CHAPTER 1

INTRODUCTION
What is statistics?
Statistics is a group of method used to collect, analyze, present and interpret data and to make
decisions.

Example:
Item 2000 2010
Malaysia Population 23,494,900 28,908,795
Source: Jabatan Perangkaan Malaysia 2011

Types of statistics
Descriptive Statistics : Consists of methods for organizing, displaying, and describing data by
using tables, graphs and summary.
Example: The test scores of students enrolled in a statistics class. The data set usually large, thus
manage the data set by using frequency tables, drawing graphs or calculating measures (e.g.
averages).

Inferential Statistics: Consists of method that use sample results to help make decisions or
predictions about population.(As a “best guess” of a population value based on sample
information)
Example: Do you feel pressured by your boss to come to work?

Yes,
38%

No,
Not 61%
, 1%sure
Population versus Sample
Population is a collection of all individuals, objects, or measurements of interest. A survey that
includes every member of the population is called a census.

Sample is a portion or subset of the population selected for study.A sample selected must be
represents the characteristics of the population. The technique of collecting information from a
portion of the population is called sample survey
Types of data
Primary data Secondary data

Definition Data collected from primary sources Data collected from other sources
or from samples (raw data)

Data collected from annual

A researcher collects data on report, magazine, library, internet,
Example customer satisfaction toward counter
etc.
service at Maybank Segamat

More accurate and consistent, and in

Advantages Easily access, inxpensive and
many cases, the data needed by less time to collect.
decision-makers are not available
from secondary data.

Lack accuracy due to the

measurement procedure and the
Disadvantages Requires more time, manpower and methodof data collection are not
a higher cost to collect. explained by the previous
researchers.
Types of variable

Qualitative variable is known as attribute variable. It is measured on non-numerical scale.

Quantitative variable is known as numerical variable. It is measured on numerical (number)
scale. Quantitative data can be divided into two groups; discrete and continuous.
 Discrete data are countable data (integer 1, 2, 3 …) and are collected by counting.
 Continuous data are collected by measuring and are expressed on a continuous scale.

Example:
Variable Type of variable
Number of students in a class
Number of football players in a team
Height of a child
Monthly income of a manager
Gender
Ethnicity
Favorite reality TV shows
Level of measurement

Example:

Variable Level of measurement

Gender
Weight of 30 coconuts
Grade of lung cancer
Number of children
Socio-economic status
Dress size as 10, 12 and 14
Ethnicity
Favorite reality TV shows
Temperature of a room
Sampling Techniques
Students must first understand the following terms.
 Sampling is a process of selecting a sample from a population.
 Sampling techniques is a scientific methods of selecting samples from populations.
 Sampling frame is a list of all individuals or items (taken from population) from which the
sample can be obtained.
Why we need to do sampling?
 Cost– sample can be studied at much lower cost.
 Time – sample can be evaluated more quickly than population.
 Accuracy– sample can provide data set which is small enough to monitor carefully.
 Reliability/Feasibility – A high level of reliability can be achieved because fewer units
are surveyed in a sample than a full survey (census).
 Scope of information– when evaluating a smaller group, it is sometimes possible to gather
more information on each unit evaluated.

Types of sampling techniques

Sampling

Nonprobability Probability
sampling sampling
techniques techniques

Convenience Simple random Systematic

sampling sampling sampling sampling

Snowball Quota Cluster

sampling

Multistage
sampling
Nonprobability Sampling Techniques

1. Convenience Sampling
This technique attempt to obtain a sample of convenient elements. The selection of
sampling units is left primarily to the interviewer. Often, respondents are selected because
they happen to be in the right place at the right time.

2. Judgemental Sampling
A form of convenience sampling in which the population elements are purposely selected
based on the judgement of the resecher.

3. Snowball Sampling
Snowball sampling method begins by identifying a small number of individual (at least
one) who meet the criteria for inclusion in a study. We then ask them to recommend others
they knowwho also meet the selected criteria. This technique is appropriate when we are
trying to get a group of specific respondents who are difficult to reach or find.

4. Quota Sampling
Quota sampling involves non-random selections of individuals based on some
predetermined (category). Before selections made, we must decide what characteristics
that we want to use as the base of the quota such as gender, age, religion, etc.

Probability Sampling Techniques

1. Simple Random Sampling (SRS)

A sampling technique which a sample drawn randomly from a sampling frame so that each
item in a population has the same probability of being selected. There are 2 methods of
using SRS technique:
a) Lottery method/ Lucky draw method
b) Random number table. Generate from statistical package such as SPSS, MINITAB or
Excel.

Example:
Suppose 5 students from a class of 40 students are selected as a sample. Use simple
random sampling to select the sample.

Lottery Method
Step 1 : Have a list of 40 student’s names.
Step 2 : Write each of the names/ numbers on a separate piece of paper.
Step 3 : Place all 40 names/ numbers in a box and mix them.
Step 4 : Draw one name/number randomly from the box and repeat this experiment for
4more times.
Step 5 : The corresponding 5 names/numbers in a sampling frame is a sample.
 Random number table
Step 1 : Have a list of 40 student’s names
Step 2 : Generate5 different random numbers.
For example, the generated random numbers are 15, 31, 27, 02 and 11.
Step 3 : Thus the corresponding 5 random numbers in a sampling frame is a sample

2. Systematic sampling
In systematic random sampling, we first randomly select one member from the first k units.
Then every kth member, starting the first selected member, is included in the sample.
N
Range k  , where N is a population size and n is a sample size.
n

Example:
Suppose a population consists of 1800 employees of BigMoney Industries. A sample of
300 employees is to be selected from that population. Use systematic sampling to
select the sample.

Step 1 : Have a list of 1800employee’s names

Step 2: Calculate range k:
N
k
n
1800
 6
300
Step 3 : Select a starting point from the first k rangei.e. 1, 2, 3, 4, 5 and 6. Use lucky draw
method to select 1 number, say 4.
Step 4 : Therefore, the sample chosen from 1800 employees are 4 th, 10th, 16th, 22nd,
… until a sample size of 300 employees is obtained.
Step 5 : Thus the corresponding number in a sampling frame is a sample
3. Cluster Sampling
A population is divided into (geographical) subgroups called cluster. Then clusters are
randomly selected and a sample is collected by randomly selecting elements from each
cluster.Each cluster is representative of the population.

Segamat

Divide
Muar
Segamat
sample
Batu pahat
Mersing

Kluang Select all

J.bahru

Mersing
Select at random using
SRS/systematic
K.Tinggi

Pontian

J. Bahru

Population Cluster Sample

Example:
Suppose a company wants to make a survey on the number of households in the state of
Selangor. Steps in using cluster sampling are as follows:

Step 1 : The whole state of Selangor is divided into 12 districts which are called
clusters.

Step 2 : Then, select cluster at random using SRS, let say 5 districts from 12 districts.

Step 3 : For each of 5 selected districts, a sample is drawn using SRS.

4. Stratified Sampling
In this method, a population is divided into subgroups called strata and a sample is a
sample is collected by randomly selecting elements from each stratum.
Population can be divided on the basis of any characteristic such as gender, education,
race, income or family size.
Differences between strata should be heterogeneous.
 Accountancy, Finance and etc are diploma program
Element is each stratum should be homogeneous.
 Accountancy program – all students taking account subject.

Divide according to
program

Accountancy
Diploma
programM Finance

Computer Science

Select all Mathematics Select at random

using SRS/systematic

Strata Sample
Population

Example:
A company has 2000 employees, 55% of the employees work in manufacturing
department, 35% in painting department and 10% in inspecting department. A human
resource manager of the company wanted to conduct a survey on number of
absentee by selecting a sample of 120 employees. Use stratified sampling to obtain
120 employees from 2000 employees.

Step 1 : The employees are divided according to departments which are called strata
such that:
Department Number of employees
Manufacturing 55%x120  66
Painting 35%x120  42
Inspecting 10%x120 12
Total number of employees 120
 Step 2 : Then, 66 employees are selected at random from manufacturing department,
42 employees from painting department and 12 employees from inspecting
department as a sample.

5. Multistage Sampling
This method is designed to reduce time and cost when working with samples from very
large population.

Example:
Suppose we need a random sample of 2000 residents from Malaysian population. Since
Malaysia consists of 12 states with many districts within eah state and many villages within
each district, we could apply the multi-stage sampling technique.

Step 1: Select district at random, let say 4 states was selected. (using SRS)
Step 2: Then choose 5 districts randomly from each 4 selected states.(SRS)
Step 3: Finally we select 100 individuals at random from each of the 20 districts chosen to
make up our sample of 2000 (4states x 5 districts x 100 people)

Advantages and disadvantages of Probability Sampling

Techniques
Non-probability sampling

Techniques Strengths Weaknesses

Convenience sampling Less expensive, less time Selection bias, sample not
consuming, most convenient representative, not
recommended for descriptive
or causal research

Judgemental sampling Low cost, convenient, not Does not allow

time consuming generalization, subjective

Quota sampling Sample can be controlled for Selection bias, no assurance

certain characteristics of representativeness
 Probability sampling

Simple Random sampling Easily applied. Results can Difficult to obtain sampling
(SRS) be projected on population frame, expensive,
sometimes no assurance of
representativeness

Systematic sampling Easier to implement than Can decrease

simple random sampling representativeness if certain
patterns exist in sampling
frame

Stratified sampling Includes all important Difficult to select relevant

subpopulation, precision is stratification variables, not
improved feasible to stratify on many
variables, expensive

Cluster sampling Easy to implement, Imprecise, difficult to

compute and to interpret
and work is reduced results.

Data collection methods

Face-to-face An interviewr ask the question from a questionnaire
interview
- Personal interview
- High response rate, note the reactions of the respondents, can
clarify terms that might confused respondent.
- Expensive, error in recording response.

Telephone Ask the question from a questionnaire over the telephone

interview
- Short in duration, Less expensive
- Restricted only to individuals who can be reached by telephone.
- Response rate lower than face-to-face interviews.
- A few question can be asked.

Direct the researcher will greet respondents and explain briefly his intention
questionnaire before giving the questionnaire to the respondent. The researcher will wait
until the questionnaire has been completed.
 Mail -cheaper than personal interviews, the coverage is wider, No interviewer
questionnaire influence and More time to think of proper response

- Low response rate, might be bias, only simple questions can be asked,
nobody is on hand to explain the question.

Direct -record the real situation, not influenced by what people say or think,
observation
-require researcher with high skills and unbiased

Other -internet survey, sms, email etc.

Methods

Questionnaire Design and Surveys Management

The main questions are:
1. What is he purpose of the survey? (objective)
2. What kinds of questions the survey would be developed to answer? (data)
3. What sorts of actions is the administration/company considering based on the results of
the survey? (Recommendation/Suggestion)

Step 1: Planning Questionnaire Research

Step 2: Conducting Item Try-Outs and an Item Analysis (pilot study)
Step 3: Preparing a Questionnaire for Administration
Step 4: Selecting a sample of Respondents
Step 5: Preparing Statistical Tables and Figures
Step 6: Describing Averages and variability
Step 7: Describing Relationships
Step 8: Writing Reports
Guideline in Designing Questionnaires

The following points should be observed while designing questions:-

 Question should be short and simple
 Begin with simple and less controversial questions first
 Questions should not be biased
 Avoid sensitive questions
 Questions should be neither irrelevant nor too personal
 Question should unambiguous
 Avoid leading questions
 Avoid using vague words and bombastic words
 Avoid giving questions which requires calculations

Source of Errors
 The use of inadequate frame
 A poorly designed questionnaire
 Recording and measurement errors
 Non-response problems