This document describes how to use the statistics
capabilities of the Texas Instruments TI-81 calculator to
perform a linear regression ; i.e., to find the slope and Y - intercept of a line
of best -fit through a set of data. If you
want to know more, refer to Chapters 7 and 3 of the TI-81 Guidebook that came
with the calculator. Key presses
are indicated using the same convention used by your Guidebook.
Set-Up
First, configure your calculator according to the
fol lowing two figures . The key to the left of each screen
describes how to get there. For each figure, make your calculator match the
illust ration . Remember to use 
to
enter -5. Next, press 
[ DRAW ] <1:ClrDraw>
Finally, press
and then press
until any
existing formulas have been deleted.
Entering the Data
A student has obtained 11 sets of temperature readings in
the Quantitative
Measurements lab. These are tabulated at left. To de termine the relationship
between the two, you will plot °F vs °C and obtain an ex pression for the line of
best fit (Note: °F vs °C means °F goes on the Y-axis and °C goes on the X-axis.)
To get the data into the calculator, you enter the numbers as a sequence of (X,Y)
pairs.
To clear any old (X,Y) values , press
[STAT] <2:ClrStat>
. Then,
press [STAT] <1:Edit…> and start entering the
data: 0 27.1
10.1 ….
Continue until all 11 data have been entered.
Doing the Calculation
Press [STAT] <2:LinReg>
. Your screen should now look like the one
shown at the right. The line of best-fit has a Y-intercept of 29.783 and a slope
of 1.815.
Therefore, °F = 1.815 °C + 29.783 (remember that the equation for a straight
line
is y = bx + a).
Finally, let’s graph your result . You will plot the
regression line (i.e. the line of
best-fit), and then superimpose your data as a scatter plot. For
the regression line, press 
<EQ>. You should
see
the first screen at right. Select <4:RegEQ> to store the equation
for the regression line (second screen at right). For the scatter
plot, press 2nd [STAT] <DRAW> <2:Scatter> .
This last
step also displays the graph
