ROSTER is a gradebook software program on the mainframe computer. The program accesses a specially created Names file which contains student names and ID numbers and can store up to 19 scores. The commands in ROSTER control the way scores are stored, types of output produced, and making changes to the names file. The program receives input directly from the GRADER program or from data that has been entered into a file from a computer.
The Names file is a database file. It contains student names, id numbers, and the reference number of the class in which the student is enrolled. The names file must be setup by the Test Scoring Coordinator at Computing and Network Services in Nichols Hall. It is stored as a data file on a mainframe direct-access storage device. Computing and Network Services obtains data from the Registar's office that contains names and id numbers of all students who are enrolled in each reference number specified in the Schedule of Classes. The data is received several times each semester. The first is at the end of the first week of classes and then each week until the official 20th day of classes. This database can only be accessed by the GRADER/ROSTER programs.
The user must remember that when the names file is created it contains only the students who were enrolled at the time the data was received from the Registar. If students have dropped or added the class the users must make those changes to the names file using the ROSTER program change commands.
When GRADER and ROSTER are run together, the GRADER program will pass the scores to the ROSTER program for storage and accumulation. Since the two programs can be run seperately or together, there are a number of ways the programs can be used.
Computing and Network Services will set up and process exams. A charge is billed to the department. To avoid charges made for the processing service, users are urged to do their own test scoring. Help is available from this handout "Getting Started with GRADER/ROSTER", the GRADER/ROSTER Users Guide, and consulting from the Test Scoring Coordinator at Computing and Network Services in Nichols Hall.
If you are not familiar with the KSUVM environment, there is a handout available from the Test Scoring Coordinator. It explains the basics of Xedit, OSJ, DEPRINT, QUERY READER ALL (Q RDR ALL), and how to print, etc.
An OMR scanner reads the test scoring answer sheets and converts them into a text data file containing one record for each answer sheet read.
A record in the text data file will look like this:
123456789, 111--, '14253423142534231425342314253424231425342314253423' |...+....1....+....2....+....3....+....4....+....5....+....6....+....7...The first line shown here is the actual data. The second line is a scale line to help determine the actual column number where data is located. The line preceding the scale is referred to as the "current line." In the record a hyphen (-) appears when a field was not marked or marked too lightly to be read and an asterisk (*) appears when more that one option in a field was marked or was not erased properly. Generaly only one mark per field is allowed. There are cases when multiple marked are allowed. This m ultiple option is selected at the time the answer sheets are scanned.
Columns 1-9 contain the student ID number, followed by a comma in column 10.
Columns 12-14 contain data from the optional fields. Column 12 is for the test version number, column 13 is for the answer sheet number, column 14 is for the permission to post score. Columns 15 and 16 are blank. This section is also followed by a comma.
Column 22-71 contain the responses to the exam question 1-50. This section starts in column 21 with an apostrophe, followed by the responses, and the section ends with an apostrophe.
The key record contains the correct answers for the exam. The weight record contains individual weighting of questions.The type record is used for multiple response questions.
Columns 1-9 (ID field) of the answer sheet is blank (---------).
Column 12 of the answer sheet is used to identify which parameter is being used. A "1" is used to identify a key record, a "2" is used to identify a weight record, and a "3" is used to identify a type record.
Column 13 of the answer sheet is used to identify which parameter record it is. There can be up to four records used for each parameter. (There can be up to 200 questions for the exam, four records of 50 questions each.) A "1" is used to identify the first fifty questions (1-50), a "2" is used to identify the second fifty questions (51-100), a "3" for the third fifty questions 101-150, and "4" for the fourth fifty questions 151-200.
The following is an example of key records for an exam with 100 questions (2 answer sheets). The first two lines are the acutual exam key records. The third line is a scale line to help determine the actual column number where the data is located.
---------, 11---, '14253423142534231425342314253424231425342314253423' ---------, 12---, '25241342425242214425243425241324231423414251424152' |...+....1....+....2....+....3....+....4....+....5....+....6....+....7...This is an example for exam weights for an exam with 100 questions. Questions 1-25 are worth 1 point each, questions 26-51 are worth 2 points each, questions 51-75 are worth 3 points each, and question 76-100 are worth 4 points each.
---------, 21---, '11111111111111111111111222222222222222222222222222' ---------, 22---, '33333333333333333333333444444444444444444444444444' |...+....1....+....2....+....3....+....4....+....5....+....6....+....7...The type parameter is only used with multiple reponse questions, when a student can mark more than one correct response. This parameter is not use very often. More detailed use of this parameter is found in the GRADER/ROSTER User's Guide.
1. Scan the exam parameter answer sheets and student answer answer sheets into two seperate data files (one containing the exam parameter data and the other containing the student answer sheet data). Use the NCS OpScan5 scanner in the hallway next to Room 17 in the basement of Nichols Hall. Upload the two data files to your user ID account on the KSUVM mainframe computer.
2. Logon to KSUVM at another computer terminal. Public labs, with mainframe terminals and high speed printers, are available.
3. Check the data files for errors. (The students and the scanner are not perfect.) Edit the data files with Xedit, the text editing program on KSUVM. To edit the data file, enter the command: X filename filetype
The filename and filetype are the names of a data file. All data files on KSUVM will have a filename and a filetype. Look for data not penciled in properly (that is, lightly marked), poor erasures, and unmarked fields. Data not detected or not correctly read will be represented by a hyphen (-). Multiple marked that are read will be represented by an asterisk (*). To help identify the question number in this field enter the command: COLUMN OMR
This will change the regular column scale to a special answer number scale:
123456789, --1--, '14253423142534231425342314253424231425342314253423'
Answer Number: ....+....1....+....2....+....3....+....4....+....5
When finished, to save the changes and exit the Xedit program, enter
the command: FILE
4. Modify the program. Edit the program file with Xedit and change parameters as needed on the commands and the filename and/or filetype in the "include" records for the new exam. Enter the command: X filename filetype
Here filename and filetype are the names of the program file. Make the needed changes. To save the changes and exit the Xedit program, enter the command: FILE
5. To send the program file to the computer for execution, enter the command: OSJ filename filetype
Here filename and filetype are the names of the program file that you just finished editing. After the job has finished running, print the results.
|...+....1....+....2....+....3....+....4....+....5....+....6....+....7... //*++ TIME ,09 VMMSG ENQ OMRGRADE EXPAND //OMRGRADE EXEC OMRGRAD,ACTNOF4=044,IDNO=xxxx,PRM='GRADER,ROSTER' //OPTS DD * ITEMS 50 ID 9,3,0 LISTS 1,1 POSTS 1,3 ALPHA 1 TITLE 'COURSE NAME - SEMESTER - EXAM NUMBER' //*$$ INCLUDE filename filetype (OMR PARAMETER DATA FILE) //OMRCARD DD * //*$$ INCLUDE filename filetype (STUDENT DATA FILE) //MANAGE.CONTROL DD * GRADER 1 COURSE 'COURSE NAME - SEMESTER' SCORE 1,,'EXAM NAME' LIST 1,ALPHA LIST 1,NUM ANALYSIS 1,0,HIST /* (END OF PROGRAM FILE)Lines that begin with a slash (/) are Job Control Language (JCL). They tell the computer operating system (OS) what to do. All other lines are GRADER or ROSTER program commands. GRADER commands follow the "//OPTS DD *" line and ROSTER commands follow the "//MANAGE.CONTROL DD *" line. You must replace the xxxx on the EXEC line with a 4 digit number. That number will be given to you by the Test Scoring Coordinator and will let the program access the names file created for the class.
There are two lines that begin with //*$$ INCLUDE. The first occurrence is for the file containing the exam parameter data. The second occurrence is for the file containing the student answer data. Enter the filename and filetype of the respective files that were uploaded from the OMR scanner.
Data enclosed by apostrophes on the TITLE, COURSE, and SCORE commands is information used as titles on the printout to give it identity. The descriptions provided in those fields is determined by the person running the exam.
This is a brief description of each GRADER command used above:
The ITEMS command is the number of the questions on the test.
The ID command always stays the same. The "9" refers to the nine columns of the student ID field. The "3" refers to the three optional fields. (See step 3 in the preceding section.) The "0" means to always sort on the ID field.
The LISTS command defines how many unsorted and sorted (by ID number) lists of scores to create.
The POST command defines how many lists for posting to create.
The ALPHA command defines how many alphabetical lists (Name, ID, and Score) to create.
Five (5) is the maximum number of any kind of list that can be requested.
The TITLE command will put a heading at the top of each page of the GRADER printout. The data that goes in the heading is enclosed within the apostrophes. Course name, semester and exam number are common data used.
This is a brief description of each ROSTER command used above:
The GRADER command tells the program that a list of student ID numbers and scores will be prepared by the GRADER program to be used by the ROSTER program. The number on the command is the actual column (1-19) of the names file that the scores will be entered into.
The COURSE command will put a heading at the top of each page of the ROSTER printout. The data for the heading is enclosed within the apostrophes. Course name and semester are common data used.
The SCORE command will give a title to each column used. The number is the column used and a title description is enclosed in apostrophes. The exam number is commonly used for the title description.
The two LIST command will create one alphabetical and one numeric (ID) list that gives the score for each column used and an accumulative score.
The ANALYSIS command will create a score analysis and histogram of the accumulative score.