STA104 / QMT181

CHAPTER 1:
INTRODUCTION TO
STATISTICS

Prep a red by:

Fadila Amira Razali
Faculty of Computational Science and
Mathematic
Universiti Teknologi MARA
Pahang Branch, Jengka Campus
 OBJECTIVES
To define the terms in statistics
To define the types of statistics
To identify method of collecting data
To identify types of various sampling
techniques

Prepared by: Fadila Amira Razali

 WHAT IS STATISTICS?
Statistics is a branch of mathematics dealing with
collecting, organizing, analyzing, and making
inferences(drawing conclusion) from data.
Example:
1) The number of cars accident in Johor.
2) Trend of student’s enrollment at UiTM Pahang.

Prepared by: Fadila Amira Razali

 BASIC TERMS IN STATISTICS

• Also known as experimental unit.

ELEMENT • Objects either people or thing on which measurement is taken.

• A set or collection of all possible observations of

POPULATION some specific characteristic.

• A subset or subgroup selected from the population

SAMPLE

Prepared by: Fadila Amira Razali

 BASIC TERMS IN STATISTICS (cont…)

• Measures the characteristics of the

VARIABLE population that the researcher wants to study

• A study conducted to measure a variable for every

CENSUS item in the population of interest.

• A study involves a subgroup or sample

SAMPLE of a population.
SURVEY

Prepared by: Fadila Amira Razali

 BASIC TERMS IN STATISTICS (cont…)

• The characteristic value of an element.

DATA

• A summary measures for the entire

PARAMETER population

• A summary measures for a sample

STATISTIC

Prepared by: Fadila Amira Razali

TYPES OF STATISTICS

Prepared by: Fadila Amira Razali

SOURCES OF DATA

Prepared by: Fadila Amira Razali

 ADVANTAGES & DISADVANTAGES

PRIMARY DATA

ADVANTAGES:
1. Primary data are more accurate, reliable and up-to-date
2. If the data needed by decision makers are not available from other
sources (secondary data) then the primary data has to be gathered.
3. Primary data usually satisfy the objectives of a research.

DISADVANTAGES:
1. Data gathered from primary sources are very costly, time consuming
and require a lot of man power

Prepared by: Fadila Amira Razali

 ADVANTAGES & DISADVANTAGES
SECONDARY DATA

ADVANTAGES:
1. It is less time, less cost, and less effort required

DISADVANTAGES:
1. May contain errors due to error in printing and also due to transcription
from the primary sources.
2. Secondary data may not be able to fulfill the objectives of a research.
3. Individuals who use secondary data do not know the conditions under
which the data were collected and summarized. Therefore, the
intended user must first determine whether the data is relevant or not.

Prepared by: Fadila Amira Razali

 VARIABLES

response in non-numerical form

example: gender, religion and race

DISCRETE
Can be measured precisely by
counting such as number of flower,
number of pen and number of students
response in numerical form
example: age, weight and
CONTINUOUS
height.
response that can only be approximated
to some accuracy using measuring
devices such as temperature and time

Prepared by: Fadila Amira Razali

 LEVEL OF MEASUREMENT (VARIABLE SCALE)
If the different
between data A scale in which
values are the numbers or
meaningful but letters are
cannot be assigned to
manipulated with objects serve as
multiplication and labels for
division such as identification or
temperature classification

A scale that is can

Starting point be arranged in
cannot be ranking, orders or
negative value rating. For example,
(absolute/true education level,
zero point) such rate of services and
as body weight perception level
and income

Prepared by: Fadila Amira Razali

Summary for Variables & Measurement Scale

Prepared by: Fadila Amira Razali

EXERCISES
1. A researcher from the Mathematics Department of a local university wanted to
determine the perception on the effectiveness of teaching algebra using a special
technique among Form Five Mathematics teachers. He selected 20 schools from a
northern region area and gave training on the special technique to these teachers. At
the end of the training session, all the teachers responded to a set of questionnaire. The
variable of interest in this study were years of experience in teaching Mathematics,
perception on the special technique and duration of time to solve algebra problem
using special technique.

a. Identify the population and sample of the study.

b. Identify any TWO (2) variables of interest and its level of measurement.
c. Identify what kind of data sources that researcher used for this study?

2. A researcher is interested in studying students’ satisfaction towards hostels services.

There are 1000 registered students in Gemilang Hostel and 200 students are randomly
selected. Questionnaire are disseminated among students to gather information such
as student’s background and hostel services.
a. State the population and sample for the above study.
b. State the variable of interest for the above study.
Prepared by: Fadila Amira Razali
EXERCISES
3. State if the following variables are qualitative or quantitative:
3. Waiting time at a clinic
4. Qualification of candidates for a particular job
5. Number of workers at Firm ABC
4. State the scale of measurement for each variables:
a) Models of computer
b) Daily temperature of the city
c) The time recorded on runners in 400-metres sprint
d) Numbers of accidents in a month
e) Colour of shirts
f) The top five best students in a class
g) The distances jumped by athletes in the long-jump event

Prepared by: Fadila Amira Razali

EXERCISES
5. State whether the following statements are TRUE or FALSE:
a) Data that can be obtained from the published articles, magazines and
reports is a primary data
b) A summarized value taken from a population is a parameter
c) The variable marital status is an examples of qualitative variable
d) The weight of papaya is considered to be a continuous variable

6. For each of the following situations, state whether the area of statistics
used is descriptive or inferential
a) Obesity has become a major problem in Malaysia. It is found that 35% of
primary school children in the country are obese. Of these, 58% of the
children are males
b) A study on “Obesity Among Primary School Children” conducted by
Professor Kamariah, which involved a random sample of 1500 primary school
children, indicated that these children are obese because they preferred
eating at fast food restaurants and playing video games rather than sports
activities.
Prepared by: Fadila Amira Razali
 METHODS OF COLLECTING DATA

TELEPHONE MAILED
DIRECT
QUESTIONNAIRE
INTERVIEW INTERVIEW S

DIRECT
ONLINE
OBSERVATIO
SURVEY
N

Prepared by: Fadila Amira Razali

 DIRECT INTERVIEW
- Trained interviewer will get the information from respondents depending
on the objective of the survey and based on questionnaires.
- It is commonly used in marketing research and social survey

Prepared by: Fadila Amira Razali

 TELEPHONE INTERVIEW
- It is possible to use telephone interview as a personal interview
- It is conducted by telephone services line and respondents are asked to
respond to the questionnaire

Prepared by: Fadila Amira Razali

 MAILED QUESTIONNAIRES
-Questionnaires are mailed to the respondent included from the sample
- A carefully constructed questionnaire will encourage participation by the
respondents
-Does not require interviewers

Prepared by: Fadila Amira Razali

 DIRECT OBSERVATION
-Observation is used in work-study and in organization
- A quality control exercise in a factory making light bulbs in which the data are
collected by taking bulbs from the production line and observing the number of
hours they last.

Prepared by: Fadila Amira Razali

 ONLINE SURVEY
Nowadays, online questionnaire is a very popular method to
collect data.

Advantages Disadvantages
• cheap way to reach respondents • Respondent may not answer the
globally questionnaire
• Response rates are higher than • Respondent might not understand
mailed questionnaire method the question very well.

Prepared by: Fadila Amira Razali

 DESIGNING A QUESTIONNAIRE
o Questionnaire is a written instrument that contains a series of questions or items that
attempt to collect information on a particular subject.
o May be handed out personally by the researcher or posted mail.
o The researcher should construct a questionnaire to encourage participation from the
respondent
o Several steps to be considered when designing a questionnaire
1. Specify the information to be collected
2. Questions should be as short as possible
3. Questions must be kept simple and phrased to imply the same meaning to all
respondents.
4. Avoid leading questions
5. Questions should be organized systematically
6. Avoid questions that need calculation
7. Avoid questions on sensitive issue or may be confidential to the respondents

Prepared by: Fadila Amira Razali

 SAMPLING

o Many times it is not practical to study an entire population. Thus, inference

about a population can be made based on information contained from
sample. This technique is called sampling.

Prepared by: Fadila Amira Razali

 TERMINOLOGIES IN SAMPLING
PROBABILITY SAMPLE (RANDOM SAMPLE)
-A sample in which every element in the population has a known probability of being
included in the sample.
Eg: simple random sampling, systematic sampling, stratified sampling, cluster sampling and
multi-stage sampling

NON-PROBABILITY SAMPLE (NON-RANDOM SAMPLE)

-A sample that does not have the feature of probability sample
Eg: Quota sampling, judgemental sampling, convenient sampling

SAMPLING FRAME
-A list of all population members
Eg: List of all students registered in UiTM Raub

SAMPLING UNIT
-The element listed in the frame

Prepared by: Fadila Amira Razali

 TERMINOLOGIES IN SAMPLING
SAMPLE SURVEY DESIGN
-The procedure for selecting the sample

PILOT STUDY
-Small exploratory exercise conducted on a small number of respondents before the actual
survey is done.
-Objectives: to improve questionnaire, to identify problems that occur during the survey, to
predict cost, time and workforce needed.

SAMPLING ERROR
-Error that arises because a sample cannot give complete information on a population

NON-SAMPLING ERROR
-Error that occur from the survey due to non-response from respondent, faulty measuring
devices, and respondents giving false information or errors in writing and analyzing data

Prepared by: Fadila Amira Razali

 1. SIMPLE RANDOM SAMPLING (SRS)
- A sample is selected in such a way that every item in a population has an equal chance to
selected.
- This method involves in the numbering of every individual in the population
- SRS is used when the items are from a group of the same characteristic
- Not suitable if the population is divided into several heterogeneous groups because biasness m
occur

HOW TO SELECT THE SAMPLE???

1. Small population ( n < 300 )  using ‘lottery method’

2. Large population ( n > 300 )  using random number table

Prepared by: Fadila Amira Razali

Example of Random Numbers:

Prepared by: Fadila Amira Razali

 2. SYSTEMATIC SAMPLING
- A random sample in which the items are selected from the population at a ‘uniform interval’ of a
listed order providing that there is no regularity within the list.
- This method involves complete sampling frame
- Method: using equal interval method or constant skip method I = N/n where I = interval, N =
population, n = sample size
- Eg: N=300, n=30, I=300/30=10
- For every 10 item, only one item is selected.
- The first item will be selected at random between 1 to 10, e.g No.5
- The next item will be 5+10, 15+10…
- 5, 15, 25, 35….295 (until we reach the sample size of 30)

Prepared by: Fadila Amira Razali

Example of Systematic Sampling:
A sampling frame consists of 2000 names of MyTV subscribers in Georgetown area. The
names are arranged alphabetically. A sample of 150 subscribers was chosen from the
sampling frame to estimate the average monthly payments per subscriber.

Solution Steps:

(1) Give the number to all MyTV subscribers.

(2) Calculate the interval, I
I= 2000/150 = 13.3 ≈ 13th
(3) Select first sample from 1 to 13 using SRS. For example: 5
(4) Select the remaining sample until we get 150 samples.
5, 18, 31, …

Prepared by: Fadila Amira Razali

 3. STRATIFIED SAMPLING
- This method is used if the population is divided into suitable sub-populations where between
groups are heterogeneous and within groups are homogeneous
- All the groups will be selected and item from each group will be selected randomly using simple
random sampling or systematic sampling
- The same proportion of each group will be represented in correct proportion using Proportional
Stratified Sampling
Ni
ni = ×n
N
- Where ni = sample size needed for each group
Ni = population size for each group
N = population size
n = sample size

Prepared by: Fadila Amira Razali

Example of Stratified Sampling:

A total of 200 students of a school are grouped according to their race. The sample size
needed is 60.

 (n = 60 , N = 200)
𝑵𝑵𝒊𝒊
Groups Number of student, Ni Sample: 𝒏𝒏𝒊𝒊 = × 𝒏𝒏
𝑵𝑵
A 20 6
B 60 18
C 80 24
D 40 12
TOTAL 200 60

Next, elements are selected from each group by using a random procedure, usually SRS.

Prepared by: Fadila Amira Razali

 4. CLUSTER SAMPLING
- This method is used if the population is divided into sub population where between groups are
homogeneous and within groups are heterogeneous
- The groups will be selected at random
- All the items in each group will be chosen
- Eg: Consumer survey in large cities often employ cluster sampling

ADVANTAGES DISADVANTAGES
1. Reduce cost, time and workforce 1. The sample might not be
since only a few clusters formed are representative of the population as a
selected as sample whole since nothing is known of the
cluster not sampled
2. It is preferable to divide the
population into a large number of
small clusters than a small number of
large clusters.

Prepared by: Fadila Amira Razali

Example of Cluster Sampling:

A research was conducted to study the 'Effectiveness of a Computer for Every

Home’. The study focused on dwellers or household of housing estates in town. Five
housing estates were randomly selected from a total of 60 housing estates. Every
household from the selected housing estates was studied on the effectiveness of the
campaign.

Solution:
This is a cluster sampling technique because those selected as the samples are all
household of five housing estates out of 60 housing estates. Those who are not in the
selected housing estates will not be selected as samples. The sampling frame is the
list of 60 housing estates.

Prepared by: Fadila Amira Razali

 5. MULTISTAGE SAMPLING
- This method can cut down the survey area and reduce the survey costs from very large
population
- Eg: an immunization survey of schoolchildren in a particular state might entail 5 stages

Prepared by: Fadila Amira Razali

Example of Multistage Sampling:

E.g: Suppose we need a random sample of 2,000 residents from the Malaysian population.
Since Malaysia consists of 14 states, with many districts within each state, and many villages
within each district, we could apply the multistage sampling technique.

Select four states at random (SRS)

State

choose five districts randomly

District from each selected state.

Select ten town at random (SRS)

Town from selected district

Select 20 village from selected

Village town, Finally select 100 individuals
from selected village

Prepared by: Fadila Amira Razali

 6. CONVENIENT SAMPLING
- The most convenient items are selected for the sample
- Example: A market research class has decided to sample student opinion on a several subjects
for a class project. The team leader assigns each student a sample of 10 people and requests
that interviews must be completed in one week.
- This could be a convenient sample because each student will presumably sample the 10 most
convenience people in the absence of any instruction.

Prepared by: Fadila Amira Razali

 7. QUOTA SAMPLING
- Samples are selected on the basis of specific guidelines about which items and / or how many
should be drawn.
- Quotas are set and interviewer is allowed to select sample according to quota.
- It is not a method of random sampling and no sampling frame is needed.
- Example: An account officer is interested in knowing the types of people who invested at the
Kuala Lumpur Stock Exchange (KLSE) and the kinds of stock they invested in. he selected
respondents by interviewing twenty persons per day at a brokerage firm nearby for ten days. The
respondents were from many races where Malay, Chinese, and others are represented by 45%,
50% and 5% respectively. The interviews were done between 10.00 a.m till 4.00 p.m.
- This is a quota sampling because those selected as the samples are only those who are at the
brokerage firm between 10.00 a.m till 4.00 p.m. Those who are not in the brokerage firm between
those hours will not be selected as a samples.

Prepared by: Fadila Amira Razali

 8. JUDGEMENTAL SAMPLING
- The researcher’s judgement is used to decide which population items will be included in the
sample
- Example: Angkasa Board of directors has decided to sample opinions of other companies
around the country regarding the economic forecast for next year. The company’s president
chooses those persons to be interviewed by telephone.
- This is a judgemental sample because the specific individuals to be called aren’t selected
randomly; rather the President’s judgement is used in the selection process. If the company
president is experienced and personally selects a representative group to a sample, a good
sample should result. On the other hand, if the president talks only to friends who are not in a
good position to assess economic conditions, a misleading picture of the company’s operating
environment could result.
ADVANTAGES DISADVANTAGES
1. Good judgement will give a good 1. The sampler must use good
result judgement in choosing items to be
sampled.

Prepared by: Fadila Amira Razali

EXERCISES:

QUESTION 1

Ali Travel Agency, a nationwide travel agency, offers special rates on Phuket-Penang
cruises to Malaysian citizens. A researcher of Ali Travel Agency wants to do a research on
the ages of those people taking the cruise. 100 customers taking a cruise last year was
selected as a sample.

(a) State the population for the above research.

(b) State the variable of interest and identify its type.
(c) State the most appropriate sampling technique for the above research.

Prepared by: Fadila Amira Razali

EXERCISES:

QUESTION 2

A group of researchers plan to carry out a survey on the number of vehicles bought in Town
Y from January to June 2017. These vehicles can be categorized according to the types.
Numbers of authorized
Type
dealer
Motorcycle 50
Car 85
Heavy Vehicles 20

In order to save cost and time, they plan to survey only 40 of these authorized dealers.
a) State the population.
b) Which is the most appropriate sampling technique that may be used by the researchers?
Give ONE reason for your answer.
c) Calculate the sample size that represent each type of vehicles using proportional
sampling technique.
d) Which is the most suitable method of data collection? Give TWO advantages of this
method.

Prepared by: Fadila Amira Razali

EXERCISES:
QUESTION 3
A manager of a company wishes to estimate the average number of days of vacation
taken by the employees. The company has six branches located throughout the country.
The following table shows the number of employees in each branch:

Branch Raub Segamat Kuala Banting Tawau Pasir Mas

Pilah
No. of
140 200 120 130 160 250
Employess

The manager then selects randomly a number of employees from each branch. He decides
to select at random a total of 400 employees from 1000 employees.

a) State the appropriate sampling frame for the above study.

b) State the variable on interest in the above study.
c) What is the sampling technique used? Give TWO advantages of using this technique.
d) Explain briefly the process of selecting employees from branch Banting using systematic
sampling technique.

Prepared by: Fadila Amira Razali