Multiple Linear Regression: Julia Hartman

Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 56

Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman

1 1
An Interactive Tutorial for SPSS 10.0 for
Windows

by
Julia Hartman
Multiple Linear Regression
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
2 2
Multiple Linear Regression: Multiple Linear Regression:
Introduction
Introduction
Next
Regression Analysis is the
estimation of the linear relationship between a
dependent variable and one or more independent
variables or covariates.
(from SPSS Help)
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
3 3
Multiple Linear Regression: Multiple Linear Regression:
Introduction
Introduction
This tutorial will use multiple regression to
determine if the year of matriulation, total
umulati!e undergraduate "#As, and
$CAT sores an %e used to predit how
medial students will sore on the &'
$edial (iensing )*am +#art 2,-
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
. .
Multiple Linear Regression: Multiple Linear Regression:
Introduction
Introduction
/
0ndependent !aria%les

matyr1 $atriulation year

totum1 &ndergraduate "#A +Total Cumulati!e,

nmtot11 $CAT Total 12223present, most reent


/
4ependent !aria%le

usmle21 &'$() 'tep 2


Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
5 5
Multiple Linear Regression: Multiple Linear Regression:
Starting the Procedure
Starting the Procedure
/
0n the menu, li6
Analyze
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
7 7
Multiple Linear Regression: Multiple Linear Regression:
Starting the Procedure
Starting the Procedure
/
0n the menu, li6 on
Analyze
/
#oint to
Regression
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
8 8
Multiple Linear Regression: Multiple Linear Regression:
Starting the Procedure
Starting the Procedure
/
0n the menu, li6 on
Analyze
/
#oint to
Regression
/
#oint to inear
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
9 9
Multiple Linear Regression: Multiple Linear Regression:
Starting the Procedure
Starting the Procedure
/
0n the menu, li6 on
Analyze
: and click-
/
#oint to
Regression
/
#oint to inear
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
2 2
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
Choose the !aria%les for
analysis from the list in
the !aria%le %o*-
$o!e !A"R#$%A"&'
(A") * +)AR, whih is
already highlighted, to the
%o* la%eled
#ndependent(s) %y li6ing
the arrow-
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
10 10
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
'roll down the
!aria%le list,
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
11 11
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
'roll down the
!aria%le list, point to
the !aria%le la%eled
%'()R,RA( ,PA
"&"A * $%!%A"#-)
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
12 12
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
:and click-
'roll down the
!aria%le list, point to
the !aria%le la%eled
%'()R,RA( ,PA
"&"A . $%!%A"#-)
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
13 13
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
$o!e totc/m to the
#ndependent(s) %o*
%y li6ing the arrow-
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
1. 1.
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
A few !aria%les down
the list, point to nmtot0,
whih is la%eled
!$A" "&"A 0112*
PR)S)'"3 !&S" R)$)'"
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
15 15
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
:and click-
A few !aria%les down
the list, point to nmtot0,
whih is la%eled
!$A" "&"A 0112*
PR)S)'"3 !&S" R)$)'"
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
17 17
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
$o!e nmtot0 to the
%o* la%eled
#ndependent(s) %y
li6ing the arrow-
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
18 18
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
'roll further down
the list and point to
the !aria%le /smle2,
whih is la%eled
%S!) S")P 2,
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
19 19
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
:and click-
'roll further down
the list and point to
the !aria%le /smle2,
whih is la%eled
%S!) S")P 2 "&"A,
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
12 12
Multiple Linear Regression: Multiple Linear Regression:
Selecting Variables
Selecting Variables
$o!e /smle2 to the
%o* la%eled
(ependent %y li6ing
the arrow-
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
20 20
Multiple Linear Regression: Multiple Linear Regression:
Requesting Statistics
Requesting Statistics
Re;uest desripti!e
statistis %y li6ing
the %utton la%eled
Statistics
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
21 21
Multiple Linear Regression: Multiple Linear Regression:
Statistics
Statistics
0n '#'', you an
display information %y
right3li6ing, %ut
right3li6ing doesn<t
wor6 in this tutorial-
To see what right3
li6ing would show,
left3li6 the %o* for Part and partial correlations-
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
22 22
Multiple Linear Regression: Multiple Linear Regression:
Statistics
Statistics
=or this tutorial, li6
the he6%o* for
Part and partial
correlations.
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
23 23
Multiple Linear Regression: Multiple Linear Regression:
Statistics
Statistics
'tatistis for the !odel
fit and )stimates for
Regression $oefficients
will %e produed %y
default-
Cli6 the he6%o* for
(escriptives.
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
2. 2.
Multiple Linear Regression: Multiple Linear Regression:
Statistics
Statistics
This tutorial re;uests
no additional
statistis, so li6 the
$ontin/e %utton.
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
25 25
Multiple Linear Regression: Multiple Linear Regression:
Requesting Plots
Requesting Plots
>ou an also
re;uest se!eral
different plots-
Cli6 the Plots
%utton-
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
27 27
0n the %o* la%eled
Standardized
Resid/al Plots3 first
li6 the he6%o*
for Histogram,
Multiple Linear Regression: Multiple Linear Regression:
Standardized Residual Plots
Standardized Residual Plots
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
28 28
then li6 the %o*
for 'ormal probability
plot-
0n the %o* la%eled
Standardized
Resid/al Plots3 first
li6 the he6%o*
for Histogram,
Multiple Linear Regression: Multiple Linear Regression:
Standardized Residual Plots
Standardized Residual Plots
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
29 29
This tutorial re;uests
none of the other
plots a!aila%le in the
(inear Regression
proedure- Cli6 the
$ontin/e %utton-
Multiple Linear Regression: Multiple Linear Regression:
Standardized Residual Plots
Standardized Residual Plots
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
22 22
The independent !aria%les an
%e entered into the analysis
using fi!e different methods-
This tutorial inludes the Enter
and Stepwise methods-
Choose one of the following1
Multiple Linear Regression: Multiple Linear Regression:
Method of Variable ntr!
Method of Variable ntr!
4
)nter
4
Stepwise
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
30 30
The independent !aria%les an
%e entered into the analysis
using fi!e different methods-
This tutorial inludes the )nter
and 'tepwise methods-
Choose one of the following1
Multiple Linear Regression: Multiple Linear Regression:
nter Method
nter Method
4
)nter
4
Stepwise
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
31 31
)nter is the default
method of !aria%le
entry- Cli6 the &5
%utton to run the
$ultiple (inear
Regression proedure-
Multiple Linear Regression: Multiple Linear Regression:
nter Method
nter Method
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
32 32
The independent !aria%les an
%e entered into the analysis
using fi!e different methods-
This tutorial inludes the )nter
and 'tepwise methods-
Choose one of the following1
Multiple Linear Regression: Multiple Linear Regression:
Step"ise Method
Step"ise Method
/
)nter
/
'tepwise
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
33 33
To selet the stepwise
method of !aria%le entry,
li6 the drop3down arrow
for !ethod +ne*t to )nter,
whih is the default
method,-
Multiple Linear Regression: Multiple Linear Regression:
Step"ise Method
Step"ise Method
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
3. 3.
#oint to Stepwise
Multiple Linear Regression: Multiple Linear Regression:
Step"ise Method
Step"ise Method
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
35 35
Multiple Linear Regression: Multiple Linear Regression:
Step"ise Method
Step"ise Method
#oint to Stepwise
and li6-
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
37 37
Multiple Linear Regression: Multiple Linear Regression:
Step"ise Method
Step"ise Method
Cli6 the &5 %utton to
run the $ultiple (inear
Regression proedure-
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
38 38
Multiple Linear Regression #utput: Multiple Linear Regression #utput:
$escripti%e Statistics
$escripti%e Statistics
The la%els and format
of your output may %e
somewhat different-
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
39 39
Multiple Linear Regression #utput: Multiple Linear Regression #utput:
&orrelations
&orrelations
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
32 32
Multiple Linear Regression nter Method #utput: Multiple Linear Regression nter Method #utput:
Variables ntered
Variables ntered
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
.0 .0
Multiple Linear Regression nter Method #utput: Multiple Linear Regression nter Method #utput:
Model Su''ar!
Model Su''ar!
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
.1 .1
Multiple Linear Regression nter Method #utput: Multiple Linear Regression nter Method #utput:
(N#V(
(N#V(
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
.2 .2
Multiple Linear Regression nter Method #utput: Multiple Linear Regression nter Method #utput:
&oefficients and &orrelations
&oefficients and &orrelations
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
.3 .3
Multiple Linear Regression nter Method #utput: Multiple Linear Regression nter Method #utput:
Residuals Statistics
Residuals Statistics
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
.. ..
Multiple Linear Regression nter Method #utput: Multiple Linear Regression nter Method #utput:
Residuals )istogra'
Residuals )istogra'
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
.5 .5
Multiple Linear Regression nter Method #utput: Multiple Linear Regression nter Method #utput:
Plot of Standardized Residuals
Plot of Standardized Residuals
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
.7 .7
Multiple Linear Regression #utput: Multiple Linear Regression #utput:
$escripti%e Statistics
$escripti%e Statistics
The la%els and format
of your output may %e
somewhat different-
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
.8 .8
Multiple Linear Regression #utput: Multiple Linear Regression #utput:
&orrelations
&orrelations
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
.9 .9
Multiple Linear Regression Step"ise Method #utput: Multiple Linear Regression Step"ise Method #utput:
Variables ntered
Variables ntered
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
.2 .2
Multiple Linear Regression Step"ise Method #utput: Multiple Linear Regression Step"ise Method #utput:
Model Su''ar!
Model Su''ar!
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
50 50
Multiple Linear Regression Step"ise Method #utput: Multiple Linear Regression Step"ise Method #utput:
(N#V(
(N#V(
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
51 51
Multiple Linear Regression Step"ise Method #utput: Multiple Linear Regression Step"ise Method #utput:
&oefficients and &orrelations
&oefficients and &orrelations
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
52 52
Multiple Linear Regression Step"ise Method #utput: Multiple Linear Regression Step"ise Method #utput:
xcluded Variables
xcluded Variables
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
53 53
Multiple Linear Regression Step"ise Method #utput: Multiple Linear Regression Step"ise Method #utput:
Residuals Statistics
Residuals Statistics
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
5. 5.
Multiple Linear Regression Step"ise Method #utput: Multiple Linear Regression Step"ise Method #utput:
Residuals )istogra'
Residuals )istogra'
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
55 55
Multiple Linear Regression Step"ise Method #utput: Multiple Linear Regression Step"ise Method #utput:
Plot of Standardized Residuals
Plot of Standardized Residuals
Next
Copyright 2000, Julia Hartman Copyright 2000, Julia Hartman
57 57
(n Interacti%e *utorial for
(n Interacti%e *utorial for
SPSS +,-, for .indo"s
SPSS +,-, for .indo"s

:
:
Multiple Linear Regression
Multiple Linear Regression
/
Repeat this tutorial
Cli6 one of the following1
/
Return to the hoie of methods of
!aria%le entry
/
Return to the list of tutorials

You might also like