Domain D Flashcards
(29 cards)
the dependent variable is the __________
target behavior / behavior of interest
the independent variable is the __________
treatment/intervention
think: IV=intervention
extent to which an experiment strongly shows that changes in behavior are a direct result of the independent variable
internal validity
extent to which the results of a study can be generalized to other settings, behaviors, or subjects
external validity
is internal or external validity more important?
internal is a priority over external validity
When the way a study was designed, conducted, and analyzed gives a trustworthy answer to the research question
internal validity
can only be assessed by replicating the study
external validity
2 major types of scientific replication:
- Direct Replication
- Systematic Replication
explain each
Direct replication has 2 types:
-Intrasubject direct replication= exact replication of a study, including participants used. (which strengthens reliability of a functional relation)
-Intersubject direct replication= exact replication of study, but different participants
Systematic Replication= Researchers intentionally change one or more features of a prior experiment (e.g. participant demographics, settings, aspects of IV or DV)
when unknown or unintended variables have uncontrolled influence on a study; extraneous variables, confounding variables
threats to internal validity
(extraneous=environment; every aspect of the environment must be controlled, confounding=factors you have no CONtrol over
Measurement confounds, independent variable confounds, subject confounds, and setting confounds are __________
threats to internal validity
any aspect of the environment other than the IV that must be held constant to prevent unplanned environmental variation
e.g. lighting, room temperature, noise level
extraneous variables
any uncontrolled factor (extraneous variable) known or suspected to influence the DV that is unexpected, not considered in planning, or unable to be controlled for
things an experimenter cannot control e.g. parents fought in car on the way to school; client didn’t sleep well
confounding variables
threats to internal validity
measurement confounds
-observer drift
-observer bias/expectations
-reactivity
independent variable (IV) confounds=anything that may muddle effects of the IV. more likely to occur when IVs are complicated
subject confounds= participant related variables that may muddle results;
-maturation=growth
-history=environment/world
-attrition=reduction in size of participant group; client turnover
-practice effects= during baseline bx improvement resulting from the opportunity to practice
-adaptation=reduction in responding due to repeated presentations
-warm-up effects= during baseline responding tends to be weaker at first
setting confounds= uncontrolled variables in tx setting
-bootleg reinforcement= reinforcement that is accessed without meeting the response requirements of the contingency ; aka reinforcement that isn’t part of the treatment
Anything that may muddle measurement of outcomes therefore the internal validity of an experiment; observer drift, reactivity, observer bias/expectations
measurement confounds
Any participant-related variables that may muddle results of a study; maturation, history, attrition, practice effects, adaptation, warm-up effects
subject confounds
Anything that may muddle effects of the independent variable
Independent variable confounds
Uncontrolled variables in the treatment setting that could impact the outcome of a study
setting confounds
specific arrangement of conditions in a study, created to compare relevant relations/effects of the presence, absence, or change in values of the IV; comparing how interventions impact behavior
experimental design
when a predicted change in behavior (DV) is reliably produced by systematically manipulating the environmen (IV)
experimental control
used to detect a reliable and convincing functional relation between DV and IV
analysis
behavior changes when and only when the IV is introduced
functional relation
- Research Question
- Participants
- DV
- Setting
- measurement and analysis of results
- IV
- Experiment
components of an experimental design
a type of experimental design in which each individual participant serves as their own control and the effects of treatment are compared to the participant’s own baseline data
single-case experimental design (SCD)
- Data collection- must be accurate and reliable
- Controlling setting events- identify extraneous variables that may have effect on behavior
- Ongoing visual inspection and analysis- to make predictions about future of target behavior and see whether treatment is working; inspect level trend, variability
defining features of single-case experimental designs (SCD)