Psychology 1000 Chapter Two (Studying Behaviour Scientifically)
Festinger-Cognitive Dissonance ($1v $20)
Scientific inquiry (question) usually follows five steps:
Identifying a question of interest
Gathering information and creating a hypothesis.
Testing the hypothesis through experiments, research, etc.
Define independent/dependant variables. (Independent changes dependant variables,
independent is manipulated)
Analyzing the data, drawing tentative conclusions and report findings
Build a body of knowledge (create theories, create new hypotheses, etc.)
Two Approaches to Understanding Behaviour
There are two approaches to understanding behaviour: hindsight and the step-by-step approach. Most
scientists choose to ignore hindsight and stick with the step-by-step approach.
The problem with hindsight is that multiple reasons could be used (in hindsight) to explain a
possible outcome of any event.
There is also no way of determining :
a) if any alternate outcomes are possible, and
b) which one is correct
Scientists prefer using the step-by-step approach because:
They can control certain variables that may cause an unexpected outcome
They can predict what will happen, given they understand the cause of a certain
Theory development is the strongest test due to:
It organizes all necessary information in a meaningful way
It is testable
Predictions made by the theory are supported by new findings from research
It conforms to Occam’s Razor (The simplest explanation is usually the most correct one)
Defining and Measuring Variables
A variable is a factor within an experiment that is different from individual to individual, could change and
alter the outcome.
o Many variables that scientists study cannot be observed directly, such as IQ, stress, happiness, etc.
o Since scientists cannot directly observe certain variables, scientists must use operational
definitions to define a variable; translating abstract variables into something that can be
o All external variables must be controlled, or else the experiment is confounded.
Due to the fact that psychologists study complex and varied processes, they use a variety of methods to
measure such variables:
o Self-reports (asking the individual to report on things such as beliefs, feelings, experience,
However, self-reports may be altered due to social desirability bias, the desire to
conform to social norm, instead of answering what the individual would; how she is truly
feeling or behaving, etc.
To counter this, questionnaires may be administered to people who know the individual
o Overt behaviour (directly observable behaviour) Observers use a coding system so that the measurements stay constant and reliable.
Because organisms may act differently when they know they are directly being observed,
researchers may disguise their presence or use unobtrusive measures to prevent the
individual from observing the researcher.
Researchers may also use research that already exists (archival measures) to conduct
Psychologists create tests to measure many variables, e.g. an IQ test. As well, psychologists measure
physiological effects as well, such as perspiration, heart rate, as well as brain function.
Methods of Research
Descriptive research tries to find out how/why an individual behaves, usually in its natural settings;
providing information about the variety of behaviour someone/thing chooses to exhibit and could bring out
cause-effect relations that could be used later in research.
Case studies are in-depth analyses of things, usually an individual, event or group.
o In-depth study of one individual (e.g. Luria’s ‘S’ The Mind of a Mnemonist, Breuer’s ‘Anna O’)
Naturalistic observation (observational study) is where the researcher (duh) observes behaviour in a
natural setting, attempting to avoid influencing behaviour in any way possible.
o However, natural observation does not provide any conclusions to why an individual exhibits the
o Sometimes, the researcher gets involved (Festinger’s When Prophesy Fails)
Survey research (survey method) is where (duh again) a survey is given, asking questions to individuals,
usually given to a large sample population at once. (e.g. Kinsey’s report of sexual behaviour)
o Population is the group of individuals that a researcher is interested in studying, where a sample
is a small group of individuals chosen from a population
o While surveys are a good way of massively conducting research, one must be careful of the
Internet based research is challenging due to the fact that individuals taking the
questionnaire can manipulate/lie about information, providing researchers with flawed
Also, surveys cannot be used to conduct cause-effect conclusions.
To truly reflect upon the important characteristics of a group, a representative sample is often used, and
random sampling is used to ensure that the research being conducted is truly random; where every
member of a population has the same odds of being chosen.
Many scientists also preform correlation research (correlation method) to determine if a has any
relationship to b. (e.g. relation between hours of TV viewing and grades)
o The researcher first chooses two variables s/he wishes to observe and measures both of them. Then,
through statistics, they determine whether the two variables have a correlation between them.
o However, correlation research cannot pinpoint a cause-effect relationship, just like survey based
methodology, e.g. If a certain study finds that happy people have high IQs, the study cannot
determine (by itself) which one