CHAPTER 1 Objectives

„ Demonstrate knowledge of statistical terms.

„ Differentiate between the two branches of

statistics.
The Nature of Probability and
Statistics „ Identify types of data.

„ Identify the measurement level for each

variable.

1-1 1-2
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004

Objectives (cont’d.) Introduction

„ Identify the four basic sampling techniques. „ Statistics is the
„ Explain the difference between an science of
observational and an experimental study.
conducting studies
„ Explain how statistics can be used and to collect, organise,
misused.
summarise, analyse,
„ Explain the importance of computers and and draw
calculators in statistics.
conclusions from
data.

1-3 1-4
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004
 Some Examples Use of Statistics in Business

„ IRELAND FACTS „ Top Internet search sites „ Quality control. Statistical quality-control procedures
• Population: 4 million User in millions for March,2002 assure high product quality and enhance productivity.
(official figure, 2003)
• Capital: Dublin „ Product planning. Statistical methods are used to
• Major languages:
English, Irish analyse economic factors and business trends and to
• Major religion: prepare detailed budgets, inventory-control systems,
Christianity
and realistic sales quotas.
• Life expectancy: 74
years (men), 80 years
(women) (UN)
„ Forecasting. Statistics are used to predict sales,
productivity, and employment trends.
(BBC Website) (USA Today, 2002)

1-5 1-6
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004

Descriptive and Inferential Statistics Basic Vocabulary

„ Descriptive statistics „ Inferential statistics „ Probability is the chance of an event

consists of the consists of generalising occurring.
collection, from samples to
populations, performing „ A population consists of all subjects that are
organisation,
estimations, hypothesis being studied.
summarisation, and testing, determining
presentation of data. relationships among
„ A sample is a group of subjects selected from
variables, and making a population.
predictions.

1-7 1-8
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004
 Populations and Samples Some examples

You use inferential statistics to form conclusions about a large „ The average cost of an airline meal in 1993 was $4.55;
group – a population – by collecting a portion of it – a sample. (Source: Ever Has Its Price, Richard E. Donley, Simon and Schuster)

The analyst decides what the population is. Typically, the „ “More than 1 in 4 United States children have cholesterol levels
population is so large that it would be nearly impossible to of 180 milligrams or higher”;
obtain information about every item in it. In stead, we obtain (Source: The American Health Foundation)
information about selected members and attempt to draw a
„ “Every 10 minutes, 2 people die in car crashes and 170 are
conclusion about all members. In other words, we attempt to
injured”;
infer something about the population using information about
(Source: American National Safety Council estimates)
only some of the members of this population.
„ “The average amount spent per gift for Mom on Mother’s day is
$25.95”;
(Source: The Gallup Organisation)

1-9 1-10
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004

More Examples Variables and Data

„ 9 out of 10 on-the-job fatalities are men; „ In order to gain knowledge about seemingly
(Source: USA Weekend)
haphazard events, statisticians collect
„ Expenditures for the cable industry were $5.66 billion on 1996;
(Source: USA Today)
information for variables that describe the
events.
„ The median household income for people age 25-34 is $35,888;
(Source: USA Today)
„ A variable is a characteristic or attribute that
„ “Allergy therapy makes bees go away”;
can assume different values.
(Source: Prevention)

„ Drinking decaffeinated coffee can raise cholesterol levels by 7%; „ Data are the values that variables can
(Source: American Heart Association)
assume.

1-11 1-12
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004
 Variables and Data (cont’d.) Variables and Types of Data

„ A data set is a collection of data values. „ Qualitative variables can be placed into
distinct categories according to some
„ Each value in the data set is called a data
characteristic or attribute.
value or a datum.
„ Quantitative variables are numerical in nature
„ Random variables have values that are and can be ordered or ranked.
determined by chance.

1-13 1-14
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004

Some Examples Variables and Types of Data (cont’d.)

„ Qualitative „ Quantitative Quantitative variables can be further classified

-- Colours of cars -- Number of pages in the into two groups.
statistics textbook
-- Gender „ Discrete variables assume values that can be
-- Capacity of students in a
-- Religious preference classroom
counted.

-- Ethnic group -- Weights of fish caught today „ Continuous variables can assume all values
by fisher men
-- Nationality between any two specific values.

1-15 1-16
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004
 Variables and Types of Data (cont’d.) Some Examples

„ Rounding of the continuous values:

Usually, continuous data are rounded to the

nearest given unit. For example, heights
might be rounded to the nearest inch. Hence,
a recorded height of 73 inches could mean
any measure from 72.5 inches up to but NOT
including 73.5 inches, which would be
rounded to 74 inches.
1-17 1-18
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004

Measurement Scales Levels of Measurement

In addition to being classified as qualitative or „ Nominal—classifies data into mutually

quantitative, variables can be classified by how they exclusive (nonoverlapping), exhausting
are categorised, counted or measured. For example,
categories in which no order or ranking can
can the data be organised into specific categories,
such as area of residence (rural or urban)? Can the
be imposed on the data.
data values be ranked, such as heights or IQs? This For example, Zip code, Gender (M or F), Eye
type of classification - i.e. how variables are
colour (blue, brown, green, hazel), Political
categorised, counted, or measured - use
measurement scales, and four common types of
affiliation, Religious affiliation, Study field
scales are used: nominal, ordinal, interval and ratio. (mathematics, computers, etc.), Nationality.
1-19 1-20
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004
 Levels of Measurement (cont’d.) Levels of Measurement (cont’d.)

„ Ordinal—classifies data into categories that „ Interval—ranks data, and precise differences
can be ranked; however, precise differences between units of measure do exist; however,
between the ranks do not exist. there is no meaningful zero.

For example, Grade (A,B,C,D,F), Judging (first For example, IQs, Temperatures
place, second place, etc.), Rating scale (poor,
good, excellent), Ranking of tennis player.

1-21 1-22
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004

Levels of Measurement (cont’d.) Data Collection and Sampling Techniques

„ Ratio—possesses all the characteristics of „ Surveys are the most common method of
interval measurement, and there exists a true collecting data. Three methods of surveying
zero. are:

For example, Height, Weight, Time, Salary, ¾ Telephone surveys

Age. ¾ Mailed questionnaire surveys
¾ Personal interviews

1-23 1-24
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004
 Sampling Methods Sampling Methods (cont’d)

Investigating the whole population is often impossible due to „ Random samples are selected using chance
expense, time, or size of population etc. Using samples saves
time and money and, in some cases, enables the researcher to methods or random methods.
get more detailed information about a particular subject.
Samples cannot be selected in haphazard ways because the
information obtained might be biased.
One example is to number each subject in the
To obtain samples that are unbiased - i.e. give each subject in
population. Then place numbered cards in a bowl,
the population an equally likely chance of being selected -
statisticians use four basic methods of sampling: random, mix them thoroughly, and select as many cards as
systematic, stratified, and cluster sampling. needed. The subjects whose numbers are selected
constitute the sample.

1-25 1-26
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004

Sampling Methods (cont’d) Sampling Methods (cont’d)

„ Researchers obtain systematic samples by „ Researchers select stratified samples by

numbering each subject of the populations dividing the population into groups according
and then selecting every kth number. to some characteristic that is important to the
For example, supposed there were 2000 subjects in the study, then sampling from each group.
population and a sample of 50 subjects were needed. Since
2000/50=40, then k=40, and every 40th subject would be For example, suppose the president of a 2-year college wants to
selected; however, the first subject (numbered between 1 and 40) learn how students feel about a certain issue. Furthermore, the
would be selected at random. Suppose subject 12 were the first president wishes to see if the opinions of the first year students
subject selected; then the sample would consist of the subjects differ from those of the second-year students. The president will
whose numbers were 12, 52,92, etc. until 50 subjects were select students from each group to use in the sample.
obtained.
1-27 1-28
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004
 Sampling Methods (cont’d) Other Sampling Methods

In addition to the four basic sampling methods, researchers use other

„ Researchers select cluster samples by using
methods to obtain samples. One such method is called a convenience
intact groups called clusters. sample. Here a researcher uses subjects that are convenient. For
example, the researcher may interview subjects entering a local mall to
Suppose a researcher wishes to survey apartment dwellers in a
determine the nature of their visit or perhaps what stores they will be
large city. If there are 10 apartment buildings in the city, the patronising. This sample is probably not representative of the general
researcher can select at random 2 buildings from 10 and customers for several reasons. For one thing, it was probably taken at a
interview all the residents of these buildings. Cluster sampling is specific time of day, so not all customers entering the mall have an
used when the population is large or when it involves subjects equal chance of being selected since they were not there when the
residing in a large geographic area. survey was being conducted. But convenience samples can be
representative of the population. If the researcher investigates the
characteristics of the population and determines that the sample is
representative, then it can be used.

1-29 1-30
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004

Observational and Experimental Studies Uses and Misuses of Statistics

„ In an observational „ In an experimental „ Suspect samples

study, the researcher study, the researcher
¾ Very small samples
merely observes what is manipulates one of the
happening or what has variables and tries to ¾ Bias sample selection
happened in the past determine how the
¾ Volunteer samples
and tries to draw manipulation influences
conclusions based on other variables. „ Ambiguous averages (median, mean, mode)
these observations.
„ Changing the subject ( 3% vs. 6,000,000)

1-31 1-32
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004
 Uses and Misuses of Statistics (cont’d) An Misleading Graph

„ Detached statistics (compared to what?)

„ Implied connections (may help, suggest)

„ Misleading graphs

„ Faulty survey questions

1-33 1-34
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004

Computers and Calculators Statistical Packages

„ In the past, statistical „ Excel, MINITAB, SPSS

calculations were done
with pencil and paper. „ Students should realise that the computer
However, with the and calculator merely give numerical answers
advent of calculators, and save time and effort of doing calculations
numerical computations by hand.
became easier.

1-35 1-36
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004
 Summary Summary (cont’d.)

„ The two major areas of statistics are „ Data can be classified as qualitative or
descriptive and inferential. quantitative.

„ When the populations to be studied are large, „ The four basic types of measurement are
statisticians use subgroups called samples. nominal, ordinal, interval, and ratio.

„ The four basic methods for obtaining samples „ The two basic types of statistical studies are
are: random, systematic, stratified, and observational and experimental.
cluster.

1-37 1-38
© Copyright McGraw-Hill 2004 © Copyright McGraw-Hill 2004

Conclusion

„ The applications of statistics are many

and varied. People encounter them in
everyday life, such as in reading
newspapers or magazines, listening to
the radio, or watching television.

1-39
© Copyright McGraw-Hill 2004