Coventry University 102ACC: Wage Determinants Analysis Report

This report presents a comprehensive data analysis of wage determinants using the 'wage.sav' dataset. It begins with an introduction outlining the objectives and scope, followed by descriptive statistics of key variables such as wage, age, experience, and tenure. The report creates a new variable using the logarithm function and explores the relationship between marital status and wage through scatter plots. Regression analysis is performed, including the explanation of model summary, ANOVA, and coefficients tables, to identify the nature and strength of the relationship between the dependent variable (lnwage) and independent variables. The results are discussed, highlighting the direct and positive relationships between independent variables and wage levels. Furthermore, diagnostic tests are conducted to check for multicollinearity, with detailed methodology and results presented. The report concludes with a summary of findings and references.

Data Analysis

Contents
INTRODUCTION...........................................................................................................................1
Descriptive statistics....................................................................................................................1
Graphs..........................................................................................................................................2
Creating a new variable ln(wage) and explaining the use of the function logarithm..................3
REGRESSION ANALYSIS AND DISCUSSION OF THE RESULTS........................................4
The regression model...................................................................................................................4
Explaining the model summary table..........................................................................................4
Interpreting the results for the ANOVA table.............................................................................4
Interpreting the results for the coefficients table.........................................................................5
Discussing the results..................................................................................................................6
DIAGNOSTIC TESTS....................................................................................................................6
Checking for the existence of the multicollinearity issue, explaining the methodology and
discussing the results...................................................................................................................6
CONCLUSION..............................................................................................................................11
REFERENCES..............................................................................................................................12

INTRODUCTION
Data analysis is a process of analysing a data set in order to make relevant decisions
(Chambers, 2018). The main aim of this report is to analyse the data set having information
about wage and factors which influence the wage of workers. In this report, descriptive statistics
of all variables will be interpreted along with creating a new variable using logarithm. In this
report, regression and correlation tests using scatter plots will also be conducted. Lastly, data set
will be diagnosed by checking the existence of multicollinearity issue.
By investigating, there are various factors which impacts people’s wage. The most important
factors which are identified are the skill and capability. Due to variation in skills and abilities of
labours, their wage also gets varied. Another essential factor is experience; if an individual has
valid experience then the individual acts as an asset due to which he or she is viable to get higher
wage than others (Tian and Yu-mei, 2017). Few other factors which are identified are geographic
region, industry, age of the employee, gender, personal responsibilities, hazardous working
conditions, absence of union and many more.
DESCRIPTIVE STATISTICS, GRAPHS AND CORRELATION TEST
Descriptive statistics
Statistics
The wage of the
interviewee
Age at the time of
the interview
Total years of
experience
Number of years in
the current company
Weekly
working hours
N Valid 28534 28510 28534 28101 28467
Missing 0 24 0 433 67
Mean 2036.5150 29.05 6.2153 3.12 36.56
Std. Error of
Mean 9.51723 .040 .02754 .022 .058
Median 1677.4410 28.00 5.0577 1.67 40.00
Mode 959.51 24 1.00 0 40
Std. Deviation 1607.65071 6.701 4.65212 3.751 9.870
Variance 2584540.804 44.898 21.642 14.073 97.409
Skewness 8.876 .264 .858 1.940 -.896
Std. Error of
Skewness .015 .015 .015 .015 .015
Range 77553.83 32 28.88 26 167
Minimum .00 14 .00 0 1
1

Maximum 77553.83 46 28.88 26 168
Sum 58109919.80 828076 177347.83 87783 1040741
The descriptive statistics of five variables are computed using SPSS in above table.
Description of these statistics for each variable is given below:
 Wage – The mean value for this variable is 2036 dollars which means on an average the
salary of an individual for the population is 2036 dollars. Another important descriptive
statistics is mode which is 959 that implies majority of the people earns 959 dollars.
 Age – The mean value for this variable is 29.5 which imply average age of the entire
population is 29. By analysing the minimum and maximum value it can be said that the
most least aged individual has age of 14 and most aged individual has age of 46.
 Experience – By analysing the mean value for this variable, it can be said that average
experience of entire population is 6.2 years with standard error of (Skewness) .015 years.
 Tenure – The mid point of tenure data set is 1.67 years with the range of 26 years.
 Hours – The average weekly working hours of entire population is 36.56 hours with
standard deviation of 9.870 which implies every worker in this industry works at least 36
hours ± 9 hours.
Graphs
2

From the above scatter plot, it has been analyzed that there is a positive relationship between
marriage and wage of the interviewee. Among the both marital status, it has been analyzed that
married labors are tend to have higher wage.
Creating a new variable ln(wage) and explaining the use of the function logarithm
Logarithm function is an inverse function to exponentiation that is used to solve exponential
equations. There are various types of logarithms, one of which is common logarithm which has
the base 10. Using this logarithm, a new variable is created using variable of “wage” and log 10
(Jha, 2020).
3

REGRESSION ANALYSIS AND DISCUSSION OF THE RESULTS
The regression model
This type of model is a statistical measure which helps in analysing the relationship between
a dependent variable and an independent variable (Gasso, 2019). The regression model is
considered as more effective than correlation test as regression helps to not only analyse the
relationship but also helps in identifying the nature and strength of the relationship (Cleveland,
Grosse and Shyu, 2017). In this situation Lnwage is considered as dependent variable and all
other are regarded as independent variable.
Explaining the model summary table
Model Summary
Model R R Square
Adjusted R
Square
Std. Error of the
Estimate
1 .828a .686 .686 .14925
a. Predictors: (Constant), Weekly working hours , The wage of the
interviewee, Dummy variable, if married=1 and 0 otherwise, Dummy
variable, if union member=1 and 0 otherwise, Age at the time of the
interview, Dummy variable, if college graduate=1 and 0 otherwise,
Number of years in the current company, Total years of experience
The model summary table provides two values which are “R” and “R square”. The R
value shows the simple correlation which is .828 that is considered as large linear correlation
between dependent (Lnwage) and independent variables (wage, age, msp, collgrad, union,
experience, tenure and hours). On the other hand R square value shows the total variation in
dependent variable due to independent variable which is .686 or 68% that means all independent
variables largely impacts Lnwage.
Interpreting the results for the ANOVA table
ANOVAa
Model Sum of Squares df Mean Square F Sig.
1 Regression 920.996 8 115.124 5168.363 .000b
Residual 422.242 18956 .022
Total 1343.238 18964
a. Dependent Variable: lnwage
4

b. Predictors: (Constant), Weekly working hours , The wage of the interviewee, Dummy variable, if
married=1 and 0 otherwise, Dummy variable, if union member=1 and 0 otherwise, Age at the time
of the interview, Dummy variable, if college graduate=1 and 0 otherwise, Number of years in the
current company, Total years of experience
In the Anova table, the value of significance is considered as the indicator of significant
relationship. If the significance or p value is less than 0.05 then there is a significant relationship.
As the p value in this case is .000 which less than 0.05, it is implied that all independent
variables have significant positive and strong relationship with Lnwage.
Interpreting the results for the coefficients table
Coefficientsa
Model
Unstandardized Coefficients
Standardized
Coefficients
t Sig.B Std. Error Beta
1 (Constant) 2.963 .008 355.080 .000
The wage of the interviewee .000 .000 .733 159.554 .000
Age at the time of the
interview -.004 .000 -.085 -14.815 .000
Dummy variable, if
married=1 and 0 otherwise .010 .002 .018 4.271 .000
Dummy variable, if college
graduate=1 and 0 otherwise .055 .003 .082 18.857 .000
Dummy variable, if union
member=1 and 0 otherwise .035 .003 .056 13.433 .000
Total years of experience .007 .000 .126 17.887 .000
Number of years in the
current company .003 .000 .052 9.715 .000
Weekly working hours .001 .000 .047 10.959 .000
a. Dependent Variable: lnwage
The coefficient table are used to develop regression equations. As all the significance or p values
are less than 0.05, its means all the independent values has direct relationship with Lnwage.
Regression equation for one of these independent variables is developed below:
Lnwage = 007(experience) + 2.963
5

Discussing the results
By analysing the above regression results, it has been analysed that every independent
variables has direct, positive and strong variable with dependent variable which means every
independent variables can be used to explain people’s wage level.
DIAGNOSTIC TESTS
Checking for the existence of the multicollinearity issue, explaining the methodology and
discussing the results
Coefficientsa
Model
Collinearity Statistics
Tolerance VIF
1 The wage of the interviewee .787 1.271
Age at the time of the
interview .503 1.988
Dummy variable, if
married=1 and 0 otherwise .963 1.039
Dummy variable, if college
graduate=1 and 0 otherwise .885 1.130
Dummy variable, if union
member=1 and 0 otherwise .949 1.054
Total years of experience .335 2.984
Number of years in the
current company .568 1.760
Weekly working hours .908 1.101
a. Dependent Variable: lnwage
Coefficientsa
Model
Collinearity Statistics
Tolerance VIF
1 Age at the time of the
interview .498 2.009
Dummy variable, if
married=1 and 0 otherwise
.963 1.039
6

Dummy variable, if college
graduate=1 and 0 otherwise .876 1.141
Dummy variable, if union
member=1 and 0 otherwise .940 1.064
Total years of experience .330 3.031
Number of years in the
current company .565 1.769
Weekly working hours .913 1.095
lnwage .737 1.358
a. Dependent Variable: The wage of the interviewee
Coefficientsa
Model
Collinearity Statistics
Tolerance VIF
1 Dummy variable, if
married=1 and 0 otherwise .969 1.032
Dummy variable, if college
graduate=1 and 0 otherwise .875 1.142
Dummy variable, if union
member=1 and 0 otherwise .941 1.063
Total years of experience .564 1.774
Number of years in the
current company .573 1.745
Weekly working hours .925 1.081
lnwage .318 3.145
The wage of the interviewee .336 2.976
a. Dependent Variable: Age at the time of the interview
Coefficientsa
Model
Collinearity Statistics
Tolerance VIF
1 Dummy variable, if college
graduate=1 and 0 otherwise
.869 1.151
7

Dummy variable, if union
member=1 and 0 otherwise .941 1.063
Total years of experience .330 3.031
Number of years in the
current company .565 1.769
Weekly working hours .924 1.082
lnwage .315 3.178
The wage of the interviewee .336 2.977
Age at the time of the
interview .501 1.996
a. Dependent Variable: Dummy variable, if married=1 and 0
otherwise
Coefficientsa
Model
Collinearity Statistics
Tolerance VIF
1 Dummy variable, if union
member=1 and 0 otherwise .940 1.064
Total years of experience .330 3.029
Number of years in the
current company .566 1.766
Weekly working hours .910 1.099
lnwage .320 3.123
The wage of the interviewee .339 2.954
Age at the time of the
interview .501 1.996
Dummy variable, if
married=1 and 0 otherwise .962 1.040
a. Dependent Variable: Dummy variable, if college graduate=1
and 0 otherwise
Coefficientsa
Model
Collinearity Statistics
Tolerance VIF
8

1 Total years of experience .332 3.017
Number of years in the
current company .575 1.741
Weekly working hours .905 1.105
lnwage .317 3.151
The wage of the interviewee .336 2.979
Age at the time of the
interview .497 2.010
Dummy variable, if
married=1 and 0 otherwise .963 1.039
Dummy variable, if college
graduate=1 and 0 otherwise .869 1.151
a. Dependent Variable: Dummy variable, if union member=1 and
0 otherwise
Coefficientsa
Model
Collinearity Statistics
Tolerance VIF
1 Number of years in the
current company .772 1.296
Weekly working hours .923 1.084
lnwage .320 3.128
The wage of the interviewee .336 2.976
Age at the time of the
interview .850 1.176
Dummy variable, if
married=1 and 0 otherwise .963 1.038
Dummy variable, if college
graduate=1 and 0 otherwise .870 1.149
Dummy variable, if union
member=1 and 0 otherwise .946 1.057
a. Dependent Variable: Total years of experience
Coefficientsa
9

Model
Collinearity Statistics
Tolerance VIF
1 Weekly working hours .904 1.106
lnwage .316 3.165
The wage of the interviewee .336 2.979
Age at the time of the
interview .504 1.984
Dummy variable, if
married=1 and 0 otherwise .962 1.040
Dummy variable, if college
graduate=1 and 0 otherwise .870 1.149
Dummy variable, if union
member=1 and 0 otherwise .956 1.046
Total years of experience .450 2.222
a. Dependent Variable: Number of years in the current company
Coefficientsa
Model
Collinearity Statistics
Tolerance VIF
1 lnwage .316 3.161
The wage of the interviewee .340 2.945
Age at the time of the
interview .510 1.962
Dummy variable, if
married=1 and 0 otherwise .985 1.015
Dummy variable, if college
graduate=1 and 0 otherwise .876 1.142
Dummy variable, if union
member=1 and 0 otherwise .943 1.061
Total years of experience .337 2.968
Number of years in the
current company .566 1.766
a. Dependent Variable: Weekly working hours
10