home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Wildcat Gold - The Optical BBS
/
Wildcat Gold - The Optical BBS (The Golden ROM Series)(Volume 4 Number 1)(The Digital Publishing Company)(1992).ISO
/
sdn
/
mtg21.sdn
/
MTG21DOC.EXE
/
MTG21.DOC
Wrap
Text File
|
1991-12-05
|
135KB
|
3,184 lines
MAKING THE GRADE
Version 2.1
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.
Twelve reports are available and can be printed in draft or
near letter quality mode. Report types include a progress report,
a gradebook printout, 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, CGA 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 great with VGA, EGA, and CGA video, but will not work with
Hercules, Olivetti, and MDA (mono) adapters (the program makes use
of four separate video 'pages', which are not all present in these
adapters). MS DOS or PC DOS 3.0 is the minimum version of DOS sup-
ported. (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.
_______________________