Copyright 2003 Cengage Learning 1.

1
Chapter One
What is Statistics?
Copyright 2003 Cengage Learning 1.2
What is Statistics?
'Statistics is a way to get inIormation Irom data.
Copyright 2003 Cengage Learning 1.3
What is Statistics?
'Statistics is a way to get inIormation Irom data
Data
Statistics
InIormation
DeIinitions: OxIord English Dictionary
Statistics is a tool Ior creating new understanding Irom a set oI numbers.
Copyright 2003 Cengage Learning 1.4
Example 2.6 Stats Anxiety
student enrolled in a business program is attending the Iirst
class oI the required statistics course. The student is somewhat
apprehensive because he believes the myth that the course is
diIIicult.
To alleviate his anxiety the student asks the proIessor about
last year`s marks.
The proIessor obliges and provides a list oI the Iinal marks,
which is composed oI term work plus the Iinal exam. What
inIormation can the student obtain Irom the list?
Copyright 2003 Cengage Learning 1.S
Example 2.6 Stats Anxiety
65 81 72 59
71 53 85 66
66 70 72 71
79 76 77 68
65 73 64 72
82 73 77 75
80 85 89 74
86 83 87 77
67 80 78 69
64 67 79 60
62 78 59 92
74 68 63 69
67 67 84 69
72 62 74 73
68 83 74 65
Copyright 2003 Cengage Learning 1.6
Example 2.6 Stats Anxiety
Typical mark"
Nean (average mark)
Nedian (mark such that S0 above and S0
below)
Nean = 72.67
Nedian = 72
!s this enough information?
Copyright 2003 Cengage Learning 1.7
Example 2.6 Stats Anxiety
Are most of the marks clustered around the mean or
are they more spread out?
Range = Naximum - minimum = 32S3 = 33
variance
Standard deviation
Copyright 2003 Cengage Learning 1.8
Example 2.6 Stats Anxiety
Are there many marks below 60 or above 80?
What proportion are A, B, C, D grades?
A graphical technique -histogram can provide us
with this and other information
Copyright 2003 Cengage Learning 1.3
Example 2.6 Stats Anxiety
Histogram
0
10
20
30
50 60 70 80 90 100
arks
F
r
e
q
u
e
n
c
y
Copyright 2003 Cengage Learning 1.10
Descriptive Statistics
Descriptive statistics deals with methods oI organizing,
summarizing, and presenting data in a convenient and
inIormative way.
One Iorm oI descriptive statistics uses graphical techniques,
which allow statistics practitioners to present data in ways that
make it easy Ior the reader to extract useIul inIormation.
Chapter 2 introduces several graphical methods.
Copyright 2003 Cengage Learning 1.11
Descriptive Statistics
nother Iorm oI descriptive statistics uses numerical
techniques to summarize data.
The mean and median are popular numerical techniques to
describe the location oI the data.
The range, variance, and standard deviation measure the
variability oI the data
Chapter 4 introduces several numerical statistical measures
that describe diIIerent Ieatures oI the data.
Copyright 2003 Cengage Learning 1.12
Case 12.1 Pepsi's Exclusivity Agreement
large university with a total enrollment oI about 50,000
students has oIIered Pepsi-Cola an exclusivity agreement that
would give Pepsi exclusive rights to sell its products at all
university Iacilities Ior the next year with an option Ior Iuture
years.
In return, the university would receive 35 oI the on-campus
revenues and an additional lump sum oI $200,000 per year.
Pepsi has been given 2 weeks to respond.
Copyright 2003 Cengage Learning 1.13
Case 12.1 Pepsi's Exclusivity Agreement
The market Ior soIt drinks is measured in terms oI 12-ounce
cans.
Pepsi currently sells an average oI 22,000 cans per week (over
the 40 weeks oI the year that the university operates).
The cans sell Ior an average oI 75 cents each. The costs
including labor amount to 20 cents per can.
Pepsi is unsure oI its market share but suspects it is
considerably less than 50.
Copyright 2003 Cengage Learning 1.14
Case 12.1 Pepsi's Exclusivity Agreement
quick analysis reveals that iI its current market share were
25, then, with an exclusivity agreement,
Pepsi would sell 88,000 (22,000 is 25 oI 88,000) cans per
week or 3,520,000 cans per year.
The proIit or loss can be calculated.
The only problem is that we do not know how many soIt
drinks are sold weekly at the university.
Copyright 2003 Cengage Learning 1.1S
Case 12.1 Pepsi's Exclusivity Agreement
Pepsi assigned a recent university graduate to survey the
university's students to supply the missing inIormation.
ccordingly, she organizes a survey that asks 500 students to
keep track oI the number oI soIt drinks they purchase in the
next 7 days.
The responses are stored in a Iile on the disk that accompanies
this book. Case 12.1
Copyright 2003 Cengage Learning 1.16
!nferential statistics
The inIormation we would like to acquire in Case 12.1 is an
estimate oI annual proIits Irom the exclusivity agreement. The
data are the numbers oI cans oI soIt drinks consumed in 7 days
by the 500 students in the sample.
We want to know the mean number oI soIt drinks consumed
by all 50,000 students on campus.
To accomplish this goal we need another branch oI statistics-
inferential statistics.
Copyright 2003 Cengage Learning 1.17
!nferential statistics
InIerential statistics is a body oI methods used to draw
conclusions or inIerences about characteristics oI populations
based on sample data. The population in question in this case
is the soIt drink consumption oI the university's 50,000
students. The cost oI interviewing each student would be
prohibitive and extremely time consuming. Statistical
techniques make such endeavors unnecessary. Instead, we can
sample a much smaller number oI students (the sample size is
500) and inIer Irom the data the number oI soIt drinks
consumed by all 50,000 students. We can then estimate annual
proIits Ior Pepsi.
Copyright 2003 Cengage Learning 1.18
Example 12.S
When an election Ior political oIIice takes place, the television
networks cancel regular programming and instead provide
election coverage.
When the ballots are counted the results are reported.
However, Ior important oIIices such as president or senator in
large states, the networks actively compete to see which will
be the Iirst to predict a winner.
Copyright 2003 Cengage Learning 1.13
Example 12.S
This is done through exit polls, wherein a random sample oI
voters who exit the polling booth is asked Ior whom they
voted.
From the data the sample proportion oI voters supporting the
candidates is computed.
statistical technique is applied to determine whether there is
enough evidence to inIer that the leading candidate will garner
enough votes to win.
Copyright 2003 Cengage Learning 1.20
Example 12.S
The exit poll results Irom the state oI Florida during the 2000
year elections were recorded (only the votes oI the Republican
candidate George W. Bush and the Democrat lbert Gore).
Suppose that the results (765 people who voted Ior either Bush
or Gore) were stored on a Iile on the disk. (1 Gore and 2
Bush) Xm12-05
The network analysts would like to know whether they can
conclude that George W. Bush will win the state oI Florida.
Copyright 2003 Cengage Learning 1.21
Example 12.S
Example 12.5 describes a very common application oI
statistical inIerence.
The population the television networks wanted to make
inIerences about is the approximately 5 million Floridians who
voted Ior Bush or Gore Ior president.
The sample consisted oI the 765 people randomly selected by
the polling company who voted Ior either oI the two main
candidates.
Copyright 2003 Cengage Learning 1.22
Example 12.S
The characteristic oI the population that we would like to
know is the proportion oI the total electorate that voted Ior
Bush.
SpeciIically, we would like to know whether more than 50
oI the electorate voted Ior Bush (counting only those who
voted Ior either the Republican or Democratic candidate).
Copyright 2003 Cengage Learning 1.23
Example 12.S
Because we will not ask every one oI the 5 million actual
voters Ior whom they voted, we cannot predict the outcome
with 100 certainty.
sample that is only a small Iraction oI the size oI the
population can lead to correct inIerences only a certain
percentage oI the time.
You will Iind that statistics practitioners can control that
Iraction and usually set it between 90 and 99.
Copyright 2003 Cengage Learning 1.24
Key Statistical Concepts
#6pulati6n
a 5o5ulation is the group oI all items oI interest to
a statistics practitioner.
Irequently very large; sometimes inIinite.
E.g. ll 5 million Florida voters, per Example 12.5
Sample
sam5le is a set oI data drawn Irom the
population.
Potentially very large, but less than the population.
E.g. a sample oI 765 voters exit polled on election day.
Copyright 2003 Cengage Learning 1.2S
Key Statistical Concepts
#arameter
descriptive measure oI a 5o5ulation.
Statistic
descriptive measure oI a sam5le.
Copyright 2003 Cengage Learning 1.26
Key Statistical Concepts
Populations have Parameters,
Samples have Statistics.
Parameter
Population
Sample
Statistic
Subset
Copyright 2003 Cengage Learning 1.27
Descriptive Statistics
are methods oI organizing, summarizing, and presenting
data in a convenient and inIormative way. These methods
include:
Graphical Techniques (Chapter 2), and
Numerical Techniques (Chapter 4).
The actual method used depends on what information we
would like to extract. re we interested in
W measure(s) oI central location? and/or
W measure(s) oI variability (dispersion)?
Descriptive Statistics helps to answer these questions
Copyright 2003 Cengage Learning 1.28
!nferential Statistics
Descriptive Statistics describe the data set that`s being
analyzed, but doesn`t allow us to draw any conclusions or
make any interIerences about the data. Hence we need
another branch oI statistics: inferential statistics.
InIerential statistics is also a set oI methods, but it is used to
draw conclusions or inIerences about characteristics oI
5o5ulations based on data Irom a sam5le.
Copyright 2003 Cengage Learning 1.23
Statistical !nference
$tatistical inference is the 5rocess oI making an estimate,
prediction, or decision about a population based on a sample.
Parameter
Population
Sample
Statistic
!nference
What can we infer about a Population`s Parameters
based on a Sample`s Statistics?
Copyright 2003 Cengage Learning 1.30
Statistical !nference
We use statistics to make inIerences about parameters.
ThereIore, we can make an estimate, prediction, or decision
about a population based on sample data.
Thus, we can apply what we know about a sample to the
larger population Irom which it was drawn!
Copyright 2003 Cengage Learning 1.31
Statistical !nference
#ationale:
W Large populations make investigating each member impractical
and expensive.
W Easier and cheaper to take a sample and make estimates about the
population Irom the sample.
owever:
Such conclusions and estimates are not always going to be correct.
For this reason, we build into the statistical inIerence 'measures oI
reliability, namely confidence level and significance level.
Copyright 2003 Cengage Learning 1.32
Confidence S Significance Levels
The confidence level is the proportion oI times that an
estimating procedure will be correct.
E.g. a conIidence level oI 95 means that, estimates based on this
Iorm oI statistical inIerence will be correct 95 oI the time.
When the purpose oI the statistical inIerence is to draw a
conclusion about a population, the significance level
measures how Irequently the conclusion will be wrong in the
long run.
E.g. a 5 signiIicance level means that, in the long run, this type oI
conclusion will be wrong 5 oI the time.
Copyright 2003 Cengage Learning 1.33
Confidence S Significance Levels
II we use u (Greek letter 'alpha) to represent signiIicance,
then our conIidence level is 1 - u.
This relationship can also be stated as:
ConIidence Level
SigniIicance Level
1
Copyright 2003 Cengage Learning 1.34
Confidence S Significance Levels
Consider a statement Irom polling data you may hear about
in the news:
This poll is considered accurate within 3.4
percentage points, 19 times out of 20.
In this case, our conIidence level is 95 (19/20 0.95),
while our signiIicance level is 5.
Copyright 2003 Cengage Learning 1.3S
Statistical Applications in Business
Statistical analysis plays an important role in virtually all
aspects oI business and economics.
Throughout this course, we will see applications oI statistics
in accounting, economics, Iinance, human resources
management, marketing, and operations management.