CHAPTER FIVE

The Research Design

Learning Objectives
After reading this chapter, you should be able to:

● Understand the types of research design and its elements.

● Determine the depth of interference a researcher is allowed.
● Identify the units of analysis.

1
Research Methodology
Halilah Haron, March 2014
5.1 Types of Research Design

Research Design

Experimental Designs Non-experimental Designs

 Exploratory Study

 Descriptive Study

 Correlational Study

5.2 Non-experimental Design

Exploratory Research

Designed to better understand the nature of the problem since very few studies
might have been conducted in that area.

Example 1: How developed are we Malaysians in terms of statistical usage

compared to the countries in the West?

Descriptive Research

 Designed to describe the characteristics of the variables of interest.

 Can be used for estimating, predicting, and examining associative
relationships (correlational studies).
2
Research Methodology
Halilah Haron, March 2014
  Examples include case-studies, surveys.

Types of Descriptive Research

 Cross-Sectional Study

 Also known as one-shot studies.

 Typically involves conducting a survey of a sample of population
elements at one point in time, perhaps over a period of days, weeks or
months.
 Useful because it provides a quick snapshot of what‟s going on with
the variables of interest for our research problem.

 Longitudinal Study

 An investigation that involves taking repeated measures over time.

 Can see patterns of change.
 Take more time and costs.
 Useful for conducting trend analysis, tracking changes in behavior
over time (e.g., brand switching, levels of awareness, turnover) and
monitoring long-term effects of marketing activities (e.g., market share,
pricing effects).

Some examples of descriptive study are shown in Examples 2 – 5.

Example 2:

The earliest recorded case of dengue fever in Malaysia was described by Skae in
1902. The first major outbreak occurred in 1973 (Shekhar, 1992). Since then
dengue has become a major public health problem in the country. The table and
graph below show the number of reported dengue cases in Malaysia between
1995 and 2005.

Table 1. Number of reported dengue cases in Malaysia between 1995 and

2005. (Ministry of Health, Malaysia)

Year 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Total 6543 14255 19429 27381 10146 7103 16386 32767 31545 33895 39654

Table 2 shows that students are not immune to dengue fever or dengue
haemorrhagic fever. Further, there is an obvious 5-fold difference between the
victims in urban area and sub-urban area. Of the 200-student sample, only 2% of
3
Research Methodology
Halilah Haron, March 2014
 students from sub-urban school have contracted dengue fever while 10% of them
from urban school have suffered the disease.

Table 2. Data on students or their family members who contracted dengue

fever and noticed fogging in housing area

No. of Family
students member Fogging
who who done in
contracted contracted housing
dengue dengue vicinity
fever fever
Victoria Institution 20 (10.0%) 49 (24.5%) 147 (73.5%)
Bandar Baru Bangi
4 (2.0%) 22 (11.0%) 136 (68.0%)
Secondary School

The number of known cases among family members of students in urban school
is also significantly higher. In fact the data shows that the frequency of
occurrence is more than twice in that direction. Majority of the students noted
that fogging had been carried out by government agencies in both rural and urban
area.

On the susceptible age, of the total of 24 dengue cases, 3 of them contracted

dengue fever at the tender age of 6 (Table 3). From the spread of distribution, we
can conclude that our school children at all ages are at risk from dengue virus
infection.

4
Research Methodology
Halilah Haron, March 2014
 Table 3. Distribution of dengue cases by age

Age 6 7 8 9 10 11 12 13 14 15 16
Victoria
3 0 0 1 2 1 1 2 5 4 1
Institution
Bandar Baru
Bangi
0 0 0 0 2 0 0 0 0 1 1
Secondary
School

Field Study – conducted in a natural environment i.e. SMK Bandar Baru Bangi
& SMK Victoria, KL

Example 3:

ACCOUNTABILITY IN THE PUBLIC SECTOR IN MALAYSIA

Abstract

In the recent years, we have witnessed a worldwide movement towards better

accountability in the public sector. It was considered both timely and appropriate
to introduce empirical evidence by examining the financial reports of government
entities in Malaysia. A financial accountability approach in the context of the
conceptual framework was undertaken in conducting this evaluation.

62 government entities from federal, state and local level were examined to
evaluate if their financial reports reflect financial accountability and if size, long
term debt and the form of government affects the level of disclosure reflecting
financial accountability.

The study provided empirical evidence to support the research question that
financial reports by government entities in Malaysia lacks financial
accountability. Significant variations were found in the level of disclosure
reflecting accountability by government entities.

From the regression test undertaken, size in terms of assets and revenue were
found to be insignificantly associated with the level of disclosure reflecting
accountability. On the other hand, long term debt and form of government in
terms of level of government were found to be significantly associated with the
5
Research Methodology
Halilah Haron, March 2014
 level of disclosure reflecting accountability. On the whole, the results provide
support for relationships between long term debt and form of government
determinants and the level of disclosure reflecting accountability.

Example 4:

A descriptive analysis of IT adoption in Malaysian small software

firms

This paper reports the progress of a research that focusses om information technology (IT)
adoption in Malaysian small software firms. The research provides insights into the
characteristics of small software firms and the types of technology adopted in such firms …..

Example 5:

Consumer Perceptions Towards Mobile Entertainment

Abstract

Mobile commerce is widely applied in the age of technological advancement today due to the
widespread use of the mobile devices. Perception among the users of mobile devices towards
mobile commerce and adoption of mobile commerce vary from country to country. This study
concerns about the perspective and expectation of Malaysian mobile device users towards
mobile commerce. The study is important for the development of the literature regarding
mobile commerce in Malaysian context as well as contributing to the improvement of service
for companies that adopting mobile commerce.

The study reviews the literature on consumer perception of mobile commerce with the 4
classification of concept in our proposed theoretical framework. A conceptual framework is
built by linking the variables that have connection with the consumer perception of mobile
commerce. The variables are perceived security and privacy, sense of „always on‟ and
convenience, ability of mobile devices to support mobile application and familiarity of
consumers. An empirical study is done on the various articles that related to this topic. 4
hypotheses generated from these relationships are to be tested with the data collection and
analysis.

The nature of the study is descriptive research. Data will be collected using self-administered
Internet survey. Convenience sampling method will be used for data collection. Variables are
measured using the interval and ordinal scale. Descriptive analysis will be the data analysis
method of this study. The characteristics of mobile users will be drawn through inferential
statistics and chi-square will be used to analyse the differences of characteristics between age
groups and genders.

6
Research Methodology
Halilah Haron, March 2014
More examples of descriptive research can be seen on page 98 of textbook under
EXAMPLE.

Consumer

Example 6:

An IRPA study on “Emerging Youth Culture” conducted in 1996-1997 by UPM found

evidences of emerging cultural values and patterns (i.e., communication behavior and media
habit, leisure activity, health care habit, individualism and materialism value, religious
orientation, orientation toward foreign culture, identity, patriotism, and career aspiration) along
with the economic progress (see e.g., Azimi, Turiman & Ezhar, 1999). The patterns and trends
of youth cultures are of fundamental concern to policy makers and development agents as it has
consequences on the attainment of the 2020 vision. As such, the country must continuously and
accurately monitor the changes in the cultural values and behavior of Malaysian youths through
systematic indexing of cultures of the various youth groups, and accordingly use the database
to identify challenges and, hence, plan appropriate policy and development of intervention
programmed for youth.

A Objective of this study is:

1. To identify the cultural values and behavioral patterns of Malaysian youth,
2. To determine similarities and differences among youth groups with regard to their
cultural values and behavioral patterns,
3. To identify trends and challenges in contemporary youth culture and
4. To identify the implications of the contemporary youth culture on the approaches of
youth development towards the attainment of vision 2020.

This study will be longitudinal study employing quantitative research methodology. A survey
questionnaire will be developed which is actually a refinement and an expansion of the
instrument used in the “1996-1997 Emerging Youth Culture Study”. The instrument will be
pre-tested with 5% of the sample size to determine its reliability. The responses from the pre-
test will help the further refine the instrument before it is use to collect data from the sample.
Penang, Selangor, Pahang and Johor, representing 4 zones (North, East, West and South of
Peninsular Malaysia). Purposive sampling will be used to ensure sample representativeness in
terms of age group (15-19 year, 20-25 year and 26-30 year), locality (urban and suburban),
race and gender. The data will be collected using self-administered questionnaire, and the
questionnaire will be administered through group approach. There will be two phases of data
collection. The second phase of data collection will be collected a year after the first phase of
data collection. Data will be analyzed after each phase of data collection. SPSS will be used
for data analysis.

M
Question: What makes this study longitudinal?
le Entertainment

7
Research Methodology
Halilah Haron, March 2014
Other Study Designs

Cohort Study

 Also known as panel study.

 A form of longitudinal study used in medicine and social science because the cohort is
followed through time to assess their later outcome status.
 A cohort is a group of people who share some common characteristics or experience
within a defined period. Examples of cohorts: those who are born in 1988 form a birth
cohort, those who leave school in 1976, etc
 Design of choice for studying the risk factors.

Example: A cohort study would be the investigation of a cohort of smokers and non-smokers over
time to estimate the incidence of lung cancer.

Example: Cigarette Smoking as a Risk Factor for Stroke Death in Japan –

… A total of 9638 men and women aged 30 years and older without a history of
cardiovascular disease (CVD) at baseline in 1980 were followed-up for 14 years…
(Ueshima, et al, 2004)

Case-control study

 Subjects with the disease (case) is matched with subjects without the disease (controls)
according to matching variables such as age, location. This design is very common in
clinical trials.

 A form of longitudinal study.

Example: Association Between Dietary Fibre and Cancer: A Case-Control

Study in Malaysia – A case-control study to determine the association of dietary fibre
and cancer was conducted among 100-newly diagnosed cancer patients admitted to the
Radiotherapy and Oncology Ward, Hospital Kuala Lumpur. A total of 100 controls
matched with the cases for age, sex and ethnic origin were selected. …

Example: A Case-Control Study on The Association Between Environmental

Factors And The Occurance of Acute Leukemia Among Children In
Klang Valley - … A case-control study was conducted for cases from the Klang Valley
who received treatment at the Hospital Universiti Kebangsaan Malaysia (HUKM) and
Hospital Kuala Lumpur (HKL). The main objective was to determine any association with
environmental factors. 128 case subjects were children aged below 15 years and diagnosed
with acute leukemia in HUKM and HKL between Jannuary 1, 2001 and May 30, 2007. 128
control subjects were children aged below 15 years who were diagnosed with any non-
8
Research Methodology
Halilah Haron, March 2014
 cancerous acute illnesses in these hospitals. All factors were analyzed using unmatched
logistic regression. The analysis showed that the occurance of acute leukemia was strongly
determined by a number of factors including the distance of the house from a power line.
(Abdul Rahman et. al., 2008)

Before-After studies

The outcome is observed on the same individual on two separate occasions, under different
exposure (or treatment) circumstances or using different methods.

Pair-match studies

The units considered in the study are not individual items but pairs of individuals selected
according to common characteristics such as age, gender, etc. The pairs may be in the form
of natural pairs such as twins and siblings. Pair-matching is common in cohort studies.

5.3 Experimental Design

Causal Research

 Also known as causal study.

 Designed to provide information on potential cause-and-effect relationships.
 Most practical in marketing to talk about associations or impact of one
variable on another.
 May be Field Experiment or Laboratory Experiment. Field experiments are
studies conducted to establish cause-and-effect relationships in a natural
(non-contrived) environment. Laboratory experiments are experiments
conducted to establish cause-and-effect relationships in an artificial
(contrived) environment.

Example 7a: A field experiment

Mosquito trial to cut dengue infection gets go-ahead in Malaysia

12 November 2010

9
Research Methodology
Halilah Haron, March 2014
The Malaysian government has granted approval to release genetically modified sterile
mosquitoes into the wild in an open field trial. Supported by a Wellcome Trust Translation
Award, Oxford-based biotechnology company Oxitec is hoping that its method of controlling the
mosquito population will cut the spread of dengue infection.

Dengue is a potentially fatal disease that is passed on from the bite of an infected female mosquito
of the ....

Approval for the Malaysian trial follows the success of a smaller-scale trial conducted in the
Cayman Islands earlier this year, which resulted in a significant reduction in the local mosquito
population. .....

Scientists released batches of the sterile male mosquitoes in cages, three times a week over a 16-
hectare area between May and October of this year. By August, mosquito numbers in the trial area
had dropped by 80 per cent compared with a neighbouring area where no sterile mosquitoes were
released …….

Example 7b:

Jan 4, 2011
Malaysia delays landmark GM mosquito trial after protests

The Aedes albopictus mosquito that spreads dengue fever. -- ST PHOTO

KUALA LUMPUR - MALAYSIA has delayed a landmark field trial to release genetically
modified mosquitoes designed to combat dengue fever, an official said on Tuesday, following
protests from environmentalists.

In the first experiment of its kind in Asia, 4,000-6,000 male Aedes aegypti mosquitoes were
originally scheduled to be released by last month in a bid to fight dengue, which killed at least 134
people last year in Malaysia alone ….

10
Research Methodology
Halilah Haron, March 2014
Example 7c:

Stop risky GM Aedes test

Athi Shankar
| January 17, 2011

Two NGOs warn the government that releasing the genetically modified mosquitoes can court
disaster.

GEORGE TOWN: Two Penang-based NGOs have called on the National Biosafety Board to
revoke the approval issued to the Institute for Medical Research (IMR) to test run the genetically
modified (GM) male Aedes aegypti mosquitoes.

In making this call, president of Consumer Association of Penang (CAP) and Sahabat Alam
Malaysia (SAM), SM Mohamed Idris said the field experiment project, which could involve some
24,000 deadly insects to be released in natural environment, was a risky and dangerous enterprise
…..

Example 8: A lab experiment

Some factors influencing student performance in laboratory experimental

work in physiology with implications for curriculum deliberations and
instructional design
(1) The Department of Physiology, The Medical Faculty, University of Malaya, 59100 Kuala
Lumpur, Malaysia

Abstract A class of 1st-year medical students was divided into a control group who performed a
laboratory experiment for themselves, and an experimental group who were given a
demonstration data interpretation exercise. Prior to the practical session the experimental group
were given a programmed text covering the underlying theoretical principles, to use in their
private study. The control group were instructed to do their private study in the usual way using
only their lecture notes and textbooks. Prior knowledge and the gain resulting from practical

11
Research Methodology
Halilah Haron, March 2014
instruction were assessed using pre- and post-tests.

The conventional laboratory method was ineffective and did not facilitate conceptual
understanding. The post-test performance and gain of the experimental group was significantly
better than that of the control group, but this result did not appear to be dependent upon greater
prior knowledge. The pretest performance of both treatment groups revealed that many students
were inadequately prepared for the practical exercise. Many students did not appear to study the
programmed text conscientiously, probably because they were accustomed to the traditional
method of teaching and adopted a surface, reproductive approach to all learning. Furthermore, the
students had difficulty linking their prior knowledge to the experiment, suggesting the need for
pre-laboratory exercises designed to achieve closer integration between theory and
experimentation.

The differences between field study, field experiment and lab experiment can be seen in
on pages 100 and 101.

5.4 Extent Of Researcher’s Interference

Minimal - Correlational studies which are conducted in natural settings require minimal
researcher interference.

Moderate & Excessive – Studies conducted to establish cause-and-effect relationships

involved varying degrees of the researcher interference with the natural or controlled
settings. This is mainly due to the manipulation of the variables during the process of the
experimentation.

5.5 Unit Of Analysis

The unit of analysis refers to the level of aggregation of the data collected during the
subsequent data analysis stage. Issues of units of analysis will evolve from the research
questions.

12
Research Methodology
Halilah Haron, March 2014
 Unit of
analysis

Individuals Groups Dyads Organization

s

Example 9: Unit of analysis – individuals

Refer to Example 8. Each medical student in the control and experimental group from.

Example 10: Unit of analysis – groups

Refer to the example as above. Let‟s say that the study is interested to compare if there
are differences in organizational agility between small & medium manufacturing firms and
large manufacturing firms.

Unit of analysis – small & medium manufacturing firms and large manufacturing firms in
Malaysia ( two groups)

Example 11: Unit of analysis - dyads

Building Quality International Channel Partner Relations – A Dyadic

Perspective
Kathy Hastings, Les Brown, Eric Ng, Frances Cassidy, University of Southern Queensland

Abstract

This paper investigates how Australian exporters and importers in Japan and Malaysia rate
their current business relationships and areas that need improvement. These areas where
investigated in terms of the characteristics of the five states of business-to- business
relationship development from the literature. A series of 25 in-depth interviews with Asian
importers and Australian exporters about what constituted satisfactory trading relationship
development from a dyadic perspective revealed that unless the early stages have been
properly completed then consequent dissatisfaction in such relationships can be attributed
primarily to communication problems emanating from neglect in the early stages.
13
Research Methodology
Halilah Haron, March 2014
 Conversely, excellent relationships exhibited characteristics such as trust, commitment,
share goals and co-operation.

Becker and Useem (1942) said that two persons may be classified as a dyad when
intimate, face-to-face relations have persisted over a length of time sufficient for the
establishment of a discernable (distinguish) pattern of interacting personalities. Hence
dyadic studies analyses interpersonal relationships. Dyads can be in the form of husband-
wife, supervisor-subordinate,

Example 12: Unit of analysis – organizations

The relationship between information technology acceptance

and organizational agility in Malaysia

Abstract
We examined the influence of information technology (IT) acceptance on
organizational agility. The study was based on a well-established theoretical
model, the Technology Acceptance Model (TAM). We attempted to identify the
relationships between IT acceptance and organizational agility in order to see how
the acceptance of technology contributes to a firm's ability to be an agile
competitor. Structural equation modeling techniques were used to analyze the data.
Results from a survey involving 329 managers and executives in manufacturing
firms in Malaysia showed that actual system or technology usage had the strongest
direct effect on organizational agility. Meanwhile, perceived usefulness and
perceived ease of use of IT influenced organizational agility indirectly through
actual systems or technology use and attitudes towards using the technology.
The results have several implications for IS management.

Keywords: IT acceptance; IT adoption; Organizational agility; Structural equation

models; Malaysia

ASSIGNMENT:

1. Read the EXAMPLE given on page 105 of the textbook.

2. Do EXERCISE 6.1 on page 108 of the textbook.

14
Research Methodology
Halilah Haron, March 2014