Research Methods Flashcards
Descriptive Research
-Data collection tries not to interfere with how data arises in the real world
-Describes the characteristics of a population or a phenomenon
-Includes naturalistic observation, surveys, case studies, and self-reports
Correlational Research
-Looks for relationships between variables
-Uses descriptive research methods to obtain data on variables
Experimental Research
Manipulates variables in a controlled manner to isolate causes of some phenomena
Naturalistic Observation
-Recording behaviour in real world settings without trying to manipulate the situation
-Have high external validity
-Cannot make casual inferences
-Observation can affect behaviour (reactivity)
External Validity
The extent to which we can generalize findings to real-world settings
Case Studies
-A research design that examines a person/people in depth over an extended period of time
-Useful for existential proof (demonstration that a psychological phenomena can occur)
-Provides information about rare phenomena that cannot be studied in the lab
-Very subjective
-Cannot explain the causation of a phenomena
-Relies on indirect statements (prone to biases and inaccuracies)
Anecdote
A short, often interesting personal experience
ie. “I knew a woman who did yoga daily for three weeks and hasn’t experienced depression since!”
-Does not explain causation of phenomena
-Inaccurate/biased portrayal of events
-May not represent other aspects that come into play
-May ignore contradictory claims
-NOT evidence
Surveys/Self-Reports
-Using questionnaires or interviews to gather information about aspects of a participant’s background, experiences, and behaviour
pros:
-easy to collect large amounts of data on numerous topics/factors
cons:
-does not establish causation between variables
-wording of the questions can influence responses
-assumes that the participant understands the question
-assumes that the participant has insight into their personality characteristics and are reporting it honestly
-positive impression management: tendency to make oneself appear better than they actually are
-malingering: tendency to make oneself psychologically disturbed to achieve some goal
Random Sampling
-Ensures every person in a population has an equal chance of being chosen to participate
-Required for generalizable results in any form of research
Reliability
The consistency of measurement
Test-retest reliability: when a test/measure is re-administered, it should produce the same result as when it was administered the first time
Interobserver reliability: multiple people using the same test/measure should yield the same conclusion
Validity
The extent to which a measure assesses what it claims to measure
Correlational Designs
-Used to examine the relationships between variables
-Relies on the correlation coefficient (the degree to which two variables are related)
Scatter plots: each point is a specific pair of data (ie. x-coordinate is the variable making the prediction and y-coordinate is the variable we want to predict)
Positive correlation (ie. taller fathers predict taller sons)
r=0 —> no correlation r=1—> perfect correlation
Negative correlation (ie. taller fathers predict shorter sons)
r=0—> no correlation r=-1—> perfect correlation
Illusory Correlation
Perception of a statistical association between two variables where none actually exists
Confirmation Bias
The tendency to seek out information that supports a hypothesis while denying, ignoring, or distorting information that contradicts the hypothesis
Availability Heuristic
When you estimate the likelihood of an occurrence based on the ease which it comes to mind
The Third Variable Problem
Correlation DOES NOT EQUAL Causation
-Two variables may be correlated to one another only because they are both casually related to a third variable
Observational Study
Independent variable is not under the control of the researcher
Experimental Study
Independent variable is under the control of the researcher
Hazards in experimental design
Random assignment is NOT used
-random assignment refers to randomly sorting subjects into experiment groups
Confounding variable(s) present
-a confounding variable refers to differences between the experimental and control groups (those that are not the independent/manipulated variable)
-makes it impossible to determine whether the group differences were due to manipulation or the presence of a confound
Blinding
-subjects are NOT blinded to the condition they are in
-blinding is when a subject is unaware of what group (experimental or control) they are in
-double blinding: neither the researchers or subjects know who is in each group
Experimenter expectancy effect
-when the experimenters hypotheses lead them to unintentionally bias the outcome of the study
Subjects are using demand characteristics
-demand characteristics are cues that participants pick up from a study that allow them to generate guesses regarding the researcher’s hypotheses
-may be done to try to disguise the purpose of the study
Placebo and Nocebo effect
Placebo
-an inert substance that provokes perceived benefits
-If a person has an expectation that a (fake) medication will be beneficial, their body will be more likely to react in ways that are consistent with an improvement of some kind
Nocebo
-an inert substance causes perceived harm
-if a person has an expectation that the (fake) medication will produce some adverse effect, their body will respond accordingly and be more likely to produce adverse symptoms
