Psych Midterm 1 Flashcards
Identify, gather, test, analyze, build
The Scientific Method
why use the scientific method?
reduces bias and helps provide better theories
we know what happens after it occurs
hindsight bias
we only consider evidence that is present
present/present bias
basing judgements off of what comes to mind first
availability heuristic
objective statement usually based on direct observation that observers agree us true
fact
hypothetical account of how/why a phenomenon occurs
theory
a specific testable prediction made by a theory
hypothesis
what makes a good theory?
comprehensive - explain info we already know
testable/falsifiable
simple
generative - generates new ideas and research
simpler theory is preferred
law of parsimony
conceptual definition of a variable
what it means (one per variable)
operational definition of a variable
how it is measured/manipulated in context of a study (can be multiple per variable)
self-report
people report about themselves
easy but not always accurate (social desirability bias)
observing behaviour
use frequency, occurrence, or timing of natural occurrence
knowing you’e being observed may affect behaviour
decrease of observation-based effects over time
habituation
recording info while participants are unaware (sometimes unethical)
unobtrusive measures
using pre-existing records/documents
good for large scale studies
archival measures
looking at what the body is doing
ex. heart rate, sweat, neural imaging
physiological measures
projective psychological tests
based on assumption of projected hypothesis
performance-based psychological tests
how confident you are in certain areas
ex. memory-based
who we want to make a conclusion about
population
smaller group, represents population
sample
random sampling
every member of population is equally likely to be chosen
representative sample
reflects important aspects of populations
convenience sampling
based on availability; not random
systematic study plan used to turn research questions into study
research design
qualitative research
use non-numerical data
quantitative research
uses numerical data
most common
research that seeks to identify how humans and animals behave
ex. case study, naturalistic observation, surveys
descriptive research
in-depth analysis of individuals/groups/events
case study
behaviour is observed as it occurs in a natural setting
naturalistic observation
collecting info through questionnaires/ interviews
survey research
examines/measures relationship between 2+ variables
quantitative, allows us to make predictions
correlational research
correlation coefficient
indicates direction/strength of relation
-0 is no relationship, +1 or -1 is stronger
-positive: both variables increase
-negative: one variable increases, other decreases
can determine causation and rule out alternative explanations
experiments
3 characteristics of experiments
-manipulate variables
-measure whether manipulation influences other variables
-attempt to control extraneous factors
manipulated/controlled variable (cause)
independent variable
measured and may be influence by independent variable (effect)
dependent variable
3 criteria for causation (in experiments)
covariance
temporal precedence
internal validity
as IV changes, DV changes (there is an effect)
covariance
cause happens before effect
temporal precedence
does experiment support clear causal connections
internal validity
experimental design in which participants only experience one level of IV/manipulation
between-subjects
experimental design in which participants experience all levels of IV/manipulation
repeated measures
experiments with multiple IVs test effect of IV on the DV varies depending on levels of another IV
interaction
tendency for a measure to produce the same result whenever it’s used to measure the same thing
reliability
consistent assessment at different points in time
test-retest reliability
consistent assessment from different researchers/evaluators (ex. Olympic judges)
integrator reliability
consistent items within an assessment
internal reliability
accuracy of a measure
allows you to draw accurate inferences from observation
validity
measure appears to assess what it is supposed to (can tell by looking at it)
face validity
measure assesses all parts of a defined construct (ex. exam)
content validity
measure correlates (+ or -) strongly with constructs
integrates other measures
convergent validity
measure correlates weakly/has no relationship to things it should relate to
discriminant validity
variables researchers are not including/ considering in a study but could be affecting results
confounding variables
change in symptoms/behaviour based on expectation/belief of receiving treatment
placebo effect
substance with no pharmacological effect
ex. sugar pills, saline solution
placebo