3.

Multivariate statistical Analysis

• Many statistical techniques focus on just one or two variables

• Multivariate analysis (MVA) techniques allow more than two

variables to be analysed at once
• Multivariate analysis (MVA) is defined as the simultaneous
analysis of more than two variables

July 1, 2024
 Multiple Regression Analysis
• Multiple Regression is a statistical method for
estimating the relationship between a dependent
variable and two or more independent (or
predictor) variables.
• Multiple linear regression analysis is used to examine
the relationship between two or more independent
variables and one dependent variable.
• The independent variables can be measured at any
level (i.e., nominal, ordinal, interval, or ratio).
• The dependent variable MUST be measured at the interval- or
July 1, 2024
 Simple vs. Multiple Regression
• .
• One dependent variable Y
predicted from one • One dependent variable Y

independent variable X predicted from a set of

independent variables (X1, X2
….Xk)
• One regression coefficient
• One regression coefficient for
each independent variable

• r2: proportion of variation • R2: proportion of variation in

in dependent variable Y dependent variable Y predictable
by set of independent variables
predictable from X
(X’s)
July 1, 2024
 Basic Assumptions
• Independence: The data of any particular subject are

independent of the data of all other subjects

• Normality: in the population, the data on the dependent variable

are normally distributed for each of the possible combinations of

the level of the X variables; each of the variables is normally

distributed

• Homoscedasticity: In the population, the variances of the

dependent variable for each of the possible combinations of the

levels of the X variables are equal.

• Linearity: In the population, the relation between the dependent

variable and the independent variable is linear when all the other
July 1, 2024
independent variables are held constant
 Model Check

•Examine collinearity diagnostics to check for

multicollinearity
•Examine residual plots to check error variance
assumptions (i.e., normality and homogeneity of
variance)
•Write the final regression equation and interpret the
coefficient estimates

July 1, 2024
 Multicollinearity - 1
• Multicollinearity is a problem in regression analysis that occurs
when two independent variables are highly correlated,

• e.g. r = 0.90, or higher.

• The relationship between the independent variables and the

dependent variables is distorted by the very strong relationship
between the independent variables, leading to the likelihood that
our interpretation of relationships will be incorrect.

• In the worst case, if the variables are perfectly correlated, the

regression cannot be computed.

• SPSS guards against the failure to compute a regression solution

by arbitrarily omitting the collinear variable from the analysis.
July 1, 2024
 Multicollinearity - 2

• Multicollinearity is detected by examining the tolerance for each

independent variable. Tolerance is the amount of variability in one

independent variable that is no explained by the other independent

variables.

• Tolerance values less than 0.20 indicate collinearity.

• If we discover collinearity in the regression output, we should reject

the interpretation of the relationships as false until the issue is

resolved.

• Multicollinearity can be resolved by combining the highly correlated

variables through principal component analysis, or omitting a

variable from the analysis.

July 1, 2024
 The Multiple Regression Model building
Idea: Examine the linear relationship between
1 dependent (Y) & 2 or more independent variables

Multiple Regression Model with k Independent Variables:

Y-intercept Population slopes Random Error

Yi  β0  β1 X 1i  β2 X 2i    βk X ki  ε
Interpretation of the Slopes: (referred to as a Net Regression
Coefficient) b1=The change in the mean of Y per unit change in X1,
taking into account the effect of X2 (or net of X2)
July 1, 2024
b0 Y intercept. It is the same as simple regression.
 Vignette

• A researcher wants to test some hypotheses

regarding the relationship between size and age
of a firm and its performance in a particular
industry.
• Size was measured by the number of employees
working in the firm, age was the number of years
for which the firm has been operating,
• and performance was measured by return on
equity. A sample of 30 firms was selected at
July 1, 2024
random. Data on these variables is given in .
 Exercise

Ho : Performance of a firm is not related to its size

Ho : Performance of a firm is not related to its age

[Link] the values of the variables into SPSS.

2. Determine the multiple regression equation

for the data.

3. At the 5% significance level, determine if the

model is useful for predicting the response.

[Link]
1, 2024
the hypothesis.
 Procedures

• The given data is entered in the data editor and the

variables are labeled as id, perf, size, and age.

• Click on Analyze, which will produce a drop down menu,

• choose Regression from that and click on Linear

• Transfer the dependent variable into the right-hand side

box labeled Dependent.

• Transfer the independent variables into the box labeled

Independent(s).

• The DV in our example is Performance and the IVs are size

July 1, 2024
and age
 .

• We have to select the method for analysis in the box labeled Method.

• SPSS gives five options here: Enter, Stepwise, Remove, Backward,

and Forward.

• In the absence of a strong theoretical reason for using a particular

method, Enter should be used.

• The box labeled Selection Variable is used if we want to restrict the

analysis to cases satisfying particular selection criteria.

• The box labeled Case Labels is used for designating a variable to

identify points on plots.

• After making the appropriate selections click on the Statistics button.

• This will produce a dialog box labeled Linear Regression: Statistics as

shown in
July 1, 2024
 .

•The Estimates option gives the estimate of regression

coefficients.

•The Model fit option gives the fit indices for the overall
model.

•The Descriptives option gives the descriptive

statistics of the selected variables.

 The Collinearity diagnostics option produces

important statistics used for assessing the presence of

multicollinearity in the data.

•Besides these, the R squared change option is used to get

the incremental R Square value when the models change.
July 1, 2024
 Output
The output produced has several tables:
 Descriptive Statistics,
• Correlations,
• Variables Entered/Removed,
• Model Summary,
• ANOVA,
• Coefficients,
• Collinearity Diagnostics.

July 1, 2024
 Variables Entered/Removed
Model Variables Entered Variables Removed Method
1 age of firms, size of a firm . Enter
a. All requested variables entered.
b. Dependent Variable: performance of firms
• table
The . tells us about the independent variables and the
regression method used.
Model Summary
Std. Error of the
Model R R Square Adjusted R Square Estimate
1 .324 .105 .038 14.705
a. Predictors: (Constant), age of firms, size of a firm

Model Summary gives us the R values for assessing the overall fit of the
model.
The adjusted R Square value is 0.038. This tells us that the two IVs
in our model account for 3.8% variance in the DV—performance of
the firm
Clearly this is not a very good model as there are factors other than
age and size of a firm which should also be used to predict a firm’s
performance.
July 1, 2024
 Multicollinearity.

• .

For testing multicollinearity in the model, Collinearity Statistics gives two

values—Tolerance and VIF (variance inflation factor).
As one can see Tolerance is just the inverse of VIF.
A value of VIF higher than five (or Tolerance less than 0.2)
indicates the presence of multicollinearity.
We can conclude that there is no problem of collinearity.

July 1, 2024
 .

Now we test our hypotheses. The null hypothesis is that there is no

relationship ,i.e., the beta coefficient is not different from zero. The
p-value for beta coefficient of Size is 0.267, the same for Age is .276.
at 5% significance level, the relationship between performance and
age will not be significant. For both variables p> .05. This implies
that the influence of size and age on company
performance is insignificant.

Y= 13.08+.131x -.217x
July 1, 2024
 Writing Research Reports
Key Ideas
• What is a research report?
• Who will receive this report?
• What are the types of reports?
• How should your report be structured?
• How do you write in a sensitive and scholarly way?
• How do you evaluate the quality of your research?

July 1, 2024
 What is a research report?
• A research report is a completed study that reports

an investigation or exploration of a problem,

identifies questions to be addressed, and includes

data collected, analyzed, and interpreted by the

researcher.

July 1, 2024
 Who will receive this report? The audience for
your report

• Determine the acceptable standards

• Look in journals to learn the criteria required for
submitting articles
• Look at the literature for specific standards
• Check with the specific standards for a thesis

July 1, 2024
 What are the types of research reports?

• Research Report: a completed study that

reports an investigation or exploration of a
problem
– Theses
– Theses proposals
– Journal articles
– Conference papers
– Conference paper proposals
– Reports for policy makers and schools

July 1, 2024
 Structure of Proposal

• . Title page
 Abstract
 Introduction/ Background

(Problem Statement, Purpose/ Aims/ Rationale/

Research Questions ,Hypothesis
Significance/ Implications

 Review of the Literature

 Methodology

 Timeline, Budget, and Preliminary Chapter Outline

 References/ Bibliography

 Appendices
July 1, 2024
 Writing Research Reports
1. Introduction/Background (Operational definition, statement of the

problem, Research questions, objectives, Hypothesis , significance of the study, scope of

the study)

2. Literature Review ( theoretical review, empirical study, conceptual framework)

3. Methodology (Research design, Research approach, source of data, Data

collection methods, questionnaires design, sample design , Reliability and validity )

4. Data analysis & interpretation: Tie the results to study

purpose (hypotheses)

5. Discussions and Conclusion:

Tie discussions to the literature; recommendations for

July 1, 2024
practice; limitations of the study
How do you evaluate the quality of your research
report?
• Does it meet publication standards?
• Will it be useful ?
• Will it advance policy discussions?
• Will it add scholarly knowledge about a topic or
research problem?
• Will it help address some pressing educational
problem?

July 1, 2024
 Effective Writing
• Clear, concise, direct

• Thoughts are complete

• Attention given to details such as grammar, spelling

• Flow of thoughts

July 1, 2024
 Ways of Conducting Reports

A. Quantitative reports
• Report results by the use of tables and graphs
• Avoid first-person pronoun
• Use passive voice (It is shown / suggested that…)

July 1, 2024
 Qualitative reports
• Look for a deep and valid description (narrative
style)
• Look for well-grounded theory
• Seek contextual meaning by understanding
demographic information (different experiences)

July 1, 2024
 BIBLIOGRAPHY
• This include all materials used and reviewed by the
researcher, such as books, magazines, periodicals, journals,
thesis or dissertation (published or unpublished).
• Monographs, speeches and modules, web page or internet,
etc.
Difference between a reference list and a bibliography
• The reference list only identifies sources referred to (cited) in
the text of your assignment. You may also be required to
provide a bibliography.
• A bibliography is presented in the same format as a
reference list but it includes all material consulted in the
preparation of your assignment.
• In other words, a bibliography presents the same items as a
reference list but it also includes all other sources which you
read
July 1, 2024or consulted but did not cite.
 Citation/Bibliographic Standards
• Whenever you quote or base your ideas on another person's work, you must document

the source you used. Even when you do not quote directly from another work, if

reading that source contributed to the ideas presented in your paper, you must give the

authors proper credit.

• Citations allow readers to locate and further explore the sources you consulted, show

the depth and scope of your research, and give credit to authors for their ideas.

• Citations provide evidence for your arguments and add credibility to your work by

demonstrating that you have sought out and considered a variety of resources.

• In written academic work, citing sources is standard practice and shows that you are

responding to this person, agreeing with that person, and adding something of your

own.

• Think of documenting your sources as providing a trail for your reader to follow to see

the research you performed and discover what led you to your original contribution

July 1, 2024
 .

1. APA [Link] (APA) The American Psychological

Association's official website for APA Style.

2. MLA (Modern Language Association) Style

• Documenting Sources
Guidelines based on the MLA Style Manual and
Guide to Scholarly Publishing, Third Edition (2008)

3. Chicago Manual of style

N.B: Please refer the different types of styles in the attached

document
July 1, 2024 provided to you.
 APA-formatted references
• Journal articles
– Roy, A. (1982). Suicide in chronic schizophrenia. British
Journal of Child and Family Studies, 141, 171-177.

• Book
– Struck, W., Jr., & White, E. B. (1979). The elements of style

(3rd ed.). New York: Macmillan.

• Web page, no author, no date

– GVU’s 8th WWW user suvey. (n.d.) Retrieved September 19,
2001, from
[Link]
July 1, 2024
 Useful websites for Researchers

1. Free book down load : [Link]

hl=en&shva=1#inbox/14194f1f5762cf9f

[Link] <newsletter@[Link]

2. Journal

[Link]
ml

3. Google, write your research title, ok

4. Google, Google scholar, library, write your title

and okay.
July 1, 2024
 .
•Thank you
for your
attention
July 1, 2024