Survey
NOT method, it’s technique
able to generalize for population
people lie
wording issues
cheap
social desirability effects
easy
Naturalistic Observation (can see)
not good because cannot replicate, unnatural, can’t generalize
observer bias
when observer knows what to expect and is influenced
observer effect
know they are being watched
Case Study - take advantage of unreplicable situation
Scientific Method - observing/experimenting to justify hypothesis
Hypothesis:
specify variables (anything that can be changed)
developed from theory
look for evidence that proves and refutes hypothesis
support existing theories
falsifiable if can be proven wrong
Confirmation Bias
search for what you want to be right
Operational Definition: statement of procedure / way in which a researcher is going to measure behaviors or qualities
Biological
The body and brain enables human emotions/behaviors
Cognitive
Human Nature
People are information-processing systems
Determines Behavior/Mental Process:
Mental interpretation of our experience
Psychodynamic
Human Nature:
Driven by dark forces of the unconscious
Determines Behavior/Mental Process:
Unconscious needs, conflicts, repressed memories, childhood experiences
Humanistic
Human Nature:
Emphasizes human growth/potential
Determines Behavior/Mental Process:
Influence of self-concept, perceptions, interpersonal relationships, need for personal growth
Behavioral
Human Nature:
Behavior is primarily shaped through learning
Determines Behavior/Mental Process:
Stimulus cues and our history of rewards and punishments
Sociocultural
Human Nature:
People are social animals, human behavior must be interpreted in social context
Determines Behavior/Mental Process:
Cultures, social norms, expectations, social learning
Evolutionary
Human Nature:
Behavior is developed and adapted over time
Determines Behavior/Mental Process:
Natural Selection
only research method for investigating cause and effect relationships
Cause and Effect: manipulation of variable to cause effect
Hypothesis:
if = independence variables
experimenter will change
cause
then = dependent variable
variable measured
effect
depends on IV
Population: all individuals who can participate
cant use everyone, so use a small group (sample)
Representative Sample: sample that has characteristics similar to those in population, easy to generalize
Random Sample: used to avoid sampling bias
every member has same chance
Stratified Sampling: ensuring representative, make sure each segment of population is equally represented, easy to generalize
Convenience Sampling: selects participants based on accessibility, can’t generalize (sampling bias)
Experimental Group: has no independent variable manipulated
Control Group: comparison group, measures the dependent variable by not giving experimental treatment
Random Assignment (NOT random sample): ensures all members of sample have equal chance of being in either group
Hindsight Bias: tendency to believe, once an outcome is already known of course, that you would have foreseen it
Overconfidence: very sure of a fact although reality is different
Confounding Variables: only found in experimental research
differences (other than independent variable) that arise due to poor planning, sloppy work or bias
variables that a researcher fails to control
Single-Blind Study: participants do not know which group they belong to (control or experimental)
Double-Blind Study: neither participant nor researcher knowns which group the participants belong to
eliminates experimenter bias (expectations that may influence outcome)
Placebo Effect: real responses to an action or substance based solely on expectations, not actual properties of the action or substance
research method
Purpose of correlational research:
shows relationship between two variables
doesn’t involve manipulation of variables as in an experiment
Weakness:
correlation is not causation
cannot demonstrate cause and effect
Strength:
predict outcome if know how they are related
Correlation Coefficient: indicates the strength of the relationship between two variables (r value)
Types of Relationships:
positive (r value = 0-1)
negative (r value = -1 to 0)
illusory (r value = 0)
Directionality Problem: occurs when two variables are correlated but it not clear which causes the other
Meta-Analysis: synthesizes results of several previous independent studies on topic
could summarize multiple correlational studies or experimental studies
Outliers: data points that fall beyond the mean
Skewed: data point pile up at one end of the distribution or the other
Bimodal Distribution: types of probability distribution that has two distinct peaks or modes in a data set