Department of Computing and Information Systems

School of Science and Technology

Sunway University

LECTURE 1
INTRODUCTION TO
IST 1014/MTH1024
Introduction to Statistics
Jaya krishna

STATISTICS
RESEARCH QUESTION
Topic: Substance abuse
Thesis: What are the special treatment considerations for female substance
abusers?
 
Why is substance abuse treatment as it relates to women an important issue?
What is the necessary background information?
What is unique about women and substance abuse in terms of use?
What is unique about women and substance abuse in terms of effects?
What is unique about women and substance abuse in terms of treatment?
Based on my research, what do I think about treatment for female substance
abusers?
 
WHY DO YOU NEED TO
LEARN STATISTICS?
Why are you here?

Subjective Decision Objective Decision

Making Making
• Based on “gut • Based on various
feeling” data and information
• Does not rely on any • Rational and
data scientific approach
to making decisions
WHAT IS STATISTICS?
Statistics is the science of collecting, organizing, presenting
and analyzing raw data to obtain useful information for
decision making purposes.

Raw data

Explore /
Collect Organize Present
Analyze

Student
survey

Useful information for

decision making
VARIABLE
A variable is a characteristic that varies from one entity to
another.
Examples for humans: height, marital status, number of
siblings.
Data is the set of individual values associated with a variable.
The study and use of statistics help transform data into useful
information for decision makers.
TYPES OF VARIABLE

Qualitative Quantitative
(Categorical) (Numerical)
Variable Variable
• Take categories as • Have values that
values represent a counted
• Example: Marital or measured
status (single, quantity
married, divorced, • Can be defined as:
others) • Discrete
• Continuous
TYPES OF QUANTITATIVE
VARIABLE
Discrete
• Numerical values that arise from a counting process
• Example: Number of items purchased

Continuous
• Numerical values that arise from a measuring process
• Example: Time spent on checkout line
LEVELS OF MEASUREMENT IN
DATA
• Labels represent various • Labels represent an order • Numerical labels indicate
levels of a categorical that indicates either order and distance
variable preference or ranking between elements.
• Example: Brand names, • Example: Academic • There is no absolute zero.
gender qualification (Bachelor’s, • Multiples of measures are
Master’s, PhD) not meaningful.
• Example: Temperature (in
Fahrenheit or Celcius)

Nominal Ordinal Interval

• Numerical labels indicate

order and distance
between elements.
• There is an absolute zero.
• Multiples of measures are
meaningful.
• Example: Length

Ratio
Is the variable age qualitative or quantitative?
 Raw data

Explore
Collect Organize Present /
Analyze

Useful information for

decision making
 POPULATION & SAMPLE
Before we proceed to discuss data sources, first we clarify two important
terms in statistics: population and sample.
Population

Time constraints
Cost
Practicality

A portion or subset of the

population from which information
is obtained.
The collection of all individuals or
items of interest in a statistical study.

When we analyze data from a sample

we compute statistics.
When we analyze data from a population
we compute parameters.
DATA SOURCES

Primary Sources
• We collect our own data for analysis.

Secondary Sources
• We use data which have been collected
from someone else (an individual or an
organization).
DATA COLLECTION METHODS

Designed Experiments Observational Studies

In a designed experiment, the In an observational study, the
investigator may introduce investigator observes conditions
treatments and controls which that normally prevail in the
do not exist in any naturally population.
occurring population. The Example: A psychologist
investigator then observe observes the way children
characteristics and take interact with each other in a
measurements. playground.
Example: Clinical experiments
DATA COLLECTION METHODS

Data Distributed by Surveys Data Collected by

Organizations or Census – all members of Ongoing Business
Individuals the population are Activities
Market research included Can be collected from
companies Sample Survey – involves operational and
Investment services only a sample from the transactional systems,
population social media networks,
Print and online media online apps, website
companies services that collect
tracking and usage data
DATA CLEANING
There may be irregularities in the data values collected
 Undefined value
 Impossible value (Example: A negative value recorded for height)
 Outliers (Example: A person with height 2.0 m is recorded)
 Missing values

When we spot an irregularity in the data, we may have to

“clean” the data. Data cleaning will be discussed in further
details in IST2024 Applied Statistics.
RANDOM SAMPLING
METHODS
Simple Random Sample
• Each individual in the population has an equal chance of
being selected.
• Can be done with or without replacement.

Systematic Sample
• Divide the population size by the sample size and round the
result down to the nearest integer, m.
• Randomly choose a number, k, between 1 and m.
• Then choose the members numbered (k, k + m, k + 2m, …)
of the population as a member of the sample.
RANDOM SAMPLING
METHODS (CONT.)
Stratified Sample
• Divide the population into groups (strata).
• Select a random sample from each group. The size of the
random sample should be proportional to the size of the
group.
• Use all the members selected as the sample.
Cluster Sample
• Divide the population into individual clusters or units.
• Randomly select one or more clusters.
• Use all the members of the clusters selected, as the sample.
NON-RANDOM SAMPLING
METHODS
Convenience Sampling
• Easy, inexpensive, convenient to sample.
• Example: Workplace colleagues, friends.
• Issue of representativeness

Judgement Sampling
• The judgement of the researcher is used to determine as to who is
representative of the population.
Snowball Sampling
• Uses recommendations to find people with the specific range of skills that
has been identified as useful to the research.
• A few potential respondents are contacted and asked whether they know of
anybody with the required characteristics in the research.
• Example: Contacts through support groups.
BRANCHES OF STATISTICS
• Involves organizing, summarizing and displaying data in an
informative manner
• Includes:
• Construction of graphs and tables
Descriptive • Calculation of descriptive measures such as average, standard
statistics deviation, percentiles.

• Involves using sample data to draw and measure the reliability of

conclusions about a population.
Inferential
statistics
A survey of 1,179 adults reported that 54% thought
that 15 seconds was an acceptable online ad length
before seeing free content.

A study compiled a comprehensive report on the use

of antibiotics and antibiotics resistance in various
countries in the world.