Research Methods I
Introduction to Scientific Research
The Scientific Method
The goal of any scientist is to discover new information about the way the
Psychologists collect data to test hypotheses that help us understand human
thought and behaviour.
The scientific method provides a seven-step “recipe” for how to collect and
analyze information while trying to minimize biases, conflicts, or oversights.
Scientists begin by studying the existing collection of information about the
world, such as previous work published by other scientists.
Theory General set of ideas about the way the world works.
Hypothesis Testable statement guided by theories that make specific
predictions about he relationship between variables.
Research Method The way in which the hypothesis will be tested.
Collect Data Taking measurements of the outcomes.
Analyze Data Understand the data and discover trends or relationships
between the variables.
Report Findings Publish in scholarly journals.
Revise Theories To include new information into our understanding of the
world. Every so often, there is a particularly dramatic change in our way of thinking
called a paradigm shift.
Because science and research are dynamic processes, theories are
continually being revised to account for new information.
Conducting an Experiment
Testing A Hypothesis
A hypothesis makes a testable prediction.
You may initially look for anecdotal evidence to support or refute the
Anecdotal evidence evidence gathered from others or self experience.
Single experience might not be representative of the general result that
would occur if you tested the hypothesis multiple times.
Personal experience might not represent others. We all experience things
differently and react in different ways.
Scientists often use the experiment to measure the effect of one variable on
Experiment scientific tool used to measure the effect of one variable on
Independent variable variable manipulated by scientist.
Dependent variable variable being observed.
Using Control Groups
A well-designed control group is a critical aspect.
An experiment contains 2 groups of participants:
o Experimental Group will receive the manipulation of the
o Control Group will NOT receive manipulation.
We can then compare the dependent variable measure for both groups.
Participants in the experimental and control groups should be as similar as
possible, differing only in the independent variable.
Within Subjects Design
One way to guarantee the experimental and control groups are as similar as
This technique tests the same subject repeatedly while the independent
variable is manipulated.
This design minimizes the effect of subject differences on the dependent
measure (external variables). Problems: Because the same participant needs to be tested repeatedly, it can
be time consuming and costly. Also, the measure we are using or the subject
himself may change.
Practice effect improved performance over the course of an experiment
due to becoming more experienced.
Between Subjects Design
One group of subjects receive experimental manipulation.
One groups acts as a control group.
It’s important for those individuals being compared in a between subjects
design to be as similar as possible in every way except our manipulation of
the independent variable.
A systematic difference between the participants other than the independent
variable is a confounding variable.
You have to be careful how you select participants for the experiment if you
want to eliminate confounding variables.
The general group of people we are trying to learn about is called the
population, while the selected members of the population that we actually
collect data from is called the sample.
The sample must accurately reflect the population itself so that the results of
the experiment can be generalized.
o Choosing a sample at random from the entire population.
o This reduces the chance that our selections might be biased towards a
o Assigning subjects to either the experimental or control group at
random to avoid any biases that may cause differences between the
groups of subjects.
Conducting an Experiment
The placebo effect cannot be completely explained using modern psychology.
The placebo effect occurs when an individual exhibits a response to
treatment that, in reality, has no related therapeutic effect.
This effect can be demonstrated by some cases where patients have shown
remarkable recovery from illness when given drugs that were presented as
“miracle cures”, even when these drugs were known to have no effect. We must always consider the placebo effect when designing any experiment
where participants might know in advance the expected result of the
The placebo effect is a form of participant bias, which can intentionally, or
unintentionally, influence the results of the experiment.
Solution: Give the control group a mock treatment.
If the participants do not know whether they belong to the experimental or
control group, the experiment is said to be blind.
It is also possible that a researcher may try to influence the results of an
experiment, intentionally or unintentionally.
As experimenters know the hypothesis that they are attempting to test, it is
possible that experimenters may promote the result they hope to achieve.
We can reduce the experimenter bias if the experimenter hims