Scribd
Upload
96 views22 pages

Multiple Linear Regression: Application

1. This document discusses the assumptions that must be met when conducting multiple linear regression analysis, including that the dependent variable is continuous, there are two or more independent variables, and observations are independent. 2. Additional assumptions are that there is a linear relationship between variables, equal variances along the line of best fit, no multicollinearity between predictors, no outliers or influential points, and normally distributed residuals. 3. The document provides examples of how to test for these assumptions, such as using scatterplots, partial regression plots, and examining variance inflation factors. Meeting all assumptions is necessary for valid results from multiple linear regression.

Uploaded by

danna ibanez
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
96 views22 pages

Multiple Linear Regression: Application

1. This document discusses the assumptions that must be met when conducting multiple linear regression analysis, including that the dependent variable is continuous, there are two or more independent variables, and observations are independent. 2. Additional assumptions are that there is a linear relationship between variables, equal variances along the line of best fit, no multicollinearity between predictors, no outliers or influential points, and normally distributed residuals. 3. The document provides examples of how to test for these assumptions, such as using scatterplots, partial regression plots, and examining variance inflation factors. Meeting all assumptions is necessary for valid results from multiple linear regression.

Uploaded by

danna ibanez
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 22

MULTIPLE LINEAR REGRESSION

APPLICATION

Engr. Alan B. Alejandrino, Ph.D


Graduate School of Government and Management
University of Southeastern Philippines
ASSUMPTIONS
Assumptions

When we
When we choose to analyze our data using
Multiple Linear Regression (MLR), we have
Multiple
to ensure that the data we want to analyze
can actually be analyzed using MLR
Assumptions

Dependent Variable should be measured

1 on a Continuous Scale
 Interval Variable
 Ratio Variable

If our Dependent Variable was measured


on an ordinal scale, we will need to carry
out Ordinal Regression rather than
Multiple Linear Regression. Examples
of Ordinal Variables include Likert Items
Assumptions

We have two or more independent variables,

2 which can be either Categorical or Continuous


Variable
Categorical Variable
 Nominal Variable
 Ordinal Variable

Continuous Variable
 Interval Variable
 Ratio Variable
If one of your Independent Variables is dichotomous
and considered a moderating variable, you might
need to run a Dichotomous Moderator Variable
Assumptions

3
We should have Independence of Observations

 We can easily check using the Durbin -


Watson Statistic
 A Rule of Thumb is that test statistic
values in the range of 1.5 to 2.5 are
relatively normal. Values outside of
this range could be cause for concern
R - Multiple The adjusted R-squared
Correlation compares the descriptive
Coefficient power of regression
One of the measures models that include diverse
of the quality of the numbers of predictors.
prediction of the
dependent variable

R2 - Coefficient of Every predictor added to a model


Determination increases R2 and never decreases it.

Independent Variables In an Overfitting Condition, an


explain 67.3% of the incorrectly high value of R2,
variability of our which leads to a decreased ability
dependent variable to predict, is obtained.
Assumptions

4
There needs to be a Linear Relationship
between
a. The dependent variable and each of
our independent variables, and
b. The dependent variable and the
independent variables collectively.
There are a number of ways to check for
these linear relationships,
 Scatterplots and
 Partial Regression Plots using SPSS
Assumptions

Our data needs to show Homoscedasticity,

5 which is where the variances along the line


of best fit remain similar as we move along
the line.
To assess the Homoscedasticity of our data,
we need to plot the Standardized Residuals
against the Standardized Predicted Values.
Assumptions

Our data must not show Multicollinearity,

6 which occurs when we have two or more


independent variables that are highly
correlated with each other.
This leads to problems with understanding
which independent variable contributes to
the variance explained in the dependent
variable, as well as technical issues in
calculating a multiple regression model.
Assumptions

SPSS Statistics detect Multicollinearity through


an inspection of correlation coefficients and
Tolerance/VIF Values

< 10
Assumptions

There should be no Significant Outliers, High

7 Leverage Points or Highly Influential Points.

Outliers, Leverage and Influential Points are


different terms used to represent observations
in our data set that are in some way unusual
when we wish to perform a multiple regression
analysis.
Assumptions

These points can have a very negative effect on


the regression equation that is used to predict
the value of the dependent variable based on
the independent variables.
Fortunately, when using SPSS
Statistics to run Multiple Linear
Regression on our data, we can
detect possible outliers, high
leverage points and highly
influential points.
Assumptions

Finally, we need to check that the residuals

8 (errors) are approximately normally


distributed.

Two common methods to check this


assumption include using:
a. Histogram with a superimposed normal
curve and a Normal P-P Plot; or
b. Normal Q-Q Plot of the Standardized
Residuals.
Let’s get started with… and

SPSS
Institutional, Strategic, Leadership
Traits and Research and Extension
Capability of Presentation of Mary
Sister’s Managed Schools

Conceptual Framework
showing the relationship
between the Independent
and Dependent Variables

You might also like