The 640 MEG Shareware Studio 2

home *** CD-ROM | disk | FTP | other *** search

/ The 640 MEG Shareware Studio 2 / The 640 Meg Shareware Studio CD-ROM Volume II (Data Express)(1993).ISO / business / mtgp23.zip / MTG23.DOC < prev next >

Wrap

Text File | 1992-08-14 | 135KB | 3,184 lines

MAKING THE GRADE Version 2.3 for the IBM PC and compatibles by Rick Pedley Copyright (c)1990-1992 Published by Coffee Mug Software 95 Notch Hill Road, Suite 306 Kingston, Ont., Canada K7M 5Y6 DOCUMENTATION FOR SHAREWARE VERSION TABLE OF CONTENTS Introduction ........................................ 1 School And School Board Licenses .................... 2 Hardware Requirements ............................... 2 Shareware Version Notes ............................. 2 Getting Started ..................................... 4 Quick Start ......................................... 5 Administrative Details ............................ 6 Using Edit ........................................ 6 Printing A Report ................................ 10 Tutorial ........................................... 11 Changing The Program Setup ....................... 12 Standard Data Disk Location .................... 12 Creating A New Subdirectory .................... 12 Printer Setup .................................. 13 Letter Grades .................................. 13 Optional Categories ............................ 14 Installing A Password .......................... 14 Editing A Class .................................. 15 The Test Screen ................................ 15 The Student Screen ............................. 17 Editing The Sample File .......................... 18 A New Student .................................. 18 A New Test ..................................... 19 Determining How To Weight A Test ............... 19 Using The 'Total Marks' Method ................. 21 Group Weight Adjustment ........................ 22 Converting Total Marks To Weights .............. 25 Notes Editor ................................... 25 Using The Editor ............................... 26 Attendance ..................................... 28 Deleting Or Moving A Student Offline ........... 29 Deleting A Test ................................ 30 Deleting A Class ............................... 30 Hard Copies .................................... 31 Class Stats .................................... 31 Setting The Trend Period ....................... 32 Arranging ...................................... 32 Arranging By Trend ............................. 33 Saving A File .................................. 34 Quick-Saving A File ............................ 34 Temporary Exit To DOS .......................... 34 Quitting ....................................... 35 TABLE OF CONTENTS Using Graph ........................................ 35 Graphing A Student ............................... 35 Graphing A Test .................................. 36 Correlation Coefficient .......................... 39 Adjusting Grades ................................. 41 Graphing A Class ................................. 43 Other Options In Graph ........................... 44 Other Ways To Adjust Scores ...................... 44 More Than 40 Sets Of Scores? ..................... 44 Using Report ....................................... 46 Progress Report .................................. 47 Gradebook Report ................................. 49 Attendance Report ................................ 49 Student Information Report ....................... 50 Missing Assignment Report ........................ 50 Class Summary Report ............................. 50 Blank Attendance ................................. 51 Blank Seating Plan ............................... 51 Blank Score Sheet ................................ 51 Individual Grade Report .......................... 51 Graph Paper ...................................... 52 Lined Paper ...................................... 52 Backing Up Your Data ............................... 52 - 1 - INTRODUCTION Making The Grade is a complete system for recording and processing student grades. The program can accommodate up to forty students and an unlimited number of test scores per class. As well, as many as fifteen students can be moved 'offline', their academic records preserved for future use. Scores are entered on a customized spreadsheet-like screen, and the many built-in prompts and safeguards help to ensure straightforward, error-free record keeping. Percentages and weighted averages are calculated automatically as data is entered, so that at any given moment, all values shown are up-to-date. Provision is made for up to 18 letter grades and any passing grade you wish. MTG distinguishes between a score of zero, and a 'no score', allowing you to include or ignore missing scores when calculating test averages, student averages, and final grades. Apart from data related purely to grades, MTG records due dates for assignments, names and descriptions for each assignment, course descriptions, and provides two categories in which to record student-specific information such as student identification numbers, telephone numbers, homeroom numbers -- it's entirely up to you. MTG can also track absenteeism, truancy, and times late. Notes can be kept on any test or student by means of a built-in text editor. Any item -- a student, a test, or the entire class -- can be graphed. Simply move the cursor onto the item you wish to graph, and press the appropriate command key. Any graph (in fact, any screen) can be printed. 'Cumulative average' histograms (bar graphs) show a student's progress compared to his/her classmates over a period of time. 'Frequency distribution' histograms indicate, in a way that numerical data cannot, how test scores are distributed. Standard deviation, median, average, range, percen- tage failing, and correlation coefficient are displayed on the graph. The standard deviation and average can be temporarily changed, resulting in a new distribution of scores. A temporary set of new raw scores is also shown for comparison. Thirteen reports are available and can be printed in draft or near letter quality mode. Report types include a progress report, gradebook printouts, blank seating plan, blank attendance record, attendance report, class summary, student information report, missing assignments report, blank mark sheet, individual grade report, graph paper, and lined paper. Most reports have a variety of customizing options, such as date format, titles, leaving some data categories blank, the number of copies to be printed, etc. MTG uses many of the enhanced capabilities that most printers have today, such as headlined titles, bold print, letter quality, and - 2 - underlining. The user interface is easy to learn, and help screens are available for each of the main program segments: Edit, Graph, and Report. SCHOOL AND SCHOOL BOARD LICENSES MTG can now be configured for an individual user, a school license, or a school board license. In the case of a school or school board license, provision has been made in the Change Setup menu for a teacher to enter his/her name, and the school's name (or board name in the case of a board license) is 'hard-coded' into the various reports. License fees for schools and school boards are assessed according to the method outlined in the SCHOOL.LIC document file on the program disk. Please enquire in writing for further information at this address: Coffee Mug Software 95 Notch Hill Road, Suite 306 Kingston, Ont. K7M 5Y6 Canada HARDWARE REQUIREMENTS The minimum requirements to run MTG are a PC-compatible with 512K of RAM, any video, a single 5 1/4" or 3.5" drive, and a printer. A color monitor is preferable but not essential. Some editing in MTG can be done with a Microsoft mouse (or compatible -- Logitech's mouse works fine), but a mouse is not essential. MTG works with any video video adapter, and automatically adjusts for Hercules and MDA (mono) adapters. To force MTG to start up with mono colors (black, white, bright white), type MTG /m or MTG /b at the DOS prompt. DOS or PC DOS 3.0 is the minimum version of DOS supported. (DOS 2.1 has a known problem with the SHELL command; upgrade to 3.0 or higher if possible.) SHAREWARE VERSION NOTES This is a shareware version of Making The Grade (hereafter referred to as MTG). The shareware version is designed to allow you to try out the program for a period of time at no cost, be- fore deciding whether or not to buy the commercial version. Both versions are unprotected and can be easily backed up or copied to a hard disk. - 3 - If you decide that MTG meets your requirements and wish to continue using it, you must then register the software by sending in a completed order form (which can be printed from the Main Menu) with a check or money order for $33 US ($37 Cdn) plus $3 shipping & handling. Shareware operates on the principles of trust and honesty: you get to thoroughly test a completely functional program before purchasing; the programmer can offer his/her products at a lower price than similar commercial products because of the highly efficient way in which shareware is distributed. You are encouraged to make as many copies as you like of the shareware version of MTG - please give them to your friends. If you purchase the commercial version of MTG, you may make as many back-up copies for your own use as you wish. However, it is a violation of the copyright laws of Canada and the United States to duplicate the commercial version of MTG for other than the purchaser's personal use. The shareware version contains all the features of the commercial version, with the following exceptions: o normally, a user-defined password can be installed at the opening screen, giving some measure of pro- tection against unauthorized access to your files. This feature has been disabled in the shareware version to enable copies to be more easily passed along. o the commercial version displays the owner's name at the top of the title screen, instead of the "This is a SHAREWARE version of..." line you see there now. Aside from discouraging people from passing unauthorized copies around, this feature serves a more useful purpose: your name will automatically be inserted into any of the printed reports that require a teacher's name. o the commercial version includes a printed manual with an index, and a handy command chart summariz- ing all keyboard commands. o the commercial version can be licensed by a school or school board, while the shareware version is configured for a single user only. o the commercial version comes with a free copy of Word Challenge, a four-in-one, 12,000-word word game, a $25 value. - 4 - To order the commercial version of MTG, select 'Print an order form' from the Main Menu. When completing the order form be sure to print your full name in BLOCK LETTERS, since this is the name that will appear on the title screen and on reports (decide whether you wish to be officially known as Bill, Liz, or William, Elizabeth, etc). Please make payment by cheque or money order (personal cheques accepted). Packages are shipped the day after orders are received. GETTING STARTED The first thing you should do is make at least one copy of the program disk, and then put the original away in a safe place. Always use a copy, never the original. To make a copy using drive A:, make sure your program disk is write-protected, insert the disk, and then type: DISKCOPY A: A: (case is unimportant) If you want to run MTG from a hard disk, insert the floppy disk and type: A: (then press Enter) Then type: INSTALL The installation utility will copy all program files to a subdirectory named \MTG on C: drive, or you may specify another drive and/or subdirectory. You may then want to create another subdirectory under \MTG for storing data files, \MTG\GRADES for example, or a different subdirectory for each semester, e.g. \FALL92, \SPRING93, \FALL93,... (this can be done from Main Menu or the Change Setup menu by first choosing 'Select data pathname'). If you aren't using a hard disk, you should format a floppy disk on which to store your data files. A 360K floppy disk can comfortably store ten full data files. If you are the adventurous sort, and you have a list of students handy, and perhaps a set of scores that you would like to enter, proceed with the next section, Quick Start. Otherwise, jump ahead to the Tutorial, which uses the sample data file that has been provided, CHE3A1!. ____________________________ - 5 - QUICK START Type MTG at the prompt to run Making The Grade. Be sure to first 'log on' to the drive and pathname from which you are running MTG. For example, if you are currently logged onto drive B:, don't put the program disk in drive A:\> and type A:MTG. This will seem to work at first, until MTG looks for support files and is unable to find them. Instead, log onto drive A: first by typing A: (press Enter). Then run MTG. If you are running MTG from a subdirectory on a hard disk, be sure to add the subdirectory name to the pathname. From the Main Menu screen, select 'Change the setup' by pressing the appropriate number, or using the arrow keys and Enter. From the Change Setup menu, choose the printer you wish to use, and whether your printer needs an extra linefeed or not (say no for now). You should also check the letter grade format to see if it matches the format used by your school; change it if necessary. Two data categories are provided to store information on each student. By default, these are homeroom number and student number, but they can be changed if you wish. You can also specify where MTG should look for your data files each time you run the program. This pathname is stored in the setup file, and is displayed at the top of the Main Menu screen. At this point you may also install (or remove) a password. If you use MTG at school or you use the Notes editor, a password is advisable. When you have finished changing the setup, exit to the Main Menu. Once you have selected the drive and pathname where your files are to be stored (displayed at the top of the screen), you're ready to start a new class. Select 'Start a new class' from Main Menu. This brings up the following: 1. Start from scratch 2. Duplicate an existing roster 'Duplicate an existing roster' is handy if you teach the same (or nearly the same) students more than one subject. From here, you can select an already existing file. MTG will use the list of students from this file, all student information, and such things as the name of the school, and discard all test data. Students can then be added to or deleted from this duplicated roster, and a new class (file) is created. The old file remains completely unchanged. This is also how you can continue another file with more than 40 sets of scores in it; see 'More Than 40 Sets Of Scores?' for more information. - 6 - To begin a completely new class, select 'Start from scratch'. This takes you directly to the Administrative Details screen. ADMINISTRATIVE DETAILS Fill in each of the four categories. The course code serves as the file name. Note the restriction of eight characters for a filename; a list of legal DOS filename characters appears to the right. Once a course name has been entered, MTG checks to see if a file with the same name already exists and warns you of the possibility of over-writing the old file. MTG also checks your data disk for files and automatically grabs the school name and telephone number from the first file it finds on your data disk and fills these in for you on the Administrative Details screen. You may change them at this point, but usually you are saved some extra typing. NOTE: School-licensed copies of MTG already have the school name entered for you. All of the information you enter appears later in printed reports, so enter it accurately and make appropriate use of upper and lower case letters. When you've completed all the information, proceed to the Edit screens. USING EDIT There are two kinds of Edit screens: Student screens and Test screens. A Student screen shows one student's name at the top, along with homeroom, student number, letter grade, and average (SAvg). Below are listed all the tests he has written ('test' is used generically throughout to indicate any assignment, essay, lab, quiz, etc.). A Test screen shows one test name at the top with the maximum score (what the test is out of), a description, date due/marked/assigned, weight, and test average (TAvg). You can work on either type of screen. To switch to the other screen, position the cursor on the item (test or student) you wish to zoom on, and press Alt-Z. That item now appears at the top, with all the relevant information below. NOTE: Pressing Alt-B on an Edit screen cycles the cursor through seven different colors; the default is black. To save a new cursor color, just be sure to visit and exit the Change Setup menu before quitting the program; the change is saved in your SET_UP.MTG file. One color in particular, white (so you'll have bright white lettering on a grayish background), may show up better on mono- chrome monitors and some laptop LCD's. - 7 - The other way to view data is to page through successive Student or Test screens by pressing Pg Dn or Pg Up (or F9 and F10). Currently, there is no data in the file, so there isn't much to see. Load the sample file to better see the effect of 'zooming' and 'paging'. You should decide at this point whether the new file represents one term/semester, or a whole year. Most teachers surveyed indicated that 40 assignments is more than sufficient for a whole year, but in cases where students are continually evaluated (perhaps three times or more a week), two or more files may be necessary. See the section entitled 'More Than 40 Sets Of Scores?' to find out how MTG can accommodate an unlimited number of tests/assignments. It is recommended that the primary school teacher start a new file for each subject; use 'Duplicate a Roster' to avoid retyping the names. If you decide to record a whole year's marks in one file, determine roughly how the weights for each term or semester should be split (you can always change them later). After you have decided to assign 50% of the weight to the first semester, for example, then you can begin to weight each test accordingly. Many teachers are uncomfortable with weighting assignments versus using a total marks scheme, mostly because of the extra arithmetic involved. With MTG, the weighting system is easy to use but you can also use the total marks method if you wish. In general, keep the weight for each assignment small, say 2% to 4%, with larger weights reserved for more important assignments. The total weight cannot be more than 100%, but more importantly, never needs to total 100% at the end of the year. What matters is the relative weight of one assignment to another, not the total weight. If the final exam is assigned a weight of 40%, that leaves up to 60% for the rest of the semester. A weight can be as small as 0.1% (one decimal place is allowed). Actually, a weight can be zero, which removes the test from any calculations. Weights can be changed at any time, even after all the scores have been entered. Everything else is immediately recalculated to reflect the change. If you run out of room (i.e. you reach 100% before the end of semester), use Group Weight Adjustment (Alt-W) to help you scale down current weights to make more room. Since it's the relative weight that counts, not the total, no harm is done. Here's a typical scenario: at the beginning of the semester, you assigned 40% of the total weight to the final exam, and 60% for everything else. At the end of the semester however, you find that the weights for all tests and assignments add up to 35%, not 60%. At first glance, it seems you must increase the weight for every test by some ratio so that the weights total 60% -- a lot of work. But since the total weight need not be 100%, it's much simpler to - 8 - adjust the exam weight so that the ratio remains 60:40. Letting Y represent the new exam weight, here is the simple arithmetic you must do: 60:40 = 35:Y 60 x Y = 40 x 35 60Y = 1400 . Y = 23.3 In other words, 60:40 is the same ratio as 35:23.3; enter 23.3 as the exam weight. Or if you prefer, use Group Weight Adjustment to convert the 35% total to 60%; all of the arithmetic is done for you. Press Alt-Z until you are looking at a Test screen. Using the arrow keys (or press Home), move the cursor to the top of the screen just to the right of 'Test:' and enter a test name. You can enter information in two ways: 1. Simply begin typing. A row of dots appears, indicating how much room there is. To back up and correct something, use the backspace key; to start over, press Esc. To exit the cell and restore whatever was there before, press Esc once more. To finish editing a cell, press Enter or one of the four arrow keys. Depending on how fast your computer is, there may be a slight pause as some calculations take place. 2. If you aren't sure what goes in a particular cell or column, press the space bar. Whenever the first character typed is a space, this signals to MTG that you want an 'edit box'. An edit box is a small window that opens on the screen with a line for data input, but more importantly, with a brief description of the type of information you are expected to enter. Enter data in the same way as in method 1. Mouse users: move the mouse cursor to the cell you want to edit and click the left mouse button once. If you don't want an edit box, begin typing. Click once more for an edit box. Once you have entered some data, click once if you want to clear the line, and once more to exit the cell and restore its former contents. The second Edit help screen covers all mouse functions. A test name can be any six-character name you want; in MTG you are not restricted to pre-set categories. I suggest some sort of naming system though -- something like Lab 1, Lab 2, Test 1, Quiz 1, etc., just so there is consistency, but it's totally up to you; MTG doesn't care what you call them. The large blank space to the - 9 - right of the test name is for a test description. Enter a description of what material the test covers, or what kind of assignment it is. Try not to leave these cells blank; they will be very useful later on. Move to the second row and enter a weight (the edit box tells you how big the weight can be, based on the current total weight), a maximum score (must be a whole number), and a date (the edit box brings up today's date, or you can enter something else). Next, move down to the third row, in the upper left corner (press the space bar here, to be sure you're in the student name column). You're now ready to begin entering the list of students. The preferred format for names is: Smith, John L. Whatever format you use, try to be consistent throughout. These names will later be printed on reports, so neatness is important. Don't worry about the order in which you enter the names; the program can sort them for you later. Move down the left side, and if there are more than twenty names, continue in the upper left corner on the other side. After you enter all the names, you may also want to enter the homeroom numbers and student numbers if you have the data handy (or whatever you have assigned to those categories). When you're all done, press Alt-A (for Arrange) and select '1. Student name' to sort the names alphabetically. While you're doing this, note the other options for sorting. One of the options inverts the list after you have done any one of the other sorts. This is a good time to save the file. Make sure your data disk is in the drive and press Alt-S. Save your work periodically during a session to avoid suddenly losing a lot of work if the power goes off. Almost all commands in MTG are selected from menus or by pressing the Alt key in combination with something else. Press Alt-H (or F1) to look at a help screen for Edit, or refer to the included command chart. At some point, you should try all the commands listed to become familiar with them. There are equivalent help screens for Graph and Report. Notice the use of Home, End, PgUp, PgDn, and their function key equivalents. The function key equivalents allow you to have Num Lock on so you can enter data with the numeric keypad, and still be able to zoom and page (this system works only with keyboards that have a second set of cursor keys, or if you position the cursor exclusively with a mouse). The Tab key functions the same as PgUp. Now you're ready to enter some scores. Move to any cell directly below the maximum score, and enter the 'raw score' (abbreviated as 'RawS' on the Student screen). A raw score can be zero, no score (enter an '*'), or any number up to nine times as large as the maximum, to one decimal place. It is much more useful - 10 - to enter a 'no score' instead of a zero, and as you have noticed, this is the default value when a new test is entered. The useful- ness will be apparent later on, and remember it is a simple matter to change it to a zero at the end of the semester. In the mean time, the student's average is calculated based only on tests for which there are scores. In other words, 'no scores' don't count as zeros 'til you say so. As a raw score is entered, the percentage appears immediately to the right (rounded to the nearest whole number), and in the next column, the student's over-all average. In the second row at the top, the test average (TAvg) is constantly updated as new scores are entered. When all scores have been entered, press Alt-S again to save the file. PRINTING A REPORT It's a good idea to print a gradebook report after each test. The gradebook report serves two purposes. You can print the data for every test given so far. This report contains every piece of data necessary to re-construct a file in the unlikely event the file is damaged and becomes unreadable. It is your 'hard copy' backup of irreplaceable data. The second function of the gradebook report is to print the results of a single test or range of tests, so that they may be posted on a classroom bulletin board. This report gives each student his result on this test, the test average, so he can see how the rest of the class did, his current over-all average, and the class over-all average (the underlined figure in the lower right hand corner). You can also print any range of tests between first and last. NOTE: Some school boards prohibit the posting of marks by name or any other means. Check the policy in your particular board. Press Alt-R to go to the Report menu and print a gradebook report. Notice that you are asked if you want to print names or student numbers on this report. Numbers are handy if you intend to post the results; they preserve anonymity. On the student screen in Edit, one of the data categories is labelled 'SN' for 'student number'. MTG looks for this category, and if you haven't changed it to something else, asks you this question whenever you print a gradebook report or a missing assignment report. Either assign a small number to each student (from 1 to however many students there are) or use something like the school's identification number for each student. The last step in a data entry session is to make a copy of your data disk. A good system is to keep one disk at home and one at school. Of course, the gradebook report is your third, hard copy record. - 11 - To learn more about Edit, Graph, and Report, the best thing to do is exit to the Main Menu and load the sample file: CHE-3A1!. This file contains sixteen complete sets of scores, and better demonstrates the various functions than a file with only one set of scores. TUTORIAL Main Menu is the 'control center' of MTG. The diagram below shows how the various parts of the program are linked, and the possible directions in which you can move from one part to another. Graph <-. ^ | | | | | v Change <------> Main <----> Edit <----> Notes Editor Setup Menu ^ | | | | v | Report <-' From Main Menu the following selections can be made: 1. Edit a class 2. Select data pathname 3. Start a new class 4. Select a new data drive 5. Change program setup 6. Delete a class 7. Quit Making The Grade Press a number key for immediate action or move up and down with the arrow keys to highlight a new selection and then press Enter. Most menus in MTG work in this fashion. - 12 - CHANGING THE PROGRAM SETUP If this is the first time you have used MTG, there are a number of user-definable items to install, so select 'Change program setup'. From the Change Setup menu you can change or do the following: 1. Password 2. Standard data disk location 3. Printer setup 4. Letter grade format 5. Optional categories 6. Return to the Main Menu NOTE: If you have a school or school board licensed copy of MTG, and you're using the program for the first time, item number one should be 'Teacher's Name' instead of 'Password' (if it isn't, you'll need to exit MTG and delete the file SET_UP.MTG, and then run MTG again). When you've entered your name, you'll then be able to enter a password. Be sure to get a copy of MTG from the master disk (given to the licensed school/school board) rather than a copy from a fellow teacher; that way you'll be sure of having a brand new copy without someone else's password and name installed. STANDARD DATA DISK LOCATION Select 'Standard data disk location'; MTG finds and lists all disk drives by letter (hopefully, you're familiar enough with your system to know which drive is represented by each letter). Decide where you want your data disk to be and select the drive. If you are using a hard disk, likely you will want to store data files in a subdirectory. By specifying a subdirectory, MTG will know every time you run the program, where your files are stored. CREATING A NEW SUBDIRECTORY From 'Select Data Pathname' at the Main Menu, and from 'Default Data Disk Location' at the Change Setup menu, you can create a new subdirectory. This option has been added to the list of other options in the pop-up box in the lower right hand corner, and is a convenient feature at the start of a new semester; it's quite simple to create a different subdirectory for new classes, something like \SPRING92 or \FALL92. Subdirectories help keep the various class files organized, like folders and drawers in a file cabinet. The new subdirectory is created just below the level you're at; for example if you are at C:\MTG\GRADES, then the complete path for the new subdirectory would be C:\MTG\GRADES\SPRI- NG92 (or whatever). - 13 - PRINTER SETUP Select 'Printer setup' from the Change Setup menu. The currently installed printer is highlighted. When you run MTG for the first time, a default setup file is created, and the default printer is the Epson FX-85. If you have a different printer, look for it on the list. If your printer is not on the list, try an Epson or one of the two 'Alternate' printers. MTG assumes you have a printer connected to 'LPT1:', a parallel port. If you're not certain, this is where a printer is connected 99.80486% of the time. If MTG detects two parallel ports, an additional selection can be made at the bottom of the screen; you may choose LPT1: or LPT2:. If in doubt, stick with LPT1:, the default. This feature may not work with some print spoolers and has not been tested with networks. If it doesn't work, you may have to pre-configure the parallel ports with a separate utility before running MTG. The next paragraph tells you that your printer currently sends an extra line feed after a carriage return, or does not, and then asks which you would like. Select 'No' for now -- you can always change it later if your printouts are double-spaced, or if everything prints on one line. You can also move a DIP switch to change the extra linefeed status, but I advise against doing this; setting and resetting DIP switches for each software package leads to insanity. NOTE: A highlighted response is a 'default' selection; to accept the default, just press Enter. To select a different response, type the letter or use the arrows and press Enter. LETTER GRADES Next, from the Change Setup menu, select 'Letter grade format'. MTG lets you define a passing percentage between 0% and 100%. You can also define as many as 18 letter grades. If you are satisfied with the default settings, press 'N'. To make changes, press Enter. First, enter a passing percentage (a whole number between 0 and 100). Then, you are asked for a one- or two-character symbol for the highest letter grade. Any visible characters except spaces may be used. Use the left arrow to back up, Esc to start over, and Enter to advance a line. Enter the lower boundary represented by the letter grade; then press Enter. The next line appears, and the process continues, until you enter a zero for the lower boundary, or until you have used up the 18 letter grade categories (why 18, you ask? Some schools use every position between A+ and F-). - 14 - When you're done, MTG displays the new letter grade scheme and passing percentage, and again asks if you want to change it. Answer No to finish and return to the Change Setup menu. TIP: If you don't use letter grades at all, do this: enter a neutral character in the first interval (a '-', or '*' for instance) and then type a zero as the lower boundary of the first interval. OPTIONAL CATEGORIES From Change Setup, select 'Optional categories'. Two categories (fields if you like) are provided to store miscellaneous information pertaining to each student. The default definitions are student number and homeroom, but these can be changed to anything you want. Each definition consists of two parts: a three character acronym or abbreviation, which appears on the Edit screen, and a twelve character description, which serves as the prompt in an edit box, and also shows up on some of the printed reports. Twelve characters in each category are provided for the information itself. INSTALLING A PASSWORD A password offers simple protection against unauthorized tampering. It can also prevent authorized tampering if you forget it, so it is important that you work only with a copy of the original program disk. That way, you can always start with a fresh copy if you forget the password, or your work copy is lost or stolen. If a password has been installed previously, you will first be asked to enter the old password, before being permitted to enter a new one. A password can be any combination of 6 to 20 visible characters. There are over two billion six-character passwords possible, so there is no need to use words like ACCESS, PASSWORD, or SECRET. Lower case letters are automatically converted to upper case. After entering the new password, MTG will write it to the program disk. The password is stored directly into the program file MTG.EXE, which is another reason you should use only a copy of the original disk. If you use MTG at school, or you use the Notes editor, you should install a password; otherwise you probably needn't bother. RETURN TO MAIN MENU If any change to the setup has been made, MTG saves the new setup to the program disk. These changes are now in effect each time you run the program. - 15 - EDITING A CLASS From the Main Menu, choose 'Select data pathname'. When the files are listed, you should see the drive and pathname at the top, and the data file CHE-3A1! listed below. The menu in the lower right corner lists your options at this point. IMPORTANT: Notice also, that the amount of free memory on your data disk is displayed, to show you how full the disk is. One K of data consists of 1024 characters of data (letters and numbers), which is equivalent to about one double-spaced typewritten page. A completely full MTG data file requires 28K bytes of disk space. When there is less than 28K of free space left on the disk, MTG will warn you by flashing the amount of free space and beeping. At this point, you should exit to DOS, format a fresh data disk, and copy some files over to the new disk to free up some room. If you elect to ignore the warning, and keep adding more data to the disk, you may at some point enter a new set of scores, and then be unable to save the new data to disk. If you have a blank disk handy, you could exit temporarily to DOS (Alt-T) and format a new data disk, but it's best to avoid this situation. From the menu, choose 'Accept the pathname above'. From Main Menu, select 'Edit a class', and then 'Select a class to edit'. The cursor jumps to the top of the list of files (as many as 90 files may be listed here). Press Enter on the file CHE-3A1!, and it will load into memory. THE TEST SCREEN In MTG, all assignments, essays, labs, homework assignments, quizzes, exams, etc. are referred to as 'tests'. After pressing Alt-Z, look at the top line on the screen, and you'll see the name of the class displayed, the name of a test (TEST:Quiz.1), and a description of the test (TOPIC: Matter). Each test is given a name up to 6 characters long, and a description up to 43 characters long. This particular file uses a simple numbering scheme for tests, e.g. Lab.1, Lab.2, Quiz.1, but you may label tests any way that makes sense to you. There is no attempt to force you to group marks in specific categories, as some other programs do, which allows maximum flexibility. The description may also contain information about the structure of the test, for example: '2000 words on My Summer Vacation', or '20 multiple choice on Plant Structure'. On the next line, the weight for this test is shown (Wgt:3%). All tests are given a weight between 0% and 100%, and can be as small as 0.1%. If you are not familiar with weighted scores, don't panic; weighting with MTG is very simple to implement, and above - 16 - all, is flexible. If the weight you entered for a test three weeks ago turns out be higher or lower than you want, it can easily be changed to something else, with no extra work on your part. If you have been using the 'total marks' method, this is in fact just another form of weighting, but is less precise. The Test Average (TAvg:87%) shows the class average, as a percentage, on this test. This average is displayed and updated even as you enter the test scores, so that at any point during data entry, you know exactly how the class is doing. The next figure on the second line tells you what maximum the test is scored out of, in this case 25; this number represents the actual number of marks, not a percentage. The column of numbers below the Maximum shows the 'raw score' obtained by each student on the test. For example, Paul Ackerman scored 22 out of 25 on Quiz.1. The next heading (Per) shows the percentage calculated by dividing the raw score by the maximum, and multiplying by 100. As you can see, Ackerman's 22/25 results in a percentage of 88%. Percentages are calculated automatically as you enter the raw scores. The last column, to the right of the percentage figures, shows each student's overall average under the heading 'SAvg'. This figure is also immediately updated as a new score is entered, meaning you always know a student's average -- no guessing. NOTE: the total weight need not be 100% to give accurate student averages; more on this later. 'TDate' can be the date a test is due, the date you marked it, or the date it was assigned. The date may be entered any way you wish: 91-09-10, Sept 10, etc. MTG handles a maximum of 40 students, and they can all be displayed on this screen. 23 spaces are allowed for each name. The names in this file are entered in 'last name, comma, space, first name' format, but you can enter anything here. When first entering the names, it is not necessary to put them in any particular order; MTG can sort them for you later, if you wish. MTG also prevents you from entering the same name twice -- use an initial to dis- tinguish between the two Ernest Pringle's. NOTE: 15 students can also be stored 'offline' in addition to the 40 that are 'online'; see 'Deleting A Student'. - 17 - THE STUDENT SCREEN You may be wondering: 'Where are the other tests?'. Earlier, MTG was referred to as a 'modified spreadsheet'. Rather than scrolling across and down a cumbersome 40 x 40 spreadsheet, MTG organizes all its data into Test screens and Student screens. A Test screen, as you have seen, displays the pertinent information for one test, and the scores obtained by each student on that test. The procedure for handling more than 40 sets of scores is discussed later on. Using the four arrow keys, move the cursor around the screen until you have highlighted 'Watson, Kimberley' (actually, the cursor can be on the student's name, raw score, percentage, or average). Then press Alt-Z. Mouse users: To zoom on any item, move the mouse cursor onto the cell and click the right mouse button. When you press Alt-Z, MTG 'zooms' to the test or student positioned under the cursor. The Test screen you were looking at is replaced by a Student screen. The name: Watson, Kimberley is now displayed at the top of the screen, along with other information about that student, and all tests are listed below. To the right of the student's name on the first line, are the two 'optional categories', in this case set to the defaults: Homeroom and Student Number. On the second line is the student's overall average (SAvg:90%), and his/her letter grade (LG:A). The heading 'Max' tops the column of maximum scores for all tests. 'RawS' heads the column of raw scores on all tests for Kimberley Watson. To the right is the percentage each raw score represents. Further to the right is the weight (Wgt%) assigned to each test, as a percentage. The last column (TAvg) shows how the class performed on each test, also given as a percentage. Looking at the 'Per' column and the 'TAvg' column, you can compare a student's results on any test with the rest of the class. Finally, the heading 'TotWgt' indicates the total of all weights assigned so far, which cannot exceed 100%, or 'TotMks' if you are using the total marks method. After you've zoomed around a few times, press the Pg Dn and Pg Up keys (or F7 and F8). This moves you to the next test or student, depending whether you're on a Test screen, or a Student screen. By paging repeatedly, you will move past the last test (or student), back to the top of the list. Mouse users: To page backwards, move the mouse cursor to the left margin of the screen (on the double vertical line) and click the left mouse button. To page forward, click on the right margin. - 18 - HELP SCREENS Each of the three modules: Edit, Graph, and Report has one or more Help screens, which give you a summary of all commands. To look at a Help screen, press Alt-H or F1. To print a Help screen, press Shift-* (Prt Sc). To print the vertical and horizontal lines properly, your printer must be in IBM mode. The included command chart summarizes the help screen information. EDITING THE SAMPLE FILE To practice entering data before starting your own file, work through the following exercise. The changes you make have no effect on the original sample file stored on disk unless you save the file. A NEW STUDENT 1. Let's add a new student to the list. Move the cursor to the blank line below Zebedee. A blank line is always reserved here for a new student, until all 40 spaces have been filled. Enter the name: Cruthers, Sandy (or your own name, if you like). 2. Now press Alt-A (for Arrange), and select: Arrange by Student Name. MTG will sort all names alphabetically, which is one good reason to enter the names in 'Last, First' order. A lower case 'a' sorts ahead of an upper case 'A', so be sure to capitalize where necessary. You can also sort by Student Average, Total Absences, Unauthorized Absences (all highest to lowest), by Homeroom, Student Number (sorted alphabetically), Times Late, Trend, Random Selection, Results of currently selected test, or by inverting the list. 3. Notice that asterisks appear in the three columns that contain the raw score and percentage for this test, and the student's average. The asterisk indicates there is no score available, rather than a score of zero. In fact, 'no score' is the default value whenever a new student or test name is entered, and remains thus until you enter a number. 4. Move the cursor to the 'raw score' column (the next column over) and enter a raw score for this new student (the maximum score is shown at the top), and you'll see that a percentage is calculated, and also a student average. In this case, the student average is just the score obtained on this one test. A raw score can be specified to one decimal place, may not be smaller than zero, and may be as large as nine times the maximum score. - 19 - 5. Press Alt-Z anywhere on this line, and the new student's name appears at the top of the Student screen. On the list of tests, you can spot the score you just entered, and note that all other scores are 'no scores'. 6. Move to the top line and enter a homeroom and student number; remember that these two categories can be defined to be anything you like. On the second line, note the letter grade based on the student average. 7. Just to see what happens, edit the raw score you just entered, and type an asterisk. You'll notice that the student average and letter grade appear as asterisks. Now enter a zero for the raw score, and again note the changes. A NEW TEST 1. Stay on the Student screen, and move down to the blank line just below 'Exam'. Enter the name of a new test, which we'll call 'Essay1'. 2. Move right to the next column, and press the space bar. Today's date will pop up, which you can either accept by pressing Enter, press Tab to choose a different format, or reject by press- ing any other key and entering your own, in any format you wish. NOTE: today's date will not come up if your computer's clock isn't set correctly; this must be done from DOS. 3. You can now enter a maximum score, say 30, in the next column. A maximum score must always be a whole number, and cannot be zero. If you change a maximum score after a set of raw scores has been entered, the new maximum score can be no smaller than 1/9 of the largest raw score entered; more on this later. 4. Next, move to the 'Wgt%' column. You can see that the total weight (TotWgt) is less than 100%, so you can enter a weight for this test; let's make it 2% (just type a '2'). The edit box always shows you the maximum weight that may be entered, by first subtracting any weight that might already be in the cell you are editing. DETERMINING HOW TO WEIGHT A TEST When you assign a weight to a test, you determine how much the test will contribute toward a student's final grade, compared with all other tests the student writes. MTG does not require that weights add up to 100% at the end of the year -- only that they do not exceed 100%. Accurate student averages can be calculated at - 20 - any time during the school year, even though not all tests have been written, and weights do not sum to 100%. The arithmetic used to do this is very simple. The raw scores are first converted to percentages. Then each percentage is multi- plied by its weight, which has been converted to a decimal. The results are added and this number is divided by the total weight thus far. Here are the calculations for two tests with weights of 4% and 5%: Raw score: 20/25 = 80% --> 80% x .04 = 3.2 Raw score: 4/10 = 40% --> 40% x .05 = 2.0 --- 5.2 Now the total 5.2 is divided by the total weight (.04 + .05): 5.2 --- = 58% --> student's CURRENT average .09 rounded to a whole number If the tests had been weighted evenly (both at 4% for example), the student's average would be simply (80% + 40%)/2, or 60%. The uneven weighting gives more emphasis to the lower score in this example, to yield a lower overall average, even though this test was out of fewer marks. When a student misses a test, and a 'no score' is entered, the weight for that test does not contribute toward (or detract from) the student's average. Similarly, a test with zero weight is removed from all calculations. Weights are always entered as percentages, and can be specified to one decimal place. There are several advantages to using weights, rather than 'total marks', or some other scheme. First, it no longer matters what a test is out of. If you really want to ask 30 questions of equal value, but your 'total marks' scheme dictates that the test should only count for 10 marks worth out of the whole term's total, you can relax. Score the test out of 31 or 67, and assign a weight based on what you think the test is really worth. Secondly, weights give the teacher a better feel for how much a test is worth, than a total marks scheme. See if you can quickly figure out what percentage of the total is contributed by the last maximum score on the list below: - 21 - 65 80 25 40 55 10 100 75 --> 15 The answer is 0.03%. Over the course of a semester, it is easy to over or under-estimate the significance of a test in context with others. On the other hand, if you can ignore what a test is marked out of, and know it is worth 3%, this is a more relevant number, and there is never any doubt as how much it will contribute toward a student's final mark. A third reason weighted scoring is used, is that it is relatively easy to change a weight at any time, especially with a spreadsheet program. Redistributing the 'worth' of a test under the total marks method on the other hand, would presumably involve adjusting the maximum score, and then all of the raw scores, to maintain correct percentages. In conclusion, assign any weight you feel appropriate when you enter new test data. Nothing is carved in stone, and you can always adjust the weights later on, either individually, or with Group Weight Adjustment. Remember it is the relative weight of one test to another that is important. USING THE TOTAL MARKS METHOD Having said all that, beginning with version 2.0, the total marks method can be used instead of weights. Using this method is extremely simple -- don't enter any test weights! If you enter even a single test weight, MTG reverts to the weighting method. If you are using weights, weights can still be zeroed as before (to remove a test from all calculations), but if you zero all tests, 'total marks' kicks in. All other information (mean, median, standard deviation, etc.) is calculated according to whichever method you're using. Switching from one method to the other in mid-semester will most likely affect averages by at least a few percent, so it's a good idea to decide which method you're going to use and then stick to it. When you begin a new file the default is 'total marks' until a test weight is entered, and you'll see the heading 'TotMks' instead of 'TotWgt'. Those who prefer 'total marks' invariably want to use weights at the end of the semester because they want term work to count for 70% of the final mark for example, and the - 22 - final exam to count for 30%. To switch to weights at the end of a semester, follow these steps: a. Finalize all term marks; you need a completed data file except for the final exam. Decide whether 'no scores' should remain as is, or convert them to zeros. b. From the Main Menu, select 'Start a new file' and then 'Duplicate a roster'. Select the file you have just completed and if the class is named ENG2G0 for example, name the new class something like ENG2G0F (for 'final') so you'll still have the original file. c. Reply 'yes' to carry forward student averages (and attendance data if you are using this feature). d. The new file contains a 'test' named 'So.Far' and the test scores are in fact the student averages from the previous file. The weight for this test is currently zero, so change this to whatever you want term work to represent,say 70%. e. Now you can enter the marks for the final exam in a new column and assign this a weight of 30%. Term work and final exam are now correctly represented by weight. 5. To continue with the tutorial, zoom on the test you have just entered (anywhere in the row). The test name is now at the top of the screen, and you can fill in the TOPIC (test description): 1500 words on 'Molecules I Have Known'. 6. Notice that all raw scores and percentages for this new test are 'no scores', which is the default until scores are entered. There are good reasons for leaving a score as 'no score' until a final report is due. These reasons will be discussed in Report. GROUP WEIGHT ADJUSTMENT Although the total weight need never sum to 100 percent, the Group Weight Adjustment feature allows you to do precisely that, among other things. Here are a couple of scenarios in which this feature can be useful: a. At the start of the term, you decided that term work will be worth 60% and the final exam 40%. At the end of the term however (before the exam), the total weight is greater than 60% (or less than 60%, which is no problem). Group Weight Adjustment lets you reduce a total weight of say 75%, to 60%, distributing the reduced amount in the correct proportion among individual test weights. This is done precisely as you would do by hand with a calculator. You can now allot exactly 40% for the final exam. - 23 - b. You wish to adjust the weights of only some tests. For example, you notice that the six labs you have assigned have a total weight of 30%, and you would prefer (perhaps because they were so poorly done) to reduce this total to 20%. Group Weight Adjustment lets you select any group of tests and then specify a new total weight for that group. Obviously, group weight adjustment is unavailable if you are using 'total marks'. USING GROUP WEIGHT ADJUSTMENT Before doing any adjustments, the current file should be saved, including any new data that has been entered. You should then make a hard copy of both screens (any Test screen and any Student screen) by pressing Prt Sc. Then move to a Student screen (which shows all test weights) and press Alt-W: Current total weight is: 75% -- adjust ALL weights? YES NO In a case like that in (a), where the total weight is to be adjusted, answer YES. You are then prompted for a new weight. There are a few things to consider when choosing a new weight: 1. The program requires that there be a large enough difference between the old total weight and the new total weight you have entered. If the difference is too small, a significant rounding error tends to accumulate and it becomes very difficult to maintain the correct ratio of one test weight to another. A rounding error results whenever you must round a newly calculated weight to one decimal place, which is the maximum precision you are allowed. To prevent gross inaccuracies, MTG requires the difference between new total weight and old total weight to be 0.2 x the number of test weights being adjusted. For example if there are 12 tests in the group of weights to be adjusted and the current total weight for that group is 75%, then the new total weight must be either greater than 77.3% or less than 72.7%. Don't be overly concerned about this -- MTG will alert you if there's a problem and will not attempt to make the adjustment. 2. Remember that weights of 0% will be unaffected by a group weight adjustment because 0 x any weight adjustment ratio is still 0. 3. If the new total weight you have entered is much smaller than the original weight, there is a possibility that one or more test weights will reduce to zero through rounding. MTG does the calculations and then warns you of this situation. You then have the option of continuing or backing out. - 24 - 4. After MTG has done the calculations, there is often a small, cumulative rounding error, typically between 0.1% and 0.3%, positive or negative. MTG reports this error and then asks permission to make up the difference by adding or subtracting the error from the largest weight in the group you have selected. Making a corrective adjustment to the largest weight results in the least amount of error (you can see at a glance that one penny is missing from a pile of five, but not from a pile of one hundred). With permission, MTG makes the changes and then advises that you may wish to make further minor adjustments by hand (e.g. take that +0.3% error, split it in three and distribute the parts more evenly). Or you can back out of the operation and try something else. Note that many of your tidy, whole number weights will become decimal fractions after a weight adjustment. 5. Another consequence of rounding error is that a weight adjustment is usually irreversible. In other words, if you change the total weight from 75% to 60% and then back to 75%, it's unlikely you'll have the same individual test weights you began with. Rounding errors are made and compensated for in both directions so that the adjustment is made to a different set of numbers each time. This is why you are advised to get a hard copy of the Student screen -- if necessary the original weights can be entered by hand, or you can simply reload the file. 6. When all weights are adjusted, student averages should remain exactly as before. Check them against the hard copy you made of a Test screen however -- rounding error can sometimes changes one or two slightly. If you answered NO to 'adjust ALL weights?', you can now move the cursor around, pressing the space bar to select each weight you want to change; the cumulative total of these weights is shown at the top of the screen. The same restrictions and cautions apply as those outlined above, with this additional caveat: when changing only a group of weights from among many, student averages will almost certainly change, often by an appreciable amount. Be sure to compare the newly calculated averages with the old (from your screen hard copy) to see whether the desired result was achieved. Remember that adjustments are also applied to students that are offline. Finally, if you are uncertain of the principles and arithmetic involved in weight adjustment, experiment with the sample file CHE-3A1! provided on the program disk. If you're still hesitant to use Group Weight Adjustment, it's probably best that you don't use it; misused, it can cause problems. - 25 - CONVERTING TOTAL MARKS TO WEIGHTS If you have been using the total marks system, the total weight is currently zero. If you'd like to see how those tests are weighted, based solely on the maximum score you have assigned, Group Weight Adjustment can be used to convert maximum scores to their equivalent weights. First, be sure any new data is saved. Then press Alt-W, select 'Yes' to change all weights, and enter whatever total weight you want. At the end of a semester, you may choose a total weight of 60% or 70%. MTG will now calculate the equivalent weight for each maximum score, scaled to the total weight you have selected. This conversion is especially useful when you wish to assess the 'impact' a particular test will have on the class average; when using only the total marks method, it is easy to misjudge the over- all significance of a set of scores. To revert to total marks again, activate Group Weight Adjustment and set the total weight to zero. NOTES EDITOR On either a Test or Student screen, move the cursor to a test name or student name (or anywhere in the row) or onto the class name in the upper left hand corner, and press Alt-N ('N' for 'Notes') indicating you wish to write notes on this particular item. There are many possible uses for such an editor; it's a convenient way to record consultations with parents, guidance counsellors and students, to record medical disabilities, reasons for absence, details of domestic problems. Many teachers of special education and elementary school may be required (or may wish) to keep detailed logs of student behavior and progress. The editor may also be used to write up a course outline or a test. Like many other features in MTG, some teachers will have a need for the editor and some won't. When a test name (or the class name) is selected, MTG searches for a file name in the data subdirectory composed of the class name and the extension .TST; for example, if the class name is ENG- 2A.MTX, then the Notes file for tests is named ENG-2A.TST. If the file is not found, you are advised that you are beginning a new file and are asked how many pages per test you wish to allocate, between one and eight pages. The data you write on each test is stored in this file in database format and requires a considerable amount of drive space. As you increase the number of pages allocated per test (using the right arrow), you are shown the total drive space that a full file will occupy (forty tests' worth) which can range from 151K to over a megabyte, depending on the number of pages. The default number of pages per test is two; decide how many pages you'd like according to the amount of drive space you - 26 - can afford to allocate. I highly recommend that you use Notes only on a hard drive, or at least on a high density 3.5"/5.25" floppy disk. In the latter case, you should plan on using one floppy per class if you use Notes extensively. Once you have selected the number of pages you wish to allocate per test, it cannot be changed for that file. Data 'keyed' to the class name is stored in the test Notes file, so that you can use Notes to write up a course outline. In similar fashion, if you press Alt-N on a student name, a file named ENG- 2A.STU is loaded or created as necessary, and you then specify the number of pages you wish to allocate per student (which need not match the number you have specified for tests). All of this may seem needlessly complicated, but I feel it is better to offer some flexibility with regard to file size rather than taking a 'one size fits all' approach. In actual use, much of this will be transpar- ent. MTG checks to see if there are already notes on the selected item (test/student/class) and if so, loads this data into the editor; you can then edit the data or add new information. If there are no previous notes on this item, the editor starts up with a blank page except for the header it creates at the top of the page. The two-line header consists of such things as the test or student name, the name of the class, and your name, and cannot be edited or removed; MTG uses this information internally to identify the block of data and to store and retrieve it correctly on disk. Below the header, MTG inserts today's date and time if the selected item is a student, and the cursor appears a couple of lines below this. If the item is a test, that test's maximum score, weight, and description is inserted for you (any of which can be deleted if you wish). USING THE EDITOR The Notes editor is a text editor, and is distinguished from a true word processor mostly by the fact that it doesn't support word-wrap, i.e. each line ends in a carriage return (press Enter). If you have used EDIT in DOS 5.0, or TED, or the editor in PC Tools, then you are already familiar with this type of editor. It is very simple to use, and is entirely adequate for the task at hand, but it is not Word Perfect. If you need fonts, automatic hyphenation, hanging indents with bullets, multi-column pages, and spell-checking, this editor is not for you. Each page consists of 58 lines of text, with a top and bottom margin of 4 lines. The left margin is set for 0.7 inches (seven characters in from the left edge), and the line width is set for 65 characters. Page breaks are indicated by arrows at the left and right margins and the page number is shown beside each page break. - 27 - As in Word Perfect (for example), text entered vertically adjacent to the page break prints with correct top and bottom margins. Pages are numbered automatically when you print, beginning with page 2. Some commands are listed at the bottom of the screen. Press Alt-S at any time to save current data. When data is saved, it is automatically encrypted (another reason for the header) to prevent someone else from easily reading the file (COPY ENG-2A.STU PRN or TYPE ENG-2A.STU will display gibberish). This is not DES encryption and it won't stop someone who is determined and knowledgeable; and of course if your copy of MTG isn't protected by password, anyone can run it and consequently read your files. Alt-P prints either all data for the currently selected item or all marked (highlighted) data. Note that a marked block prints out with a correct top and bottom margin even if the block is somewhere in the middle of a page -- no need to adjust the page in the printer. Press Alt-Q or Esc to exit the Notes editor; you are warned if any changes have been made. A pair of numbers in the lower right corner of the screen indicate the present cursor location by row and column; note that the cursor is in column 7 at the extreme left edge, the left margin already accounted for. On occasion you may wish to 'export' some data from the Notes file and save it to an ASCII or .TXT file. This type of file can then be loaded into a word processor and be edited further, or you may simply want to store it separately from the Notes file. Exported data is no longer encrypted so be sure to keep this data in a safe place if it's sensitive. You may export either all data for a particular test or student, or only data in a block you have marked (use Shift-down arrow to mark a block of data). If you export all data, it is automatically saved in a file with the extension .EXS (for student data) or .EXT (for test data). In the case of a class named ENG-2A for example, exported data on a student is written to a file named ENG-2A.EXS. Each time you export data in this manner, it is appended (added) to the end of this file. When all data on an item is exported, the editor screen is cleared and you can then start with a fresh file on that test or student. If you save the file at this point (blank screen), the data on disk for that item is also cleared from the Notes file (e.g. ENG-2A.STU). This is a way to circumvent the limitation of having allocated, say, only two pages per student. When that two pages is used up, export it to an ASCII file and continue with two now empty pages. If you export only a marked block of data on the other hand, nothing is cleared from the screen or the Notes file; a copy of the marked block is simply written to an ASCII file, and you continue from there with all data intact. Note that in this case you can also specify another file name for the exported data other than the - 28 - default by pressing 'F' at the prompt. The chart below summarizes the different types of files and data: __ Class _ (large Notes files, (small files, / \ database format) ASCII text format) / \ Alt-E Alt-N ---- Test ------> ENG-2A.TST ---------------> ENG-2A.EXT \ \ Alt-E -- Student ---> ENG-2A.STU ---------------> ENG-2A.EXS Alt-H displays a list of other editing commands, most of which are intuitive and self-explanatory. In addition to edit commands, you can also display the 'Class Stats' box with Alt-C, and if you're writing notes on a student, Alt-X brings up the Attendance data for that student. NOTE: If you're using a school or school board licensed copy of MTG, the teacher's name can be changed at the Change Setup menu. If you happen to change the spelling of your name, you'll no longer be able to access Notes files created under the old name; the program will consider you an 'intruder'. ATTENDANCE 1. MTG can keep track of the number of days absent, unauthorized absences, and times late for each student. This means adding up the totals perhaps once a week, and incrementing the counters that the program uses to keep track of these totals. Select a student from the list and press Alt-X. 2. Select which category you want to edit with the up/down arrows, and change the numbers with the left/right arrows. The figures always begin in normal text. As soon as one is increased or decreased, it changes to white on black (or whatever your cursor color is). If you back up and pass the original setting again, it will revert to normal text. This feature is designed to remind you what the original figures were, before you started playing 'one-armed bandit'. 3. Pressing Esc once resets all figures to original values. Press Esc again to close the box, or press Enter to make changes permanent. All counters increment from 0 to 999 and then return to zero. 4. Most teachers have their own method of recording attendance, such as the time-honored 'tick-mark on the seating plan' system, and may not wish to use the one in MTG, which is fine. If you do - 29 - elect to use this feature, attendance data can be automatically inserted into progress reports, attendance reports, and individual grade reports in the Report module. Report can also print up blank attendance sheets (with students' names listed). The attendance function can also be invoked when using the Notes editor (Alt-N). Pressing Alt-X while in the editor brings up the attendance data for the selected student. DELETING OR MOVING A STUDENT OFFLINE If a student leaves your class, you may decide to erase the student and all score data from the file. Before you do this however, it is wise to get a 'hard copy', or printout of that student's data. Chances are you already have all this information in a gradebook report, but an individual grade report is more useful because it acts as a transcript, and also summarizes attendance data. An alternative is to move the student 'offline'. Fifteen extra positions are available to save marks and attendance data. While a student is offline, he is not involved in any calculations, graphs, or reports, with these exceptions: when you enter a new test, delete a test, change a test weight, or alter a maximum score. Offline students are updated accordingly, so that data for online and offline students 'match'. Students who are offline are assigned 'no score' for any new tests you enter. 1. Highlight a student in the sample file and press Alt-D (as usual, putting the cursor anywhere in the row identifies that student). You are asked: Move Smith, John offline, keeping all data. Delete Smith, John and all relevant data. If you pressed Alt-D by accident, just press Esc; otherwise, select one of the options. If in doubt, it's better to move the student offline. If you decide to delete a student, you are asked: Delete Smith, John and all relevant data? 2. Continue by pressing 'Y'. MTG will now ask: Are you really sure? (Y/N) Again, you can back out by pressing Esc. - 30 - 3. Press 'Y' again. The message: Deleting: Smith, John appears and MTG will delete the student's name, raw scores, average, attendance data, and both optional categories containing student information. To bring a student back online, press Alt-O on either screen and then select the student. DELETING A TEST When a test is deleted, the weight, maximum score, test name, test date, and test description are all erased, and the total weight is adjusted accordingly. Then all student averages are re-calculated, based on the remaining scores. Usually a test is a candidate for deletion if the teacher feels the class results indicate that the test was in some way inappropriate: too difficult, poorly designed -- or perhaps the class will be re- tested on the same material, and you wish to discard the old scores. NOTE: while you cannot give two students exactly the same name, you can give two tests the same name, which is convenient if one is a re-write of the other. There is a simple way to keep these test results, while at the same time preventing them from influencing the class average: just change the weight to zero. This removes the test from all calculations and student averages are immediately updated. You now have the option of using the scores again later on, perhaps with a different weight. Deleting a test follows the same steps used to delete a student. There are two opportunities to cancel the deletion. Once deleted, the data is permanently erased from memory. IMPORTANT: A deletion does not affect the file on disk, until you SAVE the file to disk. If you have accidentally deleted a test or student, you can retrieve the lost data from disk by immediately re-loading the file from Main Menu. DELETING A CLASS Deleting a class can be done only from the Main Menu, but this seems an appropriate place to discuss it. First, consider keeping class files indefinitely on floppy disk after the course is over. They don't take up a lot of room (see the discussion in Notes - 31 - Editor on archiving files), and they might be useful later on. For example, you might want to compare the academic performance of two classes who were taught the same course, or you may need to refer to Notes files on particular students at a future time. It seems a shame to throw away such a rich source of statistical data when it's a simple matter to copy the files to a floppy and put them away. You should at least have a final Gradebook report printout before proceeding. To delete a class, select 'Delete a class' from the Main Menu. MTG reads part of the file and lists some information about the class: how many students, tests, course description, and the first five students' names. There are two further prompts before the class is actually erased. At the top of the screen, you are warned that data in any Notes files for that class will be exported to plain, unencrypted ASCII text files, and the .TST and .STU files will be deleted. Once the Notes data has been exported and the class file deleted, you can no longer edit that data with the Notes editor, so be certain that you've finished with it before deleting the class. The ASCII text files can be loaded into most word processors and editors for further editing and printing. HARD COPIES Sometimes it is useful to print the current screen display. This is a handy way to give a student a record of all his/her scores, average, and letter grade, for example. Simply put the student's name at the top of the Student screen, switch your printer on, and press Prt Sc. Any screen can be printed in this manner, including graphs and help screens, and it is sometimes useful to print an Edit screen immediately following a sort. To reproduce the box lines and bars in the graphs correctly, your printer must be in IBM mode. Consult your printer manual to find which DIP switch must be moved. CLASS STATS Press Alt-C for a quick class summary. Find out how many students are listed, how many are offline, how many tests have been assigned, the over-all class average, the median average, the standard deviation, attendance totals, percentage/number of 'no scores', and the percentage/number of students failing, the current trend, and range of averages. This summary is also available in Graph, Report, and while in the Notes editor. - 32 - SETTING THE TREND PERIOD By default, Class Stats (Alt-C) gives the class trend for the last three tests. This trend is calculated by first finding the class average three tests before the set of scores you last entered, and subtracting that from the current class average. For example, if the last test you entered scores for was Test.6, and the three tests previous to that are numbered Test.5, Test.4, and Test.3, then the class average after Test.3 is subtracted from the class average after Test.6. This results in a positive, negative, or zero change, indicating how the class performed as a whole during this period. The default trend period is three, but can be anything from the last two to the last nine tests. Press Alt-P (for trend Period) in Edit or Graph, and use the arrow keys to select a new trend period. If you select a trend period that is too large for the number of tests in the file, the trend value is simply unava- ilable. The new trend period is not saved between sessions; it will be 3 again next time you run MTG. The same trend period is used when you graph a student (Alt-G). In the lower right hand corner is the caption 'Last 3 tests:' (or whatever trend period you have set). Below this caption is the trend for both student and class during this period. If there are more than eight cumulative averages on the graph, you can scroll with the up/down arrows to see the rest of the graph. When you scroll the graph, the trend is updated as well, calculat- ing from whichever test is displayed at the top of the graph. Note that test dates are also updated. Experiment with this feature to see how it works (it's easier to see than to explain). ARRANGING The student roster can be arranged (sorted) on a variety of criteria. Press Alt-A to bring up the Arrange menu. Most of the selections are self-explanatory. Arranging by 'inverting the list' can be used after any other sort to reverse the order. For example, sort by student average first, which ranks students from highest to lowest based on overall average. Inverting the list then positions the lower averages at the top. When a sort is complete, the criterion on which you sorted (student number for example) is temporarily displayed on screen in sorted order, giving you the opportunity to do a Prt Sc or use a utility like SNIPPER to capture this text to a file (SNIPPER.COM is a free PC Magazine utility commonly found on BBS's). Pressing Esc restores the usual display. - 33 - To arrange by 'Results from TEST.1', first zoom on TEST.1 (or whatever test you wish to sort on) so that test name appears at the top of the Test screen. Then select the sort. The 'Random Selection' simply scrambles the roster in no particular order, useful if you want to post test results by student number, and avoid giving away someone's identity by his position on the list. You should decide how a roster is to be sorted before printing a report. Often it makes no difference, but sometimes it is useful to have the names sorted alphabetically or by average for a report (see Class Summary and Gradebook reports concerning ranking of scores). ARRANGING BY TREND This sort arranges students by recent trend, highest to lowest, using the trend period you have set (or the default of 3), and based on the test currently displayed at the top of the Test screen. To use this sort effectively, do the following: a. Set the trend period you want with Alt-P. b. Zoom on the test (Alt-Z) you want at the end of the trend period, so that it appears on the Test screen. Usually, this will be the most recent set of scores, but you can go as far back as you like. c. Press Alt-A for the Arrange menu and you'll see on the list (for example): Trend: Test.2 - Quiz.5 Select this item after making sure this is the trend period you wanted. The selection will be unavailable if your trend period goes back further than there are scores for. When the sort is complete, all positive trends are displayed on a green background, negative trends on a red background. You can do a 'print screen' (Prt Sc) at this time or press any other key to clear the trend indicators. From these indicators, you can quickly identify students who are beginning to 'lose it', or those who have dramatically improved. You may also wish to graph certain students (Alt-G) for a more detailed view of their performance. - 34 - SAVING A FILE Pressing Alt-S saves the file to the current data subdirectory. During a data entry session, you should save the file every fifteen minutes or so; that way, if the power is suddenly shut off, you won't have to re-enter as much data. If you have pressed Alt-S and suddenly realize you don't want to save the file, press Esc to cancel the save; the file on disk remains as it was, and you are returned to the spreadsheet. (If you are using a fast hard disk, you'll have to press Esc in a hurry!) QUICK SAVING A FILE Occasionally, you may wish to save a class data file to a different drive or path from which the file was loaded. For example, you have loaded the file from data path C:\MTG\GRADES on your hard drive, and while the file is in memory you'd also like to save it to a backup floppy in drive B:; or you are getting a drive error on drive C: (disk full, perhaps) and rather than lose the work you've done so far, you'd like to save the file to a floppy disk in drive A:. To do this, instead of pressing Alt-S, press Ctrl-F1 to save to drive A:, or Ctrl-F2 to save to drive B:. If a path was previously set on the floppy drive (e.g. A:\FALL92), then the file will be saved here; otherwise it will be saved to the root directory A:\. If you need to set a path on the floppy disk before saving, exit to DOS first (Alt-T) and set the path (e.g. type A:, and then type cd FALL92); then exit from DOS back into MTG and press Ctrl-F1 -- the file will now be saved to A:\FALL92. The 'Changes have been made to this file' reminder will not be cleared when you use Quick Save, only when you have pressed Alt-S and saved to the usual data directory. TEMPORARY EXIT TO DOS If you need to copy some files or format a disk, you can exit to DOS by pressing Alt-T (for Temporary exit) on an Edit screen. As a precaution, you are warned to save the file you are working on if any changes have been made. This is because there is a possibility that you will forget that you have left MTG to perform these DOS functions, and may switch off the computer without having re-entered MTG to save your file. To re-enter MTG, type EXIT at the DOS prompt. Before shelling to DOS, MTG checks that COMMAND.COM is available somewhere on the PATH defined in your AUTOEXEC.BAT file, since a second copy of the command processor must be loaded - 35 - whenever you 'shell' to DOS. If you booted with a system disk in drive A:, you'll be prompted to re-insert the system disk before shelling can occur. This situation will most likely arise with laptops, many of which don't have hard drives. QUITTING To quit entering data, press Alt-Q or Esc. MTG keeps track of whether you have entered any more data since the last save to disk, and asks if these changes should be saved before quitting. Press Esc to cancel the operation and return to the spreadsheet. Press 'N' if you want to quit without saving recent changes, and 'Y' or Enter if you wish to save the changes and then quit. If there have been no changes since the last save, this menu is displayed: 1. Return to Main Menu 2. Quit Making The Grade Returning to Main Menu allows you to work on another file, go to Change Setup, etc. You now have enough information to start a new file and enter your own class data. Jump directly to Quick Start at this point, or continue to follow the tutorial, which next looks at Graph and Report. USING GRAPH With MTG you can produce a graph for a student, a test, or the whole class. To graph an item, follow these steps. First, select the file you wish to edit from Main Menu, and load the file. If you are following the tutorial and have just quit to the Main Menu, re-load the sample file CHE-3A1!. GRAPHING A STUDENT 1. Move the cursor to: Thurston, Sheila and press Alt-G. NOTE: If Alt-G doesn't do anything, look closely at what you are trying to graph. MTG will not graph a student if there is insufficient data. This type of graph is called a 'cumulative average' bar graph. The vertical axis is labelled with from one to eight test names. If more than eight tests have been assigned so far, the last eight will be displayed on the graph, and a white arrow appears next to the start and end dates in the lower right corner. By pressing the - 36 - up and down arrows, or by clicking with a mouse on the white arrows, you can scroll right back to the first test. The horizontal axis shows the student's over-all average as a weighted percentage. It is a cumulative graph, because each red bar shows a student's average at that time, in other words, just after that particular test was entered. The green bar represents the class average with that student's average subtracted out, which makes for some interesting comparisons. Under 'Last 3 tests:' (or whatever trend period you have set) you can see how much a student has lost or gained, compared with his/her average 4 tests ago, and compared with the rest of the class, indicating the recent trend. You can see from this graph for instance, that Sheila Thurston remained at least 10% below class average throughout the school year, and generally her average declined even further toward the end of the year, compared with the rest of the class. It seems almost no one benefits from writing a final exam. If a student's average is greater than 100% for any reason, a '>' at the end of the bar shows that it extends beyond 100%. 3. To print the graph, press Prt Sc. If your printout is double-spaced, or prints all on one line, you need to go to Change Setup, and get your printer to remove or add an extra linefeed. (Pressing the space bar hides those white up/down arrows beside the test dates. 4. To graph each student in turn, press Pg Up and Pg Dn, just as you did in Edit. Mouse users can click on the left and right edge of the screen. 5. To graph a specific student without paging your way through the list, you can do one of two things: a. Press Alt-G or Alt-E. You can now select another student by moving the cursor to that student. Then press Alt-G again. b. Or, from the graph, press Esc. This brings up a menu of selections: 1. Graph another item. 2. Go to Report. 3. Return to Edit. GRAPHING A TEST Select any test you wish from the list, and press Alt-G. This graph is a 'frequency distribution'. Each interval on the horizontal axis contains 10 percentage points: 0-9, 10-19, and so on (the last interval cheats a little and includes 100, for a total - 37 - of 11 points.) The vertical axis shows the frequency -- in this case, how many test results fall within a particular grade interval. For example, percentages of 51, 54, 55, and 59 all fall within the interval 50-59, and the frequency is 4. If a frequency is greater than 13, the actual frequency is shown at the top of the bar. The title includes the weight for that test. NOTE: If Alt-G doesn't graph anything, look closely at the test you are trying to graph. MTG will not graph a test if there is insufficient data. To the right are listed some useful statistics: the average on this particular test, the median, standard deviation (SD), range of scores, number of scores, and percentage/number who failed the test, and the correlation coefficient. If you are familiar with all of these statistical terms, please skip to the next section: Adjusting Grades. The average (also known as the mean) has the same value as 'TAvg', the test average. Remember that 'no scores' are not included when computing this average, but zeros are. The median represents the score right in the middle, if you were to sort all the scores from lowest to highest. If the number of scores is even, the two scores in the middle are averaged to calculate the median. The median tends to be close to the mean, except when the distribution of scores is exaggerated (or skewed) toward one end or the other, so a difference of 5% or more between mean and median generally indicates a lopsided distribution of scores. The average of a set of scores is useful, but it tells us little about how the scores are dispersed. The three sets of scores below, for example, all have the same mean, but obviously they represent three completely different distribution profiles: A: 20 20 50 50 80 80 B: 50 50 50 50 50 50 C: 5 5 5 95 95 95 The standard deviation is a number that tells us something about the degree of dispersion in a set of data. To find the standard deviation you must: 1. Subtract each score from the mean, and square the difference. 2. Add all the results from step 1, and divide this by the number of scores. - 38 - 3. Take the square root. (If you stop after step 2, you have calculated the variance.) Although the standard deviation by itself has little meaning (it is not a percentage, and has no unit of measure), it has significance when we know the possible range of scores (0% to 100% in our case), and what the mean is. Measurements of naturally occurring phenomena tend to produce a bell-shaped curve when the data is plotted on a frequency distribution graph. Such a curve would most likely result if you were to measure the height of each student in the class, for example. One bell curve with interesting properties is called the normal distribution curve. The normal curve is symmetrical, and the mean falls exactly in the middle. It turns out that 68% of all scores on a normal curve fall within one standard deviation on either side of the mean. In other words, if the standard deviation has a value of 10 (from the steps outlined above), and the mean is 70%, then 68% of the class scores between 60% and 80%. Further, 95% of all scores tend to fall within 2 standard deviations on either side of the mean (between 50% and 90% in our example), and 99% of all scores lie between +3 and -3 standard deviations (between 40% and 100%). If you page through several of the test graphs, you'll see all kinds of distributions. In general, tests are designed to distinguish between those who know what they're doing, and those who don't. If the test is too easy, scores are bunched together at the upper end, the mean is too high, and the standard deviation is too small, i.e. less than 10. When the test is too difficult, scores will be bunched together at the other end. Sometimes a distribution is flat; that is, scores seem to be evenly spread across the spectrum, a possible indication that the test design encouraged students to guess more or less randomly on a series of multiple choice questions, for example. On this kind of graph, the standard deviation will typically be 25 or greater. You will find various examples of each type in the sample file, which, by the way, is based on actual test data. Ideally, we'd like about 2/3 of the class to be clustered between 15% and 20% on either side of what we consider a good, average mark, with the remainder falling somewhere at either end of the curve -- some always seem to score above 90%, and a few always seem to fail a given test. For this reason, the normal dis- tribution curve is a good model on which to base any 'grading on the curve' adjustments you might wish to make; more on this later. The standard deviation will tend to be greater than 10 however, and anything between 15 and 20 is fairly typical for sets of test scores. - 39 - The range is simply the difference between the lowest and highest scores, and provides another measure of dispersion. If the highest score is 83% and the lowest score is 41%, then the range is 42. Note that a singularly high or low score with no other scores nearby, distorts our perception of the spread, as shown in the diagram below. This is why it is better to have several statistical measurements, rather than relying on just one. | # | # | # # | # # # | # # # | # # # # |--+--+--+--+--+--+--+--+--+--+ <----- range -----> of scores CORRELATION COEFFICIENTS When you graph a test, the statistical data includes the correlation coefficient for those test results versus student averages before the test (abbreviated as 'C.Coeff'). The Pearson product-moment correlation coefficient is a unitless index telling you (in this application) how close a relationship exists between students' averages before the test was given and their marks on the test you have selected to graph. The statistic only becomes available after the fourth test has been entered, so that student averages have had time to stabilize. Normally, one would expect a positive linear relationship between past performance and future performance. That is, students with high averages will tend to score high on the next test and those with low averages to score low. So this statistical index answers the question: How close are these test results to the results you would have predicted, based on past performance? Another way to put it is: How consistent are these test results with results obtained previously? Did students on the whole, perform typically on this test compared with previous tests? In statistical work, a correlation coefficient of +0.7 or higher indicates a strong linear relationship. Such a relationship is typically found between height and weight, or between those having blonde hair and blue eyes. A coefficient less than +0.3 indicates a very weak relationship, the kind of relationship you might find were you to compare students' grades in Math with those in Art. A coefficient of 1.0 indicates an exact correlation. Notice from the three data samples in Figure 1 that an exact correlation doesn't necessarily mean that each student's test score - 40 - is exactly the same as his average, only that there is perfect consistency. In other words, if the test was more difficult than previous tests, a coefficient of 1.0 shows that each student scored lower than usual by the same amount. A correlation coefficient of zero indicates that there is no linear relationship whatsoever. Such a relationship might be found between shoe size and preference for Coke or Pepsi. A coefficient less than zero (to as low as -1.0) indicates an inverse linear relationship. In Making The Grade, a negative coefficient would result if high-average students did poorly on a particular test, and low-average students did well on that test. (See Figure 2) (COULDN'T REPRESENT THESE GRAPHS IN ASCII TEXT) Figure 1: Each group of points is called a scatter plot. The straight line that best represents the relationship is called the line of regression. In these examples, r (the correlation coefficient) remains exactly 1.0, even though the test average is different in each case. Figure 2: A shows no apparent relationship, B a strong positive relationship, C a weak positive relationship, and D a negative relationship. (COULDN'T REPRESENT THESE GRAPHS WITH ASCII TEXT) WHAT ARE THE IMPLICATIONS? A low correlation coefficient (less than 0.3) should alert you to the fact that for whatever reason, this test did not produce typical results. Students may express surprise, some saying they did much better then they expected, and some worse than they expected. Remember that if everyone performed worse than usual, or everyone did better than usual, the strong linear relationship is maintained; a low coefficient indicates unexpected or scattered results. By studying the responses to individual questions on the test, you may discover that certain questions are worded ambiguously. Students have interpreted the questions in a number of ways, each believing he is answering the question correctly. Or perhaps the test has many multiple choice questions and everyone did a lot of guessing. The test is certainly suspect if you give the same test to different classes, or in different semesters, and get similar correlation coefficients. You may want to change or eliminate some questions. - 41 - Another possibility is that a particular group in the class, independent of scholastic ability, scored poorly on certain test items. For example, several students who happen to be in the school band, or are on the football team, all missed a class in which an important topic was covered. This may account for the discrepant test results. A test with many questions (and out of many marks) generally produces a more valid correlation coefficient than a test with only a few questions, because the distorting effect of 'fluke' responses is minimized when there are many questions; the testing procedure is more thorough and students tend to perform more closely to their true capabilities. Also, student averages become more stable as the semester progresses and are less subject to change (as is a goalie's 'goals against average' for example). As a result of these two factors, final exams tend to yield correlation coefficients of 0.8 or greater. A five-mark quiz near the beginning of the semester, on the other hand, cannot be expected to yield a coefficient of great significance. A certain amount of skill is needed to interpret correlation coefficients in a useful way. To minimize the effect of a strongly atypical set of test scores, you can either lower the test weight so it is worth less, or even eliminate the test by setting its weight to zero. Alternately, you may wish to experiment with 'grading on the curve' (press Esc from the Test graph screen) to try and 'smooth out the wrinkles'. (The latter strategy actually has only a small effect -- the linearity of the relationship is already well established, and shifting and stretching the curve doesn't alter this relationship significantly.) In conclusion, the underlying assumption throughout this discussion has been that students perform in a predictable fashion, and that results from tests administered over a long period of time should reflect this assumption. The correlation coefficient therefore, either confirms that everything is pretty much as you expected, or signals to you that something may be amiss. A more thorough mathematical treatment of correlation coefficients (and the closely related linear regression) can be found in any first year university statistics or psychology textbook. ADJUSTING GRADES Every teacher at one time or another has probably had to adjust a set of test results, or a set of final grades. One technique for adjusting scores is known as 'grading on the curve', while another assigns letter grades according to a formula: so many A's, so many B's, so many C's, etc. All methods of adjustment tend toward a common goal: center the scores about an ideal mean, i.e. - 42 - a figure the teacher (or administration) selects to represent the average student, and secondly, control the amount of spread between highest and lowest scores, or in other words, adjust the standard deviation to our requirements. In a typical class, the scores are almost always dispersed widely, compared with a normal distribution curve. Take a moment to look through some of the test graphs from the sample file, and note how many standard deviations exceed 15 and even 20. Compared to a normal distribution, many students' scores fall at the low and high ends of the curve. Recall that in a normal distribution of scores between 0 and 100, one standard deviation has a value of approximately 10. In practice, test scores hardly ever have a standard deviation of 10, and generally will be between 15 and 20. To adjust a set of scores therefore, you need to do two things: move the scores toward a new mean, and either narrow or broaden the curve. The formula for doing this correctly is as follows: New score = New mean + ((Old score - Old mean) x New SD / Old SD) MTG can do these calculations for you. Follow the steps below to practice adjusting scores. 1. Begin by graphing one of the sample tests. Then press Esc to bring up the 'Options' menu. 2. Select 'Try another mean and SD' from the menu. Another box pops up, and from here you can either accept the default standard deviation and mean (17 and 70%), or use the left and right arrows to change them. Press Enter when finished, or Esc to back up a step. 3. MTG will graph the new data according to the mean and standard deviation you have set, adjacent to the original graph for comparison. The new mean, median, SD, range, number failing, and correlation coefficient are also displayed. NOTE: This will not alter the original scores, either in memory, or on disk. The new scores resulting from the changes you have made are only temporary and are eventually discarded. Thus, you can continue to try other means and standard deviations, without fear of losing important data. TIP: If you teach statistics in a Math course, Graph can be used to teach some of the basic concepts in a highly visual and interactive way. - 43 - 4. You can go a step further at this point, and see the actual scores that would result, were you to make the adjustments permanent. Press Esc again to bring up the 'Options' menu. Select 'Look at the adjusted data'. The first column of figures shows the original percentage scored by each student on the test. The second column shows the new percentage, and the third column shows the raw score a student would need to produce the new percentage. Students who gain marks are highlighted in green, those who lose marks in red. This lets you quickly see the over-all effect of the standard deviation and average you have selected. 5. If you would like to print this screen, switch the printer on first, and then press Prt Sc. Esc will return you to the graph. NOTE: I purposely left out the capability to automatically insert these new raw scores into a file. I feel that this capability would encourage a certain amount of quick and careless adjustment of marks; a student's academic record is too important for idle tampering. The hard copy however, allows you to go back and enter the adjusted raw scores by hand on the Edit screen (not a huge task, but enough to discourage casual adjustments), or to simply enter the adjusted final marks on a report form. Some students gain from an adjustment, and others lose. One modification to grading on the curve is to not allow any student's score to be lowered, known as 'beating the curve'. Note however, that this practice alters the target mean and standard deviation you have set. To implement this method in MTG (if you decide to enter the adjusted scores), only change a raw score if the adjusted score is higher. GRAPHING A CLASS To graph the class, go to either the Test or Student screen, move the cursor over the class name (CHE-3A1! in the sample file), and press Alt-G. The frequency distribution graph shows the distribution of student averages, and is similar to the test graph. The mean and standard deviation can be adjusted here as well, and in fact, this is where adjustments are usually made, at the end of the semester or school year. The correlation coefficient is unavailable here, because no comparison is being made. NOTE: if Alt-G doesn't graph the class, it is because of one or more of the following: - there is no test data - there are no students - there are no available student averages - there are one or more tests with a maximum score set to zero - 44 - OTHER OPTIONS IN GRAPH To go back to Edit from Graph, press Alt-E. Alternately, bring up the 'Options' menu by pressing Esc, and select 'Return to Edit'. You can also go to Report from here, or quit by pressing Alt-Q. You can save a file while in Graph (and Report). MTG remembers whether or not you changed anything in Edit mode, and will remind you if you try to quit without saving. Press Alt-H or F1 to bring up a help screen of available commands, or refer to the command chart. Often, it is useful to look at Class Stats (Alt-C) while in Graph, especially when graphing a test -- the two sets of data appear side by side. OTHER WAYS TO ADJUST SCORES MTG makes it easy to adjust scores globally, or on an individual basis. To lessen the over-all effect of a particular test, simply reduce its weight. To remove it entirely, either delete the test, or set its weight to zero. If for some reason you wish to make the maximum score higher or lower, this can be done even after the raw scores have been entered; a maximum score can be as large as 999. A raw score can be as big as nine times the maximum (to a possible 8991), which allows extra credit to be awarded. This feature is also useful if, for example, a student has missed a test, or has just joined the class and marks are needed to generate a grade for assessment. By doubling his/her raw score on another test of equal worth, you can compensate for missing data. Some teachers prefer to drop each student's lowest score at the end of the course. This is easily done by entering an asterisk, which converts the low score to 'no score'. CAUTION: to be fair, only scores of equal weight should be dropped. If a test worth 10% is dropped for one student, and a quiz worth 2% is dropped for another student, based solely on these being their lowest scores, the first student gains an advantage over the second. The same principle applies when using the 'total marks' scheme. MORE THEN 40 SETS OF SCORES? When you 'Duplicate an existing roster' (under 'Start a new class'), you have the option of carrying forward all student averages and attendance data from the first file. This is useful if you have more than 40 sets of scores in a semester, or if you - 45 - want to carry averages and attendance data forward to the next term. When the roster to be duplicated has loaded, you are asked: Carry student averages forward into the new file? Carry attendance data forward into the new file? If you choose to keep the student averages, the following happens: a. A new test name called 'So.Far' is created in the new file. The maximum score is 100, the weight is the total weight from the other file, and the raw scores are the student averages from the other file. The test average is calculated from these student averages, so it also represents the class average at this point! A test description is entered for you: Student averages from CHE-3A1! (or whatever the previous file was named). Change the test name (TERM.1 for example) and the description to whatever you like, and you may also want to enter a date for the transfer, under TDate. If this file is merely a continuation of the same term or semester, you will probably want to leave the weight as is; it represents the total weight so far. If you're beginning a new term however, and the previous term only contributes (say) 30% toward the final grade, now is the time to adjust the weight of this 'test'. In this way, your new file accurately reflects the year's progress for each student. If you like, temporarily make the weight for 'So.Far' zero. Now you can see how students are progressing just in this term, and the weight for the first term can be re-entered later on. b. You must enter a new name for the class, to distinguish it from the old file. This name can be anything you like, but it MUST be different from the name of the other file. For example, if the other file is called CHE-3A1!, call the new file something like CHE-3A1X or CHE-3A12. If you were to save the new file with the same name as the old file, the old file would be erased. This is nothing new -- just follow the same procedure you have used in the past: two files in the same directory cannot have the same name. CAVEATS Before carrying the student averages forward, the old file should be as complete as possible. That is, all 'no scores' should be resolved; are they zeros or 'no scores'? The weight and maximum score you assigned for each test should also be finalized. The reason is simple: these data all affect the student averages, which you are carrying forward into another - 46 - file. If for some reason, you must go back and change something in the old file, and this change alters one or more student aver- ages, you have to be sure to copy these changes by hand into Test 1 in the new file. Hint: in Edit, bring up a Test screen, which lists all student averages in a column, and press Prt Sc. This gives you a handy reference to update the new file, in case any late changes have to be carried forward. One advantage of the above method is that you know a student's over-all standing at any time, and not only his standing for a particular term. If the arithmetic seems a little convoluted, practice with the sample file on side two of the program disk to familiarize yourself with how it works. You may still prefer to keep each term completely in its own file, but this means the final grades will have to be calculated by hand at the end of the year. Since a weight can be as little as 0.1%, and the total weight can be reduced when you start the new file, an unlimited number of scores can be accommodated, although it's unlikely you'll ever need this many. The method of carrying averages forward, outlined above, can be applied to several files in succession. The file at the end of the 'chain' will still accurately reflect the year's work. Just be certain that any retroactive change you make is entered manually into successive files. Keep attendance data? Decide whether you want to carry forward absences, unauthorized absences, and lates into another file. If the new file is a continuation of the same term or semester, it makes sense. If the new file starts a new term, it may make more sense to maintain a discrete record for each term. USING REPORT To use Report, first select a file to edit from Main Menu (use the sample file CHE-3A1! for practice). When you reach the Edit screen, press Alt-R, which takes you to Report. You can also go to Report from Graph, either by pressing Alt-R, or by selecting 'Go to Report' from the 'Options' menu, and then 'Prepare a report'. Follow the steps below to produce each type of report. IMPORTANT: Report expects you to have at least entered a class list. This enables you to print attendance reports, blank attendance charts, student information reports, and a few others. Other reports, such as the gradebook, missing assignment, and progress reports, require at least one set of scores to work with. - 47 - You should also sort the student roster (Alt-A) if you want names to be printed in any particular order, before going to Report. As well, you should check the printer setup before preparing a report to see that your printer is correctly installed. To do this, go to Change Setup from Main Menu, and then select 'Choose a printer'. PROGRESS REPORT 1. You are first asked: Print a report for every student? The default reply is 'Y', so just press Enter if you wish to do the whole class. 2. If you reply 'No', the class list is displayed. Use the up and down arrows to move to a student, and press the space bar to select the student; a marker appears to the left of the name you have selected. To deselect a name, again press the space bar. To deselect all names and start over, press Esc. Press Enter when you're finished. 3. Next, each student you have selected is displayed in turn at the bottom of the screen, and you can pick from one to four comments for each student. Press the space bar to choose a comment, or Enter for no comment, in which case a blank line will be printed on the report. Progress Report allows comments in the COMMENTS.TXT file to be 55 characters long. If longer than 36, the comment is truncated on screen with a small right arrow at the end of it, but will appear in its entirety at the bottom of the screen if selected, and when printed. The third option is to type in a comment. To do so, just begin typing; a cursor will appear and several editing functions are available. Press Enter when done. An incomplete sentence can be continued on the next line. To flip through the four pages of comments, press Pg Up or Pg Down. Page 4 simply gives instructions on modifying the COMMENTS.TXT file so it contains your own comments. Follow these rules to modify the comments correctly: a. Load the file into your word processor as an ASCII text file, NOT as a word processing file. Also, save it as an ASCII text file (DOS text file), sometimes called 'printing to disk'. - 48 - b. Any comment longer than 55 characters will be truncated when MTG loads the file, so use the column position indicator to keep track of the line length. c. The file must contain only ASCII text, no control characters for underlining or indenting, etc. d. Check your speling and grammer careful to avoid embarissment. 4. After comments have been selected for all students, press Pg Up/Pg Down to select the date format you wish to use and press Enter. 5. From the two columns, select the title you wish to display above 'PROGRESS REPORT'. Choose 'Blank' from both columns if you don't want any title. Press Esc to start over. 6. MTG then asks you which attendance data should appear in the report. The defaults are YES for 'Days absent', NO for 'Unauthorized absences', and NO for 'Times late'. Select any combination you wish. MTG uses information from the attendance records entered during Edit. If you haven't been using this feature, but still wish to show 'Days absent', answer NO to all three. MTG will then print a 'Days absent' column with underscored blanks, so you can enter your own data by hand afterwards. 7. Decide whether you wish to print the numeric grade, letter grade, or both. 8. Decide whether you want 'Near Letter Quality' or 'Draft'. (This option will appear only if your printer supports NLQ.) 9. Select the number of copies you want to print. This does not refer to how many individual reports, but to how many times you wish to print the entire set of reports. 10. Lastly, you'll see this message: Turn printer on, with print head at perforation. Press Enter when ready, or Esc to cancel. The perforation should be even with the top of the print head (the part that goes back and forth). MTG inserts its own linefeeds where necessary, so don't worry about starting too close to the top. Two reports are printed per page, which can then be cut in half (saving a few trees). Pressing Esc exits Progress Report and returns to the Options menu. - 49 - GRADEBOOK REPORT This report gives a typical spreadsheet-like printout of all students, tests, scores, percentages, weights, etc. Nine tests can be printed per page. The column to the extreme right shows each student's current average, and the underscored number in the lower right hand corner is the current class average. 1. You are asked if you want to print every test. Since they are printed in groups of nine, you may only need to print the last page of tests. For example, if your file contains 13 tests so far, and you have printed gradebook reports before, you need only print from test #10 to #13, and add this page to your printed records (unless you have made adjustments to earlier tests since the last printout). 2. If you don't want to print every set of scores, you are then asked to select a range of tests. Press the space bar to select a starting point, move to the finish point, and press the space bar again. To print one test, select the same test for start and finish. If you select only one test, the test description is printed along with the test name, which makes it more useful when posting test results. Press Esc if you need to start over. The Gradebook report now also prints each student's ranking (1 = highest mark, 2 = second highest mark, etc.) under the following conditions: you have selected only one test to print, and the students have already been sorted on the results of this test at the Edit screen (Zoom onto Test.6, press Alt-A, followed by 'Arrange By Results Of Test.6', for example). That's all there is to it. You should print this report after each data entry session. It provides a paper-based record of all vital data, and supplements the backup copy of your data disk. If your data disks were somehow erased tomorrow, you could re-con- struct the file from this printout. ATTENDANCE REPORT If you have been using the built-in attendance feature, and have kept it up to date, the attendance report gives a complete summary of days absent, unauthorized absences, and times late for each student (or any combination of these three). By answering NO to all three, a 'days absent' column will be printed with underscored blanks, so you can enter the figures from your own records by hand if you wish. - 50 - STUDENT INFORMATION REPORT This report will list the contents of the two optional categories (the default is 'Homeroom' and 'Student Number'). You can list either, or both. If you specify neither, the two headings will be printed, with underscored blanks where the data usually appears. TIP: this blank report is useful at the start of the year for gathering data; you can later enter the information into MTG. MISSING ASSIGNMENTS REPORT This report lists each student who has missing tests, along with the name of each test. MTG determines that a test is missing if a raw score is a 'no score', another good reason for keeping the 'no score' status as long as possible, rather than immediately entering a zero. At the end of the listing, a separate page lists each test name, and the test description beside it. This report also shows the student what his/her average will be if all 'no scores' are converted to zeros, which may be just the incentive a student needs to hand in those missing assignments. One way to use the report is to cut the page into strips, so that each student has his/her own list. Or, post the entire list on a bulletin board, using student numbers rather than names. CLASS SUMMARY REPORT This report lists all students and their current averages, which may be shown as a percentage, a letter grade, or both. The number of students in the class is included, and also the number/percentage who are failing. The Class Summary report also checks to see whether or not you are using the attendance tracking feature in Edit. If any non-zero figure is found for Total Absences, the Class Summary report will automatically include an extra column listing the total absences for each student. This feature is handy when filling in report cards, where the grade and number of absences are both required. Some schools, at the end of the school year, raise a grade that is within 3 marks of passing to a pass, and also raise a bare pass by one mark. For example, if the passing grade is 50%, scores between 47% and 49% are raised to a 50%, and a legitimate score of 50% is raised to a 51% (to distinguish it from a raised pass). MTG will do these manipulations if you wish (based on whatever passing grade you have installed); a raised pass prints in bold face, and is underscored. These changes are made in the report only; the data file is unaffected. - 51 - The Class Summary report now allows you to print student number rather than student name, similar to the Gradebook and Missing Assignment reports. As well, if the students have first been sorted by over-all average at the Edit screen, each student's ranking is printed to the left of his name (highest mark is #1, second highest is #2, etc.). BLANK ATTENDANCE This report is simply a class list with a blank for the month at the top, and a grid for recording absences, lates, etc. There are five blank lines at the bottom for new students, and also a blank line to record totals each day. This blank report can easily be used as an order form or a tally sheet. BLANK SEATING PLAN Specify how many seats across the front, and how many seats deep, and MTG will print a blank seating plan that fits the size of your classroom. The plan can be dimensioned from 2 x 5 up to 12 x 12 (after an exhaustive survey of area teachers, the author concluded that hardly anyone needs a 1 x 1 seating plan). The blank seating plan is also useful any time you need a grid or game board of a particular size, or for making a bunch of paper strips, which you can then cut out. NOTE: The 'front' of the class will be printed down the right side of the page, because this gives the most convenient rectangular dimensions for entering names, most of the time. You may need to switch the dimensions (say from 7 x 8 to 8 x 7) to find the best arrangement. BLANK SCORE SHEET This is similar to the blank attendance report, but has wider columns for entering raw test scores. Normally, it is recommended that scores be entered directly into Edit from the test papers (less chance of an error). The mark sheet is handy when work is marked right in class, and can also be used as a general purpose tally sheet. INDIVIDUAL GRADE REPORT From the Report menu, individual grade reports can be printed for selected students or for the entire class, for a selected range of tests or for all tests. Each report includes test names, scores, weights, test averages, the student's over-all average and - 52 - attendance figures, and class average. This report is suitable for including with a progress report or as a transcript if a student transfers to another class or another school. Many teachers also like to give each student one of these for their own records. GRAPH PAPER Graph Paper prints a reasonable facsimile of Cartesian graph paper in a 55 x 42 grid -- somewhat finer resolution than the usual graph paper and perhaps more suitable for certain tasks. LINED PAPER Lined Paper, as the name implies, prints lined paper, 28 lines per page, with slightly more space between lines than the usual 3- hole binder paper. These last two 'reports' have been added mostly for convenience if you happen to run out of either type. BACKING UP YOUR DATA Your gradebook, whether on paper or on disk, is irreplaceable, and the thought of losing it causes any teacher to break into a sweat. With a few simple precautions, you need never worry about losing data; in fact, if you follow the suggestions below, you'll have several duplicates of all vital information, something few teachers have if they enter all scores by hand into a conventional gradebook. 1. First, use only a COPY of the MTG program disk. If you are unfamiliar with everyday disk and file maintenance, get someone to show you how. Using the original program disk is definitely getting off to a bad start. 2. There should be two copies of each data disk. 5 1/4" disks are as little as $0.50 apiece, and 3.5" disks are about $1.00 apiece -- this is a minimal cost. Use one copy to actually work with and save the files to (write WORK COPY on the disk label). The other is your back-up copy (write BACK-UP on the label). After each data entry session, copy each work disk onto its corresponding back-up copy. If you don't know how, have someone show you. Hard disk users should copy all data files to a floppy disk once a week, including Notes files. 3. No one should ever trust magnetic media or computers completely, so immediately after a data entry session, print out a gradebook report from the Report module, two if it makes you feel - 53 - better (one for school, one for home). This report will allow you to reconstruct the file if necessary. 4. Finally, an up-to-date Student Information report (and an Attendance report if you use the built-in attendance keeper), provides a hard copy of other data unrelated to marks. Happy grading! THE LEGAL STUFF While every precaution has been taken to ensure the correctness of the software and its accompanying manual, Coffee Mug Software cannot assume any responsibility or liability for any damage or loss caused by the software. Coffee Mug Software reserves the right to make changes and improvements to the manual and software at any time without notice. It is the responsibility of the user to make the necessary backups for his/her data and program diskette. _______________________