MODULE FOUR

RESEARCH DESIGNS
 OVERVIEW
• Explain the term research design
• Outline the basic components of a research design
• Discuss selected research designs associated with quantitative
• Discuss selected research designs associated with qualitative
approaches to research

• Determine an appropriate sample size for a research project

 WHAT IS A RESEARCH DESIGN

• A research design is a plan, structure and strategy of investigation

so conceived as to obtain answers to research questions or
problems.

• A set of advance decisions that make up the master plan

specifying the methods and procedures for collecting and
analyzing the needed information.
 PURPOSE

• The identification and/ or development of procedures and

logistical arrangements required to undertake a study

• Ensuring that the procedures are adequate in quality to obtain

valid, objective and accurate answers to the research questions.
 MAIN COMPONENTS

• A description and justification of the research design

• Research Method
• Population and Sampling Strategy
• Settings for the research study
• Data Collection Techniques and Tools
• Procedures
• Ethical Considerations
 BROAD CATEGORIES

Quantitative Qualitative
RESEARCH DESIGN BY PURPOSE

Descriptive Research Design

Correlational Research Design

Experimental Research Design

Diagnostic Research Design

Explanatory Research Design

RESEARCH DESIGN TYPES BY GROUPING

Cross- Longitudinal
sectional Cohort study study
study
 What is my research problem?

DETERMINE Does my research questions fall within the

qualitative or quantitative paradigm?
THE
RESEARCH Which tradition of the paradigm best answers
my question?
DESIGN
Who are my participants and what are their
demographics?
What might be the best method to collect data
for this problem and group of participants?
 TYPES OF RESEARCH METHODS
Correlational study

Experimental

Research and Development (R & D)

Survey

Action research

Case study

Ethnographic

Historical research
 QUANTITATIVE QUALITATIVE
• Correlational study • Action research
• Experimental • Case study
• Research and • Ethnographic
Development (R & D) • Historical research
• Survey
QUANTITATIVE
RESEARCH
METHODS
 SURVEY
Generally used to describe a phenomenon
▪ Sample survey – data collected from a sample of the
population.
▪ Census survey – data collected from each member of the
population

• In surveys, the group of participants is important.

• Quantitative approaches tend to be used: - structured
questions, scales.
• No manipulation of variables.
• Data are analysed statistically.
 TYPES OF SURVEYS
CROSS-SECTIONAL SURVEY
Involves collecting data from people in a single period.

To carry out a cross-sectional study:

• Identify, define and operationalise the characteristics your want to
study;
• Acquire an instrument that would help you to collect the data you
need;
• Identify and select appropriate sources of data; (Sampling)
• Administer the instrument;
• Compile, analyse the data and interpret the findings;
• Draw conclusions.
 Types of Surveys: Longitudinal Survey

Trend Survey Cohort Survey Panel Survey

• Collects data • Collects data • Collects data

from samples from a sample from the same
from the general from a specific sample over time
population at population at
different periods. different times.
The same
individuals are
not used, but
ones with similar
characteristics.
 CORRELATIONAL STUDY

Correlational studies explore associations between two or more

variables, without any attempts to manipulate them

The researcher is looking for the degree to which these variables

are related

Relationships may be strong or weak; positive or negative

 RESEARCH AND DEVELOPMENT

R & D involves two major types of research by purpose.

• It combines systematically both basic and applied research and aims at

discovering solutions to problems or creating new educational products.

• It involves researching the market and the learners’ needs and developing new
and improved products and services to fit these needs.
 EXPERIMENTAL
Concerned with the manipulation of conditions to investigate cause-effect relationships.

• The researcher manipulates one condition (called the independent variable) and observes
the effect on some other characteristic (called the dependent variable).

• To ensure that the observed effect is caused by the manipulation of the independent
variable, the researcher has to strictly control other conditions (extraneous variables) that
may also have an effect on the dependent variable.

• When extraneous variables cannot be controlled, the researcher may either build them into
the study OR find a means of factoring them out.

• Usually involves hypothesis testing.

Characteristics of Experimental Research

• Manipulation of at least one variable

• Measurement of outcomes on another variable

• Use of equivalent groups

• Random assignment of participants to groups (often also random selection)

• Comparison of two or more groups or sets of conditions

• Heavy reliance on statistical procedures (inferential statistics)

• Maximum control of extraneous variables (participant and environmental)

Validity of Experimental Research

INTERNAL VALIDITY
Experimental research is valid if the results obtained can be
attributed to the manipulation of the independent variable

EXTERNAL VALIDITY
the results can be generalised to people or contexts outside of
the experimental setting
Threats to Internal Validity (Campbell & Stanley, 1963)
History – Occurrence of events not part of the experiment that can affect the outcome

Maturation – Physical or mental changes in the participants over time that may affect the outcome

Testing – improved scores on post-test as a result of participants being pre-tested (test wiseness)

Instrumentation – unreliability of measuring instruments can produce invalid assessment data

Mortality (attrition) – participants fall out of the study in ways that may alter the characteristics of the
experimental group

Interaction of selection and other factors – if existing groups are used, one group may profit more
from (or less) from a treatment or have an initial advantage (or disadvantage) because of maturation,
history, or testing factors.
Threats to External Validity

Pre-test treatment interaction – participants perform or react differently to the

treatment because of having been pre-tested

Post-test sensitisation – treatment effects influenced by the pre-test (pre-test

provides info that affects post-test results

Multiple-treatment interference – when same participants receive more than one

treatment in succession and carry over effects from an earlier treatment influence
results of a later treatment

Specificity – treatment variables not properly operationalised making it unclear to

whom the results may be generalised

Experimenter’s bias or expectations – when the researcher affects participants’

behaviour or is unintentionally biased when scoring different treatment groups.
 • Pre-experimental: There are no controls for threats
to internal validity.

• True Experimental: involve random assignment

TYPES OF of participants to treatment groups. Control for
EXPERIMENTAL threats to internal validity (may also involve
DESIGNS random selection)

• Quasi Experimental: There is no random

assignment to groups. Intact groups are involved.
Control for some threats to internal validity.
PRE EXPERIMENTAL DESIGNS

One-shot Case Design

• One group exposed to treatment and post-tested (X – O).
No threats controlled.

One-group Pre-test – Post-test Design

• One group is pre-tested, then exposed to treatment and
post-tested (O – X – O). No threats controlled.

Static-group Design
• At least 2 groups;
• one receives new (unusual) treatment,
• both groups post-tested.
• No random assignment, no pre-test data; difficult to
determine how equivalent the groups are.
X = Treatment condition
O = Observation (data collected)
TRUE EXPERIMENTAL DESIGNS

Post-test Only Control Group Design

R X1 O
R X2 O
Pre-test – Post-test Control Group Design
R O X1 O
R O X2 O

Solomon Four-group Design

R O X1 O
R O X2 O
R X1 O
R X2 O

R indicates Random Assignment of participants to the treatment or control groups

QUASI-EXPERIMENTAL DESIGNS (NO RANDOM ASSIGNMENT)

Non-equivalent Group Designs

O X1 O
O X2 O

Time Series Design – One group

OOO XOOO

Counterbalanced Design – Several groups; all groups receive all

treatments but in a different order

X1 O X2 O X3 O
X3 O X1 O X2 O
X2 O X3 O X1 O
SINGLE-SUBJECT DESIGNS
Used to study changes in behaviour as a result of some
treatment.

A–B [O O O O | X O X O X O X O ]
A B

A – B – A [O O O O | X O X O X O X O | O O O O ]
A B A

A = Baseline measurement period

B = Treatment period
 VALIDITY
&
RELIABILITY
VALIDITY - the extent to which a concept is accurately measured

in a quantitative study

RELIABILITY - the accuracy of an instrument. The extent to

which a research instrument consistently has the same results if it is

used in the same situation on repeated occasions.

Source:
[Link]
thodology/reliability-vs-
validity/
Source:
[Link]
thodology/reliability-vs-
validity/
 QUANTITATIVE SAMPLE SIZE

There are a few variables to be aware of before working out the right sample

size for your project.

• Population size

• Confidence interval (margin of error)

• Confidence level

• Population variability

• Project scope