Vous êtes sur la page 1sur 12

Faculty of Applied Engineering and Urban Planning

Civil Engineering Department

Data Collection
(Quantitative Methods)

Research Methodology

UP Copyrights 2012

Quantitative Research

The following are the main features of Quantitative


research:

Deductive:
deductive

quantitative
nature

as

research

it

tests

has

theory

or

hypotheses to provide evidence for or against


this theory.

Objective:

as

quantitative

research

seeks

precise measurement and analysis of target


concepts, e.g., uses surveys, questionnaires
etc.

Quantitative research usually begins with prespecified

objectives

focused

on

testing

Quantitative Research Methods

Major types of quantitative research


methods are:

Experimental.

Quasi-experimental.

Surveys.
3

Experimental

Experimental studies are quantitative research


methods used to study a certain phenomena,
mainly by examining cause and effect.

Experiments usually designed to examine the


influence

of

an

independent

variable

(e.g.

treatment) on a dependent variable.

The

main

experimental

issue
study

when
is

to

conducting
control

as

an
many

aspects of the situation as possible; such as


participants and the environment in which the
experiment is conducted.

Laboratory studies are examples of this type of

Design of experiments
Design of experiments is the design of
any task that aims to describe or
explain the variation of information
under conditions that are hypothesized
to reflect the variation.

Experimental Parameters
Fixed

All the
parameters of
the experiment
except the
ones that are
of concern

The
Controlled Measured

The
influencing
parameters
to be
studied

parameters
that are
dependent
on the
controlled
ones

Example: What is the effect of fiber glass on the


flexural strength of concrete

Experimental Parameters
Fixed

Temperature
Water/Cement
Ratio
Aggregate
Additives
Beam size
etc ...

Controlled Measured

Ratio of
Flexural
fiber
Strength
glass

Fisher's principles

Comparison
it

is

not

possible

to

have

independent

measurements to a traceable metrology standard,


and often compared against a scientific control or
traditional treatment that acts as baseline.
Example: Improvement of groundwater due to the application
of infiltration methods

Randomization

Such as in Quasi-experimental (will be discussed


later on)
8

Fisher's principles

Statistical replication

Measurements are usually subject to variation


and measurement uncertainty; thus they are
repeated and full experiments are replicated
to help identify the sources of variation, to
better

estimate

treatments,

to

the
further

true

effects

strengthen

of
the

experiment's reliability and validity, and to


add to the existing knowledge of the topic.
9

Fisher's principles

Blocking

Blocking is the arrangement of experimental


units

into

groups

to

exclude

unnecessary

parameters.

Factorial experiments

Use of factorial experiments instead of the onefactor-at-a-time method. These are efficient at
evaluating

the

possible interactions of
(independent variables).

effects
several
10

and
factors

Quasi- Experimental

Quasi-experimental studies are similar to


experimental

studies

except

that

the

researcher does not have full control over


the situation in which the experiment is
conducted.

Quasi-Experimental methods of research


are commonly applied in health related
studies; where it is difficult to allocate
11

participants to different groups due to

Surveys

The most popular method of data collection used in


quantitative research is the survey.

The survey technique involves the collection of


primary data about subjects through a questionnaire.
The survey is usually conducted on a particular
population to collect certain type of data.

Surveys could be conducted;

via telephone,

self-administration, or

face-to-face surveys.

12

Vous aimerez peut-être aussi