 000 000 000 500 000 000 000 000 000 000 000 410 000 780 000 208 000 221 000 400 000 382 000 161 000 199 000 430 000 244 000 446 000 1340 000 448 000 475 000 423 000 505 000 499 000 424 000 atere at et Don sat uta tul 66060000000 50 DECEMBER 1990 Data File Example Standard fil Standard fil is an input data file created by the user and stripped of any descriptive screen labels ASSESS can read but cannot generate such a file All entries must start in column 1 The number e g 1 that appears on each line is used as a line number reference and does not appear in the actual file For descriptions of each line referenced by the line number follow the example Figure A 3 Explanation of standard fil data file s a a e atv 560 0500 0000 0000 0000 anatara a aa a a n n a a etat 0000 0000 0000 1 9999 000 9999 000 9999 000 448 000 505 000 9999 000 9999 00
20. e ASSESS or Hpplot at the DOS prompt For example to start program ASSESS type ASSESS enter ASSESS uses most of the available memory If an error message occurs after typing ASSESS try to free other existing memory resident programs and type ASSESS to restart the program ASSESS plots graphics directly on the screen This approach is used to provide a quick look at data or program results Such graphics displays may be printed on a dot matrix printer When a graphics screen is displayed the program will wait for a key to be pressed Pressing Q will cause an interactive screen and menu to be displayed Pressing P will produce a hard copy of the screen on a graphics printer It is important to make sure that a graphics printer is connected to your computer if you choose this option or the program will lock up Also make sure t hat the printer is turned on and on line If the program locks up the computer will probably have to be re started The graphics displays may also be written to a metacode file when the Save Plot menu option is selected A metacode file is a file of device independent plotting instructions This file can be converted to a HPGL Hewlett Packard Graphics Language formatted file using the program Hpplot refer to section 1 4 1 ASSESS Data Files concerning Hpplot exe The advantage of using a metacode file is that higher quality graphic output can be obtained on a pen plotter or other graph
21. e provided for Batch Field Duplicates BFD and their locations WordPerfect is a registered trademark of WordPerfect Corporation ASSESS 1 0 2 1 DECEMBER 1990 The data may be in free format which means that in a given line in the file values must be separated by at least one space or a single comma For readability columns of numbers should line up although this style is not required Variable values must be numeric with no embedded blanks Whenever a value could not be obtained for a variable a special value of 9999 0 will be assigned to it This value referred to as missing will not be included in the calculations The file standard fil contains the historical data and 23 batch samples Examine the contents of this file using the DOS Type command or your editor Note Refer to Appendix A for further discussions of the different data file formats INTERACTIVE SCREENS 2 2 1 Screen Format This section will show the basic components of screens used in ASSESS A menu tree of the entire program is presented to observe the flow of the program s interaction ASSESS initially displays an introduction screen Upon pressing any key the first interactive screen is displayed ASSESS is composed of several screens for selection of program options and display of results Figure 2 1 displays the first interactive screen Below is a description of common omponents Figure 2 1 Example Interactive Screen eee apte
22. ered 3 12 DECEMBER 1990 3 3 DATA INPUT THROUGH THE KEYBOARD AN EXAMPLE ASSESS 1 0 As discussed earlier the second and third screens in ASSESS are used in reviewing or revising the historical data These two screens may be skipped without any effects on the computational aspect of ASSESS In this example we will allow the user to access the Evaluate Data menu input pair data carry out computations and sketch plots and finally save input data at anytime during the calculation phase Obtain a hardcopy of the file smdata fil The data in this file will be used as input to the program At the DOS prompt in the appropriate subdirectory type the command ASSESS enter After observing the introductory screen the screen of Figure 3 1 will be displayed In the previous section the sequence of actions needed to display various results were explained in a long hand notation where every keystroke was thoroughly explained To simplify our explanations an abbreviated notation for the sequence of events will be used general formula exists for each option Initiate the option then take one or more actions each of which may result in a screen field taking a particular value In order to get to the Evaluate Data menu without a data file use the following set of actions Option Action Field Value Data Enter Data File Evaluate Data Enter Number of Rows 13 Revise Enter Enter the values from the file smdata fil
23. ered into numeric fields some numeric fields will only accept integer non decimal numbers The program will respond to any erroneous keystrokes such as alphabetic keys with a low pitched error tone An example of numeric fields in the screen of Figure 2 1 are the desired accuracy and desired precision fields Only integers may be entered for these two fields Toggle Fields A toggle field is a special type of field which contains a list of 2 or more preset choices Only one of these choices is displayed in the field The space key is used to change the displayed choice and the enter key is used to make the selection Two examples of toggle fields in the screen of Figure 2 1 are the Data field and the Alternative field Once the data option is selected the toggle field will contain a Yes and No in response to the prompt Do you want to specify a data file use space bar gt The toggle field will be highlighted and each time the space key is pressed Yes or No will appear Press the enter key to make a selection If Yes is selected the toggle fields Input with labels and Standard with no labels may then be selected Alternative option allows for two toggle fields Yes and No in response to the prompts regarding the sample locations Yes No prompts prompts for additional information These prompts are for information which will not be displayed permanently on the screen They w
24. f not using FES and ELES values in the study The sampling and handling variance is equal to zero This variance is the result of subtracting the error estimate for the preparation splits 5 from the error estimate for the field duplicates amp 8 tg For this example Of 0 2 793 27 188 24 395 ASSESS will report a value of zero when the measured variance 1 a negative number Follow the sequence of keystrokes Option Action Plot Enter The resulting graph is displayed in Figure 3 13 Figure 3 13 Alternative Design Error Bar Graph a y vtt RI E Notice the large confidence interval for the preparation split samples Perhaps more preparation split samples could be taken to reduce this large interval Option Action lt Q gt Enter Var Estimates Enter Press any key Enter ASSESS 1 0 DECEMBER 1990 The resulting table is displayed in Figure 3 14 Figure 3 14 Variance estimates for the Alternative Design The displayed data in this figure is a summary of the computed error estimates for the alternative design Note that the estimate for the batch field duplicates is the one computed for condition a This condition is appro priate when the batch field duplicates are all taken from one sampling location You may end this session by the following given keystrokes or proceed with the data transformation as previously discussed Option Action Field Value Press any key Enter lt Q gt Ente
25. features as the Unary Operation option and allows you to perform multiplication division addition subtrac tion exponentiation and retain the original data Save Plot This option saves the generated plot through the Transform option The plot that will be saved is the plot just observed prior to reaching this screen layout Conceptually it saves the plot that is no longer visible by the user though it still exists in order to save the plot just viewed select this option and enter Scatter met for the Meta File prompt Quit This option will return to the Quality Assessment Data screen Select this option to return to the Quality Assessment Data screen Execute This option will display the results obtained by applying the equations in table 5 of the Rationale Document The primary purpose here is to estimate measurement error variance components The measured lead concentra tions in soil in mg kg are given for 10 Preparation Spilt PS pairs and for 10 Field Duplicate FD pairs The amount of data used has been kept small for readability and to illustrate the use of the computer program to calculate variances Select the Execute option The following screen will be displayed 3 9 DECEMBER 1990 ASSESS 1 0 Figure 3 8 The Results Screen 5 rie en 5 Figure 3 8 shows a summary of the results Note that when enough samples do not exist to assess any of the variances Insufficient sa
26. ics device The HPGL format is directly supported by WordPerfect 1 4 DECEMBER 1990 Example To importa 1 Generate a metacode file called Metacode met using ASSESS metacode file generated by Either of the two graphs generated by ASSESS may be saved to a ASSESS into metacode file Note that the name Metacode met is a generic name WordPerfect 5 0 perform the and it may be referred to by any other name The met extension is following steps used as a means of distinguishing these files and the regular ASCII text files 2 Run the program HPPLOT Enter Metacode met as the input file and Metacode pit as the output file HPPLOT will convert Metacode met to an HPGL formatted filecalled Metacode plt which is supported by WordPerfect 3 If you have a VGA adapter the HPGL formatted file may require one more conversion using the WordPerfect conversion program Graphcnv exe To convert call GRAPHCNV and then enter Metacode pit as the input file and Metacode wpg as the output file GRAPHONV will generate Metacode wpg which can be successfully imported into WordPerfect on PC s using a VGA adapter 1 6 ERROR AND RECOVERY PROCEDURES Normal Error Processing ASSESS performs error checking on such items as file existence file Input Output and bounds checking on numeric parameters When errors of these types are encountered in programs error messages are displayed on the message line at the bottom
27. ill appear temporarily on the message line A Yes No prompt will typically have the form Question lt Y N gt To respond Yes press the Y key to respond in the negative press any other key A typical Yes No prompt is the Do you really want to quit lt Y N gt prompt which is displayed after the Quit terminate program option is selected Another example is when you attempt to write over an existing file ASSESS typically requires input from data files and through interactive screens These program inputs are arranged in a hierarchy of functionally related groups Each group or individual program parameter value is accessed through a menu of choices Some choices will lead to other menus while some will lead to prompts for groups of one or more inputs ouch an arrangement can be represented in a menu hierarchy as illustrated in Figure 2 2 The menu tree representation of program options provides a road map for ASSESS which summarizes the functional capabilities of a program You may explore the hierarchy of options by traversing the menu tree and reading the descriptive messages which appear at the bottom of each screen ASSESS initially displays an introduction screen Upon pressing any key the first interactive screen is displayed The information entered or displayed on the succeeding three screens namely the Data Quality Objectives screen the Sampling Considerations screen and the Historical Assess
28. ld The following line is historical information about each batch of data maximum of 10 such lines is permitted the program will report an error in reading the data Oocupies columns 1 7 Occupies columns 9 23 Occupies columns 25 39 This tells Assess that the last line batch data descriptions eg line 13 has been written An E or e the first column of this line is sufficient Historical Assessment screen 6 4 26 6 4 26 6 4 2 6 4 26 6 4 26 6 4 26 6 4 26 4 6 4 26 6 4 33 34 Source 26 35 Subsampling no contamination F 6 4 36 Subsampling contamination F 6 4 37 Source A 26 38 Analytical no contamination F 6 4 39 Analytical contamination F 6 4 40 Souroe A 26 41 Data Handling F 6 4 42 Souroe A 26 Quality Evaluation Data 43 This is the first row of quality evaluation data There are eight columns The first column represents the batch number and is an integer The remaining seven columns represent the routine subsample RS the 4214 Auntiinsate anhaaunla f i See n the nranarati an anl 4t arrima da fma 4 gt gt 104 LAA WAR Dese SP first field evaluation sample FES1 the second evaluation sample FES2 the first external laboratory evaluation sample ELES1 and the second external laboratory evaluation sample ELRS2 respectively 11 seven s
29. log base e of the original data After selecting this option you will remain in this screen In order to view the newly created plot select the Quit option Notice the message at the bottom of the screen when the cursor bar is moved to the Quit option This option will return you to the Quality Assessment Data screen Notice the data displayed on the screen The values are log base e of the original data Select the Transform option and the RS amp FD toggle field Press enter to view the following graph Figure 3 7 The Scatter P et of the Log Transformed Data ASSESS 1 0 3 8 DECEMBER 1990 ASSESS 1 0 The now fall along a straight horizontal line The variance of the data is now said to be stabilized Press the letter Q and select the Quit option to return to the Quality Assessment Data screen Notice that the screen shows the log transformed data of the input data In order to explore the remaining options of the Transforms menu select the Transform option Select the RS amp PS toggle field and view the graph Press the letter lt Q gt to quit the graph and enter the Transform screen The Unary Operation option was discussed above It may be used to retain the original data so that additional transforms may be applied to the untransformed data Binary Operation This option allows data transformation using the given binary operators on the screen This option has the same
30. mples exist for this variance will be displayed For example sample collection variance can not be calculated because the error estimate calculated from external laboratory evaluation samples ELES does not exist Refer to Table 5 of the Rationale Document 1 for the formulas Also note that the analytical variance could not be oomputed since the estimate from external laboratory evaluation samples ELES pairs can not be obtained This screen also displays a summary of the number of QA samples present Note that the number of ELES values is equal to zero The options on this screen will now be discussed Change Data This option will return the user to the Evaluate Data menu to make additional changes The user will examine the result on the screen and decide if changes are necessary This option will not do anything computationally but is only a means for accessing the data screen You may select this option if there is a need for additional changes This option will take you back to the Quality Assessment Data screen After making changes the Execute option will bring you back to the Results screen This option will be selected after explaining the following options Var Estimates This option will display a screen showing the values of various variance estimates Select the Var Estimates option The following screen will be displayed 2 1990 Figure 3 9 ssassn E 0 The Intermediate Re
31. nsity media and are ready for read write operations when Read Save or Write options are selected The following actions are guaranteed to create a lock up situation WordPerfect is a registered trademark of WordPerfect Corporation ASSESS 1 0 1 5 DECEMBER 1990 rying to print text or graphics screen when the printer is not connected or on line If a printer is connected make sure it is turned and is ready to accept output from the computer on line If a printer is not connected to your system it may be necessary to boot the computer Accessing a file on a floppy diskette drive when the disk drive door is open or no diskette is present In some cases DOS may respond with a message Device not ready Abort Retry Ignore Insert a diskette and press R for Retry If this does not work you must re start the computer ASSESS 1 0 1 6 DECEMBER 1990 2 1 2 1 1 ASSESS Data Files Section 2 System Operation ASSESS employs a particular format for its data files An ASSESS data file is an ASCII text file and is different from the Metacode files used by the HPPLOT program The ASSESS data files may be created with any text editor For instance WordPerfect files may be converted to ASCII text files before being supplied to the program Make sure your data files are compatible with the input format or the program will not be able to read them ASSESS will also operate without an inp
32. of the screen These messages are displayed in a black on white format reverse video and are accompanied with a buzz sound To return to the interactive screen after such a message is displayed press any key Problems Although ASSESS has been tested and debugged it is still possible that there are situations which will cause ASSESS to crash or fail terminate prematurely or to lock up pause indefinitely with no response If ASSESS locks up the computer must be re started The problem may be due to a bug or due to a printer or disk drive problem see below ASSESS may crash when a binary operation in the Transform menu is chosen which would produce a very large or very small number An example would be the operation 1 0 X or 10 to the power of X when X is a very small or very large 1 000 1 E1 000 This type of program error cannot be trapped or handled by ASSESS In such cases the result cannot be produced due to hardware limitations in the precision of the numeric coprocessor or floating point emulation software Since there is no remedy for this situation the only solution is to avoid such operations There are several error conditions which the program was specifically not designed to check for These involve checking to see if disk or printer peripherals are connected and ready for data transfer The user should ensure that printers are attached and on line or that disk drives have the correct de
33. only 10 are shown on the screen Select the Scroll option In the message frame you will notice the message Use arrow key to scroll or Q to Quit As you push the right arrow key you will notice that additional columns of input will appear on the screen Note that missing values are represented by blanks on the screen thus improving readability The down arrow key will let you scroll and observe additional batches if any After scrolling through the screen press the letter lt Q gt to get back to the Scroll option 1990 Figure 3 5 Scatter plot of the QA data ASSESS 1 0 Revise This option is the same as previously discussed You may alter any of the input values by a sequence of arrow keys to get to the correct position and ping over the old values Note that the Save Data option must be selected o save any new changes before proceeding to transformations of the data or execution of the program For the time being do not alter any of the data Proceed to the following option Transform The Transform option is used as a tool to determine if the data needs to be transformed and to carry out various data transformations The Transform option plots a scattergram of the data When the Transform option is selected you will have a choice as to the type of data to be plotted Select this option and you will be prompted by Specify plot of which concentration to use use space
34. ot on an HP plotter Notice that you may also press the letter P to produce a hardcopy of the plot Make sure that the printer is on line For more information on how to port this file into WordPerfect refer to section 1 5 2 and consult your WordPerfect manuals Report This option generates a report of the data and its results in three formats The output may go to a file with input format a file with report format and to the printer Select this option and select the file report format toggle field Type in results fil after the Report File prompt Quit This option will return the user to the main menu In order to return to the Evaluate Data menu select the Change Data option You are now located at the Quality Assessment Data screen Save Data This option allows the user to save the contents of the screen in a data file with the input format Select this opt ion and type in Saved1 fil in response to the File prompt Quit This option will return the user to the main menu This is a good time to take a break Select the Quit option of the Quality Assessment Data screen The resulting screen is the main screen of ASSESS Select the Quit option to end this session of ASSESS We will return to ASSESS to demonstrate input from keyboard Now the basic concept of ASSESS has been studied two shorter examples dealing with input data from the keyboard and using alternative QA design will be off
35. otter instruction file may be produced using the program HPPLOT with the metacode file as input The historical information measurement error variances and data set maybe written in a report format to either an output file or sent to a printer The printer must be either an Epson or IBM compatible dot matrix printer HP Laserjet or HP Deskjet ASSESS may be used to provide a foundation for answering two basic questions How many and what type of samples are required to assess the quality of data in afield sampling effort How can the information from the QA samples be used to identity and control sources of error and uncertainties in the measurement process Once the analytical results are received bias and precision values will be computed Note that ASSESS will only compute the precision An alternative QA design that does not employ field evaluation samples FES and external laboratory evaluation samples ELES is also discussed and error variances are computed ASSESS uses two temporary files called scratch files to store and process data read from an ASSESS data file These scratch files are assigned the names XXXXXXXX XXX and ZZZZZZZZ ZZZ The concentrations units of the samples may be specified for notational purposes but unit conversions cannot be made by ASSESS 1 1 1990 ASSESS 1 0 It is recommended that the EPA publication A Rationale for the Assess ment of Errors in the Sampling of Soils l
36. ples error estimate between batches External laboratory evaluation samples error estimate Batch field duplicates error estimate Total measurement error variance Sampling error variance Between batch variance Subsampling variance Handling variance Analytical variance sampling and handling variance Analytical and subsampling variance Number of sample pairs Number of batch field duplicate samples Number of sample locations for the alternative design Ca DECEMBER 1990
37. ppy Diskette 14 1 4 4 Using ASSESS from DOS sss 1 4 1 5 1 4 1 5 1 On Screen Graphics 14 1 5 2 Metacoda Based 1 4 1 6 ERROR AND RECOVERY PROCEDURES 15 2 1 DATA sss 2 1 2 1 1 ASSESS Data Files rrr 2 2 INTERACTIVE SCREENS nemen 2 2 2 2 1 SCREEN For unreal 2 2 2 Types of Screen input 2 3 2 2 3 The Menu Tree ias 2 4 2 2 4 ASSESS Screens and Menus 2 4 31 T 31 OVERVIEW 3 2 THE EXAMPLE SET wissen 3 3 DATA INPUT THROUGH THE KEYBOARD AN EXAMPLE 3 4 ALTERNATIVE DESIGN AN EXAMPLE ssn 3 15 Appendix A DATA FILES nrn M Appendix B REFERENCES coccion B 1 Appendix C NOMENCLATURE anne 1 iii DECEMBER 1990 Number List of Figures Page 2 1 2 2 3 1 3 2 3 3 3 4 3 5 3 6 3 7 3 8 3 9 3 10 3 11 3 12 3 13 3 14 Example Interactive Screen wn The Menu mmn ASSESS Screen susie i ipn pierde ein Sampling Considerations Screen occ Historic Assessment Screen ga Quality Assessment Data Screen Scatter Plot of the QA Data neni 977 Transforms Screen nennen The Scatter Plot of
38. ption in Data File Examp e Standard fi concerning the format for the quality assessment data line 43 These 9999 values permit ASSESS to read the data and convert the 9999 s to empty fields screen labels precede the entries making the file self explanatory Figure 1 Example fil data file subsampling no contamination 0000 subsampling contamination 0000 DATA QUALITY OBJECTIVES BPO Bite Palmerton Penn Phase I analytical no contamination 0000 Method cored analytical oontamination 0000 Analyte source Analytical Method data handling 0000 souroe Desired Accuracy 20 Desired Precision 20 SUMMARY Of RESULTS Desired Confidence Ranges 95 number of R8 20 Bias 08 number of FD 10 Precision 0 number of 10 number of FEB 3 BAMPLING CONSIDERATIONS numher of BLES 0 Number of Samples 300 Batches 20 Total Measurement Error Variance 5639 338000 Costs for Sample Collection Variance Insufficient number of samples Sampie Collection 00 Between batch Variance 748 083400 Analysis Subsampling Variance Insufficient number of samples Handling Variance Insufficient number of samples Batch Data QUALITY EVALUATION DATA Transform None Wumber Sampling Crew Analytical Lab Concentrations mg kg 1 single single END BATCH DATA Batch RS FD PS FESi FES2 ELESi 21152 1 9999 000 9999 000 9999 000 448 000 505 000 9999 000
39. r lt Q gt Enter Prompt Yes The Alt2 fil is also included in the distribution diskette This file contains data necessary to carry out the alternative design where Field Duplicate Samples are taken from different locations Table 6 of the Rationale Document l is used for the calculations Note that for this design the error estimate for batch field duplicate 35 5 is calculated sing the equation denotated for part b ASSESS 1 0 DECEMBER 1990 ASSESS 1 0 Data Files The data files are simple ASCII text files which may be created with any text editor and can be printed using the DOS print command ASSESS can produce two types of output files both of which contain the historical information for notational purposes only displayed on the screens the summary of results and the quality assessment data One of these output files can be read in as a data file by ASSESS see Example fil next page whereas the other output file uses a report like format see Report of Results which follows and is not readable by ASSESS A third file see Standard fil which follows produced only by the user is readable by ASSESS but lacks the descriptive labels included in the other two output files The third file type is provided so that the user may edit an input file without having to enter all the data via ASSESS this method is faster in cases where the number of quality assessment data is large
40. s before select 2 9 DECEMBER 1990 ASSESS 1 0 Figure 3 1 ASSESS Main Screen a particular field by pressing the enter key You are then prompted for a data file name Notice the cursor jumps to the Datafile field on the screen Type the name of the input data file which in this case is Standard fil and press the enter key A short tuned noise will be generated and the message ERROR data file not found press any key will be displayed if an incorrect or a non existent file is entered In such a case select the Data option and proceed with correct entries The following screen will be displayed Figure 3 1 This option allows the user to select the alternative QA design without the FES and ELES values Since the first example does not employ the alternative design DO NOT select this option Use your cursor keys to move o the Revise option Revise This option is used to revise the screen parameters Note that the parameters on the second part of the screen displaying the site method etc are only for historical notations and have no bearing on the result of the analysis Note also that the displayed information is the result of the input given in the file standard fil You may proceed to the next option by not selecting the Revise option thus accepting the listed parameters How ever for illustration purposes select the Revise option move the cursor key to the Analyte field and t
41. ssment program was implemented to assess variability from the collection handling and analysis of the samples 3 1 DECEMBER 1990 3 2 THE EXAMPLE DATA SET ASSESS 1 0 The standard fil data set is an ASCII file in the ASSESS format It contains data grouped into 23 batches The file structure is described in Section 2 The first few lines of data without the historical information are as follows 9999 000 9999 000 9999 000 448 000 505 000 9999 000 9999 000 2 9999 000 9999 000 9999 000 475 000 488 000 9999 000 9999 000 3 9999 000 9999 000 9999 000 423 000 424 000 9999 000 9999 000 4 389 000 9999 000 430 000 9999 000 9999 000 9999 000 9999 000 Each row represents a batch The first through the third rows indicate that only Field Evaluation Sample pairs FES1 and FES2 exist and all other samples are unknown i e missing The fourth batch indicates that only a routine sample RS and a preparation split sample PS exist Notice that the ASSESS data files for the regular design consist of the following columns in the given order RS FD PS FES1 FES2 ELES1 ELES2 Assuming that you have already copied the software and data into a directory called ASSESS on your hard disk and have used the command CD ASSESS to access the directory you can run ASSESS by typing the command ASSESS enter When the program begins execution it first displays a screen with intro
42. sults enn 5 5 7 x t xS Screen 3 The table shows the variance estimates given in Table 5 of the Rationale Document l Press any key to return to the Results screen Plot This option is used to display a plot of the error bar graph Note the message on the message line Select the Plot option and press any key to view the following plot Figure 3 10 The Error Bar Graph DPI PP Tete eT era poat 6 3 e ASSESS 1 0 DECEMBER 1990 ASSESS 1 0 The plot illustrates the range in which the estimates of the various variance components may be expected to occur within a 95 confidence interval It is clear from the length of the line for sbfes Field Evaluation Samples estimate that greater use of field evaluation samples would have improved the assessment of between batch variability as well as total measurement error variance Table 3 of the Rationale Document 1 is used to determine the 9596 confidence intervals for variances based on the degrees of freedom Press the letter Q to return to the Results screen Save Plot This option saves the error bar chart of the plot just viewed Select this option and type in barchart met in response to the Mets File prompt After the metacode file has been written press any key to return to the menu bar A Metacode file is a device independent file used by the program HPPLOT to produce a hardcopy of the pl
43. tes required with Hpplot exe The files denoted as Used for Comparison will be created by ASSESS as you run through the program Therefore these files may reside somewhere 1 3 DECEMBER 1990 1 4 2 Hard Disk Installation 1 4 3 Using ASSESS on Floppy Diskette 1 4 4 Using ASSESS from DOS 15 GRAPHICS 1 5 1 On Screen Graphics 1 5 2 Metacode Based Graphics ASSESS 1 0 outside the directory where ASSESS resides and thus will not be overwrit ten The source code is written in FORTRAN 77 for the Microsoft Microsoft Corporataion Redmond WA FORTRAN compiler version 4 01 With the exception of slightly modified proprietary Graflib Version 1 0 Sutrasoft Sugarland TX subroutines used for generating screen graphics the source code is also in the public domain To install the system on a fixed disk a subdirectory should first be created for example ASSESS Copy ASSESS exe and any other optional files into the subdirectory For more information on creating subdirectories and copying files from a diskette into a subdirectory refer to your DOS user s manual ASSESS exe is too large to fit on a 360 kilobyte diskette This means that if only a 360 kilobyte disk drive and no fixed disk is available then ASSESS cannot be used Either a 3 5 disk drive or a 1 2 megabyte or larger 5 25 disk drive is needed to run both ASSESS and the optional program Hpplot exe with Hershy bar To run ASSESS or Hpplot from DOS typ
44. ut data file in which case the screen layout for data input will be self explanatory Three example data files included with the distribution diskette will be used as examples in this manual Copy these files into the subdirectory where the Program ASSESS resides They are called standard fil altl and smdata fil Below is an explanation of the standard fil data file The format of the other two files is the same and will be discussed later It is helpful to obtain a print out of these files using the DOS print command before proceeding to the next section Line 1 to 42 These lines of data represent the historical information The data file includes information about the site methods desired accuracy precision etc Note that these input lines are inconsequential to the data processing phase of ASSESS and are merely supplied to keep track of sampling considerations and historical assessments Line 43 to End of File The Data Entries This is where the data are stored Each row represents a batch sample consisting of a pair of QA samples For the regular design these QA samples are stored under the following columns Routine Sample RS Field Duplicate F D Preparation Split PS Field Evaluation Sample Pairs FES1 and FES2 and External Laboratoy Evaluation Sample pairs ELES1 and ELES2 For the Alternative QA design only FES1 and ELES1 values used for detection of bias are included and two new columns ar
45. ype in lead over Pb and press the lt enter gt key Pressing the enter key till the end of the screen parameters or the left arrow key at anytime will take you back to the Revise option on the menu fields Sampling Considerations This option displays the Sampling Considerations menu It is designed only for historical notations and is merely a summary of the contents of parts DECEMBER 1990 of the data file NOT this option yet We will discuss this after examining the remaining options on the screen Evaluate Data This option displays the Evaluate Data menu Through this menu ASSESS does all of its computing work It computes and displays measure ment error variances It also performs data transformations produces reports and plots data This option may be selected immediately after the specification of data input type and will bypass the sampling consideration and historical assessment screens Experienced users will select this opt ion first if quick data evaluation is required thus bypassing the historical creens uit This option is used to exit the program Using the analogy of the menu tree the Quit option also allows you to move up one level in the tree The Quit option will also appear in other successive screens When this option is used from the main menu of the program a Yes No prompt is issued Do you really want to quit lt Y N gt The lt Q gt key is typically use
Download Pdf Manuals
Related Search