Descriptive Statistics Part I: Numerical Description

1. DESCRIPTIVE STATISTICS
Part I: Numerical Description
In this chapter, we will learn how to describe a set of data using numerical methods. This is the
first of two chapters that together will aim at providing methods of descriptive statistics. In
descriptive statistics, which is the use of graphical methods to display data and explore key
statistics.
1

2. What are the basic features of a data set?
A data set is a collection of data representing a particular variable.
Examples of data sets are given below.
Data Sets:
• Students’ grades in a calculus test:
65, 85, 70, 75, 85, 80, 82, 85, 90, 78, 81, 82, 67, 80
• Property tax of a sample of houses:
$5000, $4500, $4000, $7200, $5000, $3800, $4100, $5000
• Driving distance to work of a group of employee (miles):
1.2, 2.0, 2.2, 15.0, 11.0, 5.0, 3.7, 4.9, 15.2, 16.0
• Ages of all students in a college:
18, 19, 21, ……………………..…, 22, 18, 19, 21
2Notes:
…………………………………………………………………………………………………………………………………
………………………………………………………………………………………………………………………………….

3. In general, establishing a data set requires consideration of a number of key questions:
Notes:
…………………………………………………………………………………………………………………………………
………………………………………………………………………………………………………………………………….
3
Data Set Key Questions:
•Are the data qualitative or quantitative?
•What levels of measurement do the data exhibit? (nominal, ordinal, interval, or
ratio)
•What is the source of data?(the population)
•What is the appropriate sampling technique that should be used to collect the
samples? (random or stratified)
•What is the appropriate minimum sample size?

4. 4
Data Set Types: (1) Univariate, (2) Bivariate, (3) Multivariate
Data Set Variable Typical Tasks
Univariate One Histograms, Descriptive Statistics, Frequency tallies
Bivariate Two Scatter plots, correlations, simple regression
Multivariate
More than two
variables Multiple regression, data mining, modeling
Person # Weight (lb)
1 150
2 120
3 130
4 125
5 155
6 134
7 150
8 140
9 160
10 200
11 180
12 140
Person
#
Years at
work
Annual Salary
($)
1 5 50,000
2 20 73,000
3 10 65,000
4 5 55,000
5 8 60,000
6 10 60,000
7 15 68,000
8 15 69,000
9 20 68,000
10 20 69,000
11 18 68,000
12 10 62,000
13 3 48,000
UnivariateDataSet
BivariateDataSet
Case Name Age
Income
($) Position Gender
1Frieda 45 67,100 Consumer Analyst F
2Stefan 32 56,500
Operations
analyst M
3John 55 88,200 Marketing VP F
4Donna 27 59,000 Statistician F
5Larry 46 26,000 Security guard M
6Alicia 52 68,500 QC Director F
7Alec 65 95,200 Chief executive M
8Jaime 50 71,200
Human
Resources M
Multivariate Data Set
Notes:
…………………………………………………………………………………………………………………………………
………………………………………………………………………………………………………………………………….

5. Time-series data set
5
Cross sectional Sample
Notes:
…………………………………………………………………………………………………………………………………
………………………………………………………………………………………………………………………………….
Data Sets in the Context of Sampling:
• Cross sectional data set
• Time-series data set

6. 6
Notes:
…………………………………………………………………………………………………………………………………
………………………………………………………………………………………………………………………………….

7. 7
Working Problem 2.2: Explain what is inheritance tax. What is the difference between inheritance tax and Estate tax?
What is the level of measurement for each of the following variables: State, Income tax, sales tax, and inheritance
tax. Why do some states have a wide income tax range?
http://portal.kiplinger.com/tools/slideshows/slideshow_pop.html?nm=TaxUnfriendlyStatesRetirees
State Income
Tax (%)
U.S. States
Sales Tax
(%)
Inheritance
Tax (%)
Alaska 0.0 0.0 NO
Wyoming 0.0 4.0 No
Michigan 4.4 6.0 No
Pennsylvania 3.1 6.0 YES
Colorado 4.6 2.9 NO
Delaware 4.6 0.0 NO
Hawaii 1.4 to 11 4.0 NO
Georgia 1.0 to 6.0 4.0 NO
South Carolina 3.0 to 7.0 6.0 NO
Alabama 2.0 to 5.0 4.0 NO
California 1.25 to 10.55 8.3 NO
Rhode Island 3.75-9.9 7.0 NO
New Jersey 1.4 to 8.97 7.0 YES
Vermont 3.55-8.95 6.0 NO
Iowa 0.36 to 8.98 6.0 YES
Nebraska 2.56 to 6.84 5.5 Yes
Wisconsin 4.6 to 7.75 5.0 NO
Oregon 5.0 to 11.0 0.0 YES
Indiana 3.4 7.0 YES
North Dakota 1.84-4.86 5.0 NO

8. 8
Working Problem 2.3:
Identify the following data sets as ‘Cross-Sectional Data’ or ‘Time-Series Data’:
(a) Two weeks before the 56th quadrennial United States presidential election, which was held on November 4, 2008, a sample of
people taking randomly from undecided states revealed that Democrat Barack Obama is expected to earn 54% of the
popular votes and John McCain is expected to earn 46% of the votes
Cross Sectional ( ) Time-Series ( )
(b) A survey of 1000 students from a university of 10,000 students, revealed that 65% of the students do not prefer weekend
classes
Cross Sectional ( ) Time-Series ( )
(c) The U.S. City average price per gallon of unleaded regular gasoline from 2000 to 2009 was as follow:
Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2000 1.301 1.369 1.541 1.506 1.498 1.617 1.593 1.51 1.582 1.559 1.555 1.489
2001 1.472 1.484 1.447 1.564 1.729 1.64 1.482 1.427 1.531 1.362 1.263 1.131
2002 1.139 1.13 1.241 1.407 1.421 1.404 1.412 1.423 1.422 1.449 1.448 1.394
2003 1.473 1.641 1.748 1.659 1.542 1.514 1.524 1.628 1.728 1.603 1.535 1.494
2004 1.592 1.672 1.766 1.833 2.009 2.041 1.939 1.898 1.891 2.029 2.01 1.882
2005 1.823 1.918 2.065 2.283 2.216 2.176 2.316 2.506 2.927 2.785 2.343 2.186
2006 2.315 2.31 2.401 2.757 2.947 2.917 2.999 2.985 2.589 2.272 2.241 2.334
2007 2.274 2.285 2.592 2.86 3.13 3.052 2.961 2.782 2.789 2.793 3.069 3.02
2008 3.047 3.033 3.258 3.441 3.764 4.065 4.09 3.786 3.698 3.173 2.151 1.689
2009 1.787 1.928 1.949 2.056 2.265 2.631 2.543 2.627 2.574 2.561 2.66 2.621
http://data.bls.gov/cgi-bin/surveymost
Cross Sectional ( ) Time-Series ( )

9. 9
Notes:
…………………………………………………………………………………………………………………………………
………………………………………………………………………………………………………………………………….
What are the different elements of descriptive statistics?
Two types of descriptive statistics:
(1) Numerical measures of data, and
(2) Graphical displays of data.
The Focus of this Chapter

10. 10
Numerical measures of descriptive statistics consist of two types of measures:
• Measures of central tendency (mean, median, and mode)
• Measures of dispersion (range, standard deviation, and variance)
• Combined measures (coefficient of variation, signal-to-noise ratio, and standardized variable)
Measures of Central Tendency
10.0 12.0 13.0 14.0 14.0 14.0 24.0 24.0 24.0 26.8 27.0 27.0 29.0 30
Mean
Mode
Median
Measures of Dispersion
Range
Standard Deviation
Variance

11. 11
Key points to perform good statistical analysis
1. Identify your objectives:
 What questions do you really need to answer?
 What variable do you need to examine?
 What population you are about to evaluate?
2. Collect the appropriate samples and data to address your questions:
Do you have access to the entire population?
 Would a selection of a sample from the population be easier to access, less costly,
and less destructive than an evaluation of the whole population?
 Remember ‘GIGO’ or garbage-in, garbage-out. If the samples are not representative
of the population, and the data collected is not accurate and precise, the conclusions
drawn from the analysis will be meaningless.
3. Describe the data using the analysis of descriptive statistics :
 Do you detect data abnormality or outliers?
 Can you explore the data in such a way that will provide a clear description of data
center and data variability?
 Use descriptive statistics as a guideline for other methods of analysis
4. Perform inference :
 Can the sample statistics be used to estimate population parameters?
 Is your estimation of population parameters reliable?
 Do you have confidence in the population estimates?

12. Center Values
Measures of Central Tendency
Mean
Mode
Median
12
What are the measures of central tendency?
10.0 12.0 13.0 14.0 14.0 14.0 24.0 24.0 24.0 26.8 27.0 27.0 29.0 30

13. (1) Arithmetic Mean
Measures of Data Center
(Central Tendency)
Arithmetic Mean of Sample Observations
Arithmetic Mean of Population Observations
13

14. Example: The Table below illustrates a comparison of gas prices in some States in
September 2009 and September 2008. Determine the mean of gas prices ($ per gallon)
for each year.
State Sept- 2009 Sep-2008
California 3.099 3.75
Colorado 2.48 3.732
Florida 2.527 3.893
Massachusetts 2.597 3.582
Minnesota 2.452 3.765
New York 2.811 3.805
Ohio 2.411 3.933
Texas 2.404 3.729
Washington 2.947 3.785
Gas Prices of a number of states in September 2008, and September 2009
http://www.eia.doe.gov/oil_gas/petroleum/data_publications/wrgp/mogas_home_page.html
gallon
n
x
XMean
n
i
/636.2$
9
947.2........597.2527.248.2099.31 



gallon
n
x
XMean
n
i
/775.3$
9
785.3........582.3893.3732.375.31 



For September 2009:
For September 2008:
Comment on the Results
14

15. 15
Properties of arithmetic mean:
1. The mean of a set of data is unique and can be used as an identity
measure of the data center
2. We can determine the mean of any data set that contains ratio or
interval level data
3. We need all observation values to be able to calculate the mean
4. You know it is the correct mean value when the sum of the deviations
of each value from it is zero,

16. 16
Example: Determine the arithmetic mean of the three values of student grades:
80, 40, and 30. Using the mean value, prove that
.
Solution:
The arithmetic mean:

17. 17
Working Problem 2.4:
Calculate the mean for the following data set of minimum wage ($):
7, 8, 6, 6, 8, 5, 6, 5, 8, 8

18. 18
2. Median:
The median of a set of numbers arranged in order of magnitude is the middle value or the
arithmetic mean of the two middle values.
Example: Calculate the median of the following data set:
14, 12, 14, 16, 15, 19, 17, 17, 17
Solution:
To determine the median, we first arrange the data in order of magnitude:
12, 14, 14, 15, 16, 17, 17, 17, 19
Thus, the median is 16
Example: Calculate the median of the following data set:
8, 9, 10, 9, 8, 6, 11, 7, 12, 8
Solution:
To determine the median, we first arrange the data in order of magnitude
6, 7, 8, 8, 8, 9, 9, 10, 11, 12
Since this data set consists of an even number of observations, the middle values that split this
data into equal number of observations on both sides are 8 and 9. Thus, the median of this set of
data is (8+9)/2 = 8.5

19. 19
3. Mode:
The mode is that value which occurs with the greatest frequency. Interestingly, mode is a
French word that means fashion; perhaps, it is popular and common fashion.
Example: Calculate the mode of the following observations:
80, 87, 90, 82, 78, 74, 80, 77, 80, 91, 81, 80
Example: Calculate the mode of the following observations:
5, 7, 8, 9, 9, 9, 10, 11, 12, 14, 14, 14, 15
Solution:
The mode of this set is 80
Solution:
This set exhibits two modes 9, and 14, and is called bimodal.

20. Working Problem:
Calculate the mean and the mode and the median for the following data set
of minimum wage ($)
7, 8, 6, 6, 8, 5, 6, 5, 8, 8
20
Answer:
Mean = $6.7
Median =$6.5
Mode = $8

21. 21
Working Problem 2.6:
Calculate the median and the mode for the following data set of
minimum wage ($):
7, 8, 6, 6, 8, 5, 6, 5, 8, 8

22. Geometric Mean, G
n
nxxxxG ......321
Example: If the return on investment earned by a manufacturer of a
sport car for four successive years was: 20 percent, 15 percent, -40
percent, and 100 percent. What is the geometric mean rate of return on
investment?
1344.1656.1)2)(6.0)(15.1)(2.1(...... 44
321  n
nxxxxG
Accordingly, the average rate of return, which is essentially a compound
annual growth rate, is 13.44%.
22

23. 23
Example: Suppose the inflation rates for the last 5 years in a certain country are 5%, 4%, 2%, 8%,
and 6%, respectively. What is the mean rate of inflation over this five-year period?
Accordingly, the average rate of inflation over the five-year period is 4.9%
Geometric Mean, G
Solution:
At the end of the first year, the price index will be 1.05 times the price index at the beginning of
the year; at the end of the second year, the price index will be (1.04)(1.05); at the end of the
third year, the price index will be (1.02)(1.04)(1.05) and so on. Thus, the mean of 1.05, 1.04,
1.02, 1.08, and 1.06 is:

24. 24
Working Problem 2.8:
The percent increase in sales for the last 4 years at X-L Company were:
9.91, 10.75, 13.12, 26.6
(a) Find the geometric mean percent increase.
(b) Find the arithmetic mean percent increase.
(c) Is the arithmetic mean equal to or greater than the geometric mean?

25. 25
What are the ‘dispersion’ or variability measures?
 Range
 Mean deviation
 Standard deviation
 Variance
10.0 12.0 13.0 14.0 14.0 14.0 24.0 24.0 24.0 26.8 27.0 27.0 29.0 30
Measures of Dispersion
Range
Standard Deviation
Variance

26. What are ‘Dispersion’ or Variability measures?
 Range
 Mean deviation
 Standard deviation
 Variance
minmax XXR 
Example: Calculate the range of the following set of data:
200, 205, 204, 202, 207, 208
26
The Range = R = 208 - 200 = 8

27. 27
Properties of range:
• The range represents the most commonly used statistic after the arithmetic
mean
• It is simple as it relies on two values, the maximum value and the minimum
value
• It is easy to understand: the higher the range, the higher the variability
• Since the range relies on two values (maximum and minimum), a mistake in
any one of these two values or a presence of an outlier can result in a
misleading value of range
minmax XXR 

28. What are ‘Dispersion’ or Variability measures?
Mean deviation



n
i Xx
n
MD
1
1
28
Example : Calculate the mean deviation of the
following ten observations of metal sheet thickness (mm):
83, 90, 70, 90, 90, 60, 70, 70, 90, 100
Solution: Step 1: Calculate the Mean
X

= (83 + 90 + 70 + 90 + 90 + 60 + 70 + 70 + 90 + 100) / 10 = 81.3 mm
Step 2: Subtract each observation value from the Mean value, and add up the absolute
differences
Thickness (mm)
83 (83-81.3) =1.7 1.7
90 (90-81.3) = 8.7 8.7
70 (70-81.3) = -11.3 11.3
90 (90-81.3) = 8.7 8.7
90 (90-81.3) = 8.7 8.7
60 (60-81.3) = -21.3 21.3
70 (70-81.3) = -11.3 11.3
70 (70-81.3) = -11.3 11.3
90 (90-81.3) = 8.7 8.7
100 (100-81.3) = 18.7 18.7
Mean = = 81.3 Sum = 110.4
)(

 Xx |)(|

 Xx

X
Mean
Deviation =
110.4/10 =
11.04 mm

29. What are ‘Dispersion’ or Variability measures?
 Range
 Mean deviation
 Standard deviation
 Variance
For a Population:
 



N
i
N
x
1
2


For n < 30, we use (n-1) in the denominator
For a Sample:




n
i
n
Xx
s
1
2
)(
29

30. Standard deviation
Thickness (mm) 83 90 70 90 90 60 70 70 90 100
X

= (83 + 90 + 70 + 90 + 90 + 60 + 70 + 70 + 90 + 100) / 10 = 81.3 mm
For n < 30, we use (n-1) in the denominator
Example: Calculate the standard deviation of the following ten observations of metal sheet
thickness




n
i
n
Xx
s
1
2
)(
Thickness (mm)
83 (83-81.3) =1.7 2.89
90 (90-81.3) = 8.7 75.69
70 (70-81.3) = -11.3 127.69
90 (90-81.3) = 8.7 75.69
90 (90-81.3) = 8.7 75.69
60 (60-81.3) = -21.3 453.69
70 (70-81.3) = -11.3 127.69
70 (70-81.3) = -11.3 127.69
90 (90-81.3) = 8.7 75.69
100 (100-81.3) = 18.7349.69
Mean = 81.3 Sum = 1492.1
)(

 Xx 2
)(

 Xx
mms 88.12
9
1.1492

 



n
i
n
Xx
s
1
2
1
)(
30

31. What are ‘Dispersion’ or Variability measures?
 Range
 Mean deviation
 Standard deviation
 Variance
For a Population:
 



N
i
N
x
1
2
2 

For n < 30, we use (n-1) in the denominator
For a Sample:




n
i
n
Xx
s
1
2
2 )(
31

32. 32
Properties of variance:
• The variance represents the most commonly used statistic to indicate variability
• It is easy to understand: the higher the variance, the higher the variability
• Unlike the range, the variance takes into account all values of the observation
values. Therefore, it is largely insensitive to outliers
• Variance values cannot be subtracted to determine variability. It can only be added.
If U = X ± Y, Var (U) = Var (X) + Var (Y). This is the principle of analysis of variance
(Chapter 11)

33. What are ‘Dispersion’ or Variability measures?
Variance
Thickness (mm) 83 90 70 90 90 60 70 70 90 100
X

= (83 + 90 + 70 + 90 + 90 + 60 + 70 + 70 + 90 + 100) / 10 = 81.3 mm
For n < 30, we use (n-1) in the denominator
Example: Calculate the varianceof the following ten observations of metal sheet thickness
Thickness (mm)
83 (83-81.3) =1.7 2.89
90 (90-81.3) = 8.7 75.69
70 (70-81.3) = -11.3 127.69
90 (90-81.3) = 8.7 75.69
90 (90-81.3) = 8.7 75.69
60 (60-81.3) = -21.3 453.69
70 (70-81.3) = -11.3 127.69
70 (70-81.3) = -11.3 127.69
90 (90-81.3) = 8.7 75.69
100 (100-81.3) = 18.7349.69
Mean = 81.3 Sum = 1492.1
)(

 Xx 2
)(

 Xx
22
79.165
9
1.1492
mms 
33




n
i
n
Xx
s
1
2
2 )(




n
i
n
Xx
s
1
2
2 )(

34. Working Problem:
Calculate the minimum, maximum, range, standard deviation, and variance for the
following data set of minimum wage ($)
7, 8, 6, 6, 8, 5, 6, 5, 8, 8
34
Answer:
Minimum = $5
Maximum =$8
Range = $3
Standard deviation = $1.252
Variance = 1.567

35. 35
Example: Suppose X and Y are independent random variables. The variance of X is equal to 16;
and the variance of Y is equal to 9. Let U = X - Y.
What is the standard deviation of U?
•2.65 ……….
•5.00 ……….
•7.00 ……….
•25.0 ……….
•None of the above ……….

36. 36
Working Problem 2.9:
Question (1): Calculate the minimum, the maximum, the range, the mean deviation, the
standard deviation, and the variance for the following data set of minimum wage ($)
7, 8, 6, 6, 8, 5, 6, 5, 8, 8
Question (2): In two consecutive exams, the mean grade of the first test was 80 and the mean
grade of the second test was 90. The standard deviation of grade of the first test was 6 and
the standard deviation of grade of the second test was 8. Calculate the mean of the two tests
and the variance of the two tests?

37. 37

38. What are Combined Descriptive Measures?
Coefficient of Variation (C.V%)
100%.


X
s
VC
Thickness (mm) 83 90 70 90 90 60 70 70 90 100
X

= (83 + 90 + 70 + 90 + 90 + 60 + 70 + 70 + 90 + 100) / 10 = 81.3 mm
Example: Calculate the Coefficient of Variation of the following ten observations of metal
sheet thickness
Thickness (mm)
83 (83-81.3) =1.7 2.89
90 (90-81.3) = 8.7 75.69
70 (70-81.3) = -11.3 127.69
90 (90-81.3) = 8.7 75.69
90 (90-81.3) = 8.7 75.69
60 (60-81.3) = -21.3 453.69
70 (70-81.3) = -11.3 127.69
70 (70-81.3) = -11.3 127.69
90 (90-81.3) = 8.7 75.69
100 (100-81.3) = 18.7349.69
Mean = 81.3 Sum = 1492.1
)(

 Xx 2
)(

 Xx
mms 88.12
9
1.1492

%84.15100
3.81
88.12
100%.



X
s
VC
38

39. 39
Working Problem 2.11:
Calculate the Coefficient of Variation (CV%) for the following data set of
minimum wage ($):
7, 8, 6, 6, 8, 5, 6, 5, 8, 8

40. What are Combined Descriptive Measures?
Standardized Variable (the z Score)
A standardized variable is a measure of the deviation from the mean by an
individual value in units of the standard deviation:
40
Example: An instructor who has been teaching statistics for twenty years has
observed that the average grade of students is 88% and the standard deviation is 3%.
After teaching the course for two classes, one in the fall semester and one in the
spring semester of 2008, the instructor found that the average grades were as follow:
Term Mean Grade
Fall 2008 82%
Spring 2008 91%
How do these two semesters compare to the instructor’s average over the last twenty
years?

41. 41
Standardized Variable (the z Score)
Example: An instructor who has been teaching statistics for twenty years has observed that the
average grade of students is 88% and the standard deviation is 3%. After teaching the course for
two classes, one in the fall semester and one in the spring semester of 2008, the instructor found
that the average grades were as follow:
Term Mean Grade
Fall 2008 82%
Spring 2008 91%
How do these two semesters compare to the instructor’s average over the last twenty
years?
The standardized variable (z- score) is calculated for each semester as follows:
Term Mean Grade z-Score
Fall 2008 82% z82 = (82-88)/3 = -2
Spring 2008 91% z91 =(91-88)/3 = 1
From the above scores, you can conclude that the class’s grade in the Fall 2008 being 82% was 2
standard deviations below the teacher’s mean grade, while the class’s grade in the Spring 2008
being 91% was 1 standard deviations above the teacher mean grade.

42. 42
Example: The mean driving time of people living in Union City near Atlanta Georgia to CNN Center
in downtown Atlanta is 40 minutes, with a standard deviation of 10 minutes. You asked four CNN
employees who live in Union City about their driving time to CNN Center, and you get the following
answers: 38 minutes, 52 minutes, 58 minutes, and 40 minutes. Find the z-score that
corresponds to each driving time. Interpret the difference in z-scores?
Where t is the actual driving time, t is the mean driving time, and t is the standard deviation of
driving time.
At t = 38 minutes,
At t = 52 minutes,
At t = 58 minutes,
At t = 40 minutes,

43. 43
Working Problem 2.12:
The average scoring points per game (PTG) up to week 10 in the 2010 NFL football
season was 22 points and the standard deviation was 4 points. Using the z-score, compare
the following 3 teams and determine which team had a relatively better scoring season:
San Francisco 16 PTG, New England 29 PTG, Pittsburgh 24 PTG

44. 44
Working Problem 2.13:
The annual salaries of engineers in the U.S. automobile industry are normally distributed
with a mean of $100,000 and a standard deviation of $10,000. What is the z-score for the
income x of an auto-engineer who earns $85,000 annually? And what is the z-score for an
auto-engineer who earns $105,000 annually?

45. 45
Working Problem 2.14:
The annual salaries of U.S. state governors are normally distributed with a mean of $135,450 and a
standard deviation of $36,530. If in 2007, the Arkansas governor made $85,000 annual salary, and
the California governor made $206,000.
Compare the annual salaries of these two governors using the z-score.
Arnold Schwarzenegger Mike Beebe
(California) (Arkansas)

46. The Use of Computer for Performing Descriptive Statistics
Powerful Tools are available to perform statistical analyses, the focus
should therefore be on:
• Planning for sample and data selection in view of the study
or application objectives
• Gathering and organizing data in such a way that serves the purpose
of the application
• Selecting the appropriate type of analysis
• Organizing the analysis output
• Interpreting the analysis outcome
• Making a report addressing the case or application
in question
46

47. Data on Annual Tuition and Financial Aid by Different U.S. State Colleges
(http://www.ordoludus.com/costs.php, 2006)
School
In-State Out-of-State
Tuition
Total
Cost ($)
Fin.
Tuition Aid ($)
Georgia Institute of Technology $4,648 $18,990 $25,792 $8,222
University of Tennessee $5,290 $16,060 $21,270 $6,954
University of Mississippi $4,320 $9,744 $14,442 $7,532
University of Kentucky $5,812 $12,798 $18,027 $7,861
Louisiana State University $4,515 $12,815 $19,145 $8,006
University of Florida $3,094 $16,579 $22,839 $10,566
University of Virginia $7,133 $23,877 $30,266 $13,449
University of South Carolina $7,314 $18,956 $25,039 $9,501
University of North Carolina $4,515 $18,313 $24,903 $9,687
University of Georgia $4,628 $16,848 $23,224 $7,320
University of Alabama $4,864 $13,516 $18,540 $7,980
University of California (UCLA) $6,504 $24,324 $36,252 $13,462
North Dakota State University $5,264 $12,545 $17,675 $5,487
Florida State University $3,208 $16,340 $23,118 $8,269
The Use of Computer for Performing Descriptive Statistics
Example:
47

48. 1
2
Analysis of Descriptive Statistics: Steps 1 and 2 48

49. 3
Analysis of Descriptive Statistics: Steps 3 and 4
4
49

50. 5
Analysis of Descriptive Statistics: Steps 5 and 6
6
The minimum of the
largest 4 observationsThe maximum of the
smallest 4 observations
50

51. Analysis of Descriptive Statistics: Output 51

52. 52
The most critical aspect of statistics is to learn how to interpret
the results… This is not your typical Math course where all you
have to do is find answers…The true answer is not the outputs..it is
the interpretation of the outputs
Statistic
In-State
Tuition ($)
Out-State
Tuition ($) Total Cost ($)
Financial Aid
($)
Mean 5079 16550 22895 8878
Median 4756 16460 22979 8114
Mode 4515 None None None
Standard
Deviation 1269.44 4196.51 5602.14 2297.84
Sample
Variance 1611486.64 17610692.25 31383983.67 5280083.14
Range 4220 14580 21810 7975
Minimum 3094 9744 14442 5487
Maximum 7314 24324 36252 13462
Count 14 14 14 14
Largest(4) 5812 18956 25039 9687
Smallest(4) 4515 12815 18540 7532
Outputs of descriptive statistics for tuition, cost, and financial aid

53. 53
$0
$5,000
$10,000
$15,000
$20,000
$25,000
$30,000
$35,000
$40,000
NorthDakotaStateUniversity
UniversityofTennessee
UniversityofGeorgia
UniversityofMississippi
UniversityofKentucky
UniversityofAlabama
LouisianaStateUniversity
GeorgiaInstituteofTechnology
FloridaStateUniversity
UniversityofSouthCarolina
UniversityofNorthCarolina
UniversityofFlorida
UniversityofVirginia
UniversityofCalifornia(UCLA)
TotalCost($)andFinancialAid($)
School
Total Cost
Financial Aid
Total Cost and Financial Aid by School
Optimum
Choice

54. 54

55. APPENDIX 2.A Steps to Add Data Analysis to Excel 2007
55

56. 1
2
Data Analysis Add-In-Steps 1 and 2
56

57. 3
4
5
Data Analysis Add-In-Steps 3 through 5 57

58. Data Analysis Add-In-Steps 6 and 7
6
7
58

59. 8
9
Data Analysis Add-In-Steps 8 and 9
59