## You are required to carry out a research study in which you develop a Multiple Linear Regression (MLR) model, using stepwise regression and write a report about it. You are to create a problem statement as the focus of this research.

### computer science

##### Description

Instructions

You are required to carry out a research study in which you develop a Multiple Linear Regression (MLR) model, using stepwise regression and write a report about it. You are to create a problem statement as the focus of this research.

For the purpose of this research you need to obtain a set of at least 100 cases of secondary data with:

one metric dependent variable;

at least 8 metric independent variables.

All the statistical analyses must be conducted using SPSS software.

Note:   This data set will also be used for Question 2, which deals with factor analysis, so it is recommended to have at least 8 metric independent variables as stated above.

Write a report that tackles the following tasks:

(a)                Problem statement

The research study should have a problem statement which clearly defines a problem.

(5 marks)

(b)        Introduction

You should state the background of study and explain the reasons or purposes of your study. This is the ‘what’ and ‘why’ of the study. Besides, you must include at least 2 literatures or research studies which are related to yours.

(5 marks)

(c)        Research Objectives

State the research objectives of your study. Corresponding to these objectives at least three hypotheses are to be developed.

(7 marks)

(d)       Analysis Results and Interpretations

Beside other relevant analyses, you are also required to perform the following statistical procedure:

(ii)        Perform stepwise multiple linear regression between the dependent variable and the independent variables using SPSS.

(iii)       Examine the regression results to determine if there are violations of the model assumptions.

(iv)       Explain and interpret the model summary results.

(v)        Determine whether the regression model is adequate/significant.

(vi)       Conduct hypothesis test for the coefficients of regression.

You do not need to include all your outputs, but select the outputs that will support your result interpretations. All the statistical results/outputs need to be labelled and numbered for ease of reference.

(40 marks)

(e)        Conclusions & Recommendations

Briefly summarise your report. Draw conclusions and make recommendations. What are the implications of the results? Where appropriate, identify limitations and areas for further investigations/study.

(10 marks)

(f)        Appendix & Reference

Among other information, this may include the outputs of SPSS software.

(3 marks)

Question 2 (30 marks)

#### Instructions

This question assesses your mastery in Factor Analysis, by evaluating all the metric independent variables used in Question 1. You are required to perform factor analysis using all the independent variables and tackle the following tasks.

(a)        In view of your problem statement in Question 1, elaborate briefly the purpose of performing factor analysis on your set of metric independent variables.

(5 marks)

(b)        Discuss briefly why the non-metric independent variables are not allowed in factor analysis.

(4 marks)

(c)        Perform factor analysis on your metric independent variables by using SPSS. Present and label all the SPSS outputs for ease of reference.

(5 marks)

(d)       By using relevant SPSS outputs in part (c), group your metric independent variables into factors.

(4 marks)

(e)        Referring to your own factor analysis outputs:

(i)         Interpret the meaning of communality.

(3 marks)

(ii)        Interpret the meaning of eigenvalue.

(3 marks)

(f)        By assessing the relevant outputs obtained in part (c):

(i)         Explain briefly how the factorability of your dataset can be improved prior to the task of factor analysis.

(2 marks)

(2 marks)