PSYC Midterm Flashcards
Epistemology
The philosophical study of where knowledge comes from
Tenacity
It has always been accepted as true
Intuition
It feels correct
Authority
A respected source says it’s true
Rationalism
Using logical reasoning (if A then B) to dra a conclusion from premises
Empiricism
Observation through the senses (directly or indirectly)
Availability Heuristic
Things that come easily yo mind seem likely
Hindsight bias
Things seem obvious after you know the outcome
Confirmation bias
Looking for evidence that confirms our beliefs
Bias about bias
other people are biased; we aren’t
Falsifiability
Theory must lead to hypothesis that, when tested, could actually fail to support the theory
Parsimony
Preference for simplest explanation
Applied research
Done with a practical problem in mind
Basic research
Goal to enhance body of knowledge
Pseudoscience
Hypotheses that aren’t falsifiable
Claims are not directly supported by empirical evidence
Employs method of tenacity and/or authority
Not grounded in past scientific research
What question am I trying to answer?
Variables and relationships
What kind of data should I collect?
Measurement
Whom should I measure?
Sampling
How will I collect those measurements and what will they be able to tell me?
Research strategy and statistical inference
Variable
Thing that varies- must have at least two levels
constant
only has one level in the study
Frequencies
Rate or degree of a single variable
correlations
one level of a variable is associated with a particular level of another variable
Causal
One variable is responsible for changing another
Research strategy
The empirical approach used to gather data
Statistical Validity
Extent to which a study’s statistical conclusions are accurate/reasonable
Quantitative research
Statistical conclusions
Qualitative research
Draws descriptive conclusions
Descriptive research
Characterize single variable
Relational research
relate multiple variables
Correlational
Variables are NOT manipulated, NOT randomly assigned
Quasi-experimental
Participants are grouped but NOT randomly assigned to conditions
Experimental
Independent variable is actively manipulated and participants are randomly assigned to conditions
External validity
How well results of a study represent the people or context besides those in the original study
Sampling bias
Non-representative samples make it difficult to generalize population
Novelty effect
Being in a study can cause participants to behave strangely
Multiple treatment interference
Experience in previous treatment condition changes behavior I subsequent condition
Measurement timing
Results might be different if measurement was taken at a different time
Operational definitions
Results could be specific to choice of operational definition
Internal Validity
How sure are we that X causes Y
Experiment
Research design in which researchers manipulated at least one variable and measured another ; Participants are randomly assigned to a level of the independent variable
Extraneous variable
Anything other than the manipulated variable that might be influencing the observed effect ; must be controlled by researchers
Directionality problem
Don’t know whether X causes Y or if Y causes X or both
Third variable problem
Some unidentified variable is responsible for the observed relationship between two variables
Independent Variable
Any variable that the researcher intentionally manipulates
Conditions
Levels of the IV
Dependent variable
Any variable that the researcher measures as an outcome of the study
Control group
“No treatment” or neutral condition
Treatment/experimental group
what you call the other group when there is a control group
Confound
Alternative for relationship between IV and DV
Design confound
Mistake the design of the IV, such that second variable varies systematically with the IV
Selection effects
Participants in one level of the IV are systematically different from those in the other
Between subjects
Different groups of participants are placed into different conditions ; comparing scores from different people
Within-subjects
One group of participants completes all conditions ; comparing scores from the same person
Posttest-only design
participants are randomly assigned to a condition and tested on dependent variable once
Order effects
Exposure to one level of the IV influences responses to next level
Counterbalancing
Helps avoid order effect ; present levels of IV in different sequences
Time-related confounds: Maturation
Subjects naturally change physiologically or psychologically
between treatment conditions
Time-related confounds: History
An outside event occurs between treatment conditions and affects
dependent variable in condition(s) subsequent to event
Time-related confounds: Regression to the mean
Extreme scores in first treatment condition statistically likely to
become less extreme in subsequent conditions
Time-related confounds: Attrition
Subjects drop out of a study before completing all conditions
Time-related confounds: order effects
Experience in first treatment causes a change in subjects that affects performance in subsequent conditions
Time-related confounds: Instrumentation
Change in measurement instruments between treatment conditions affects measurement of dependent variable subsequent in instrument change
Threats to internal validity: Observer bias
Researchers’ expectations influence results
Threats to internal validity: demand characteristics
Participants guess study hypothesis and change behavior
Threats to internal validity: Placebo effects
Participants improve because they believe they are receiving a valid treatment
Double-blind study (Observer bias and demand characteristics)
Neither participants nor researchers now what condition they are in
Masked study (Observer bias and demand characteristics)
(participants do not know condition) when double-blind study is not possible
Active control group (Placebo effects)
control group given treatment that creates same
expectation of improvement as treatment group
Measurement
Process of assigning numbers or categories to individuals in your sample ; measurement creates observable variables from constructs
Construct (conceptual definitions: theoretical)
Theoretical variable that cannot be observed directly
Observable variable (operational definition: precise)
A quantity or quality that can be observed directly and varies across individuals
Self-report
A measure in which participants report on their thoughts, feelings, or behaviors ina survey or interview
Behavioral
A measure in which the researcher observes and records some aspect of participants’ behavior
Physiological
A measure that involves recording a physiological variable (part/aspect of the participant’s body)
Nominal
Category labels (male/female, etc)
Ordinal
Ordered scores (first, second, third, etc)
Interval
Equal units (But NO TRUE ZERO)
Ratio
Equal units and zero means none
Reliability
The extent to which the scores on a measure are consistent across
time, across observers, and across items when a measure has an
element of subjective judgment
Validity
Are you really measuring what you think you’re measuring?
Test-retest reliability
reliability across time ; If you measure the same person again, do they get a similar score?
Inter-rater reliability
Reliability across observers ; Do different observers give the same subject the same score?
Internal reliability
Reliability across items ; If you ask the same question in a different way, do they give a similar answer?
Construct validity
How accurately does the operational definition capture the construct?
Face validity
Measurement seems to capture the intended variable. Does the test appear to test what it aims to test?
Content validity
Measurement covers all aspects of the construct that it is supposed to measure. Does the test reflect all aspects of the conceptual definition of the construct?
Criterion validity
Measurement correlates with a relevant bahavioral outcome
Convergent validity
Measurement correlates with measures of similar constructs
Discriminant validity
Measurement doesn’t correlate with measures of dissimilar constructs
Population
Group of interest
Sample
Individuals observed ; should represent the population well
Representative sample
A sample that closely resembles the population
Biased sample
Some members of the population have a much higher probability of being included in the sample than other members
Probability sampling
Every individual in target population is identified; every individual has a known, non-zero chance of being selected; Selection is random, based on probabilities of being selected
Simple random sampling (Probability Sampling)
Probability sampling in which every member of the target population has an equal chance of being selected
Systematic sampling (Probability Sampling)
Order all individuals ina population, pick a random starting point, and select each nth individual
Cluster sampling (Probability Sampling)
Randomly select pre-existing groups and measure all (or randomly sample) members of each group.
Stratified random sampling (Probability sampling)
Purposefully select particular demographic categories, or strata, and then
randomly selects individuals within each of the categories, proportionate to
their membership in the population
Non-probability sampling
Not every individual in the population can be identified OR the probability of selection cannot be calculated (or is zero for some individuals) OR the selection process is not random
Convenience sampling (Non-probability sampling)
Sampling individuals who are easily accessible
Quota sampling (Non-probability sampling)
Similar to stratified random sampling, but selection is not random and all individuals may or may not be identified
Snowball sampling (Non-probability sampling)
Asking participants to recruit other participants who are similar to them
