1/75
Looks like no tags are added yet.
in experiments, researchers ___ one variable to ___ another
manipulate; measure
what is an independent variable
predictor, manipulated variable and assigns participants to at least two levels or conditions
what is a dependent variable
outcome, measured variable, records what happens via self report, behavioral observations or physiological measures
what is a control variable
any variable an experiment holds constant
experiments are the only research design that support ___
causal claim
covariance
is the cause variable related to the effect variable
temporal precedence
does the cause variable come before the effect variable
internal validity
the extent to which a research study accurately establishes a cause-and-effect relationship between an independent variable and a dependent variable, free from the influence of confounding factors
what do you need to make a causal claim
covariance, temporal precedence and internal validity
independent variables answer the question
compared to what?
IV must have ___ comparision conditions
one or more
control group
the neutral or no treatment group
treatment group
the nonneutral level or levels of the IV
placebo group
group that is exposed to an inert treatment
experiments establish___
covariance… IV is associated with DV, do changes in IV correlate with changes in DV
Experiments establish ___
temporal precedence…the cause occurs before the effect, manipulate IV before measuring the DV
well designed experiments establish___
internal validity…confidence that the IV caused the change in the DV, third variable explanations are controlled, confounding variable (alternative explanation for the outcome in the DV)
systematic variability
consistent change along with the IV, confounding variable, big deal and must control
unsystematic variability
random or haphazard differences that influence both groups, extraneous variable ( any variable other than IV and DV, not a big deal
design confounds…
provide an alternative explanation for the results, second variable varies systematically with the IV
selection effects ___
provide an alternative explanation for the results
selection effect
participants in one level of the IV are systematically different than the participants in the other level of the IV
solutions to selection effect
hold the variable constant, random assignment, matched groups
well designed experiments use ___ to support internal validity
random assignment
matched samples design
control for a variable by randomizing participants into groups based on that trait
helps with selection effects
four basic elements of experiments
manipulation, measurement, comparison, control
what are the two basic experimental designs
independent groups design and within groups design
independent groups design - experimental design
different groups of participants placed at different levels of the IV → independent groups design, between-subjects design, between-groups design
within groups design - experimental design
each participant is presented with all levels of the IV → within-subjects design, dependent samples design, repeated measures design
independent samples design
different groups of participants placed at different levels of the IV
two primary independent samples design
posttest-only and pretest/posttest
posttest independent samples design
Random Assignment → Treatment/Control → Posttest Measurement
prettest/posttest independent samples design
Random Assignment → Pretest Measurement →Treatment/Control → Posttest Measurement
repeated measures design
A repeated measures design (also known as a within-subjects design) is an experimental setup where the same participants are tested under all conditions or treatments of the study.
concurrent measures design
A concurrent-measures design is a specific type of within-groups (within-subjects) experimental design in which participants are exposed to all levels of an independent variable at roughly the same time, and a single response is measured as the dependent variable.
what are the advantages on within groups designs
participants are equivalent (fewer confounds), increased statistical power, requires fewer participants
disadvantages of within-subjects design
order effects, might not be practical or possible, demand characteristics
order effects are ___ to internal validity in within groups design
threats
order effects definition
sequence in which participants experience the experimental conditions influences their behavior or performance, independent of the treatment itself.
what are the effects catagorized as order effects
testing effect, practice effect, fatigue effect, carryover effect
counterbalancing
Instead of giving all participants the conditions in the exact same sequence , researchers systematically vary the order across different participants.
aims to reduce impacts of order effects
threats to internal validity
design confound, selection effect, attrition, placebo, maturation, history, regression to mean, instrumentation
attrition
occurs when participants drop out of a study before it is completed. It is a major threat to internal validity, particularly in longitudinal studies, pretest/posttest designs, and repeated-measures designs that require multiple sessions over time.
placebo
inert, inactive substance or sham treatment given to a control group in an experiment. It looks, tastes, and feels identical to the actual treatment being tested, except it contains no active medical or therapeutic ingredient
maturation effect
threat to internal validity that occurs when natural biological, psychological, or developmental changes inside participants over time influence the dependent variable, rather than the experimental treatment itself.
History threat
threat to internal validity that occurs when an external, historical event happens during the course of a study (typically between a pretest and a posttest) and influences participants' responses, independent of the experimental treatment.
regression to the mean
statistical phenomenon where individuals who score extremely high or extremely low on a first measurement (like a pretest) tend to score closer to the group average (mean) on a second measurement (posttest), simply due to random chance or measurement error.
instrumentation threatt
measuring instrument changes over time, changing the observation criteria
assessing construct validity
how well was the DV measured?
was there a separate pilot study that can confirm the effectiveness of manipulation?
was there a manipulation check that explicitly measures the IV?
interactions describe relationships where ___
it depends
factorial design - the effect of one variable ___ on the level of the other variable
depends
factorial designs describe interactions by ___
describing the change in one IV at each level of the other IV
factorial designs are experiments with ___
two or more IV
in factorial designs, IV are ___ and DV are ___
categorical, continuous
factorial design notation describes ___
the number of IV and their levels
factorial designs allow us to make ____
three independent examples: A+B; C+D; AC vs. AD and BC vs. BD
if there are ___ there is likely an interaction but if lines are ___ there is likely no interaction
a cross in lines or a fan shape; parallel
interpreting interactions require looking at ___ of one IV separately
levels
only __ interaction is possible with _
interaction; two IV
Interaction allows us to test for ___
moderation
what is a moderator
an IV that changes the relationship between another IV and DV
interactions look at ___
the difference in differences
what are main effects
the overall effect of one IV on the DV
when do you look at main effects
when the interaction is not significant → compare marginal means, look at the simple difference between the levels of the IV
interpretation for main effect
there is a main effect for…such that… is higher than …
dont mention the other IV
in a 2×2 factorial design, how many possible main effects are there and how many interactions
two main effects bc two IV and one interaction bc only one way variables can combine
what are quasi experiments missing to be a true experiment
groups are defined by a participant variable, no random assignment
types of quasi experiments
nonequivalent control group design (posttest), nonequivalent control group design (prettest/posttest), interrupted time series design, nonequivalent control group interrupted timeseries design
nonequivalent control group test (posttest)
quasi-independent variable, dependent variable, participants not randomly assigned, only tested after one exposure
nonequivalent control group test (prettest/posttest)
quasi-independent variable, dependent variable, participants not randomly assigned, tested before and after every intervention
inturrupted time series design
measures a variable repeatedly before, during and after interruptions caused by some event
nonequivalent control group interrupted time series design
quasi-independent variables (none controlled by researchers), dependent variable tracked before and after introduction of IV
potential threats to internal validity - quasi experimental design
selection effects, design confounds, attrition, maturation, history, regression to the mean, repeated testing, instrumentation, observer bias, demant characteristics (participants guess what study is about), placebo
benefits of quasi experiments
real world opportunities, external validity → generalizability, ethics
quasi experiments
sample of two different groups of people who are different based on a preexisting characteristic, cant draw casual conclusions, fewer internal validity threats
correlational studies
sample a group and measure teo variables, cant draw casual conclusions, more internal validity threats