A REPORT ON MULTIPLE REGRESSIONS

Course: BUS 173

Submitted to –

Dr. Zakir Hossain (ZKH3)

Department of Management
North South University

Submitted by:

Name ID
Samman Rahman 1721906630
Md Ahad Hossain Monve 1521424631

Md. Ehasan Kabir Emon 1721639030

Khirul Islam 1521468030

Md Ahad Hossain Monve 1521424631

1|Page
 BUS173 - Project (ZKH3)

Summer 2019
Submission deadline: 28 August 2019

The goal of the project is to examine the effect of education, gender, work experience and minority
classification on the beginning salary of the employees in a company. Different data sets are given
to different groups (individual) of students.

[Use SPSS/MINITAB/any other computer packages for data analysis]

1. Introduction: Write a paragraph about introduction of the project including data and
variables.
2. Find the estimated regression equation of the beginning salary on the education, gender,
work experience and minority classification.
3. Explain the meaning of the estimated regression coefficients (slope) values.
4. What is the value of the coefficient of multiple determination? Interpret the result.
5. What is the predicted beginning salary for the employee who is non-white male and has 17
years of education with 10 years of work experience?
6. Determine the 95% confidence intervals for the regression coefficients.
7. Discussion and conclusion: Write two paragraphs: one for the association between
dependent and independent variables [use sign of the estimated regression coefficients] and
the other about the significant (important) factors for determining the beginning salary of
the employees on the basis of your results obtained from the analysis [use p-values].

2|Page
TABLE OF CONTENTS
Answer to the Question no. 1.......................................................................................................... 4
Answer to the question no. 2........................................................................................................... 5
Answer to the question no. 3........................................................................................................... 5
Answer to the question no. 4........................................................................................................... 6
Answer to the question no. 5........................................................................................................... 6
Answer to the Question no. 6.......................................................................................................... 7
Answer to the Question no. 7.......................................................................................................... 9
Appendix:...................................................................................................................................... 11

3|Page
ANSWER TO THE QUESTION NO. 1

The goal of this project is to examine the effect of education, gender, work experience and minority
classification on the beginning salary of the employees in a company. A random sample of 200
students has been taken to conduct the research.

Using the SPSS software, we will find the following requirements:

1. Estimated Regression Equation

2. Estimated Regression Coefficient
3. Standard Deviation of errors, the Coefficient of Multiple Determination
4. Predicted beginning salary of a white male and has 17 years of education with 10 years of
work experience.
5. 95% confidence interval for the regression coefficient

Here,

The data set has a sample size of, n= 200

The 4 independent variables are education level, gender, work experience and minority
classification. The dependent variable is the beginning salary of an employee.

Let,

Y = The beginning salary of an employee

X1 = Education level

X2 = Gender

X3 = Work experience

X4 = Minority classification

4|Page
ANSWER TO THE QUESTION NO. 2

Estimated regression equation:

When we estimate the regression model using SPSS software, we obtain the value of a = -1464.389

Estimated Regression coefficient of X1 (Education), b1 = 621.524

Estimated Regression coefficient of X2 (Gender), b2 = -1174.979

Estimated Regression coefficient of X3 (Work Experience), b3= 61.709

Estimated Regression coefficient of X4 (Minority Classification), b4 = -778.530

The estimated regression equation, ŷ = a+ b1X1 +b2 X2 + b3 X3+b4X4

ŷ= -1464.389+ 621.524X1-1174.979X2+ 61.709X3- 778.530X4

ANSWER TO THE QUESTION NO. 3

 The value of b1 = 621.524 in the estimated regression model gives the change of salary in for
a year change in education level X1 when X2, X3 and X4 held constant. We can state that an
employee with one extra year of education (without considering gender and maturity
classification) is expected to get $621.524 more beginning salary per month.
 The value of b2 = -1174.979 in the estimated regression model when X1, X3 and X4 are constant
denotes that a female employee is expected to get $1174.979 less beginning salary than a male
employee.
 The value of b3 = 61.709 in the regression model when X1, X2, and X4 are constant means an
employee with 1-year extra experience is likely to get $61.709 more monthly salary.
 The value of b4 = -778.530 in the regression model when X1 and X2, X3 are constant means a
Non-White employee is expected to get $-778.530 less beginning salary than a White
employee

5|Page
ANSWER TO THE QUESTION NO. 4

From output of SPSS,

Standers Deviation of Errors, Se = 2441.081

Coefficient of multiple determination, R2 = 42.6%

Adjusted coefficient of multiple determination, 𝑅̅ 2 = 41.4%

Interpretation:

The value of Se = 2441.081 tells us that the average difference of the sample mean from the
population mean.
The value of R2 = 42.6% state that 42.6% of total variation in beginning salary is explained by
education, gender and minority classification, work experience. And 57.4% variation depends on
other factors.

ANSWER TO THE QUESTION NO. 5

Here,

X1 = 17 years

X2 = 0 (Male)

X3 = 10 years

X4 = 1 (Non White)

ŷ= -1464.389+ (621.524*17) -(1174.979*0) + (61.709*10) -(778.530*1)

The estimated regression model:

= $8,940.079

So, we can predict that the beginning salary for an employee who is non-white, male, has 17 years
of education and 10 years working experience is roughly $8,940.079.

6|Page
ANSWER TO THE QUESTION NO. 6

Determine the 95% confidence intervals for the regression coefficients

Determine confidence interval for education(x₁)

The confidence interval is 95%

We will use t distribution

Alpha(𝛼) is .05

Here, b1= 621.524

SB₁ = 68.023

N= 200

Degree of freedom n-k-1= 195

From t distribution table the value of t is 1.660

The 95% confidence interval for B₁ is

b1 ± t Sb₁ = 621.524± (1.660) (68.023)

508.61 to 734.44

We can state that for 95% confidence interval the beginning salary varies from 508.61 to 734.44

Depending on education level

(B) Confidence interval for x₂ (gender)

Here b₂= -1174.979

Degree of freedom n-k-1= 195

From t distribution the value of t is 1.660

95% confidence interval for B₂ is

b₂ ± Sb₂= -1174.979± (1.660) (386.530)

7|Page
= -1816.61 to -533.33

We can state that for 95% confidence interval beginning salary varies from 1816.61 to -533.33 for
different gender.

(C) Confidence interval for x₃ (Work Experience)

Here b₃ is 61.709

Degree of freedom n-k-1= 195

From t distribution the value is 1.660

95% confidence interval for b₃ is

b₃± Sb₃= 61.709± (1.660) (22.034)

= 25.13 to 98.28

We can state that for 95% confidence interval the beginning salary varies from 25.13 to 98.28
depending on work experience.

(D) Confidence interval for X4 (Minority classification)

Here b4 is -778.530

Degree of freedom n-k-1= 195

From t distribution the value is 1.660

95% confidence interval for b4 is

B4± Sb4= -778.530± (1.660) (443.517)

= -1514.76 to -42.29

We can state that for 95% confidence interval the beginning salary is very from -1514.76 to -42.29
for minority classification.

8|Page
ANSWER TO THE QUESTION NO. 7

The p-value for each term tests the null hypothesis that the coefficient is equal to zero (no effect).
A low p-value (< 0.05) indicates that we can reject the null hypothesis. In other words, a predictor
that has a low p-value is likely to be a meaningful addition to our model because changes in the
predictor's value are related to changes in the response variable.

From the Output of SPSS, we can see that the predictor variables of education, work experience
and gender of employees are significant because both of their P-values are 0.000. Minority
Classification has a P-value of .161 which makes it a relatively less significant factor.

Significant factors:
Education:
Null Hypothesis, H0: B1 = 0 (Coefficient of education is equal to zero)

Education (p-value = 0.000) is highly significant factor for determining beginning salary of the
employee at assuming 1% significance level.(P-value<∝ so we reject the null hypothesis)

Gender of employees:
Null Hypothesis, H0: B2 = 0 (Coefficient of gender of employees is equal to zero)

Gender of employee (p-value = 0.000) is highly significant factor for determining beginning salary
of the employee at assuming 1% significance level. (P-value < ∝ so we reject the null hypothesis)

Work Experience:

Null Hypothesis, H0: B3 = 0 (Coefficient of work experience of employees is equal to zero)

Work experience of employees (p-value = 0.000) is highly significant factor for determining
beginning salary of the employee at assuming 1% significance level. (P-value < ∝ so we reject the
null hypothesis)

9|Page
Minority Classification:

Null Hypothesis, H0: B4= 0 (Coefficient of minority Classification is equal to zero)

Minority Classification (p-value = .161) is not a significant factor for determining beginning salary
of the employee at assuming 1% significant level as the P-value > ∝, so we accept the null
hypothesis.

Therefore, we can conclude that, education, gender of employees, work experience can highly
affect the beginning salary of the employees in a company. On the other hand, minority
classification does not effect on the beginning salary on an employee by much.

10 | P a g e
APPENDIX:

Regression
[DataSet2] C:\Users\Didar\Desktop\group1.sav

a
Variables Entered/Removed

Variables Variables

Model Entered Removed Method

1 MINORITY . Enter

CLASSIFICAT

ION, SEX OF

EMPLOYEE,

WORK

EXPERIENCE

EDUCATION
b
AL LEVEL

a. Dependent Variable: BEGINNING SALARY

b. All requested variables entered.

Model Summary

Adjusted R Std. Error of

Model R R Square Square the Estimate

a
1 .653 .426 .414 2441.081

a. Predictors: (Constant), MINORITY CLASSIFICATION, SEX OF EMPLOYEE, WORK

EXPERIENCE, EDUCATIONAL LEVEL

a
ANOVA

11 | P a g e
 Sum of

Model Squares df Mean Square F Sig.

b
1 Regression 861920212.5 4 215480053.1 36.161 .000

Residual 1161981219 195 5958878.049

Total 2023901432 199

a. Dependent Variable: BEGINNING SALARY

b. Predictors: (Constant), MINORITY CLASSIFICATION, SEX OF EMPLOYEE, WORK

EXPERIENCE, EDUCATIONAL LEVEL

12 | P a g e
 a
Coefficients

Standardized

Unstandardized Coefficients Coefficients

Model B Std. Error Beta t Sig.

1 (Constant) -1464.389 1089.466 -1.344 .180

EDUCATIONAL LEVEL 621.524 68.023 .555 9.137 .000

SEX OF EMPLOYEE -1174.979 386.530 -.183 -3.040 .003

WORK EXPERIENCE 61.709 22.034 .161 2.801 .006

MINORITY -778.530 443.517 -.098 -1.755 .081

CLASSIFICATION

a. Dependent Variable: BEGINNING SALARY

MEANS TABLES=salbeg BY sex

/CELLS MEAN COUNT.

Means

[DataSet2] C:\Users\Didar\Desktop\group1.sav

Case Processing Summary

Cases

13 | P a g e
 Included Excluded Total

N Percent N Percent N Percent

BEGINNING SALARY * 200 100.0% 0 0.0% 200 100.0%

SEX OF EMPLOYEE

Report

BEGINNING SALARY

SEX OF EMPLOYEE Mean N

MALES 7889.49 113

FEMALES 5371.59 87

Total 6794.20 200

14 | P a g e