Textbook Notes (280,000)
CA (160,000)
McGill (5,000)
PSYC (1,000)
PSYC 215 (300)
Chapter 2

PSYC 215 Chapter Notes - Chapter 2: Hindsight Bias, Participant Observation, Dependent And Independent Variables

Course Code
PSYC 215
John Lydon

of 2
Chapter 2 The Methods of Social Psychology
Social Psychology
Hindsight bias: people’s tendency to be overconfident about whether they could
have predicted a given outcome; we feel more confident once we know the results
of a study
Hypotheses: predictions about what will happen in given circumstances; test
circumstances within a theory
Theory: a body of related propositions intended to describe some aspect of the
Studies with participant observation involve observing something from a close
range, often becoming involved in it yourself
Researchers can study patterns of behavior in the past using archives
o i.e. it was discovered that more homicides in the South were insult-related
Surveys are the most common type of study because they involve using a random
sample, when everyone in a population has an equal chance of being selected
Convenience samples, or choosing a group base on their convenience is not
random and can be biased
Cohen and Nisbett found Southerners had conflicting views on violence, but
generally thought violence in response to insult or in defense was justified- may
be due to a “culture of honor” in the South”
Correlational research: determining whether a relationship exists between two or
more variables
Experimental research: enables researchers to make inferences about how
different conditions or situations affect people’s behavior
In correlational research, causality is never assured; there could be a third variable
or reverse causation, when the opposite of what is expected is true
Correlational research cannot prove a causal relationship because of self-
selection, when the participant, rather than the researcher, selects his or her own
level on each variable being tested, therefore affecting the relationship between
Relationship between variables are on a scale from 0 (no relationship between
variables) to 1 (perfect covariation)
Longitudinal studies study subjects over a long of time
Independent variable: the variable that is manipulated
Dependent variable: the variable that is measured and hypothesized to have been
affected by a causal relationship with the independent variable
Random assignment: used to test levels of independent variables by randomly
assigning participants to groups
Control condition: the condition to which the manipulated is compared
External validity: the validity of findings of a study in real life situations and
ability for them to be generalized to such situationsnot always fatal
Best way to test external validity is to do field research, when an experiment it set
up in the real world and participants are unaware of their participation
Internal validity: essential to experimental research; the confidence that
researchers have that only the manipulated variable produced the results of the
Random assignment and debriefing participants so they believe in the study aids
in internal validity
Reliability: how consistent results yielded are
Measurement validity: the correlation between some measure and some outcome
the measure is supposed to predict
Statistical significance: a measure of the probability that a given result could have
occurred by chance
Basic science is trying to understand phenomena in the world through theories
about the nature of aspects of the world whereas applied science is concerned
with real would problem solving
Ethical concerns in social psychology are addressed by review of research by
Institutional Review Boards (IRB)
Participants must give informed consent before the study is deemed ethical, but is
not possible for some studies like deception research, in which the aim of the
study would be compromised if its true nature was revealed to participants