Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

Statistical Research Project

Understanding
Wage Characteristics
MBA (IB) 2009 – 2012 (PT)

October 7, 2009

Rajesh Kumar Garg (Roll no. 30)

Ravi Ganesh (Roll no. 33)
Saurabh Agarwala (Roll no. 44)
Siddharth Sikka (Roll no. 48)
Ujjwal Malik (Roll no. 59)

Page |1
 Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

Table of Contents

The Need for Understanding Wages – Setting the Background ......................................................................3

Data Integrity ...................................................................................................................................................3
Problem Statement ..........................................................................................................................................4
Method ............................................................................................................................................................4
Sample Characteristics .....................................................................................................................................5
Detailed Statistics.............................................................................................................................................7
Conclusions ................................................................................................................................................... 13
Recommendations ........................................................................................................................................ 13
References .................................................................................................................................................... 13

Page |2
 Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

The Need for Understanding Wages – Setting the Background

In the United States, work is often more rewarding for men than women. Men earn more – there is a gap
between wages paid to women who perform the same job as men. Additionally, the so-called marriage
tax penalizes secondary earners, usually women. Not surprisingly, women quit working at rates far
greater than those of men, especially once they have children.

Research by Alberto Alesina of Harvard University and Andrea Ichino of the University of Bologna has
shown that a solution to these gender-based labour inequities is by means of Gender Based Taxation. In
order to encourage women to work, and to even out the stubborn gender-based pay gap, these two
economists suggest taxing every woman's earnings at a significantly lesser rate than men.

Every year, the Income Tax department seeks information on wage characteristics of the population in
order to set discriminatory taxation policies. Not just gender, wage differences are common across
multiple other characteristics including years of experience, sector or nature of occupation.

This project is an attempt to develop an understanding of differences in wage characteristics across a few
variables of interest. Insights from such an analysis could come in handy to income tax policy makers or to
the various human resource consultants, who attempt to benchmark wages across sectors.

Data Integrity
The dataset that has been used comprises 534 observations on selected variables sampled from the
Current Population Survey. The Current Population Survey (CPS) is used to supplement census
information between census years. Sampled data contains information on gender, years of experience,
nature of occupation and sector of occupation in addition to wages.

Page |3
 Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

Problem Statement
Application of descriptive statistics to develop an understanding of the relationship between wage and
other characteristics of workers such as gender, years, of experience, nature of occupation and sector of
employment.

Method

1. The sample data is sorted in an ascending order on the basis of “wages”.

2. The population characteristics have been described based on the following variables:

o Gender distribution
o Sector
o Occupation
o Years of Experience (with appropriate intervals )

The results for this are highlighted in Section 1 of the “Results Section”

3. The descriptive statistics are carried out on the sample data including the 5-N summary. These consist
of :
o Minimum
o Maximum
o 1st Quartile
o 3rd Quartile
o Median
o Mode
o Standard deviation
o Mean
o Cumulative Wages
o Kurtosis
o Variance
o Range, etc.

The results for the above are highlighted in Section 2 of the “Results Section”

4. Graphical representation of data is done using Box-plots to help understand wage characteristics
across the four chosen parameters. The results for the above are highlighted in Section 3 of the
“Results Section”

5. Wage characteristics are modelled against the “Years of Experience”. The results for the above are
highlighted in Section 4 of the “Results Section”

6. Hypothesis testing is utilized to carry out 4 different tests to ascertain certain conclusions from the
data

Page |4
 Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

Sample Characteristics

Section 1: A series of the pie-charts given below have been used to describe the sample distribution
amongst the following characteristics

1. Gender : The distribution of male and females working population in the geography is 54 % and
46 % respectively

2. Sector: More than Two-thirds of the jobs in the area are concentrated in the non-
manufacturing and non-construction sectors

Page |5
Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

3. Occupation: “Other” category of options form the highest concentrated occupation followed
by “Professional” and “Sales” related occupations

4. Years of Experience: Equitable distribution of years of experience amongst different sectors.

This provides a insights into the strengths and accuracy of the sample data

Page |6
 Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

Detailed Statistics
Section 2: This section makes use of descriptive statistics to provide a view of the mean and cumulative
wage

1. Mean Wage: The graph and the table below provide a view of the descriptive statistics. The
mean wage is $9 an hour with a standard

Number of people Statistic Value

Mean 9.0
70 Mean
Standard Error 0.2
Median 7.8
60 Mode 5.0
Standard Deviation 5.1
50
Sample Variance 26.4
40 Kurtosis 5.0
Skewness 1.7
30 Range 43.5
Minimum 1.0
20 Maximum 44.5
Sum 4818.9
10 Count 534.0
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45
Wage ($ per hour)

2. Cumulative Wage: The cumulative wage graph provides the view of the wages as distributed
over the complete sample population.

Cumulative Wage Distribution

100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45
Wage ($ per hour)

Page |7
 Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

Section 3: This section makes use of box plots to plot the wages against all the different parameters
including: Gender, Sectors, Occupations and Years of Experience

1. Summary statistics for the wages across genders and sectors

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

TOTAL POPULATION

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

MALES

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

FEMALES

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

SECTOR : CONSTRUCTION

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

SECTOR : MANUFACTURING

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

SECTOR : OTHERS

2. Summary statistics for the wages across Occupations

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

TOTAL POPULATION

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

OCCUPATION: CLERICAL

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

OCCUPATION: MANAGEMENT

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

OCCUPATION : PROFESSIONALS

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

OCCUPATION : SALES

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

OCCUPATION : SERVICE

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

OCCUPATION : OTHERS

Page |8
Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

3. Summary statistics for the wages across years of experience

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

TOTAL POPULATION

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

EXP : 0-6 yrs

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

EXP : 7-12 yrs

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

EXP : 13-18 yrs

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

EXP : 19-28 yrs

-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46

EXP : 28+ yrs

4. Summary statistics for the wages of the total population- It can be surmised that majority of
the population has wages lying between $5 - $11 an hour

-10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48

Page |9
 Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

Section 4: In this section wage has been plotted against the Years of experience. A logarithmically fitted
curve provides best fit between wages and years of experience with the least average error
Wage

Experience

P a g e | 10
 Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

Section 5: In this section, hypothesis testing has been used to arrive at specific conclusions regarding the
wage behaviour of the population

TEST 1

H0: Average wage of female employees is same as overall average

H1: Average wage of female employees is not equal to overall average

Type of test z test

x bar 7.9
Mu 9.0
Sigma 4.5
N 245.0
z value -3.5

Conclusion: H0 rejected with 99% confidence. There is a difference in average wage of females from the
overall average

TEST 2

H0: Average wage of employees in manufacturing sector is same as overall average

H1: Average wage of employees in manufacturing sector is not equal to overall average

Type of test z test

x bar 9.6
mu 9.0
sigma 4.5
n 99.0
z value 1.1

Conclusion: H0 not rejected at even 90% confidence level. There is no difference in average wage of
manufacturing sector employees from the overall average

P a g e | 11
 Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

TEST 3

Test whether variance in wage of management is higher, lower or equal to that of total sample

Sample Size 55 n
Sample Variance 57.34 s2
2
Test Statistic 117.242
Null hypothesis p-value 1%
H0: σ2 = 26.41 0.0 Reject
H0: σ 2 >= 26.41 1.0
H0: σ 2 <= 26.41 0.0 Reject

Conclusion: Variance in wages of management is higher than that of total sample

TEST 4

H0: Proportion of females working in clerical / service sectors is greater than overall sample proportion by
not more than 10%

H1: Proportion of females working in clerical / service sectors is higher than overall sample proportion by
more than 10%

Type of test z test

Sample proportion 0.7
Population proportion 0.5
n 180.0
z value 3.7

Conclusion: Null hypothesis is rejected. Proportion of females is greater than 10%

P a g e | 12
 Statistical Research Project MBA (IB) 2009-2012 (PT) ¦ BUSINESS STATISTICS

Conclusions
 Sample wage distribution is positively skewed, with a mean of $9 and an IQR of $6. (Q1=$5 and
Q3 = $11).
 Average wage per hour of females is lower than that of males
 Wage difference across sectors manufacturing, construction and others is not significant
 Management wages tend to be higher and also have a higher variance
 Wages of clerical and service professions are least dispersed (IQR of $5 ($5-$10) for clerks and $4
($4-$8) for service employees).
 However, proportion of females working in clerical / service sectors is also higher than overall
sample proportion by more than 10%
 Entry level (0-6 experience) salaries are low. However, beyond that level of experience, salaries
remain more or less constant accounting for a logarithmic fit between wages and years of
experience

Recommendations
Based on the above conclusions, policy makers can design income tax policies that favour lower taxation
for females. A taxation policy based on the sector of occupation is not feasible as glaring differences are
not evident. As the variances in the management salaries are higher than most others, slabs can be
defined in income tax policies that increase the tax rate in proportion to the salary earned. Also, policies
directed at providing subsidies can be extended to entry level employees (0-6 years).

Human resource consultants can make use of the trend equation to determine appropriate level of wage
rates for years of experience that a candidate can bring to the table.

References
Data has been used from the Current Population Survey (CPS) from the Economics Web Institute -
http://www.economicswebinstitute.org

P a g e | 13