Research Strategies/Methods

Experimentation
 Defining experimentation
• Using a different route to work resulting in cutting down
travelling time by 10 minutes. Conclusion new route is
quicker; is it?
• An experiment investigates cause-effect relationships,
seeking to prove or disprove a causal relationship
between a factor and an outcome. This is a common
approach in physics, chemistry, metallurgy and is based
on positivism thinking
• Therefore an experiment is a particular kind of research
strategy that aims to isolate cause and effect by
manipulation of what is thought to be the causal or
independent variable and measurement of its effect on
the dependent variable (s)
 Characteristics of an
Experimentation
• Observation and measurement: make precise and
detailed observation of outcomes and changes that
occur when a particular factor is introduced or removed
• It is a process that involves: 1. observation and
measurement 2. manipulation of circumstances 3. re-
observation or re measurement of factors to identify
changes
• Disprove or prove relationship
• Identification of causal factors
• Explanation and prediction
• Repetition under different condition
 Planning and conducting an
experimentation
• 1.Develop a hypothesis: this must be in such a way
that it is subject to disapproval or falsification
• Dependent and independent variable must be
distinguished
• 2.Controls identified : this must be planned to deal with
variables that might affect outcome which include;

– Eliminate externuating factors from the experiment eg collecting

samples from a borehole or from a pool
– Hold the factor constant
– Use random selection of subjects
– Use control groups
– Make researchers and subjects blind; ie use of placebo in the
medical research
 Planning and designing an
experimentation cont’
• 3.Observation and measurement: Typical
things that are observed include;

– Establish pre-test condition to be compared with post-

test
– Project data maintained, person hrs, time, date, temp
etc
– Behavioural counts; the number of times a certain
behaviour occurs
– Quantify change to enable statistical analysis
 Planning and designing an
experimentation cont’
• 4.Internal validity: this implies that changes that are
observed are solely as a result of the manipulation of the
independent variables. Therefore threats to internal
validity include

– Difference between the experimental and control group

– Intervening events
– Maturation eg dealing with children
– Instrument becoming faulty
– Experimental mortality or drop out
– Reactivity and experimenter effects: power relationship between
researcher and subject or trying to help the researcher by
providing sound data
 Planning and Design
experimentation cont’
• 5.External validity: this is the ultimate goal of
research to generalise results to allow prediction
for subsequent occasion and other situation.
Careful designing and repeating the experiment
several times can achieve that; threats to
external validity are :

– Over reliance on special type of participants eg

students, volunteers or mice or chimpanzees
– Too few participants: results becoming statistically
insignificant
– Non representative participants
 Typical experiment
• Exercise: Smoking causes cancer

• Describe the variables and explain how

you would conduct the experiment taking
into consideration the ethical issues
 Wynder et al exp on smoking
• In the mid-1950s, Ernest Wynder and colleagues had an
ingenious idea: they condensed the chemicals from
cigarette smoke into a liquid and applied this in various
doses to the skin of groups of mice. The researchers
published data from a dose-response experiment of the
effect of tobacco-smoke condensate on mice (Wynder et
al., 1957). The researchers found a positive relationship
between the amount of condensate applied to the skin of
mice and the number of cancers that developed. This
study was one of the first pieces of experimental
evidence in the cigarette smoking debate, and it helped
strengthen the case for cigarette smoke as the causative
agent in lung cancer in smokers
Results of experiment
 Quasi – or Field experiment
• This compares to the true lab based
experiment where researcher has control
over the variable
• Quasi means ‘as if’ and takes the form of
field experiments
• Researcher must establish the experiment
and control group and try to observe
outcome
 Type of experimental design
• One group, pre-test and post test;
measure before apply treatment measure
post treatment
• Static group treatment: two groups,
experiment and control
• Pre-test post-test control group: pre-test to
both, and then treatment to one followed
post test to both
 Evaluating Experimentation based
Research
• Advantages
• They are well established strategy viewed by
many as scientific
• They are the only research strategy that can
prove causal relationship
• Lab experiments permit high level of precision
and measuring of outcome and analysis of data
• Lab based experiments allow researchers to
remain at their base avoiding cost of travelling to
visit field sites: however this is unavoidable in
Science
 Evaluation continued
• Disadvantages
• Lab experiments create artificial situations, which are
not commensurate with real world situation, carried
out in an environment outside the usual place where
the subject usually live
• It is often difficult or impossible to control all the
relevant variables
• It is often difficult to recruit a representative sample
• It might be necessary to conceal from the participants
the purpose of the research so that they do not
perform accordingly but deception is viewed as
unethical
• Small number of subjects can be involved
• Own perception can influence the observation hence
the results
– End