Logistic or Simple Linear Multiple Correlation Regression

¶ … Linear, Logistic or Multiple Correlation/Regression Proposal

Objective of this paper is to use the statistical test to investigate whether the middle income earners in developing countries are capable of becoming home owners. The paper uses Nigeria as a case study. Unlike the developed countries such as the United States and United Kingdom where middle income earners can become house owners, however, the issues are different in developing countries because various factors prevent middle income earners from purchasing housing units compared to middle income earners in developing countries. Typically, a level of income plays a critical role in homeownership. While different governments in developing countries have implemented various policies in assisting people to become homeowners, nevertheless, factors such as corruption, inflation, and poor implementations have made the policy non-realizable. In Sub-Sahara African countries, nearly 62% of people live in slums, and per capital income of most developing countries are less than $300.(Chica, 2008). Moreover, more than 87% of individual households fund are used to fund the housing units since there are little or no support from the governments and mortgage financial institutions. Moreover, economic and social factors also prevent people from becoming home owners in developing countries. For example, majority of households in developing countries have large families, high price of building materials and low propensity to save inhibit home ownership in developing countries.

Research Question

Can middle income earners in Nigeria able to 1 finance homeownership by using 10-year personal savings amount to N2,400,00?

The study multiple linear regression as a major statistical technique to determine whether middle income earners are able to finance homeownership by using o N2,400,000 of their personal savings.

Two types of linear regressions are available they are multiple linear regression, and simple linear regression. A simple linear regression uses the singe independent variable to predict the dependent variables. In other words, simple linear regression assists in predicting the value of a dependent variable.

A multiple linear regression refers to two or more independent variables that are used to predict a dependent variable. The difference between the linear and multiple regressions there is only one dependent variables however, the independent variables are different in both cases.(Trochim, 2006). The proposal selects the multiple linear regression for the analysis because it is able to use multiple independent variables to analyze the dependent variable. (Sukal, 2013).

The statistical notations for this research are as follows:

Ho = null hypothesis.

H1 = alternative hypothesis.

= significance level.

= probability of Type II error.

x =sample mean.

s = standard deviation of a sample.

Research Hypotheses

H1: Middle income in developing countries are likely to purchase homes using their personal savings. The dependent variable is Income, while independent variables are Employment. Status, Age Range, Housing Size, and Rent.

Ho: Middle income in developing countries are not likely to purchase homes using their personal savings. The dependent variable is Income, while independent variables are Employment. Status, Age Range, Housing Size, and Rent.

2. Methods

The study collects data from respondents, who are middle income earners using the survey method. The quantitative technique is used for data collection and analysis, and the study uses the SSPS Version 19 for the analysis. The study collects data from both male and female participants, and the age of participants are between 21 and 60. The participants will be selected based on their ages, and income level. Participants below 21 years of age will be excluded from the survey and participants earning income below N60, 000 will be excluded from the survey. Moreover, unemployed people will be excluded from the survey. (Creswell, 2014).

3. Procedures

The variables are presented in the following table:

Socio-economic Variables that Affect Homeownership

Income

Household Size

Age Range

Accommodation Type

Residential Status (Tenant or Homeowner)

Gender

Rent

Institutional Variables that affect Homeownership:

Building Type

HOUSE PRICE

Permit & Documentation

Labor

Building Materials

Deposit

Land

Location

Mortgage/LOAN

Suitability

Infrastructure

4. Results

The paper uses the chi-square to determine the results and based on the probability level alpha table, the chi square of 0.05 determines the level of significant. If the p-value is greater than 0.05 (i.e. p > 0.05), the study will accept the null hypothesis, which means the study will "fail to reject the null hypothesis." "

Probability Level Table (alpha)

Df

0.5

0.10

0.05

0.02

0.01

0.001

1

0.455

2.706

3.841

5.412

6.635

10.827

2

1.386

4.605

5.991

7.824

9.210

13.815

3

2.366

6.251

7.815

9.837

11.345

16.268

4

3.357

7.779

9.488

11.668

13.277

18.465

5

4.351

9.236

11.070

13.388

15.086

20.517

However, the study uses the Likelihood Ratio Tests for the results. Since the chi-squares of all the test carried out are greater than all probability level for (alpha = 0.05), the study rejects the null hypothesis (Ho). The output of the regression is presented in the Appendices.

Thus, the study rejects the following the null hypothesis:

Ho: Middle income in developing countries are not likely to purchase homes using their personal savings.

However, the study accepts the following the null hypothesis:

H1: Middle income in developing countries are likely to purchase homes using their personal savings.

The rejection of the null hypothesis makes the paper to conclude that middle income earners are able to finance the homeownership in Nigeria.

Reference

Chica, E.U. (2008). Obstacles to the Individual Home Ownership in Nigeria. International Journal of Housing Markets and Analysis. 1(2):182-194.

Creswell, J. W. (2014). Research design: Qualitative, quantitative, and mixed methods approaches (3rd

ed.). Thousand Oaks, CA: Sage.

Sukal, M. (2013). Research methods: Applying statistics in research. San Diego, CA: Bridgepoint Education, Inc.

Trochim, W. M. K. (2006). Research methods knowledge base (2nd ed.). Web Center for Social Research Methods. Retrieved 01 November 2015 from http://www.socialresearchmethods.net/kb/index.php

Appendices

Appendix 1: Social/Economic Variable Affecting Homeownership

Model Fitting Information

Model

Model Fitting Criteria

Likelihood Ratio Tests

-2 Log Likelihood

Chi-Square

df

Sig.

Intercept Only

1206,951

Final

806,743

400,208

76

,000

Pseudo R-Square

Cox and Snell

,600

Nagelkerke

,626

McFadden

,289

Likelihood Ratio Tests

Effect

Model Fitting Criteria