Sei sulla pagina 1di 9

Starting Compensation for College Graduates

By Adi Ekmescic
For this project we took a random sample of 50 employees salaries including retirement
contributions, insurance, etc. working in the following major fields: Business, Communications,
Computer Science, Education, Engineering, Humanities & Social Studies and Math & Science.
We are then going to take the data to analyze and compare each field to each other. Well
accomplish this by creating graphs like a Histogram and Box plot graphs to have a visual
representation of the data. The Histogram will show if the major fields follow a normal
distribution, uniform or any other type of distribution. After we analyze the histogram well
create then analyze a box plot graph. Well see of the box plots follow a normal distribution or a
t-distribution.
Histogram Graphs

Business Salaries
25
20
15
10
5
0

20000 30000 40000 50000 60000 70000 80000 90000 More

Computer Science Salaries


30
25
20
15
10
5
0

20000 30000 40000 50000 60000 70000 80000 90000 More

Engineering Salaries
25
20
15
10
5
0

20000 30000 40000 50000 60000 70000 80000 90000 More

Communication Salaries
30
25
20
15
10
5
0

20000 30000 40000 50000 60000 70000 80000 90000 More

Education Salaries
30
25
20
15
10
5
0

Humanities & Social Studies Salaries


30
25
20
15
10
5
0

Math & Science


35
30
25
20
15
10
5
0

Observations of Histogram Graphs


After analyzing the data of the Histogram graphs, it shows that the major fields of Business,
Computer science and Math & Sciences are all skewed left. While Communication, Education,
Engineering, Humanities & Social Studies are all skewed right. Only Computer Science and
Education followed a uniform distribution while the others look almost random and skewed to
the left or right side. Math & Science covers nursing under its umbrella and that is the field Im
pursuing.

Boxplot Graphs

Observations of Boxplot Graphs


After analyzing the data of the boxplot graphs. It seems that some of the major fields have
outliers when it comes to salary. Business, Computer Science, Education and Humanities &
Social studies all had some outliers in their data. While Communications, Engineering and Math
& Science didnt have any outliers. An outlier is any value that lies more than one and a half
times the length of the box from either end of the box. The outliers could be due to a new person
first starting off in the field or someone that has been in the field for a long time and has gotten
raises, bonuses, promotions, etc. The distributions are not normal since almost all of the boxplots
are skewed to the left or right.
Confidence Interval Estimates
A confidence interval is a range of values so defined that there is a specified probability that the
value of a parameter lies within it. The purpose of confidence intervals is to give us a range of
values for our estimated population parameter rather than a single value or a point estimate.
I constructed a 95% confidence interval for the mean starting compensation for students
graduating in Humanities and Social Sciences. I can say with 95% certainty that the student will

have a starting salary anywhere between a low of $36,836.19 to a high of $40,797.81. If you look
at the attached sheet, it will show my steps on how I got to the confidence interval I came to.
Hypothesis Tests
A hypothesis test is a statistical test that is used to determine whether there is enough evidence in
a sample of data to infer that a certain condition is true for the entire population. The purpose of
hypothesis testing is to make a decision in the face of uncertainty. We do not have a fool-proof
method for doing this: Errors can be made. Specifically, two kinds of errors can be made:
1. Type I Error: We decide to reject the null hypothesis when it is true.
2. Type II Error: We decide not to reject the null hypothesis when it is false.
The original claim (or null hypothesis) stated that students graduating in Education, have an
average starting salary of under $35,000. After testing the claim with 0.05 significance level, Ive
concluded that null hypothesis is to be rejected. The average starting would have to be equal to
or more than a starting salary of $35,000 in the Education field. The claim that a student with a
college degree will find a starting salary of over $40,000 is plausible. Based on the results I got
from my testing, it seems plausible that any college student with a degree can find any job that
has a starting salary of $40,000

Reflection
The conditions that need to be met for confidence intervals are the following; The sample is a
simple random sample. The population is large relative to the sample. 10n N< (N = the size of
the population) The sampling distribution of the sample proportion is approximately normal. np
10 np (1 )10. The following are the conditions that need to be met for hypothesis testing to be
possible; The sample proportion p must be obtained from a random sample. 10 np0 , where 0 p
is the assumed population proportion from H0. 1() 10 n p0 , where 0 p is the assumed
population proportion from H0. The population size is at least ten times the sample size (n). The

data I was given met all of these conditions for both Confidence Intervals and Hypothesis
Testing. One of the issues that could of happen while collecting the data is the people that
surveyed for the data. Since there were outliers they could throw off the result of confidence
intervals or even make the statistician make a Type I or Type II error with the hypothesis testing.
The sampling method can always be improved by increasing the sample size. Of course there is
the cost aspect of sampling a huge number of people but the more data there is to work with, the
more accurate your results will be.

Reflective Writing
In this math project I learned about the salaries some people make in the major job fields they
work in. I also learned how there are outliers in each of the fields when it comes to pay. I found
out this could be due to the person starting their first job out of college or that theyve been in
their field for a long time. The math skills I learned the most about in my Statistic math class was
all of the formulas and steps taken to get certain data. It will help in other math classes because I
already have practice seeing data and how it could affect the end conclusion. Following
somewhat complex formulas to achieve the desired statement or conclusion was also a big thing

Im going to use in future math classes. When we had to make a histogram and boxplot graphs
and find the 5 number summary of all the major fields. There are going to be plenty of classes
that require raw data to be organized then converted into a graph so the reader has a visual
display to go off of when reading your report or conclusion. There werent any steps on how to
do the required aspects of the project, so it forced you to do trial and error and finally the formula
fits or the results are satisfactory. This aspect of the project probably helped me the most to
develop my problem solving skills since it made you think about the data and how it fits into
everything. Before this project I knew math was in every aspect of our lives. This project
definitely supports that claim. From gathering, what seems to be random data then showing the
relations in the data on graphs and the probability.

Potrebbero piacerti anche