What%20is%20Meant%20by%20Statistics?

1
What is Meant by Statistics?
Statistics is the science of collecting,
organizing, presenting, analyzing, and
interpreting numerical data to assist in making
more effective decisions.
COPAID
2
Origin
John Graunt published an article in 1662. He
studied the obituary sections in a weekly London
church publication. They contained information on
the cause of death. Based on the sample data
collected in one locality, he reached broad
conclusions about the impact of diseases in the
general population. His analysis and
interpretation of data gave birth to the field of
Statistics.
3
Why study Statistics?
A. Numerical data is everywhere (you can create
your own instruments to measure data) B.
Statistics is used in making decision that affect
our lives (traffic improvements) C. You will
find yourself faced with decisions where an
understanding of data analysis is helpful say,
you want to open a new car dealership In order
to make an informed decision, you must know how
to 1. Determine whether the existing
information is adequate or additional
information is required. 2. Gather additional
information, if it is needed, in such a way that
it leads to good results. 3. Analyze and draw
inferences/conclusions. 4. Summarize the
information in an informative manner. Collecting
and selling data is a business by itself eg.
Credit, Priceline, Dow Jones, StandardPoor,
4
Example Applications of Statistics

• Recommending which stocks to buy sell
• Quality of production
• Economic data to predict future trends
• Law enforcement
• Evaluating a new drug

5
Types of Statistics
Descriptive Statistics
Inferential Statistics
6
Descriptive Statistics deals with methods of
Organizing, Summarizing, Presenting data
in an informative way.
Typically, descriptive statistics include Mean,
Mode, Median, Variance, Deviation,
Skewness, Charts histogram, bar, pie
(we will see these in later chapters)
past/current data but not estimated future data
7
Excel Example Output of Descriptive Statistics
8
Inferential Statistics methods used to
determine something about a population on the
basis of a sample.
Population all possible individuals, objects.
Sample part of the population of interest
9
Inferential Statistics
Example 2 The accounting department of a large
firm will select a sample of the invoices to
check for accuracy for all the invoices of the
company. (Sarbanes-Oxley Act)
Example 1 TV networks constantly monitor the
popularity of their programs by hiring Nielsen
and other organizations to sample the preferences
of TV viewers.
10
Why sample?

• Time cost are prohibitive
• Physical impossibility of checking all items in
  population
• (eg. Checking quality of product if they are made
  in the millions)
• Destructive nature of some tests
• Sample results are adequate for decision-making

11
Types of Variables
12
Types of Variables
For a Qualitative or Attribute Variable the
characteristic being studied is non-numeric. It
can only be labeled. (sometimes also called
Categorical variable)
13
Types of Variables
In a Quantitative Variable information is
reported numerically.
Balance in your checking account
Minutes remaining in class
Number of children in a family
14
Types of Quantitative Variables
Quantitative variables can be classified as
either Discrete or Continuous.
Discrete Variables can only assume certain
values -there are usually gaps between
values - usually counted Example the number
of bedrooms in a house, or the number of hammers
sold at the local Home Depot (1,2,3,,etc).
15
A Continuous Variable can assume any value within
a specified range (no gaps).
The pressure in a tire
The height or weight of students in a class.
16
Chart to remember
17
Levels of Measurement

• Nominal
• Ordinal
• Interval
• Ratio

The level of measurement dictates the kind of
calculations you can do on the data. Eg. If
one students major is Accounting and anothers
IS, we cannot calculate the average major. On the
other hand, we can average their heights,
weights, etc.
18
Nominal level Data that is classified into
categories. Can be arranged in any order.
Measurement consists only of counts.
19
In this example, Country or Region is Nominal
Level data
Other examples religion, major, gender,
ethnicity,
20
Nominal level data must be
Mutually exclusive An individual, object, or
measurement is included in only one category.
Exhaustive Each individual, object, or
measurement must appear in one of the categories.
21
Ordinal level - involves data arranged in some
order - magnitude of differences between data
values cannot be determined.
During a taste test of 4 soft drinks, Coca Cola
was ranked number 1, Dr. Pepper number 2, Pepsi
number 3, and Root Beer number 4.
22
Example of an Ordinal level variable
Also, see the example table in page 12 (Homeland
Security Advisory System)
23
Interval level - similar to the ordinal level -
amounts of differences between data values is of
equal size - there is no natural zero point.

• Eg.
• Temperature on the Fahrenheit scale.
• Difference between 10F - 15F is same as between
  50F - 55F
• 0 does not represent absence of temperature

24
Ratio level (highest level of measurement) -
zero value means absence - differences and
ratios are meaningful for this level of
measurement (A person with 2Million is twice as
rich as another with 1Million)
Monthly income 0 means did not make any money
Traveled 0 miles means did not travel at all
25
( Nerd Of India Rocks! )
N-O-I-R
26
Ethical Considerations

• Statistics can be used to mislead decision makers
  
• Dont do it!
• Keep taking different samples until you get the
  result you want
• Quote average to hide wide range of data values
• Misleading graphical outputs
• Make unwarranted conclusions on variable
  relationships

27
Ethical Considerations
The cost/year doubled in 5 years. But the graph
appears to depict more than that.
28
Ethical Considerations
By changing the x-y scale, the rate of change in
unemployment appears different.
29
Ethical Considerations

• Also, a statistical association between two
  variables does not automatically imply
  causation. More in Chapter 13.
• Eg.
• Consumption of peanuts is correlated with aspirin
  consumption (eating peanuts gives headaches)