Subject: Re: Simple regression information
From: Brian E. Smith (smithb@management.mcgill.ca)
Date: Mon Apr 10 2000 - 02:39:36 EDT
Bret,
it is a very simple exercise in EXCEL. Enter the data pairs in two
adjacent columns (with headers), highlight the dependent variable column
and then click on the chart wizard. Select XY (Scatter) as the graph type
and proceed to create the graph. When the graph is complete right click on
one of the points and you will see that all of the points are filled with a
yellow diamond and a dialog box pops up. Select "add trendline" and a new
box pops up. Select the type of trend you want (probably linear) and then
click on the options tab. Put a check mark in the "Display equation on
chart" box and in the "Display R-square on chart" if you wish. Click on OK
and the scatter diagram will now have a trend line and also the equation
and R-square value displayed. You can move the equation around by dragging
it to a convenient location. Then you can play around to make your graph
look pretty!
Have fun and let us know the reulsts - it sounds like an interesting idea!
Brian
At 12:47 AM 4/9/00, Bret Taylor wrote:
>I'm going to embark on a very small project next term. (Mainly just to
>collect and report data to my administration.)
>
>I'm going to take meticulous attendance records (to the nearest minute) and
>compare grades with attendance.
>
>I'm a math teacher, so I'm inherently lazy. :-)
>
>Does anybody know a good, simple program out there that will allow me to
>input an ordered par (%attendance, grade) and it will give a scatter plot,
>the line of best fit, and the function of the line? I know I could do it
>myself using a TI or a spreadsheet to do my calculations. But, I figure
>there must be something out there already that will do the "work" for me.
>
>Has anyone ever used LINEST with Excel?
>
>I plan on running the data after every exam. After I get several exams from
>several different classes I may try and see if there are some classes where
>attendance is a better predictor of grades than other classes.
>
>Has anybody ever done something similar? If so, what software did you use?
>What results did you find?
>
>I anticipate ending with the chicken-or-egg dilemma:
>
>Are students poor in math because they don't come to class; or do they not
>come to class because they are poor in math? Ah the things that
>administration and faculty debate. :-)
>
>
>****************************************************************************
>* To post to the list: email mathedcc@archives.math.utk.edu *
>* To unsubscribe, send mail to: mathedcc-unsubscribe@archives.math.utk.edu *
>* Archives at http://archives.math.utk.edu/hypermail/mathedcc/ *
>****************************************************************************
>
>
___________________________________________________________________________
Brian E. Smith TEL: 514-398-4038 (Work)
McGill University FAX: 514-398-3876 (Work)
1001 Sherbrooke St. West FAX: 514-482-1639 (Home)
Montreal, QC, Canada H3A 1G5 EMAIL: smithb@management.mcgill.ca
Url: http://www.management.mcgill.ca/homepage/profs/smithb
___________________________________________________________________________
No human investigation can be called real science if it cannot be
demonstrated mathematically. Leonardo da Vinci
___________________________________________________________________________
****************************************************************************
* To post to the list: email mathedcc@archives.math.utk.edu *
* To unsubscribe, send mail to: mathedcc-unsubscribe@archives.math.utk.edu *
* Archives at http://archives.math.utk.edu/hypermail/mathedcc/ *
****************************************************************************
This archive was generated by hypermail 2b28 : Sun Apr 09 2000 - 23:30:03 EDT