馃攳
Multiple Regression in SPSS - R Square; P-Value; ANOVA F; Beta (Part 1 of 3) - YouTube
Channel: unknown
[0]
In this video, we'll take a look at how
to run a multiple regression in SPSS. And
[6]
on your screen as an example we have
four variables SAT score,
[10]
social support, gender, and college GPA.
[15]
And in this example we're using the
first three variables SAT score, social
[19]
support, and gender,
[21]
to predict first year college GPA. And
here SAT score was taken in high school,
[28]
social support is a measure of how much
support a student felt that they received
[34]
from others, where higher scores indicate
greater support, and that was taken in
[39]
the first year in college, and then
gender, our dichotomous variable, where
[44]
1 is male and 2 is female,
[47]
and the variable, college GPA, was the GPA
after the first year in college. And in
[54]
regression
[55]
what we're trying to predict in this
case, college GPA, is known as our
[60]
criterion variable.
[62]
It's also known as the dependent
variable (DV).
[65]
And then the variables that we're using
to predict the criterion variable, SAT
[71]
score, social support, and gender,
[74]
those are known as are predictors or
predictor variables,
[78]
and we also refer to those as
independent variables (IV).
[83]
And those once again are SAT score, social
support, and gender. Now in multiple
[89]
regression
[90]
you always have one criterion or
dependent variable, and for it to be
[96]
multiple regression you have to have two
or more predictors or independent
[101]
variables.
[102]
if you just had one predictor or
independent variable, such as SAT score,
[107]
then that would be simple regression. But
since we have two or more, in this case
[112]
we have three once again, we're doing
multiple regression.
[117]
OK so to run multiple regression SPSS we
want to go to Analyze,
[122]
and then Regression and then go ahead
and select Linear.
[127]
And here we want to move college GPA to
our Dependent box and then we want to
[132]
select all the predictors and move those
to our Independent(s) box.
[137]
And then go ahead and click OK. And our
output opens here and the first table,
[143]
Variables Entered/Removed,
[145]
this confirms that we had the variables
gender, SAT score, and social support as
[151]
our predictors, and then our dependent
variable, or criterion variable, was
[156]
college GPA, so that looks good.
[159]
OK our next two tables, Model Summary
and ANOVA, these two tables, they're
[165]
looking at whether are predictors, once
again,
[169]
SAT score, social support, and gender,
when those are taken together as a set
[174]
or as a group,
[176]
do they predict college GPA. And the
Model Summary and ANOVA table are
[181]
getting that slightly different things,
but they're very closely related.
[185]
So let's go ahead and start with Model
Summary and take a look at that.
[188]
So for Model Summary in this video we're
going to focus on R square and then in
[194]
another video we'll talk about these
measures in more detail.
[198]
But for this general overview the most
commonly reported value in the Model
[202]
Summary table is the R square value. And
R squared, if I round this to two decimal
[208]
places and then convert it to a
percentage,
[211]
so this would round two .50 or 50%,
I could interpret R squared as
[217]
follows.
[219]
R squared once again is equal to .50
and then taken as a set the predictors
[225]
SAT score, social support, and gender,
account for 50% of the
[231]
variance in college GPA.
[234]
OK so R squared is a measure of the
amount of variance in the dependent
[240]
variable
[242]
that the independent variables or
predictors account for when taken as a
[248]
group. And that's very important,
[250]
it doesn't measure how much a given
individual predictor accounts for, but
[255]
only when we take them all as a group,
this Model Summary table says overall,
[260]
the regression model, which is what is
referred to sometimes as a model,
[265]
these three predictors predicting
college GPA,
[268]
that overall model accounts for 50%
of the variance. Which is pretty
[273]
good in practice.
[275]
OK next we have our ANOVA table
Most Recent Videos:
You can go back to the homepage right here: Homepage